Coordination Mechanism and Bio-Evidence: Reactive γ-Ketoenal Intermediated Hepatotoxicity of Psoralen and Isopsoralen Based on Computer Approach and Bioassay

Abstract

Psoralen and isopsoralen are secondary plant metabolites found in many fruits, vegetables, and medicinal herbs. Psoralen-containing plants (Psoralea corylifolia L.) have been reported to cause hepatotoxicity. Herein, we found that psoralen and isopsoralen were oxidized by CYP450s to reactive furanoepoxide or γ-ketoenal intermediates, causing a mechanism-based inhibition of CYP3A4. Furthermore, in GSH-depleted mice, the hepatotoxicity of these reactive metabolites has been demonstrated by pre-treatment with a well-known GSH synthesis inhibitor, L-buthionine-S, Rsulfoxinine (BSO). Moreover, a molecular docking simulation of the present study was undertaken to understand the coordination reaction that plays a significant role in the combination of unstable intermediates and CYP3A4. These results suggested that psoralen and isopsoralen are modest hepatotoxic agents, as their reactive metabolites could be deactivated by H₂O and GSH in the liver, which partly contributes to the ingestion of psoralen-containing fruits and vegetables being safe.

Keywords: CYP3A4; GSH depletion; coordination compound; furanoepoxide; hepatoxicity; molecular docking; γ-ketoenal intermediate.

Scheme 1

(a) The molecular mechanism of the CYP450s enzyme catalytic reaction; (b) Two spin states of Fe(II).

Figure 1

(a) IC₅₀ shift of psoralen and isopsoralen for the different CYP450s in human liver microsomes (HLM) with (+) or without (–) NADPH pre-incubation; (b) IC₅₀ shift of psoralen and isopsoralen for the different CYP450s in HLM with (+) or without (–) NADPH pre-incubation; (c) GSH was applied to interfere in the inhibition ability of psoralen and isopsralen (at IC₅₀ concentration as detected in the IC₅₀ shift assay) on CYP450s. Grey stylolitic shows the relative activity of CYP450s without adding GSH, while white stylolitic shows the restore activity of CYP450s by adding GSH. Results are presented as the mean ± SD from three independent experiments. * p < 0.05, significantly different from the corresponding incubations without cofactor GSH. NC is normal control.

Figure 1

(a) IC₅₀ shift of psoralen and isopsoralen for the different CYP450s in human liver microsomes (HLM) with (+) or without (–) NADPH pre-incubation; (b) IC₅₀ shift of psoralen and isopsoralen for the different CYP450s in HLM with (+) or without (–) NADPH pre-incubation; (c) GSH was applied to interfere in the inhibition ability of psoralen and isopsralen (at IC₅₀ concentration as detected in the IC₅₀ shift assay) on CYP450s. Grey stylolitic shows the relative activity of CYP450s without adding GSH, while white stylolitic shows the restore activity of CYP450s by adding GSH. Results are presented as the mean ± SD from three independent experiments. * p < 0.05, significantly different from the corresponding incubations without cofactor GSH. NC is normal control.

Figure 2

Proposed metabolic pathways of psoralen (a) and isopsoralen (b) in HLM and MLM: hydroxylation metabolites, hydrolysis metabolites, and furan ring oxidation metabolites (the most important metabolic pathway). The structures shown in brackets are the postulated reactive metabolites of psoralen and isopsoralen; electrophilic furanepoxides and the γ-ketoenal intermediate can also be deactivated with H₂O or GSH.

Figure 3

The fragmentation information for the psoralen and isopsoralen reactive metabolites formed in HLM and MLM. (a) The hydrolysis metabolites of furanoepoxide and the γ-ketoenal intermediate gave the protonated molecule [M + H]⁺ at m/z 221, and the MS2 spectrum showed prominent ions at m/z 203 [M + H − H₂O]⁺ and 175 [M + H − H₂O − CO]⁺; (b) the GSH-reactive metabolite adduct gave the protonated molecule [M + H]+ at m/z 510, and the MS2 spectrum showed prominent ions at m/z 492 [M + H − H₂O]⁺ and 363 [M + H − H₂O − 129]⁺; (c) (for psoralen) and (d) (for isopsoralen), multiple reaction monitoring chromatogram (m/z 221→175) of the hydrolysis metabolites of reactive metabolites; (e) (for psoralen) and (f) (for isopsoralen), multiple reaction monitoring chromatogram (m/z 510→393) GSH-reactive metabolites.

Figure 4

(a) The hepatic GSH level curve after treatment with BSO alone in 24 h, and the inserted pathological section suggested that BSO was without hazard to the liver during the period; (b) and (c), the hepatic GSH level after treatment with psoralen (or isopsoralen) in the absence (or presence) of ABT at 24 h (* p <0.05 compared with test compound group).

Figure 5

(a) The histogram shows the time course of ALT levels (IU/L), while the pathological section (b) shows the histopathological evaluation of mice after administration of vehicle (control), oral psoralen (300 mg/kg) alone, or a combination of BSO (650 mg/kg) and intraperitoneal ABT (100 mg/kg). Black arrows indicate pathological changes. * p < 0.05, ** p < 0.01 in comparison with control group. ^# p < 0.05, ^## p < 0.01 in comparison with psoralen group.

Figure 6

(a) The above histogram shows the time course of ALT levels (IU/L), while the pathological section (b) shows the histopathological evaluation of mice after administration of vehicle (control), oral isopsoralen (300 mg/kg) alone, or a combination of BSO (650 mg/kg) and intraperitoneal ABT (100 mg/kg). Black arrows indicate pathological changes. ** p < 0.01 in comparison with control group. ^# p < 0.05 in comparison with isopsoralen group.

Figure 7

Docking simulation of Ligand into CYP3A4. (a) Psoralen–CYP3A4; (b) γ-ketoenal intermediate of psoralen–CYP3A4; (c), Furanoepoxide intermediate of psoralen–CYP3A4; (d) Isopsoralen–CY3A4; (e) γ-ketoenal intermediate of isopsoralen–CYP3A4; (f) Furanoepoxide intermediates of isopsoralen–CYP3A4. The heme and iron atoms are colored with brown and green, respectively.

Figure 7

Docking simulation of Ligand into CYP3A4. (a) Psoralen–CYP3A4; (b) γ-ketoenal intermediate of psoralen–CYP3A4; (c), Furanoepoxide intermediate of psoralen–CYP3A4; (d) Isopsoralen–CY3A4; (e) γ-ketoenal intermediate of isopsoralen–CYP3A4; (f) Furanoepoxide intermediates of isopsoralen–CYP3A4. The heme and iron atoms are colored with brown and green, respectively.

Figure 8

(a) The illustration of the γ-ketoenal intermediate of the psoralen/isopsoralen–heme Fe(II) CYP3A4 coordination compound; (b) The illustration of the σ and π backbonding interaction between γ-ketoenal intermediates and heme Fe(II) of CYP3A4: the mechanism of coordination binding between γ-ketoenal intermediates and heme Fe(II) of CYP3A4.

Scheme 2

Chemical structures of docking ligands. (a) Psoralen; (b) γ-ketoenal intermediate of psoralen; (c) Furanoepoxide intermediate of psoralen; (d) Isopsoralen; (e) γ-ketoenal intermediate of isopsoralen; (f) Furanoepoxide intermediate of isopsoralen.