Species differences in sequence and activity of the peroxisome proliferator response element (PPRE) within the acyl...

In rats and mice, peroxisome proliferators (PP) cause liver enlargement, hepatocarcinogenesis and peroxisome proliferation associated with induction of enzymes such as acyl CoA oxidase (ACO). However, humans appear to be non-responsive to the adverse effects of PPs such as ACO induction. PPs activate the peroxisome proliferator activated receptor alpha (PPARalpha) that binds to DNA at peroxisome proliferator response elements (PPREs) within the promoters of PP-responsive genes. When the human ACO promoter was cloned previously (Varanasi et al., 1996. Journal of Biological Chemistry, 271, 2147-2155), it was reported to contain a PPRE (5' AGGTCA C TGGTCA 3') that bound PPARalpha and could be activated in vitro by Wyeth-14,643 (at >1 mM) or DEHP (at > 1.5 mM). In contrast, when we cloned the ACO gene promoter from a human liver biopsy, it was non-responsive to PPs and differed at three positions (5' AGGTCA G CTGTCA 3') from that reported previously (Woodyatt et al., 1999. Carcinogenesis, 20, 369-375). Subsequent to this, Varanasi et al. re-sequenced their constructs and obtained the same sequence as we have described (Varanasi et al., 1998. Journal of Biological Chemistry, 273, 30832). However, the observation that the errant sequence (5' AGGTCA C TGGTCA 3') was able to bind PPARalpha still remained since it appears that this sequence was used by Varanasi et al. (1996) to design oligonucleotides for their DNA binding analyses. Thus, if the 5' AGGTCA C TGGTCA 3' sequence did exist in some individuals, it could be active. To address this, we used site-directed mutagenesis to create a promoter fragment that contained the errant sequence. This reporter gene was transfected into NIH3T3 cells together with a plasmid expressing mPPARalpha, and assessed for its ability to drive PP-mediated gene transcription using a non-toxic concentration of Wyeth-14,643 (100 microM). This human ACO promoter was also inactive, unlike the equivalent rat ACO promoter fragment used as a positive control. Next, we used site directed mutagenesis to convert the PPRE found in the active rat ACO promoter (3' AGGACA A AGGTCA 5') to our inactive human sequence (AGGTCA G CTGTCA). This human PPRE was unable to drive PP-induced gene transcription even in the context of the rat ACO promoter suggesting that the activity of the rat promoter is conferred principally by the PPRE sequence, even though it may be enhanced by flanking sequences. These data confirm that neither the native nor the errant human ACO gene PPRE can respond to PPs. The absence of a responsive PPRE contributes to our understanding of the lack of response of humans to some of the adverse effects of the PP class of non-genotoxic hepatocarcinogens.