CYP3A4 Isoform Abundance Correlation: DHPBI (r²=0.98) Outperforms 5-Hydroxypimozide (r²=0.94) as a CYP3A4-Selective Metabolic Probe
In human liver microsomes (HLMs) from individual genotyped donors with cytochrome P450 isoform protein abundances quantified by mass spectrometry, the formation rate of DHPBI (i.e., 4-(2-keto-1-benzimidazolinyl)piperidine) demonstrated a near-perfect linear correlation with CYP3A4 protein abundance (r²=0.98), whereas the alternative CYP2D6-dependent metabolite 5-hydroxypimozide correlated with CYP2D6 abundance at r²=0.94 [1]. This 4-percentage-point superiority in coefficient of determination (r²=0.98 vs. r²=0.94) establishes DHPBI as the more reliable and isoform-discriminant probe for CYP3A4 activity in vitro, with less residual variance attributable to non-CYP3A4 factors [1].
| Evidence Dimension | Correlation of metabolite formation rate with CYP isoform protein abundance (r² value) |
|---|---|
| Target Compound Data | r² = 0.98 for DHPBI formation rate vs. CYP3A4 protein abundance |
| Comparator Or Baseline | r² = 0.94 for 5-hydroxypimozide formation rate vs. CYP2D6 protein abundance; r² = 0.86 for 6-hydroxypimozide vs. CYP2D6 abundance |
| Quantified Difference | DHPBI–CYP3A4 r² = 0.98 exceeds 5-hydroxypimozide–CYP2D6 r² = 0.94 by Δ = +0.04; DHPBI r² is 0.04 units higher |
| Conditions | Pooled HLMs from 7 genotyped individual donors; 5 μM pimozide incubated for 5 min; CYP protein abundance determined by MS/MS; UPLC-MS/MS quantification with calibration ranges 1–500 nM for DHPBI and 0.5–500 nM for 5-/6-hydroxypimozide |
Why This Matters
For laboratories performing CYP3A4 phenotyping or studying CYP3A4-mediated drug-drug interactions, DHPBI formation rate provides a 4% more explainable variance than the next-best alternative metabolite, reducing false-negative rates in CYP3A4 inhibition screening.
- [1] Chapron BD, Dinh JC, Toren PC, Gaedigk A, Leeder JS. The Respective Roles of CYP3A4 and CYP2D6 in the Metabolism of Pimozide to Established and Novel Metabolites. Drug Metab Dispos. 2020;48(11):1113-1120. doi:10.1124/dmd.120.000188. View Source
