Synthetic 1,4-anthracenedione analogs induce cytochrome c release, caspase-9, -3, and -8 activities, poly(ADP-ribose) polymerase-1 cleavage and internucleosomal DNA fragmentation in HL-60 cells by a mechanism which involves caspase-2 activation but not Fas signaling

Abstract

Synthetic analogs of 1,4-anthraquinone (AQ code number), a compound that mimics the antiproliferative effects of daunorubicin (daunomycin) in the nanomolar range in vitro but has the advantage of blocking nucleoside transport and retaining its efficacy in multidrug-resistant tumor cells, were tested for their ability to induce apoptosis in the HL-60 cell system. AQ10 and, especially, the new lead antiproliferative compounds AQ8 and AQ9 reduce the growth and integrity of wild-type, drug-sensitive, HL-60-S cells more effectively than AQ1, suggesting that various methyl group substituents at C6 may enhance the bioactivity of the parent compound. Internucleosomal DNA fragmentation, a late marker of apoptosis, is similarly induced in a biphasic manner by increasing concentrations of AQ8 and AQ9 at 24 hr. Poly(ADP-ribose) polymerase-1 (PARP-1) cleavage, an early event required for cells committed to apoptosis, is detected within 3-6 hr in HL-60-S cells treated with AQ9. In accord with the fact that the caspases 9 and 3 cascade is responsible for PARP-1 cleavage, the activities of initiator caspase-9 and effector caspase-3 are induced by AQ9 in the same time- and concentration-dependent manners and to the same maximal degrees in both the HL-60-S and multidrug-resistant HL-60-RV cell lines. Interestingly, a 1-hr pulse treatment is sufficient for AQ8 and AQ9 to maximally induce caspase-9 and -3 activities at 6 hr. The release of mitochondrial cytochrome c (Cyt c) is also detected within 3-6hr in HL-60-S cells treated with AQ9, a finding consistent with the fact that Cyt c is the apoptotic trigger that activates caspase-9. Moreover, AQ analogs induce Cyt c release, caspase-9 and -3 activities and PARP-1 cleavage in relation with their abilities to decrease tumor cell growth and integrity, AQ8 and AQ9 being consistently the most effective. Since apical caspases 2 and 8 may both act upstream of mitochondria to promote Cyt c release, it is significant to show that AQ9 maximally induces caspase-2 and -8 activities at 6 and 9 hr, respectively. During AQ8 treatment, the caspase-2 inhibitor benzyloxycarbonyl (z)-Val-Asp-Val-Ala-Asp (VDVAD)-fluoromethyl ketone (fmk) totally blocks caspase-9, -3, and -8 activations, whereas the caspase-8 inhibitor z-Ile-Glu-Thr-Asp-(IETD)-fmk does not prevent caspase-2, -9, and -3 activations, suggesting that AQ-induced caspase-2 activity is an upstream event critical for the activation of the downstream caspases 9 and 3 cascade, including the mitochondrial amplification loop through caspase-8. However, these caspase-2 and -8 inhibitors fail to alter AQ8-induced Cyt c release, suggesting that AQs might also target mitochondria independently from caspase activation. Furthermore, the antagonistic anti-Fas DX2 and ZB4 monoclonal antibodies (mAbs), which block the induction of Cyt c release and caspase-2, -8, and -9 activities by the agonistic anti-Fas CH11 mAb, and the neutralizing anti-Fas ligand (FasL) NOK-1 mAb all fail to inhibit AQ9-induced Cyt c release and caspase-2, -8, and -9 activities, suggesting that the FasL/Fas signaling pathway is not involved in the mechanism by which antiproliferative AQ analogs trigger apoptosis in HL-60 cells.