|Home | About | Journals | Submit | Contact Us | Français|
The regulation and function of autophagy in response to metabolic signals is not yet well understood. A recent study from our laboratory indicates than an intracellular parasite, Toxoplasma gondii, derives nutritive benefit from the upregulation of host cell autophagy. We discuss these and related findings suggesting that autophagy in infected cells functions as part of a metabolic futile cycle. The hypothesis is presented that endogenous autophagy-based futile cycles may operate in normal mammalian cells, providing a substrate for manipulation by pathogens.
Many studies have now shown a significant role for autophagy in cells infected by intracellular pathogens. In general, these studies have identified autophagy as a process that can contribute to the clearance of pathogen in the context of signals that activate host cell defense mechanisms.1 However, when such mechanisms are not active, host cell catabolic processes might conceivably play a pro-pathogen role by enhancing pathogen access to limiting nutrients.
A recent study from our laboratory provides evidence that the intracellular protozoan parasite, Toxoplasma gondii, is indeed able to exploit host cell autophagy for nutritive benefit.2 During its initial phase of rapid expansion, T. gondii invades many cell types and forms a specialized parasitophorous vacuole (PV), within which it replicates every 6–8 hours.3 This rapid growth is dependent on host cell nutrients, including amino acids, lipids and purines.4–6 The mechanism by which the parasite acquires host cell nutrients is not well understood, but recent evidence suggests that vesicular flow to the PV may be important. While the PV is fully isolated from fusion with host endosomes or lysosomes (provided anti-parasite functions have not been activated)7, infected cells display a remarkable concentration of host cell organelles around the PV, including endolysosomes, ER and mitochondria, the latter two tightly wrapping the PV. A recent study has shown that host lysosomes can enter the PV via PV invaginations associated with host microtubules.8 Therefore a host autophagic process that results in nutrient-rich lysosomes could conceivably serve as a pathway for nutrient flow to the parasite.
Our study supports this concept by demonstrating that parasite growth becomes dependent on host cell autophagy (assessed as dependence on host cell Atg5) only when the ambient amino acid concentration is reduced to physiological levels from the vast excess present in normal tissue culture medium. We also show that, under these conditions, host cell biomass is reduced by the parasite in an Atg5-dependent manner. Furthermore, parasite infection upregulates host autophagy and leads to a concentration near the PV of host Beclin 1 and vesicles bearing either phosphatidylinositol 3-phosphate or LC3.
The induction of autophagy under conditions of nutritional stress is thought to depend on negative regulation of the mTOR pathway, as a similar induction can be achieved in many cells with rapamycin, which inhibits the growth-promoting mTOR complex TORC1. We found, however, that T. gondii-induced autophagy does not involve mTOR inhibition, and even takes place in cells in which the mTOR pathway is hyperstimulated. Furthermore, in a separate, unpublished study, we found that if host cells are serum-deprived to reduce basal mTOR activity, Toxoplasma infection actually leads to upregulation of mTOR and protein synthesis in the host cell (Wang Y, Weiss LM, Orlofsky A, unpublished). The implication is that, in a physiological setting, T. gondii may simultaneously drive host protein metabolism in both anabolic and catabolic directions, generating (at least from the point of view of the host cell) a futile cycle.
The induction of a futile protein cycle might promote parasite growth by repositioning host amino acids into a parasite-accessible compartment (lysosomes). A second benefit might accrue to the parasite as a consequence of the considerable heat dissipation that accompanies protein synthesis,9 potentially resulting in increased ATP demand and elevated mitochondrial ATP production. The tight association of the PV with host mitochondria may facilitate parasite acquisition of the ATP produced. Thus, a futile ATP/energy cycle, secondary to the futile protein cycle, might amplify parasite capture of host resources (Fig. 1).
How is Toxoplasma able to upregulate host autophagy in the face of robust mTOR activity? Perhaps the parasite, which can export signaling proteins to the host cytosol,10 produces a factor that inhibits the (unknown) mTOR effector that regulates autophagy. However, an intriguing possibility is that the parasite exploits an endogenous host pathway that generates ‘mTOR-resistant’ autophagy, and, as a possible consequence, a natural protein futile cycle. The recent demonstration of elevated protein synthesis, protein degradation, energy expenditure and obesity resistance in mice deficient in branched-chain amino acid (BCAA) catabolism shows the enormous potential thermogenesis associated with such futile cycles.11 While this study did not provide evidence for a regulated mechanism of this kind in wild-type animals, such a mechanism is plausible, as it might provide adaptive benefits.
There is as yet little evidence for mTOR-resistant autophagy in uninfected cells. However, there is evidence for distinct modalities of anabolic mTOR function, since distinct mTOR-dependent phosphorylation sites on the mTOR effector 4E binding protein-1 are controlled through mTOR inputs from either amino acids or insulin.12 It is plausible, then, that an ‘autophagy-sparing’ mode of mTOR activation may exist. It is interesting in this respect that, while mTOR is activated in BCAA catabolism-deficient mice, S6 kinase-1, a canonical effector of mTOR, failed to be activated in certain tissues.11 A similar feature has been noted by us in T. gondii-infected cells (Wang Y, Weiss LM, Orlofsky A, unpublished).
Multiple upstream pathways connect external signals and cellular nutrient and energy status to mTOR activation. If an autophagy-sparing mode of mTOR activation exists, and functions to drive an endogenous thermogenic protein futile cycle, a logical potential upstream stimulus for this event would be amino acid elevation. Amino acids have recently been shown to activate mTOR via a mechanism dependent on calcium and Vps3413, a lipid kinase also implicated in autophagy. Whether autophagic signaling is resistant to amino acid/Vps34-induced mTOR is not yet known. Interestingly, T. gondii induces autophagy in host cells in a manner dependent on both calcium2 and Vps34 (Wang Y, Weiss LM, Orlofsky A, unpublished). It is possible to speculate, then, that amino acid elevation, coupled with an unknown endogenous signal that promotes autophagy in this setting, can generate an autophagy- and mTOR-dependent futile cycle via Vps34, and that Toxoplasma can mimic these upstream signals in order to exploit this mechanism and capture host nutrients. The use of the parasite to examine these and similar hypotheses may provide a fruitful new avenue for the elucidation of protein metabolism and autophagic signaling in mammalian cells.
This work was supported by funds from National Institutes of Health grant AI-55358 and by the Flow Cytometry Core of the Center for AIDS Research (AI-51519). The author wishes to thank Dr. Louis Weiss for helpful discussions.