Reduction of phospholipase D activity during coxsackievirus infection

Footnotes

In enterovirus-infected cells, the phospholipids that compose organelle and vesicle membranes are extensively altered. Besides a significant increase in de novo phospholipid synthesis that may result in changes in membrane lipid composition (Guinea & Carrasco, 1990), several cellular lipid-modifying enzymes may also contribute to these alterations. For example, activities of phosphatidylcholine-specific phospholipase C and phosphatidylinositol-specific phospholipase C increase, while phospholipase A₂ activity (PLA₂) decreases (Guinea et al., 1989; Irurzun et al., 1993). In these studies, phospholipase D (PLD) activity was not investigated. PLD is a ubiquitous lipid-modifying enzyme that catalyses the conversion of phosphatidylcholine (PtdCho) and phosphatidylethanolamine (PtdEtn) to phosphatidic acid (PtdOH) (Fig. 1a), and acts in various cellular signalling pathways, protein secretion and membrane trafficking (Exton, 2000; Ktistakis et al., 2003). Alteration of its activity might contribute significantly to the lipid modifications that occur during infection. Hence, we set out to investigate the role and regulation of PLD activity during coxsackievirus B3 (CVB3) infection.

Although PLD activity appeared to be dispensable for viral replication, it might be altered during enterovirus infection. To investigate this, we labelled cells metabolically overnight with [³H]myristic acid, which is incorporated in all de novo formed phospholipids. Cells were then either mock-infected or infected with CVB3 at an m.o.i. of 50 and harvested at 0, 2, 4, 6 and 8 h post-infection (p.i.). At 1 h prior to harvesting, ethanol was added to the medium to a final concentration of 1 %. Using this approach, PLD activity was directly reflected by the amount of phosphatidylethanol (PtdEth) produced. Subsequently, lipids were extracted as described previously (Bligh & Dyer, 1959), separated by TLC, and the amount of [³H]PtdEth produced from the total [³H]-labelled phospholipid pool was determined as described previously (Bosch et al., 1999). Because of the low basal activity of PLD, we applied a short-term (1 h) treatment of the cells with phorbol-12-myristate-13-acetate (PMA), which greatly potentiates PLD activation (Chen & Exton, 2004; Lee et al., 1997; Whatmore et al., 1994). The results showed that PLD activity decreased as early as 2 h p.i., and continued to decrease down to 20 % of the level of mock-infected cells by 8 h p.i. (Fig. 2a). This decrease in PLD activity was, most likely, not due to reduced protein levels or proteolytic degradation of PLD, nor of its activator protein kinase C alpha (PKCα), as only minor decreases in PLD1 and PKCα protein levels were seen at these time points and no cleavage products were observed (Fig. 2b).

What then may be the cause of the reduced PLD activity? No reduction in PLD or PKCα protein levels was seen, ruling out the option that decreased PLD activity is the result of inhibition of gene expression or degradation of these proteins. Our results show that the reduction in PLD activity is not caused by inhibition of Arf1 by the 3A protein. This finding, together with the observation that the level of active Arf1 increases during poliovirus infection (Belov et al., 2007), strongly suggests that enteroviruses inhibit PLD activity at a step downstream of Arf1. Long-term disassembly of the Golgi complex, rather than Arf1 inhibition per se, has been suggested to be the underlying cause of decreased PLD activity (Guillemain & Exton, 1997). Enteroviruses cause a profound membrane reorganization, leading to the accumulation of vesicles at which viral RNA replication takes place, and, ultimately, the disassembly of the Golgi complex (Cho et al., 1994). The viral 2BC protein has been shown to play a major part in this membrane reorganization (Cho et al., 1994) and thereby may affect PLD activity. In our hands, however, ectopic expression of 2BC protein was rather inefficient and therefore we could not test the possible effects of 2BC protein on PLD activity. To date no viable mutant viruses have been described expressing a defective 2BC protein because of the important role of 2BC protein in viral RNA replication. For these reasons, we were unable to determine the involvement of 2BC protein in the downregulation of PLD activity. Although 2BC-mediated effects on the Golgi complex may offer a plausible explanation for the decrease of PLD activity in infected cells, other explanations, which may be non-mutually exclusive, cannot be ruled out.

What would be the consequence of a reduced PLD activity for the virus? PLD is a ubiquitous enzyme that generates PtdOH, an important second messenger (Exton, 2000). PtdOH in turn can be used by phosphatidic acid phosphohydrolase and diacylglycerol lipase to generate arachidonic acid (AA) (Serhan et al., 1996), a very potent second messenger that can be converted to several eicosanoid hormones, like prostaglandins, which play important roles in sustaining inflammatory responses and display antiviral activities against picornaviruses (Ankel et al., 1985; Conti et al., 1996, 1999). Besides PLD, PLA₂ can also generate AA directly from membrane phospholipids (Fujita et al., 1996), and it was shown that PLA₂ is strongly inhibited during poliovirus infection, severely reducing the amount of AA produced (Guinea et al., 1989). By additionally interfering with PLD activity, the virus would reduce even further the production of AA, thus helping it to evade the immune system.