|Home | About | Journals | Submit | Contact Us | Français|
The sterol regulatory element binding protein (SREBP) pathway plays a central role in the global regulation of lipid homeostasis. SREBPs are membrane-bound transcription factors whose proteolytic activation is regulated by cellular lipid levels; when demand for lipid rises, SREBP travels from the endoplasmic reticulum to the Golgi apparatus where it is cleaved by two distinct proteases. Cleavage releases the transcription factor domain of SREBP from the membrane-bound precursor and transcription of its target genes consequently rises. Previously, we isolated Drosophila mutants null for dsrebp and others lacking site-2 protease (ds2p), the second of two Golgi-resident proteases that cleave dSREBP. dScap is a protein needed to escort dSREBP from the ER to the Golgi apparatus. We recently characterized the phenotypes of dscap mutants as well. Here, we describe additional details of phenotypes arising from the inability to activate SREBP appropriately.
SREBPs are best known for their role in the transcriptional regulation of cholesterol synthesis and uptake. Mammalian cells that lack components of the SREBP processing machinery (Scap, site-1 protease (S1P) or S2P) cannot cleave SREBP. In consequence, these cells are deficient in the transcription of numerous genes required for normal lipid homeostasis. They do not survive unless free cholesterol is added to their culture medium.1 In contrast to vertebrates, most other animals (e.g., arthropods and nematodes), lack de novo sterol synthesis. They are natural cholesterol auxotrophs yet they too possess orthologues of SREBP, S1P, S2P and Scap.
The cholesterol-auxotrophic mammalian cells also require unsaturated fatty acids for growth.1 Studies in cultured cells and in mice demonstrated that SREBPs also mediate the transcription of genes involved in fatty acid and phospholipid biosynthesis.2 This role in fatty acid and phospholipid synthesis explains why orthologues of SREBP are found in all animals, even those that cannot synthesize sterols de novo.3
Analyzing mutant cultured cells is a powerful means of assessing protein function4 but it has limitations. Analysis of mutant organisms has the potential to reveal phenotypes that cannot be revealed by cells alone. Drosophila melanogaster offers substantial advantages for studying the non-sterol aspects of SREBP function. In addition to the many, oft-cited molecular and genetic tools available, flies have only a single SREBP gene and they do not make sterols. This simplifies analysis of the role of the SREBP pathway in the synthesis fatty acids and phospholipids. In order to analyze the role of SREBPs in non-sterol lipid homeostasis, we examined the SREBP pathway in Drosophila melanogaster. In flies, proteolytic processing of dSREBP is unresponsive to sterols.5 Instead, it is regulated by levels of phosphatidylethanolamine.6 Drosophila provided the first opportunity to study animals completely lacking SREBP.7 Flies lacking dSREBP die at the end of second instar but lethality can be rescued by supplementing their diet with additional fatty acid or phospholipid. Although crucial to survival during larval life, once the animals reach adulthood, dSREBP is no longer essential.8
We also isolated flies lacking dS2P and dScap.9,10 In striking contrast to mammalian cells lacking these proteins, the ds2p and dscap null flies are viable and are easily maintained as homozygous stocks. The ds2p null flies survive owing to an alternative cleavage of dSREBP by the caspase Drice.11 This cleavage occurs on the cytoplasmic side of the membrane, between the transcription factor domain of dSREBP and the first membrane-spanning helix. Unlike the situation with dS1P and dS2P, it is unknown in which compartment Drice cleaves the precursor. Cleavage of dSREBP by Drice nevertheless releases transcriptionally-active dSREBP and requires an aspartate residue at position 386 in the substrate.
Flies lacking both ds2p and drice phenocopy dsrebp null flies since they cannot cleave dSREBP. Like the dsrebp null animals, these double-mutants die at the end of second instar. As expected, the ds2p−; drice− larvae display deficits in the transcription of dSREBP target genes that are similar to the deficits observed in dsrebp− larvae (Fig. 1A). Just as for the dsrebp null animals, supplementing their diet with exogenous lipid affords substantial rescue to the ds2p; drice double mutants.11
Flies whose only copies of dsrebp harbor a site-specific mutation (Asp386 to Ala; D386A) that abrogates cleavage by Drice survive about as well as wild-type flies. In a result complementary to that from ds2p; drice flies, when dsrebp(D386A) animals also lack ds2p, they too die at the end of second instar. Therefore lethality results from inability of Drice to cleave dSREBP rather than some other aspect of dS2P or Drice function. Again, this lethality is rescued by dietary supplementation with exogenous lipid.11 Thus, Drice cleavage of dSREBP provides an alternative means of activating this important transcription factor.
While dS2P and Drice are directly responsible for the proteolytic release of dSREBP, dScap is needed to escort dSREBP to the Golgi in response to cellular lipid levels. Mammalian cells lacking Scap cannot cleave SREBP.1 In Drosophila S2 cells, dSREBP is likewise not cleaved in the absence of dScap.5 As noted above, flies harboring null alleles of dscap survive reasonably well. Like the ds2p mutants, dscap mutants continue to cleave dSREBP in a subset of tissues. Under standard fly culture conditions, they emerge at about 70% of the expected rate. By contrast with ds2p null mutants, survival of the dscap null animals does not require Drice. Larvae doubly-mutant for dscap and drice survive about as well as either single mutant alone.9 Further, they display deficits in the transcription of dSREBP target genes that are no more severe than those of dscap single mutants (not shown).
Previously, we demonstrated that flies lacking dsrebp accumulated less lipid than wild-type flies but that its relative composition was unchanged.7 Similarly, we showed that dscap, ds2p and dscap ds2p adult flies also accumulated less total lipid than wild-type flies.9 We reared these mutants on standard medium and assayed larvae for triglyceride (TG) content. TG abundance is reduced in the mutant larvae (Fig. 2), comparable to the reduction in total lipids in the adult flies. While loss of dSREBP (or the machinery needed to process it) has a pronounced effect on the amount of lipid an animal accumulates, little effect is seen on the classes of lipid accumulated.
Figure 3 shows adult emergence curves for flies homozygous for mutations in dscap, ds2p or dsrebp. The dscap mutants (Fig. 3B) begin to emerge about five days later than wild-type flies and this delay is observed for the dscap ds2p (Fig. 3D) double mutants as well. The dscap and ds2p mutants do not survive as well as wild-type flies but survive much better than do dsrebp mutants (Fig. 3E). For the single mutants and for the dscap ds2p double mutants, supplementation of the larval diet with exogenous lipid substantially restores normal survival to these animals. This indicates that reduced survival of the mutants is largely owing to lipid deficiency under standard culture conditions.
Even on supplemented medium, the mutants continue to emerge later than wild-type flies. Surprisingly this phenotype is consistently more pronounced for dscap mutants (Fig. 3B) than for ds2p mutants (Fig. 3C), even though loss of ds2p has a more severe effect on survival. Thus, loss of dscap, ds2p or dsrebp confers phenotypes that are not rescued simply by supplementing the medium with exogenous lipid. This may reflect a need for endogenous lipid synthesis in some tissue(s) that cannot be compensated by dietary sources. Alternatively, the residual phenotypes may point to lipid-independent roles for components of the SREBP pathway, acting together or individually.
It remains possible that differences in feeding behavior contribute to some aspects of the mutants' phenotypes. We have verified by dye-uptake studies that larvae of each genotype do ingest food. On wet-yeasted grape juice agar plates, we observe no differences in larval movement or distribution when compared to control wild-type larvae of the same developmental stage (not shown). Further, dietary supplementation affords substantially increased survival for the dscap and ds2p mutants (Fig. 3B–D). This indicates that they feed sufficiently to support nearly normal development under conditions of fatty acid supplementation. However, we have not quantified rates of food uptake or the absolute mass consumed by the various mutants during larval life. If the mutants have reduced rates of consumption relative to wild-type animals, this could contribute to their delayed development and/or reduced lipid content.
Components of the SREBP pathway that are essential for mammalian cells in culture (S2P, Scap) are dispensable in the fly. In the absence of dS2P, cleavage by Drice produces sufficient dSREBP activity to enable larvae to survive. We do not yet know the mechanism that permits dS1P and dS2P to cleave dSREBP in larvae lacking dScap. It seems likely that these alternative means of activation operate in wild-type larvae, at least under some circumstances. In the presence of the classical processing machinery, their contribution to overall dSREBP activity appears to be modest. Their contribution does afford enough dSREBP activity to permit survival of the mutants. It will be interesting to learn whether homologous alternative ways of activating SREBP also occur in mammals.
This work was supported by grants from the National Institutes of Health (R01 GM07145701A1) and the Perot Family Foundation.
Extra View to: Matthews KA, Ozdemir C, Rawson RB. Activation of sterol regulatory element binding proteins in the absence of Scap in Drosophila melanogaster. Genetics. 2010;185:189–198. doi: 10.1534/genetics.110.114975.