The function of TAG synthesis in protection against FA-induced cell death is conserved among eukaryotes
Previous studies in yeast and mammalian model systems have firmly established that, in addition to serving as a means for efficient carbon and energy storage, TAG synthesis plays a critical role in detoxifying FFAs and other lipid metabolic intermediates (Yen et al., 2008; Kohlwein, 2010). This conclusion is based primarily on two key observations. First, deficiency of TAG synthesis genes renders both yeast (Garbarino et al., 2009; Petschnigg et al., 2009) and mammalian cells (Listenberger et al., 2003) highly sensitive to exogenous FA supplementation in terms of cell growth and survival. Second, disruption of TAG synthesis induces programmed cell death (Buszczak et al., 2002; Zhang et al., 2003; Stone et al., 2004), which has been linked to increased levels of FFAs (Fakas et al., 2011) or DAG (Zhang et al., 2003) in yeast. Although direct evidence for a role of TAG synthesis in protection against toxic lipid intermediates is still lacking in plants, it has been shown that Arabidopsis double mutants defective in DGAT1 and PDAT1 lack viable pollen, and RNA interference-mediated silencing of DGAT1 or PDAT1 in the pdat1 or dgat1 mutant background, respectively, results in defects in pollen and embryo development (Zhang et al., 2009). In addition, deficiency of DGAT1 activity in Brassica napus (Lock et al., 2009) and in Arabidopsis wild-type plants (Katavic et al., 1995; Zou et al., 1999) and mutants defective in FA breakdown (Slocombe et al., 2009) has been shown to have negative effects on plant growth and development. The results of the present study show that knockout of PDAT1 increases the sensitivity of developing leaves to exogenous FFAs. In addition, disruption of PDAT1 in the tgd1–1 mutant background causes premature cell death in developing leaves and floral organs, with concomitant increases in the levels of FFAs and DAG. Taken together, these results demonstrate an evolutionally conserved role for TAG synthesis in protection against the toxicity induced by FFAs and possibly other lipid metabolic intermediates in yeast, plants and mammals.
PDAT plays a critical role in lipid homeostasis
A working model for a critical role of PDAT1 in buffering cells against the cytotoxic effects of FFAs in non-seed tissues is shown in Figure 9. FAs are almost exclusively synthesized in the plastid (Ohlrogge et al., 1979). In Arabidopsis leaves, approximately 40% of these newly synthesized acyl chains are directly utilized in the plastid to support the prokaryotic pathway of thylakoid lipid synthesis (Browse et al., 1986). The remainder are exported outside the plastid, and first used to acylate LPC to produce PC, catalyzed by LPCATs (Bates et al., 2007, 2012; Wang et al., 2012). After 18:1 desaturation to 18:2 and 18:3 by ER-resident desaturases, the resulting polyunsaturated PC has multiple possible metabolic fates including (i) deacylation to LPC and FAs as part of the acyl-editing cycle (Bates et al., 2007, 2009; Tjellström et al., 2012), (ii) conversion to thylakoid lipids through the eukaryotic pathway, which brings approximately 40% of the exported FAs (24% of the total newly synthesized FAs) back into chloroplasts (Browse et al., 1986), (iii) conversion to polyunsaturated DAG for TAG synthesis (Bates and Browse, 2012), or (iv) transacylation catalyzed by PDAT1, resulting in TAG and LPC. In growing leaves of wild-type plants, <3% of nascent acyl chains are used for TAG synthesis (Figure S7). In the tgd1–1 mutant, a defect in the transport of PC-derived lipid precursors into chloroplasts results in a 3.8-fold increase in the rate of FAS and a threefold increase in the rate of FA breakdown, presumably through β–oxidation. In addition, there is a 2.4-fold increase in the acyl flux into TAG (Figure S7), largely via the acyl CoA-independent reaction catalyzed by PDAT1 (Figure 1a). Reacylation of LPC generated by PDAT1 using nascent FAs exported from the plastid regenerates PC and thus creates a deacylation and reacylation cycle that consumes toxic lipid intermediates such as fatty acyl chains and DAG, with net production of inert TAG.
Figure 9. Working model for the function of PDAT1 in maintaining lipid homeostasis in non-seed tissues.
In the wild-type plants, a major fraction of PC or its derivative is transported into the chloroplast for synthesis of glycolipids (GLs). This process is blocked (double lines) in the tgd1–1 mutant, which leads to (i) enhanced prokaryotic thylakoid lipid synthesis, and (ii) constitutive activation of fatty acid synthesis (FAS). FAs in excess of cellular needs are (i) degraded by β–oxidation or (ii) sequestered into inert TAG through the PDAT1-mediated acyl CoA-independent route, thereby protecting cells from detrimental effects of FFAs and possibly other lipid intermediates in the tgd1–1 mutant. Lighter arrows indicate pathways and metabolism that are enhanced in the tgd1–1 mutant. Dashed lines indicate potential sources of DAG and PC for TAG synthesis mediated by PDAT1. PL, phospholipids.
Download figure to PowerPoint
In developing leaves of the tgd1–1 mutant, disruption of PDAT1 resulted in no substantial changes in the rates of FAS or FA degradation. However, there were significant increases in the PC and PE levels in developing leaves of the tgd1–1 pdat1–2 mutant. Increases in membrane lipid content were also observed in yeast mutants defective in TAG synthesis upon FA supplementation (Garbarino et al., 2009; Petschnigg et al., 2009; Connerth et al., 2010) or even without FA supplementation (Fakas et al., 2011), and such increases have been postulated as an alternative means for protection against toxic effects of FAs. Apparently, the increased phospholipid synthesis is insufficient to fully compensate for the buffering function conferred by PDAT1-mediated TAG synthesis, which may explain why FFA levels increase in tgd1–1 pdat1–2.
Initiation and propagation of the cell death phenotype in tgd1–1 pdat1–2 were observed in rapidly growing cell types but not in mature and senescing tissues (Figure 3). The underlying reason for the observed developmental variations in phenotypes may lie in the fact that both the rates of FAS (Kannangara et al., 1973) and the proportions of de novo synthesized acyl chains exported outside the chloroplast (Andersson et al., 2001) are much higher in expanding leaves than in mature leaves. In addition, transcripts encoding key enzymes of β–oxidation, the major pathway for FA degradation in plants (Gerhardt, 1992), are normally expressed at very low levels during leaf growth but are induced during leaf senescence (Troncoso-Ponce et al., 2013), which may explain why disruption of PDAT1 in growing leaves of tgd1–1 did not lead to a further increase in the rate of FA degradation (Figure 7c). Thus, a block in the return of acyl chains from the ER to chloroplasts results in much more FAs being available for TAG synthesis, and higher accumulation of TAG in developing leaves than in expanded and senescent leaves, and disruption of both acyl chain return and TAG synthesis causes higher accumulation of toxic lipid intermediates, thereby resulting in more deleterious effects in the growing leaves than in mature leaves. A situation similar to that in young growing leaves appears to exist in floral organs, which accumulate almost nine times more TAG compared with the growing leaves of the tgd1–1 mutant, and TAG accumulation in flowers is, to a major extent, mediated by PDAT1 (Figure 1a). Consequently, disruption of PDAT1-mediated TAG synthesis in the tgd1–1 mutant background results in marked increases in levels of FFA and DAG, leading to the observed defects in the development of male and female gametophytes in tgd1–1 pdat1–2 double mutant plants. In this regard, it is worth noting that loss of both PDAT1 and DGAT1 functions in dgat1 pdat1 double mutants affects the viability of male but not female gametophytes (Zhang et al., 2009). Thus, although the amount of TAG is markedly reduced in the flowers of the tgd1–1 pdat1–2 mutant, it seems unlikely that the gametophytic developmental defects in tgd1–1 pdat1–2 are mainly due to the deficiency in TAG synthesis, as both male and female gametophytes are affected in the double mutant.
An important question arising from this work relates to the metabolic origin of DAG for TAG synthesis. In plants, two major pathways for DAG synthesis are (i) de novo synthesis through glycerol-3–phosphate acylation, and (ii) conversion of PC into DAG (Bates and Browse, 2012). In developing seeds, PC is the major DAG donor for TAG synthesis, whereas DAG derived from glycerol-3–phosphate acylation reactions is used for the synthesis of PC (Bates et al., 2009). Similarly, two lines of evidence suggest that PC may be the major source of DAG for TAG synthesis in growing leaves. First, the FA composition of both TAG (Figure 1b) (Xu et al., 2005) and DAG (Figure S4b) in leaves of the tgd1–1 or tgd1–1 pdat1–2 mutants, respectively, resembles that of PC (Figure S5). Second, disruption of PDAT1 results in increases in the levels of PC and PE in growing leaves of tgd1–1 (Figure 6). If PC is the DAG donor for TAG synthesis in developing leaves, the net result of the deacylation and reacylation cycle driven by PDAT1 and LPCAT is production of TAG at the expense of PC and acyl CoA. This is in line with the proposed function of PDAT in the regulation of membrane lipid composition (Dahlqvist et al., 2000).
Potential acyltransferases responsible for LPC reacylation to PC in leaves include LPCAT1 and 2, which have recently been shown to play a major role in PC acyl editing in developing seeds (Bates et al., 2012; Wang et al., 2012). In addition, recent genetic evidence also suggests that LPCAT2 is a key enzyme catalyzing LPC reacylation to support PDAT1-mediated TAG synthesis in seeds of the dgat1–1 mutant (Xu et al., 2012). In rapidly expanding pea leaves, substantial LPCAT activity is associated with chloroplast envelope membranes (Bessoule et al., 1995; Tjellström et al., 2012). It has been proposed that the chloroplast-associated LPCAT is involved in channeled incorporation of nascent acyl chains into PC via acyl editing (Tjellström et al., 2012). Although the exact subcellular localization of PDAT1 remains to be determined, PDAT1 activity has been shown to reside in microsomal membranes of Arabidopsis roots and leaves (Ståhl et al., 2004). Given the amphiphilic nature of LPC, its rapid partitioning between the ER and chloroplast membranes (Bessoule et al., 1995), and the close physical association between the ER and the plastid (Andersson et al., 2007; Xu et al., 2008), it is likely that the PDAT1/LPCAT cycle may be directly involved in directing newly synthesized acyl chains into TAG without mixing with the bulk cytosolic acyl CoA pool.
In summary, our results show that PDAT plays a critical role in mediating TAG synthesis in young growing tissues of plants. Loss of function of PDAT disrupts lipid homeostasis, leading to gametophytic defects and premature cell death in growing cell types. Studies in yeast and mammalian systems suggested that disruption of TAG synthesis often triggers lipoapoptosis, a type of FA-induced programmed cell death that has a distinct set of physiological and morphological features, including nuclear condensation, DNA fragmentation and elevated oxidative stress (Buszczak et al., 2002; Zhang et al., 2003; Fakas et al., 2011). Future studies using pdat1–2 and tgd1–1 pdat1–2 mutants will address how TAG synthesis deficiency affects the cell structure and function, and assess the mechanistic basis underlying cytotoxic effects of FFAs in plant model systems.