Phosphatidylserine (PS) exposure on the external leaflet of the plasma membrane is widely observed during apoptosis and forms the basis for the annexin V binding assay to detect apoptotic cell death. Current efforts to explain PS exposure focus on two potential mechanisms, activation of a phospholipid scramblase or calcium-mediated trafficking of lysosomes to the cell surface. Here, we provide evidence that apoptotic PS exposure instead reflects bidirectional trafficking of membrane between the cell surface and cytoplasm. Using a series of cell lines, some of which expose large amounts of PS during apoptosis and some of which do not, we demonstrate that accumulation of plasma membrane-derived cytoplasmic vesicles in a dynamin-, clathrin-and Cdc42-independent manner is a previously undescribed but widely occurring feature of apoptosis. The apoptotic exposure of PS occurs when these vesicles traffic back to cell surface in a calcium-dependent process that is deficient in a substantial fraction of human cancer cell lines. These observations provide a new model for PS externalization during apoptosis and simultaneously identify an altered step that accounts for the paucity of apoptotic PS exposure in many cell lines. Cell Death and Differentiation (2013) 20, 64-76; doi:10.1038/cdd.2012; published online 3 August 2012A common feature of apoptosis from Caenorhabditis elegans to man is the transfer of phosphatidylserine (PS) and phosphatidylethanolamine, which ordinarily reside on the cytoplasmic surface of the membranes, to the cell surface.

Since mitochondrial factors have been implicated in apoptosis, experiments were designed to assess whether or not the potent mitochondrial nuclease could be one of these factors. Nuclei isolated by two different methods were found to contain mitochondrial nuclease in masked form. This nuclease was released by treatment with the non-ionic detergent NP-40 and rendered trypsin-sensitive. It was not removed appreciably from the nuclei by washing and sedimentation of the nuclei through a sucrose cushion. Levels of the mitochondrial nuclease were followed during drug-induced apoptosis. Time courses of apoptosis in cultures of HL-60 cells were monitored by flow cytometry of propidium iodide-stained cells and by agarose gel electrophoresis of extracted DNA. Changes in the inner mitochondrial transmembrane potential were monitored by flow cytometry of chloromethyl-X-Rosamine-stained cells. Apoptosis was induced by treatment with either the chemotherapeutic agent etoposide (VP-16 at 10 microM) over an 8 h period or with the anti-rheumatic agent hydroxychloroquine (HCQ at 0.28 mM) over a 24 h period. These two drugs likely act in different pathways of apoptosis. VP-16 caused loss of the mitochondrial transmembrane potential 1.0-1.5 h before apoptosis was detected. On the other hand, treatment with HCQ caused these processes to occur in parallel possibly indicating that the mitochondrial changes are secondary events. No losses of masked mitochondrial nuclease were detected with either drug treatment during the course of apoptosis. HL-60 mitochondrial DNA was also not degraded during apoptosis induced by either agent. These observations likely explain why the mitochondrial DNA is not degraded and make it unlikely that mitochondrial nuclease plays any role in vivo in chromatin DNA fragmentation.