Correlations between 4β-hydroxycholesterol and hepatic and intestinal CYP3A4: protein expression, microsomal ex vivo activity, and in vivo activity in patients with a wide body weight range

Abstract

Purpose: Variability in cytochrome P450 3A4 (CYP3A4) metabolism is mainly caused by non-genetic factors, hence providing a need for accurate phenotype biomarkers. Although 4β-hydroxycholesterol (4βOHC) is a promising endogenous CYP3A4 biomarker, additional investigations are required to evaluate its ability to predict CYP3A4 activity. This study investigated the correlations between 4βOHC concentrations and hepatic and intestinal CYP3A4 protein expression and ex vivo microsomal activity in paired liver and jejunum samples, as well as in vivo CYP3A4 phenotyping (midazolam) in patients with a wide body weight range.

Methods: The patients (n = 96; 78 with obesity and 18 normal or overweight individuals) were included from the COCKTAIL-study (NCT02386917). Plasma samples for analysis of 4βOHC and midazolam concentrations, and liver (n = 56) and jejunal (n = 38) biopsies were obtained. The biopsies for determination of CYP3A4 protein concentration and microsomal activity were obtained during gastric bypass or cholecystectomy. In vivo CYP3A4 phenotyping was performed using semi-simultaneous oral (1.5 mg) and intravenous (1.0 mg) midazolam.

Results: 4βOHC concentrations were positively correlated with hepatic microsomal CYP3A4 activity (ρ = 0.53, p < 0.001), and hepatic CYP3A4 concentrations (ρ = 0.30, p = 0.027), but not with intestinal CYP3A4 concentrations (ρ = 0.18, p = 0.28) or intestinal microsomal CYP3A4 activity (ρ = 0.15, p = 0.53). 4βOHC concentrations correlated weakly with midazolam absolute bioavailability (ρ = - 0.23, p = 0.027) and apparent oral clearance (ρ = 0.28, p = 0.008), but not with systemic clearance (ρ = - 0.03, p = 0.81).

Conclusion: These findings suggest that 4βOHC concentrations reflect hepatic, but not intestinal, CYP3A4 activity. Further studies should investigate the potential value of 4βOHC as an endogenous biomarker for individual dose requirements of intravenously administered CYP3A4 substrate drugs.

Trial registration: Clinical.

Trials: gov identifier: NCT02386917.

Keywords: 4β-Hydroxycholesterol; CYP3A4; Drug metabolism; Midazolam pharmacokinetics; Proteomics.

Fig. 1

Hepatic CYP3A4 metrics. Correlation between (a) systemic midazolam clearance (CL) and 4βOHC concentrations^a, (b) hepatic CYP3A4 concentration^b and 4βOHC concentrations, and (c) clearance intrinsic for midazolam 1’-hydroxylation in human liver microsomes^c and 4βOHC concentrations. Spearman’s rho (ρ) is the correlation coefficient, and the p value is from the Spearman rank correlation analysis. ^aAvailable in 92 patients (patients with obesity = 74, normal to overweight individiuals = 18). ^bAvailable in 56 patients (patients with obesity = 38, normal to overweight individuals = 18). ^cAvailable in 36 patients (RYGB = 20, normal to overweight individuals = 16). Abbreviations: CL_int,u, clearance intrinsic unbound; CYP, cytochrome P450; HLM, human liver microsomes; 4βOHC, 4-beta hydroxycholesterol

Fig. 2

Intestinal CYP3A4 metrics. Correlation between (a) midazolam absolute bioavailability^a and 4βOHC concentrations, (b) apparent oral midazolam clearance (CL/F)^a and 4βOHC concentrations, (c) jejunum CYP3A4 concentration^b and 4βOHC concentrations, and (d) clearance intrinsic for midazolam 1′-hydroxylation in human intestinal microsomes^c and 4βOHC concentrations. Spearman’s rho (ρ) is the correlation coefficient, and the p value is from the Spearman rank correlation analysis. ^aAvailable in 92 patients (patients with obesity = 74, normal to overweight individuals = 18). ^bAvailable in 37 RYGB-patients. ^cAvailable in 20 RYGB-patients. Abbreviations: CL_int,u, clearance intrinsic unbound; CYP, cytochrome P450; HIM, human intestinal microsomes; 4βOHC, 4-beta hydroxycholesterol

Fig. 3

Clinical variables and 4βOHC. Correlation between (a) BMI and 4βOHC concentrations (n = 96), (b) NAFLD liver fat score and 4βOHC concentrations (n = 95), and (c) hs-CRP and 4βOHC concentrations (n = 96). Spearman’s rho (ρ) is the correlation coefficient, and the p value is from the Spearman rank correlation analysis. Abbreviations: BMI, body mass index; hs-CRP, high-sensitivity C-reactive protein; NAFLD, non-alcoholic fatty liver disease; 4βOHC, 4-beta hydroxycholesterol