Induction of cytochrome P450 1A by cow milk-based formula: a comparative study between human milk and formula

Abstract

During the treatment of neonatal apnea, formula-fed infants, compared to breastfed infants, show nearly three-fold increase in clearance of caffeine, a substrate of cytochrome P450 1A (CYP1A) and in part CYP3A4. However, human milk is known to contain higher concentrations of environmental pollutants than infant formula, which are potent CYP1A inducers. To gain insight into the mechanism underlying this apparent contradiction, we characterized CYP1A and CYP3A4 induction by human milk and cow milk-based infant formula. The mRNA and protein expression of CYP1A1/1A2 were significantly induced by cow milk-based formula, but not by human milk, in HepG2 cells. Luciferase reporter assay demonstrated that cow milk-based formula but not human milk activated aryl hydrocarbon receptor (AhR) significantly. The cotreatment of 3,4-dimethoxyflavone, an AhR antagonist, abolished the formula-induced CYP1A expression. In addition, AhR activation by dibenzo[a,h]anthracene, a potent AhR agonist, was significantly suppressed by infant formula and even more by human milk. In contrast, CYP3A4 mRNA expression was only mildly induced by formula and human milk. Consistently, neither formula nor human milk substantially activated pregnane X receptor (PXR). Effects of whey and soy protein-based formulas on the AhR-CYP1A and the PXR-CYP3A4 pathways were similar to those of cow milk-based formula. In conclusion, infant formula, but not human milk, enhances in vitro CYP1A expression via an AhR-mediated pathway, providing a potential mechanistic basis for the increased caffeine elimination in formula-fed infants.

Figures

Figure 1

Effects of cow milk-based formula and human milk on the expression of CYP1A. (a) RT–PCR quantification of CYP1A1 and CYP1A2 mRNA expression following 24 h treatment of 20% human milk, 20% cow milk-based F1 and F2, and controls. (b) The real-time PCR quantification of CYP1A1 and CYP1A2 mRNA expression following the same treatment as described in (a). All values are expressed as mean±standard deviation of at least six independent experiments. Significant differences among the groups were determined as described in Methods. *P<0.05: significantly higher than the PBS-treated group. (c) Immunoblot of CYP1A1 and CYP1A2 protein after 24 h treatment with formula or human milk. HepG2 cells were treated as described in (a). A representative of at least three independent experiments is shown.

Figure 2

Effects of human milk and formula on the AhR activity in HepG2 cells. (a) AhR activation by milk treatments. All values are expressed as mean±standard deviation of at least three separate determinations. Significant differences among the groups were determined as described in Methods. *P<0.05: significantly higher than the PBS-treated group. (b) Dose response of AhR activity to milk treatment. HepG2 cells were transfected and exposed to different concentrations of formulas and human milk for 24 h. The data are expressed as mean±standard deviation of at least three separate determinations. (c) EMSA. Cytosolic fraction of Hepa1c1c7 cells (∼20 μg μl−1) was treated with the lipid fraction of formula, DMSO, or DBA as described in Methods. An arrow indicates the substrate-inducible AhR–DRE complex. A representative of three separate determinations is shown.

Figure 3

Inhibitory effects of DMF on AhR activity and CYP1A expression in HepG2 cells. (a) Inhibition of formula-induced AhR activation by 3,4-DMF. The transfected HepG2 cells were treated with or without 10 μM 3,4-DMF for 3 h, followed by further incubation with F1, F2, PBS, or 50 nM DBA for 24 h. All values are expressed as mean±standard deviation of at least three separate determinations. Significant differences were determined as described in Methods. *P<0.05: significantly higher than the PBS-treated group. (b) 3,4-DMF inhibition of the formula-induced CYP1A mRNA expressions. The HepG2 cells were treated as described in (a) without transfection and mRNA expressions were quantified by RT–PCR.

Figure 4

Induction of HepG2 CYP1A1 and CYP1A2 by whey- and soy protein-based formula via an AhR-mediated pathway. (a) Real-time PCR quantification of CYP1A1 and CYP1A2 mRNA expression in HepG2 cells following 24 h treatment with whey-based formula (F3) or soy-based formulas (F4 and F5). (b) Western blot of CYP1A1 and CYP1A2 protein following the treatments described in (a). Lane 1: PBS; lane 2: formula treatment; lane 3: DBA. (c) Activation of AhR following the treatments described in (a). (d) Inhibitory effects of formulas and human milk on DBA-induced AhR activation. All values are expressed as mean±standard deviation. Significant differences among the groups were determined as described in Methods. *P<0.05: significant difference from the PBS-treated control group. Data shown were analyzed from at least three independent experiments.

Figure 5

Effects of infant formula and human milk on the mRNA expression of CYP3A4 in HepG2 cells. HepG2 cells were treated with 20% human milk, cow milk-based formulas (F1 and F2), whey-based formula (F3), soy-based formulas (F4 and F5), or controls for 24 h. Total mRNA was isolated and analyzed by RT–PCR for the expression of CYP3A4 and β-actin. All values are expressed as mean±standard deviation of at least three separate determinations. Significant differences among the groups were determined as described in Methods. *P<0.05: significantly higher than the PBS-treated group.

Figure 6

Effects of human milk and formula treatments on CYP3A4-XREM-Luc reporter activity with or without cotransfection of hPXR expression plasmid. HepG2 cells transfected with the CYP3A4 reporter construct with cotransfection of hPXR expression plasmid or with an empty plasmid were treated with 20% human milk or formulas for 24 h. Data are expressed as mean±standard deviation of at least three independent measurements. Significant differences among the groups were determined as described in Methods. *P<0.05: significantly different from the control group; †significantly different between hPXR and the empty vector cotransfected groups.