Disassembly of membrane-associated NSF 20S complexes is slow relative to vesicle fusion and is Ca(2+)-independent.

N-ethylmaleimide-sensitive fusion protein (NSF) and its co-factor soluble NSF attachment protein (alpha)-SNAP) are essential components of the synaptic vesicle fusion machinery and form part of a structurally-conserved 20S protein complex. However, their precise function, relative to fusion itself, is not clear. Using a UV-activated cross-linking approach, we have measured the rate at which a single round of NSF-driven ATP hydrolysis leads to 20S complex disassembly within synaptic membranes. Although this rate is substantially faster than previous estimates of NSF-dependent ATP hydrolysis, it remains much lower than published rates for fusion of synaptic vesicles. Furthermore, the stability of 20S complexes is unaffected by Ca(2+) at concentrations that elicit rapid membrane fusion. We conclude that the ATPase activity of NSF does not contribute directly to vesicle fusion, but more likely plays an earlier role in the synaptic vesicle cycle.

Human rabaptin-5 is selectively cleaved by caspase-3 during apoptosis.

We have previously shown that Xenopus rabaptin-5 is cleaved in apoptotic extracts, with a concomitant reduction in the ability of these extracts to support endosomal membrane fusion (Cosulich, S. C., Horiuchi, H., Zerial, M., Clarke, P. R., and Woodman, P. G. (1997) EMBO J. 16, 6182-6191). In this report we demonstrate that caspase-dependent cleavage is a conserved feature of rabaptin-5. Human rabaptin-5 is cleaved at two sites (HSLD(379) and DESD(438)) in apoptotic HeLa extracts. Cleavage is effected by caspase-3, since it is prevented when caspase-3 activity is either inhibited by Ac-DEVD-CHO or removed by immunodepletion. Moreover, an identical pattern of cleavage is observed using recombinant caspase-3. The action of caspase-3 is highly selective; neither caspase-2 nor caspase-7 are able to cleave recombinant or cytosolic rabaptin-5. Caspase-dependent cleavage of rabaptin-5 generates two physically separated coiled coil-forming domains, the C-terminal of which retains the ability to bind the Rab5 exchange factor rabex-5.

Bcl-2 regulates a caspase-3/caspase-2 apoptotic cascade in cytosolic extracts.

Apoptosis is accompanied by the activation of a number of apoptotic proteases (caspases) which selectively cleave specific cellular substrates. Caspases themselves are zymogens which are activated by proteolysis. It is widely believed that 'initiator' caspases are recruited to and activated within apoptotic signalling complexes, and then cleave and activate downstream 'effector' caspases. While activation of the effector caspase, caspase-3, has indeed been observed as distal to activation of several different initiator caspases, evidence for a further downstream proteolytic cascade is limited. In particular, there is little evidence that cellular levels of caspase-3 that are activated via one pathway are sufficient to cleave and activate other initiator caspases. To address this issue, the ability of caspase-3, activated upon addition to cytosolic extracts of cytochrome c, to cause cleavage of caspase-2 was investigated. It was demonstrated that cleavage of caspase-2 follows, and is dependent upon, activation of caspase-3. Moreover, the activation of both caspases was inhibited by Bcl-2. Together, these data indicate that Bcl-2 can protect cells from apoptosis by acting at a point downstream from release of mitochondrial cytochrome c, thereby preventing a caspase-3 dependent proteolytic cascade.

Formation and turnover of NSF- and SNAP-containing "fusion" complexes occur on undocked, clathrin-coated vesicle-derived membranes.

Specificity of vesicular transport is determined by pair-wise interaction between receptors (SNAP receptors or SNAREs) associated with a transport vesicle and its target membrane. Two additional factors, N-ethylmaleimide-sensitive fusion protein (NSF) and soluble NSF attachment protein (SNAP) are ubiquitous components of vesicular transport pathways. However, the precise role they play is not known. On the basis that NSF and SNAP can be recruited to preformed SNARE complexes, it has been proposed that NSF- and SNAP-containing complexes are formed after SNARE-dependent docking of transport vesicles. This would enable ATPase-dependent complex disassembly to be coupled directly to membrane fusion. Alternatively, binding and release of NSF/SNAP may occur before vesicle docking, and perhaps be involved in the activation of SNAREs. To gain more information about the point at which so-called 20S complexes form during the transport vesicle cycle, we have examined NSF/SNAP/SNARE complex turnover on clathrin-coated vesicle-derived membranes in situ. This has been achieved under conditions in which the extent of membrane docking can be precisely monitored. We demonstrate by UV-dependent cross-linking experiments, coupled to laser light-scattering analysis of membranes, that complexes containing NSF, SNAP, and SNAREs will form and dissociate on the surface of undocked transport vesicles.

Glycerolipid metabolizing enzymes in rat ventricle and in cardiac myocytes.

1. The properties and subcellular distribution of phosphatidate phosphohydrolase (PAP) were studied in rat heart. A Mg2(+)-activated activity (PAP1) which was inhibited by N-ethylmaleimide was found mainly in a 105,000 x g soluble fraction. Isolation of the membranes in a medium containing KCl increased the proportion of PAP1 that was associated. Translocation of PAP1 from these membranes occurred on subsequent incubation in a low-ionic strength medium from which KCI was omitted. Incubation of cardiac myocytes with palmitate promoted translocation of PAP activity to cellular membranes. A second activity which was insensitive to N-ethylmaleimide (PAP2) was found in the 105,000 x g membrane fraction. PAP2 was inhibited by concentrations of Mg2+ known to occur in ischaemia. Specific activities of PAP1 and PAP2 in ventricle muscle homogenates were similar. The specific activity of PAP2 in homogenates of cardiac myocytes was only 42% of that in homogenates of ventricle muscle. 2. A glycerolphosphate acyltransferase (GPAT) activity with properties similar to the GPAT found in microsomes from liver or adipose tissue was enriched in the sarcoplasmic reticulum fraction from ventricle muscle. This GPAT had a significantly higher K(m) for glycerol 3-phosphate than the GPAT found in adipose tissue microsomes. The possible physiological significance of this 'high K(m)' GPAT in heart, particularly in ischaemia, is discussed. 3. Comparisons were made of the specific activities of fatty acyl-CoA synthetase, monoacylglycerolphosphate acyltransferase, diacylglycerol acyltransferase and the mitochondrial and microsomal forms of GPAT in homogenates from cardiac myocytes and ventricle muscle.

Effects of adrenaline on triacylglycerol synthesis and turnover in ventricular myocytes from adult rats.

Ca2+-tolerant myocytes were isolated with endogenous triacylglycerol (TAG) stores prelabelled with [3H]palmitate and subsequently incubated for a 1h chase period with [14C]palmitate, 2% albumin and 5mM glucose. Measurements were then made of [14C]palmitate conversion into TAG and phospholipids, of loss of [3H]TAG, of glycerol release and of change in the total TAG content. Rates of de novo synthesis of TAG were calculated by a balance method. With 0. 5mM palmitate present, 5 microM adrenaline increased de novo synthesis of TAG by 81% and incorporation of [14C]palmitate into phospholipids by 59%. Significant increases in these processes with adrenaline were also seen with 0.08, 0.14 and 0.26 mM palmitate. The beta-agonist isoprenaline had little effect on de novo synthesis of TAG and had no effect on [14C]palmitate conversion into phospholipids. The alpha1-agonist phenylephrine mimicked adrenaline in increasing [14C]palmitate conversion into phospholipids but had no effect on de novo synthesis of TAG. Adrenaline did not significantly alter the myocyte glycerol 3-phosphate content but caused a persistent 40% increase in the activity of the form of glycerolphosphate acyltransferase found predominantly in the sarcoplasmic reticulum. With 0.5 mM palmitate present, the value [14C]TAG formed -decrease in [3H]TAG consistently exceeded the enzymically measured change in cell TAG content. From this it was suggested that the specific radioactivity of [3H]TAG pool(s) mobilized during the chase period was lower than that of the overall cell TAG. In the basal state, complete mobilization of TAG measured as glycerol release was low, but cycling of TAG to diacylglycerol or monoacylglycerol and back to TAG appeared to be high. With adrenaline present, glycerol release was increased 5-6-fold but recycling of lower acylglycerols to TAG was abolished. Glycerol release was inhibited by increasing extracellular palmitate from 0.08 to 0.5 mM. Adrenaline partially over-rode this effect.