ABSTRACT
The highly regulated process of adapting to cellular nutritional status depends on lysosomal mTORC1 (mechanistic Target of Rapamycin Complex 1), that integrates nutrients availability via the sensing of amino acids to promote growth and anabolism1. Nutrient restriction inhibits mTORC1 activity, which in turn induces autophagy, a crucial adaptive process that recycles internal nutrient stores to promote survival2. However, as successful amino acid recycling through autophagic degradation reactivates mTORC1 signaling over time3,4, it is unclear how autophagy can be maintained during prolonged starvation. Here, we show that one particular amino acid, cysteine, acts in a feedback loop to limit mTORC1 reactivation in vivo. We provide evidence that lysosomal export of cystine through the cystine transporter cystinosin fuels a metabolic pathway that suppresses mTORC1 signaling and maintains autophagy during starvation. This pathway involves reduction to cysteine, cysteine catabolism to acetyl-CoA and subsequent fueling of the TCA cycle. Accordingly, the starvation sensitivity phenotype of animals lacking cystinosin is rescued by dietary supplementation with cysteine and TCA cycle components as well as by reducing mTORC1 activity. We propose that cysteine mediates a communication between lysosomes and mitochondria to control mTORC1 signaling under prolonged starvation, highlighting how changes in nutrient availability divert the fate of an amino acid into a growth suppressive program to maintain the balance between nutrient supply and consumption.
Organisms constantly cope with variations in nutrient supply by adjusting metabolism, a process controlled by the major growth regulator mTORC1. Nutrient scarcity inhibits mTORC1 to limit growth and promote catabolic programs, including autophagy. Autophagy involves the sequestration of cytosolic material into autophagosomes that fuse with lysosomes for cargo degradation and recycling. Efficient lysosomal degradation generates new amino acids that in turn reactivate mTORC13-6. However, how mTORC1 inhibition and activation is balanced over the course of starvation remains unclear.
To study this process in vivo, we used the Drosophila larval fat body, an organ analogous to the liver and adipose tissue in mammals with important functions during starvation2,7. As in mammalian cells, prolonged starvation led to mTORC1 reactivation in fat bodies, which was dependent on autophagy induction (Fig. 1a, b and Extended Data Fig 1a). To identify signals controlling mTORC1 activity upon fasting, we focused on amino acids and performed an unbiased approach to analyze growth of fasted larvae under single amino acid supplementation (Supplementary Methods). Unexpectedly, this screen revealed cysteine as a unique amino acid associated with growth suppressive properties (Supplementary Text 1, Fig. 1c and Extended Data Fig. 1b-e). This process required mTORC1 inhibition as constitutive activation of mTORC1 in Gator1 mutants (nprl2-/-)8 partially restored growth under cysteine supplementation (Fig. 1d). Consistently, cysteine treatment partially inhibited mTORC1 activity in fat bodies (Extended Data Fig. 1f).
We reasoned that, as lysosomal efflux of nutrients promotes mTORC1 reactivation3, cysteine could limit this process. In physiological conditions, cellular cysteine can either be synthesized in the cytosol or transported as cystine from the extracellular space by the Xc- antiporter9 and by cystinosin from the lysosome10 (Fig. 1e). Thus, we analyzed whether cysteine recycling through cystinosin opposes mTORC1 reactivation upon fasting. Cystinosin is encoded by CTNS, a gene mutated in the lysosomal storage disorder cystinosis11. The Drosophila genome contains a single ortholog of CTNS (CG17119), hereafter referred to as dCTNS. Endogenous tagging of cystinosin in Drosophila confirmed its specific lysosomal localization in fat body cells, and dCTNS-/- larvae showed accumulation of cystine (Extended Data Fig. 2a-c), consistent with a role for cystinosin in lysosomal cystine transport. In fed conditions, control and dCTNS-/- larvae showed similar cysteine levels, likely reflecting dietary intake as the main source of cysteine (Fig.1f). However, upon fasting, cysteine levels dropped in dCTNS -/- larvae, and maintenance of cysteine levels required autophagy (Fig. 1f, g). Thus, we conclude that cystinosin recycles cysteine from autophagic degradation upon fasting.
Next, we analyzed the role of cystinosin on mTORC1 reactivation during fasting. dCTNS knockdown in larval fat body (lpp-gal4) did not affect mTORC1 during the early phase of starvation. However, it slightly increased mTORC1 reactivation upon prolonged starvation (Fig. 1h). Analysis of dCTNS-/- fat body clones showed that this effect was cell autonomous (Extended Data Fig. 2d) and sufficient to compromise maintenance of autophagy (Extended Data Fig. 2e). Conversely, dCTNS overexpression increased cysteine levels (Fig. 1i), which caused downregulation of mTORC1 and induction of autophagy (Fig. 1i and Extended Data Fig. 2e). Similar to cysteine treatment, dCTNS overexpression caused a developmental delay in both fed and starved conditions (Extended data Fig. 2f), in agreement with mTORC1 activity promoting growth during development. Finally, dCTNS loss of function led to starvation sensitivity in both larvae and adult animals (leading to a developmental delay in larvae, see Supplementary text 2), consistent with the importance of mTORC1 downregulation and autophagy induction for survival upon fasting2 (Extended Data Fig. 2g-i). This was dependent on cystine efflux as treatment with cysteamine (which exports cystine out of the lysosome independently of cystinosin12) rescued the starvation sensitivity of dCTNS-/- animals (Extended Data Fig. 2k). In addition, treatment with low concentration of cysteine (that did not affect development of control animals) and the mTORC1 inhibitor rapamycin, respectively, also restored resistance to starvation of dCTNS-/- animals (Fig. 1j-l), indicating that cystinosin controls mTORC1 through cytosolic cysteine. Altogether, we demonstrate that lysosomal-derived cysteine restricts mTORC1 reactivation to a threshold that maintains autophagy upon prolonged fasting.
Following reduction of cytosolic cystine to cysteine13, cysteine catabolism generates byproducts14,15. To discriminate between the effect of cysteine and these byproducts on mTORC1 inhibition, we sought to genetically suppress cysteine degradation. In mammals, cysteine degradation in liver and adipose tissues is catalyzed by cysteine dioxygenase type 1 (CDO1), a nucleo-cytoplasmic enzyme that generates cysteine sulfinate (CSA), a precursor for pyruvate and taurine synthesis14,16,17 (Fig. 2a). The Drosophila genome contains a single ortholog of CDO1 encoded by CG5493 (referred to as dCDO) that is strongly expressed in the fat body (Extended Data Fig. 3a). dCDO-/- larvae showed decreased levels of CSA and taurine, and dCDO knockdown in the fat body led to cysteine accumulation, particularly in larvae fed cysteine-enriched food, consistent with a role for dCDO in cysteine degradation (Fig. 2b; Extended Data Fig. 3b). Further, inhibition of mTORC1 by cysteine supplementation was suppressed in dCDO-/- animals (Fig. 2c), suggesting that cysteine requires degradation to CSA to control mTORC1. Accordingly, CSA supplementation had a similar effect than cysteine on growth and mTORC1 activity (Extended Data Fig. 3c, d). In addition, downregulation of dCDO in the larval fat body suppressed the developmental delay and the inhibition of mTORC1 induced by dCTNS overexpression, whereas CSA treatment restored the inhibition of mTORC1 (Extended Data Fig. 2d, e; Extended Data Fig. 3e). Consistently, dCDO knockdown in the fat body led to a stronger reactivation of mTORC1 under prolonged fasting (Fig. 2f; Extended Data Fig. 3f, g). Finally, aberrant mTORC1 reactivation following dCDO knockdown impaired fasting-induced autophagy (Fig. 2g). In sum, our data show that lysosomal cystine efflux and subsequent cysteine degradation into CSA opposes mTORC1 reactivation to maintain autophagy upon prolonged fasting.
Since amino acids and amino acid catabolism are potent activators of mTORC118,19, our finding that cysteine degradation limits mTORC1 reactivation is unexpected and prompted further analysis of the respective downstream events. As CSA is used for the synthesis of pyruvate and taurine, we first tested their effects on growth. Treatment with pyruvate but not with taurine, resembled the effects of cysteine and CSA on growth (Extended Data Fig. 4a), hinting at a role for the cysteine/pyruvate metabolic route in the control of mTORC1 reactivation. Consistently, supplementing food with pyruvate partially rescued starvation sensitivity of dCTNS-/- animals (Fig. 3a). In accordance with pyruvate function in fueling the TCA cycle, animals treated with excess cysteine showed increased levels of pyruvate and TCA cycle intermediates, particularly upon starvation (Fig. 3b, c). Accordingly, in physiological conditions (without cysteine supplementation), the level of many TCA cycle intermediates measured either from whole larvae or the fat body acutely increased during starvation, eventually reaching a steady state comparable to the levels of fed animals (Extended Data Fig. 4b-d). Essential amino acids were acutely depleted upon fasting, whereas the level of non-essential amino acids that requires TCA cycle activity was maintained. Importantly, maintenance of the level of TCA cycle intermediates required cysteine, as dCTNS-/- animals showed a pronounced depletion of pyruvate and TCA cycle intermediates upon prolonged fasting but not in fed conditions (Fig. 3d). By contrast, overexpression of dCTNS in fed animals led to opposite results (Fig. 3d). In sum, these data indicate that cysteine catabolism maintains TCA cycle activity upon fasting.
Pyruvate can contribute carbons to the TCA cycle through two metabolic routes. One serves an anaplerotic function to replenish oxaloacetate through the enzyme pyruvate carboxylase (PC), while the other contributes carbon in the form of acetyl-CoA through the pyruvate dehydrogenase complex (PDHc) (Fig. 3b). [U-13C3]cysteine labeling experiments in vivo revealed incorporation of cysteine carbons into CSA and acetyl-CoA in a dCDO-dependent manner in dissected fat bodies, indicating that cysteine/CSA provides substrate for PDHc (Extended Data Fig. 4e). However, due to very low labeling in the TCA cycle (see details in Supplementary text 3, Extended Data Fig. 5), we could not rule out a role for minimal cysteine anaplerosis through PC. Thus, we further used genetics to discriminate between PC and PDHc pathways on the regulation of mTORC1 by cysteine. Knockdown of PC (pcb/CG1516) suppressed mTORC1 reactivation upon fasting, whereas knockdown of pyruvate dehydrogenase phosphatase (pdp), which leads to PDHc inhibition20, caused a stronger reactivation of mTORC1 (Fig. 4a, b). This was supported by silencing pdha (the rate limiting enzyme of PDHc) in both entire fat bodies and in clones (Fig. 4c, Extended data Fig. 6a). Accordingly, inhibition of mTORC1 by dCTNS overexpression or cysteine supplementation was suppressed by knockdown of pdp (Fig. 4d, Extended data Fig. 6b). Finally, inhibition of PDHc impaired autophagy in larvae and triggered starvation intolerance in adult animals, similar to inhibition of dCTNS and dCDO (Fig. 4e, Extended data Fig. 6c). Thus, lysosomal cystine efflux and subsequent degradation to pyruvate opposes mTORC1 signaling reactivation through PDHc-dependent entry of acetyl-CoA into the TCA cycle.
To analyze whether the levels of particular TCA cycle intermediates downstream acetyl-CoA was relevant in the regulation of mTORC1, we screened for their effect on growth. We focused on fumarate, because it was the only TCA cycle intermediate associated with growth retardation (Extended data Fig. 6d, e). Fumarate treatment suppressed mTORC1 activity and rescued the effect of dCDO knockdown on mTORC1 in a dCTNS overexpression background (Extended data Fig. 6f, Fig. 4f). Finally, fumarate treatment partially rescued starvation sensitivity of dCTNS-/- animals (Fig. 4g), suggesting that the level of TCA cycle intermediates act downstream cystinosin during inhibition of mTORC1 activity. Alternatively, or in parallel, acetyl-CoA from cysteine could directly increase fumarate production through stabilization of cofactors required maximal activity of the fumarate-producing succinate dehydrogenase complex (complex II), as recently described21. In sum, we demonstrate that upon prolonged starvation, cysteine degradation to acetyl-CoA limits mTORC1 signaling reactivation through the TCA cycle to maintain autophagy (Fig. 4h).
Maintaining cellular homeostasis upon nutrient shortage is an important challenge for all animals. On the one hand, downregulation of mTORC1 is necessary to limit translation, reduce growth rates and engage autophagy. On the other hand, minimal mTORC1 activity is required to promote lysosomes biogenesis, thus maintaining autophagic degradation necessary for survival22. Here, we identify another critical layer of regulation that prevents reactivation of mTORC1 above a threshold that would compromise autophagy and survival upon fasting. Thus, reactivation of mTORC1 upon fasting is not passively controlled by the extent of nutrient remobilization, but instead is actively regulated by a mitochondrial pathway that maintains autophagy during starvation. Interestingly, this pathway is initiated by autophagy itself, as autophagic protein degradation controls cystine availability that further limits mTORC1 signaling through the TCA cycle.
Whether particular TCA cycle metabolites such as fumarate directly mediate mTORC1 inhibition remains unclear. Alternatively, acetyl-CoA from cysteine could promote incorporation of anaplerotic inputs into the TCA cycle at the expense of their use for biosynthesis of mTORC1-activating metabolites. Another open question concerns the crosstalk with mTORC1-activating pathways involving the TCA cycle. Our data show, for example, that knockdown of PC that replenishes oxaloacetate suppresses mTORC1 reactivation during fasting, whereas mitochondrial acetyl-CoA metabolism derived from cysteine via PDHc limits mTORC1 reactivation to a certain threshold. By contrast, cytosolic acetyl-CoA derived from leucine metabolism has recently been described to promote S6K phosphorylation through acetylation of mTORC1 regulators23. This illustrates how the source and fate of acetyl-CoA dictates its differential roles for mTORC1 activity. Further studies will therefore be required to understand exactly how different inputs in the TCA cycle interact to adjust mTORC1 activity, and how cysteine catabolism affects this system.
Author Contributions
P.J., Z.M., M.S. and N.P designed the experiments; P.J., Z.M., A.P., M.D., J.A., and I.N performed the experiments; P.J., Z.M., M.S. and N.P. participated in interpretation of data and P.J wrote the manuscript with inputs from Z.M., M.S. and N.P.
Methods
Fly stocks and maintenance
All flies were reared at 25°C and 60% humidity with a 12-h on/off light cycle on lab food. N. Perrimon’s lab food: 12.7 g/L deactivated yeast, 7.3 g/L soy flour, 53.5 g/L cornmeal, 0.4 % agar, 4.2 g/L malt, 5.6 % corn syrup, 0.3 % propionic acid, 1% tegosept/ethanol. M. Simon’s lab food: 18 g/L deactivated yeast, 10 g/L soy flour, 80 g/L cornmeal, 1% agar, 40 g/L malt, 5% corn syrup, 0.3 % propionic acid, 0.2 % 4-hydroxybenzoic acid methyl ester (nipagin)/ethanol. Density was standardized at least one generation before the experiments. For experiments, larvae were reared on freshly made food. For details on food recipes, drugs and fly stocks used, see Supplementary Methods.
Developmental timing
Three-day-old crosses were used for 3-4 hour periods of egg collection on lab food. Newly hatched L1 larvae were collected 24 hours later for synchronized growth using the indicated diets at a density of 30 animals/vial. The time to develop was monitored by counting the number of animals that underwent pupariation, every two hours in fed conditions or once/twice a day in starved conditions. The time at which half the animals had undergone pupariation is reported, +/-SEM.
Life span experiments
To generate age-synchronized adult flies, larvae were raised on ad libitum (means food available at all times with the quantity and frequency of consumption being the free choice of the animal) lab food at low density, transferred to fresh food upon emerging as adults and mated 48h. Animal were anaesthetized with low levels of CO2 and males sorted at a density of ten per vial. Each condition contained 8-10 vials. Each experiment was repeated at least 3 times and the averages values of each experiment were used for statistical analysis. Flies were transferred to fresh vials three times per week at which point deaths were scored. After ten days, deaths were scored every day.
Metabolite profiling
For whole body metabolic profiling, 25-38 mid-second instar or 8-15 mid-third instar larvae per sample were collected, snap-frozen in liquid nitrogen and stored at −80°C in extraction buffer (4-6 biological replicates/experiment). For fat body metabolic profiling, fat bodies from 35-40 larvae 96h AEL old were dissected in 25 ul PBS, diluted in 300 ul cold extraction buffer and snap frozen. Tissues were homogenized in extraction buffer using 1 mm zirconium beads (Next Advance, ZROB10) in a Bullet Blender tissue homogenizer (Model BBX24, Next Advance). Metabolites were extracted using 80 % (v/v) aqueous methanol (x2 sequential extractions with 300-600 ul) and metabolites pelleted by vacuum centrifugation. Pellets were resuspended in 20 ul HPLC-grade water and metabolomics data were acquired using targeted liquid chromatography tandem mass spectrometry (LC-MS/MS). A 5500 QTRAP hybrid triple quadrupole mass spectrometer (AB/SCIEX) coupled to a Prominence UFLC high-performance LC (HPLC) system (Shimadzu) was used for steady-state analyses of the samples. Selected reaction monitoring (SRM) of 287 polar metabolites using positive/negative switching with hydrophilic interaction LC (HILIC) was performed. Peak areas from the total ion current for each metabolite SRM Q1/Q3 transition were integrated using MultiQuant version 2.1 software (AB/SCIEX). The resulting raw data from the MultiQuant software were normalized by sample weight or protein content measured from equivalent samples and analyzed using Prism informatic software. Alternatively, 75 hours AEL old dCTNS+/- (control) or dCTNS-/- larvae were transferred either to fresh standard food (fed) or on PBS-soaked (starvation, 8h) Whatman paper. Larvae were rinsed with water, 70% ethanol and PBS to remove food and bacteria. 10-15 mid-third instar larvae per sample were collected, snap-frozen in liquid nitrogen and stored at −80°C until extraction in 50% methanol, 30% ACN, and 20% water. The volume of extraction solution added was adjusted to larvae mass (40mg/ml), samples were vortexed for 5 min at 4°C, and then centrifuged at 16,000 g for 15 minutes at 4°C. Supernatants were collected and analyzed by LC-MS using a QExactive Plus Orbitrap mass spectrometer equipped with an Ion Max source and a HESI II probe and coupled to a Dionex UltiMate 3000 UPLC system (Thermo, USA). An SeQuant ZIC-pHilic column (Millipore) was used for liquid chromatography separation24. The aqueous mobile-phase solvent was 20 mM ammonium carbonate plus 0.1% ammonium hydroxide solution and the organic mobile phase was acetonitrile. The metabolites were separated over a linear gradient from 80% organic to 80% aqueous for 15 min and detected across a mass range of 75–1,000 m/z at a resolution of 35,000 (at 200 m/z) with electrospray ionization and polarity switching mode. Lock masses were used to insure mass accuracy below 5 ppm. The peak areas of different metabolites were determined using TraceFinder software (Thermo) using the exact mass of the singly charged ion and known retention time on the HPLC column. In total, the metabolic profiling experiment was performed three times with 3-7 biological replicates per genotype.
TCA cycle isotopomer method from [U-13C3]cysteine and [U-13C6]glucose
Fed and starved early second instar animals were supplemented with the indicated concentrations of [U-13C3]cysteine, [U-13C6]glucose or vehicle in the food during the indicated time. Sample were collected (5-6 biological replicates for labelled conditions, 4 biological replicates for unlabeled condition), and intracellular metabolites were extracted using 80% (v/v) aqueous methanol. Q1/Q3 SRM transitions for incorporation of 13C labeled metabolites were established for isotopomers for 65 polar metabolites and data were acquired by LC-MS/MS. Peak areas were generated using MultiQuant 2.1 software. The mean peak areas from unlabeled conditions were used for background determination and were subtracted from each corresponding 13C labelled dataset.
Immunostaining
Tissues from 68-85 hours AEL larvae were dissected in phosphate-buffered saline (PBS) 2% formaldehyde at room temperature, fixed 20-30 min in 4% formaldehyde, washed twice 10 min in PBS 0.3% Triton (PBST), blocked 30 min (PBST, 5% BSA, 2% FBS, 0.02% NaN3), incubated with primary antibodies in the blocking buffer overnight and washed 4 times for 15 min. Secondary antibodies diluted 1:200 or 1:500 in PBST were added for 1 hour and tissues washed 4 times before mounting in Vectashield/DAPI. Rabbit anti P-4EBP1 was from Cell Signaling Technologies (CST 236B4, #2855) and diluted 1:500, rabbit anti-tRFP was from Evrogen (#AB233) and used against mKate2 to stain cystinosin-mKate2. Samples were imaged using Zeiss LSM 780 and Leica TCS SP8 SMD confocal systems with a 40x water or 40x oil immersion objective and images were processed with Fiji software.
Western Blots
Tissues from 15-30 animals were dissected in CST lysis buffer (Cat#9803) containing 2x protease inhibitor (Roche, 04693159001) and 3x phosphatase inhibitor (Roche, 04906845001), and homogenized using 1 mm zirconium beads (Next Advance, ZROB10) in a Bullet Blender tissue homogenizer (Model BBX24, Next Advance). Protein content was measured to normalize samples, 2x Laemmli Sample Buffer (Biorad) was added and samples boiled 6 min @ 95°C. Lysates were resolved by electrophoresis (Mini-PROTEAN TGX Precast Gels, BioRad; or PAGEr EX Gels, Lonza; or home-made gels), proteins transferred onto PVDF membranes (Immobilon P, Millipore), blocked in Tris-buffered saline with or without 0.1% Tween-20 buffer containing 3 or 6% BSA, and probed with P-S6K antibody (1:1000, CST 9209). After P-S6K was revealed, membranes were stripped for 5-30 min (Restore PLUS Buffer, Thermo Scientific #46430), washed, blocked in PBS Tween-20 buffer containing 5% dry milk, and probed with S6K antibody (1:10000, a gift from Aurelio Teleman25). For normalization blots were probed with GADPH antibody (1:5000, GeneTex GTX100118). Data show representative results from at least 2 or 3 biological replicates. Horseradish peroxidase (HRP) conjugated secondary IgG antibodies (1:10000) were used together with the SuperSignal West Dura Extended Duration Substrate (Thermo Scientific #34076) to detect the protein bands.
Generation of clones
For autophagy experiments, clones were generated by crossing yw,hs-Flp; mCherry–Atg8a; Act>CD2>GAL4, UAS–nlsGFP/TM6B with the indicted UAS lines. Progeny of the relevant genotype was reared at 25°C and spontaneous clones were generated in the fat body due to the leakiness of the hs-flp. For dCTNS-/- clones, autophagy was analyzed by crossing w;; neoFRT82B, dCTNS-/- to hsFlp; R4-Gal4, UAS-mCherry-Atg8a; FRT82B UAS-GFP/TM6b. For P-4E-BP1 experiments, clones were either generated by crossing hsFlp; act>CD2>Gal4, UAS nls GFP with the indicted UAS lines or, for dCTNS-/- clones, by crossing w;;neoFRT82B, dCTNS-/- to yw, hsFlp, Tub-Gal4>UAS-nlsGFP/FM6;;neoFRT82B, TubGal80/TM6,Tb,Hu. F1 embryos collected overnight were heat shocked 3 times for 30 min.
Statistics
Experiments are presented with the mean +/- SEM. P values and significance: ns, P≥0.05; *, P≤0.05; **, P≤0.01; ***, P≤0.005; ****, P≤0.0001. Life span experiments: N=1 means average of 8-10 vials per genotype and condition in 1 experiment. Fig. 1j (N≥5), Fig. 3a (N=4), Fig. 4j (N=3-4): significance was determined by a two-tailed t-test (Mann-Whitney). Pupariation assay: Fig. 1k, Fig. 1l, Fig. 2d, Fig. 3a, Fig. 4j: significance was determined by one-way ANOVA followed by a Bonferroni multiple comparisons test. Cysteine measurements (Profoldin kit): Fig. 1f, g: significance was determined by a one-way ANOVA followed by a Bonferroni multiple comparisons test. Fig. 1i: significance was determined by a two-tailed t-test (Mann-Whitney). Western blots: Fig 1h, Fig. 1i, Fig. 2e, Fig. 4i: significance was determined by one-way ANOVA followed by a Bonferroni multiple comparisons test. Fig 1a: spline curve represent the trend over multiple experiments. Fig 1b, 4a, 4b: significance was determined by two-way ANOVA followed by a Sidak’s multiple comparisons test. Fig 2c, 2f, 4d: significance was determined by one-way ANOVA followed by a Tukey’s multiple comparisons test. Metabolomics: Fig 2b, 2c: significance was determined by one-way ANOVA followed by a Tukey’s multiple comparisons test. Fig 3d: significance was determined by a two-tailed t-test (Mann-Whitney).
SUPPLEMENTARY TEXT
Supplementary Text 1: An amino acid supplementation screen reveals cysteine as a growth suppressor
To evaluate the effects of individual amino acids on growth, we developed an amino acid supplementation screen of developing Drosophila larvae and measured their development rate (Supplementary Methods). The screen identified cysteine as a strong growth suppressor (Fig. 1c), an effect that could be due to the cytotoxicity of cysteine previously reported in cell culture, yeast, and chicks1-3. However, we found that the effect of cysteine supplementation was diet dependent, with cysteine strongly suppressing growth upon starvation while having weaker effect in fed animals (Extended Data Fig. 1b-d), mitigating cytotoxicity as a unique explanation for this result. In addition, although the effect of cysteine on growth was dose dependent (Extended Data Fig. 1b), variation in cysteine intake between fed and starved conditions was not sufficient to explain the diet-dependent cytotoxicity (Extended Data Fig. 1e). Therefore, we conclude that the growth-suppressive effect of cysteine was multifactorial and decided to analyze the endogenous role of intracellular cysteine.
Supplementary Text 2: Developmental delay versus starvation sensitivity
Gain and loss of function of dCTNS have opposite effect on mTORC1 but showed similar phenotype in term of larval development: an increase in the time to pupariation. To avoid confusion between opposite processes that lead to similar phenotypes in appearance, we adapted our nomenclature accordingly. Because the loss of dCTNS caused reduced cellular cysteine, upregulation of mTOR, inhibition of autophagy, and that cysteine and rapamycin treatments rescued/accelerated the time to pupariation, we termed CTNS-/- developmental phenotypes “starvation sensitivity”. Accordingly, CTNS-/- did not affect development of fed larvae. By contrast, dCTNS overexpression increased cellular cysteine, downregulated mTORC1 and induced autophagy. In agreement with mTORC1 loss of function delaying larval growth and development, we termed the developmental phenotype of dCTNS overexpression “developmental delay”. Consistently, dCTNS overexpression retarded development in both fed and starved conditions.
Supplementary Text 3: 13C3-cysteine labeling in vivo
We used 13C3-cysteine supplementation in the food to follow cysteine carbons into the TCA cycle. We first fed animals 5 mM 13C3-cysteine for 24 hours, and then analyzed labeling in the TCA cycle following different chasing times after placing animals on PBS. This allowed us to assess the dynamics and stability of labeling in order to further adjust our protocol for experiments on dissected fat bodies. Results showed variable but overall low incorporation of cysteine carbons in different TCA cycle intermediates, and the extent of labeling progressively reduced with chasing time for most metabolites (Extended Data Fig. 5a). This indicates that cysteine carbons enter the TCA cycle and that labeling is stable enough to allow for dissection time. However, labeling with 5 mM 13C3-cysteine for 24 hours was not suitable to test the requirement of dCDO for cysteine entry in the TCA cycle in the fat body. Indeed, dissected fat bodies from dCDO mutant animals showed a >15-fold accumulation of labeled cysteine compared to control, thus mitigating conclusions regarding labeling in downstream metabolites in such different tracer concentrations (Extended Data Fig. 5b). To find a labeling protocol suitable to analyze the role of dCDO during cysteine entry in the TCA cycle, we analyzed 13C3-cysteine concentration in control and dCDO-/- animals following shorter labeling time/lower tracer concentration (Extended Data Fig. 5c). Based on these experiments, we found that labeling with 5 mM 13C3-cysteine for 3 hours was the best condition to avoid dramatic differences in tracer concentration in the fat body of control and dCDO-/- animals (Extended Data Fig. 4f, Extended Data Fig. 5c). Results showed that CDO mediates incorporation of cysteine carbons into CSA and Acetyl-CoA (Extended Data Fig. 4f). However, we could not conclude about pathway entry in the TCA cycle (PC vs PDHc) due to very low or absence of labeling in the TCA cycle using this protocol (data not shown). We further tried to force labeling in the TCA cycle intermediates in whole animals using 25 mM of [U-13C6]glucose for 4 hours. However, this also raised very limited labeling in TCA cycle (Extended Data Fig. 5d). Given these limitations, we decided to use genetics test the role of PC and PDHc in mediating the effect of cysteine on mTORC1.
SUPPLEMENTARY METHODS
Amino acid screen
We fed larvae a diet with reduced yeast extract/proteins (50% of normal diet), systematically added individual amino acids to the food (Table S1), and monitored the time to pupariation as a proxy for growth rate. The following mix was diluted 1:1 with amino acids solutions in water: 10g/L Agar; 120g/L Sucrose; 17g/L Deactivated Yeast extract; 83g/L Cornmeal; 6ml/L Propionic Acid; 20ml/L Tegosept. The amount of amino acids added to the food was determined based on those used in tissue culture growth supplements (Extended Data Table 1)4-6.
Fly stocks
lpp-gal4 was a gift from Pierre Léopold; UAS-tsc1, UAS-tsc2 a gift from Christen Mirth7; yw,hs-Flp; mCherry– Atg8a; Act>CD2>GAL4, UAS–nlsGFP/TM6B a gift from Eric Baehrecke, hsFlp; act>CD2>Gal4, UAS nls GFP a stock from Norbert Perrimon lab8, yw, hsFlp, Tub-Gal4>UAS-nlsGFP/FM6;;neoFRT82B, TubGal80/TM6,Tb,Hu a gift from Allison Bardin, nprl21 a gift from M. Lilly and hsFlp; R4-Gal4, UAS-mCherry-Atg8a; FRT82B UAS-GFP/TM6b a gift from G. Juhász. The following stocks were obtained from BDSC: UAS-wRNAi (HMS00045), UAS-wRNAi (HMS00017), attp40 (#36304), attp2 (#36303), UAS-mCherry-nls (#38425), UAS-PDHaRNAi-#2 (HMC04032), UAS-pdpRNAi (HMS01888), UAS-CDORNAi-#1 (HMJ22159), UAS-Atg1RNAi (HMS02750), UAS-Atg18aRNAi (JF02898), UAS-TSC2RNAi (HM04083), w1118, UAS-dCTNSRNAi (HMS00213). w (v60100), UAS-CDORNAi-#2 (v50637) and UAS-PDHaRNAi-#1 (v40410) were obtained from VDRC. UAS-PDHa-RB, UAS-PDHa-RC and UAS-CDO were constructed using the FlyBi project Gateway ORFs collection (http://flybi.hms.harvard.edu/). The entry plasmids were used in a LR clonase reaction (Invitrogen, 11791-020), with the destination vector pWALIUM10-roe9 or equivalent (Frederik Wirtz-Peitz, unpublished data). The plasmid was then microinjected into embryos which harbor attP2 or attp40 landing sites, as per standard procedures to create transgenic flies. When comparing the effects of RNAi knockdown, UAS-wRNAi (HMS00045), UAS-wRNAi (HMS00017), attp2 (#36303) and attp40 (#36304) were used as controls for the TRIP collection (http://www.flyrnai.org/TRiP-HOME.html), and w (v60100) for the VDRC collection.
dCDO knockout flies were generated with the CRISPR/Cas9 technology according to 10. Briefly, three sgRNA targeting the 5’ end of the first exon and the last exon of dCDO (F/R sgRNA oligos sequences: GTCGCTCGGTGTCGATCTTGGACA/ AAACTGTCCAAGATCGACACCGAG; GTCGCAGCGGCTGATAGTAGCTAG/ AAACCTAGCTACTATCAGCCGCTG; GTCGAGGGTGGACAGTATAGGTGA/ AAACTCACCTATACTGTCCACCCT) were cloned in the pCFD3 expression vector. Plasmids were injected into nos-Cas9 embryos and emerging adults crossed to Sco/Cyo. Progenies were screened by PCR for deletion of the whole dCDO locus and individual stocks were established, along with control lines that followed the same crosses scheme. Mutant stocks were sequence verified and further validated by qPCR. dCTNS knockout flies were generated with CRISPR/Cas9 technology according to11. Two sgRNA using oligos (one after the ATG start codon in exon 3: GGTGATGTCATGGGAATCGA, and the other before the translation of the first transmembrane domain in exon 4: GGGCAGTACTCGAAATCAGT) were produced by PCR and in vitro transcribed into RNA via MEGAscript(tm) T7 Transcription Kit (ThermoFisher). RNA was injected into ActCas9 flies (from Fillip Port/Simon Bullock). F0 flies were crossed to w;; TM3,Sb/TM6,Tb balancer flies and F1 progenies were screened via PCR by uisng oligos flanking the targeted genome region. Any indel difference > 3 bp was visualized in 4% agarose gel in heterozygot F1 progeny.
To generate dCTNS-mKate2 fusion allele, mKate2 open reading frame was inserted at the C-terminus of dCTNS through CRISPR/Cas9 endogenous tagging strategy using vectors kindly provided from Y. Bellaiche (Curie Institut, Paris). In brief, two 1 kb long homology arms (HR1, HR2) of the dCTNS gene flanking the sgRNA-guided Cas9 cutting site were cloned into a vector flanking the ATG/STOP-less mKate2 allele (HR1-linker-mKate2-loxP-mini-white-loxP-linker-HR2). In addition, two vectors for the expression of sgRNA (sgRNA-1: CCACCGTGACCGATGTTCAAAAT, sgRNA-2: CCGAGCGAAGTGACGACTGAGAA) targeting the C-terminal coding region of dCTNS were generated. All three vectors were injected into vas-Cas9 flies (BDSC#55821) embryos by Bestgene. Progenies were screened for the red eyes (selection marker mini-white) and crossed to Cre-expressing flies to remove the mini-white by loxP/Cre excision. For overexpression of dCTNS, dCTNS cDNA was cloned into Gateway destination vector pUASg-HA.attB (GeneBank: KC896837) according to12. The plasmid was injected by Bestgene into FlyC31 embryos (BDSC#24482) for ϕ31-mediated recombination at a attP insertion site on the second chromosome.
Fly food and starvation protocols
All amino acids and compounds used were from Sigma. In N. Perrimon’s lab compounds in solution were added to the following food mix: 60 g/L sucrose, the indicated amounts of deactivated yeast as a source of total protein (2 g/L for fast, 4 g/L for mildly fast, 20g/L for fed fly food); 80 g/L cornmeal, 0.35% Bacto Agar, 0.3% propionic acid, 1% tegosept (100g/L in ethanol). In the Simon’s lab fasting food was adapted to 6 g/L of deactivated yeast to match control fast developmental rates observed in the Perrimon. For fed food, the standard lab food was used (see protocol above). In Extended Data Fig. 1 specifically, 100% amino-acid is 17g/L deactivated yeast.
Food intake
Larvae were synchronized in L1 and reared on the indicated food types until mid-2nd instar. Larvae were then transferred on the same food type supplemented with 0.5% weight/volume erioglaucine disodium salt (Sigma) for 2 hours. Samples were homogenized in 200 µl of PBS and absorbance of the dye in the supernatant was measured at 625 nm. Results were normalized to protein content.
Growth curves/pupal weight
Synchronized, newly-hatched L1 larvae were immediately weighed or placed on the indicated food at a density of 30-50 animals/vial. Pools of 20-80 animals were weighed every 24 hours using an analytical scale (Mettler Toledo) and the weight/animal was reported +/- SEM. For pupal weight, two-day-old pupae from vials at a density of 30 animals were weighed in batches of 5-10 pupae. The weights of different batches of larvae from the same vials were averaged and counted as N=1.
Starvation resistance assay
10-day-old adult males of the indicated genotypes reared on lab food at controlled density were transferred to 1% Bacto agar in water (20-25 animals/vials) and dead animals were scored every 4 or 6 hours starting at 24 hours of starvation.
Cysteine measurement
25-40 mid-second instar animals were homogenized in cold PBS 0.1% Triton and centrifuged at 4°C. Cysteine measurement was performed in triplicate from the supernatant using the MicroMolar Cysteine Assay Kit (ProFoldin, CYS200) according to the manufacturer’s instructions. Data were normalized to protein content.
Cystine measurements
Larvae were washed three time (water, shortly in 70% ethanol, and in finally in PBS), dried on tissue paper and 10 larvae/sample were shock frozen in liquid nitrogen and stored at −80°C until lysis. Larvae were lysed in 80 μl of 5.2 mM N-ethylmaleimide, centrifuged 10 min at 4°C and 75 μl supernatant was deproteinized by addition of 25 μl of 12% sulfosalicylic acid. Protein-free supernatants were kept frozen at −80°C until analysis and centrifuged at 1,200 x g before use. Cystine quantification was performed using AccQ-Tag Ultra kit (Waters®) on a UPLC-xevoTQD system (Waters) according to the manufacturer’s recommendations. Ten µL of samples were mixed with 10µL of a 30 µM internal standard solution (stable isotope of cystine), 70µL of borate buffer and 20µL of derivative solution and incubated at 55°C for at least 10 minutes. Derivatized samples were diluted with 150µL of ultrapurified water and 5 µL of the final mix were injected in the triple quadrupole mass spectrometer in positive mode. Transitions used for derivatized cystine quantification and the internal standard were respectively 291.2>171.1 and 294.2>171.1. Cystine values were normalized by protein content using the Lowry’s method on protein pellets.
Acknowledgements
We thank Eric Baehrecke, Christen Mirth, Pierre Léopold, Aurelio Teleman, Frederik Wirtz-Peitz, Yohanns Bellaïche, Isabelle Gaugue, Sylvia Sanquer, Anne-Claire Boschat, Mary A. Lilly, the TRIP (http://www.flyrnai.org/TRiP-HOME.html), BDSC and VDRC stock centers for providing stocks and reagents. We thank Corinne Antignac, Lewis C. Cantley, Bruno Gasnier and David M. Sabatini for comments on the manuscript. We thank the Imagine Microscopy platform for assistance with microscopy. This work was funded by the Cystinosis Research Foundation (to P.J., Z.M., M.S. and N.P.), the LAM Foundation Fellowship Award LAM00105E01-15 (to A.P.), National Institutes of Health 5P01CA120964-04 (to J.M.A. and N.P.) and R01AR057352 (to N.P), the ATIP-Avenir program, the Fondation Bettencourt-Schueller (Liliane Bettencourt Chair of Developmental Biology) as well as State funding by the Agence Nationale de la Recherche (ANR) under the “Investissements d’avenir” program (ANR-10-IAHU-01) and the NEPHROFLY (ANR-14-ACHN-0013) grant (to MS.). N.P. is an investigator of the Howard Hughes Medical Institute.