PUMA promotes apoptosis of hematopoietic progenitors driving leukemic progression in a mouse model of myelodysplasia.

Myelodysplastic syndrome (MDS) is characterized by ineffective hematopoiesis with resultant cytopenias. Increased apoptosis and aberrantly functioning progenitors are thought to contribute to this phenotype. As is the case for other malignancies, overcoming apoptosis is believed to be important in progression toward acute myeloid leukemia (AML). Using the NUP98-HOXD13 (NHD13) transgenic mouse model of MDS, we previously reported that overexpression of the anti-apoptotic protein BCL2, blocked apoptosis and improved cytopenias, paradoxically, delaying leukemic progression. To further understand this surprising result, we examined the role of p53 and its pro-apoptotic effectors, PUMA and NOXA in NHD13 mice. The absence of p53 or PUMA but not NOXA reduced apoptosis and expanded the numbers of MDS-repopulating cells. Despite a similar effect on apoptosis and cell numbers, the absence of p53 and PUMA had diametrically opposed effects on progression to AML: absence of p53 accelerated leukemic progression, while absence of PUMA significantly delayed progression. This may be explained in part by differences in cellular responses to DNA damage. The absence of p53 led to higher levels of γ-H2AX (indicative of persistent DNA lesions) while PUMA-deficient NHD13 progenitors resolved DNA lesions in a manner comparable to wild-type cells. These results suggest that targeting PUMA may improve the cytopenias of MDS without a detrimental effect on leukemic progression thus warranting further investigation.

Cholesterol-dependent human complement activation resulting in damage to liposomal model membranes.

Human (but not guinea pig) complement-mediated damage. It was concluded that human complement was activated spontaneously by liposomes containing a high concentration (71 mol %) of cholesterol. This occurred in the absence of any recognizable antigen or antibody, and did not occur at a low concentration (43 mol %) of cholesterol. Activation of complement resulted in membrane damage and release of trapped liposomal glucose. The complement activity was inhibited by preheating (56 degrees C, 30 min), 10 mM Mg2EDTA3 or EGTA, and by prior adsorption with insoluble immune complexes. Almost all human sera had some reactivity, but it ranged from very low levels (less than 7% liposomal glucose release) to very high levels (greater than 50% glucose release). Complement activation appeared to be mediated by a serum factor which could be removed by adsorption and which was partially heat labile. The factor was transferred by adding heated high reacting human serum to unheated low reacting human serum, or to guinea pig serum. The serum factor, although quantitatively diminished in potency due to heat lability, caused equal activation of each of these two latter complement sources in the presence of high cholesterol liposomes. It did not cause activation of C4-deficient guinea pig complement. These data suggested that the classical complement pathway was activated. The liposomal membrane composition had an influence on this phenomenon. Activities of about half of the human sera were enhanced when galactosyl ceramide, or ceramide alone, was present in the liposomes. Activity was enhanced by longer fatty acyl chain lengths of lecithin when dimyristoyl-, dipalmitoyl-, or distearoyllecithin was employed in the liposomes. Liposomes containing sphingomyelin as the only phospholipid were not sensitive to cholesterol-dependent complement-mediated damage. It was concluded that human complement was activated in the presence of high concentrations of membrane cholesterol and that this was caused by an uncharacterized serum factor and was influenced by the lipid composition of the membrane.