Carcinogenic polycyclic aromatic hydrocarbons : role of human glutathione transferases in conjugate formation and DNA protection

Abstract

Polycyclic aromatic hydrocarbons (PAHs) are widespread mutagenic and carcinogenic environmental pollutants, which requires metabolic activation to electrophilic intermediates and subsequent covalent binding to critical targets in DNA to elicit their biological activity. Bay- and fjord region diol epoxides (DEs) have been identified as the ultimate mutagenic and carcinogenic metabolites of PAHs. The balance between metabolic activation and detoxification routes influences the extent of DE-DNA adduct- formation. The most important detoxification pathway is glutathione transferase (GST) catalysed conjugation with glutathione (GSH). The efficiency of the system is determined by the pattern and the level of the GST expressed and the intracellular GSH concentration.

In this study, human GST catalysed conjugation of stercoisomeric DEs of various carcinogenic PAHs was examined. A great difference in activity and enantioselectivity of the GSTs towards the DEs was observed. GSTP1-1 was highly selective for the enantiomer with R-absolute configuration at the benzylic oxiranyl carbon, whereas the preference of GSTM1-1 ranged from 50 to 90%. Molecular modeling, where the anti-enantiomers of benzo[a]pyrene DE (BPDE) were docked into the active site of GSTP1-1, demonstrated that the R-configured enantiomer interacted favourable with the amino acid (aa) residues in the H-site, whereas the corresponding S-enantiomer did not. These results offered an explanation for the difference in the enantioselectivity experimentally observed.

Epidemiological studies suggest that humans differing in the expression of allelic variants of the GST Pi gene differ in susceptibility to chemical carcinogens such as PAH. GSTP1-1 variants were constructed, two with the naturally occurring aa, Val or lle at position 105 in the H-site and two with Ala or Trp at this position. The catalytic efficiencies towards several stereoisomeric bayand fjord region DEs were determined. With increasing volume of the aa residue at position 105 (Alasyn-enantiomers with GSTP1-1/Trp-105. In these cases, a substantial increased catalytic efficiency were seen. The results with the pure variants demonstrated that GSTP1-1/Val-105 was generally more active than GSTP1-1/Ile-105 with the bay region DEs and the fjord region (+)-syn-enantiomers, whereas no difference was observed with the fjord region (-)-anti-enantiomers. Thus, the results indicate that individuals who is homozygous for the allele encoding GSTP1-1/V405 are more susceptible to PAH carcinogenesis due to other reasons than a reduced capacity for detoxifying DEs.

To study individual GSTs in a more biologically relevant system, V79 cells stably over-expressing different human GSTs was constructed and characterized. Furthermore, the protective effect of individual GSTs against DE induced DNA adduct formation was determined. In addition, factors governing the accessibility of lipophilic DE substrates for GSTs in the cell were investigated. In general, an increase in GST activity was concomitant with a decrease in DNA adduct formation. The selectivity of GSTP1-1 towards the R-enantiomers observed with pure enzymes was also seen in the cells. The DE of the most potent PAH so far identified, dibenzo[a,l]pyrene (DBP) showed the highest DNA binding capacity among the DEs tested. GSTA1-1 showed the highest GSH conjugating capacity and offered best protection against DBPDE induced DNA adduct formation, followed by GSTM1-1 and GSTP1-1. These results are consistent with those with pure GSTs. With BPDE, GSTP1-1 was most active in GSH conjugate formation followed by GSTM1-1 and GSTA1-1. The protection against DNA adduct formation displayed an other pattern, GSTP1-1/Ile-105 was most effective followed by GSTM1-1, GSTP1-1/Val-105 and GSTA1-1. Finally, the more lipophilic DBPDE was found to be less available for conjugation than BPDE. However, a greater proportion (up to 50%) of available DBPDE reacts with GSH relative to BPDE (1-3%).

Overall, this study demonstrates the difficulty in extrapolating data obtained with pure enzymes to the complex situation in the intact cell.