Abstract
The general consensus is that the vacuolar-type H+-translocating ATPase (V-ATPase) is critical for macroautophagy/autophagy. However, there is a fundamental conundrum because follicular lymphoma-associated mutations in the V-ATPase result in lysosomal/vacuolar deacidification but elevated autophagy activity under nutrient-replete conditions and the underlying mechanisms remain unclear. Here, working in yeast, we show that V-ATPase dysfunction activates a selective autophagy flux termed “V-ATPase-dependent autophagy “. By combining transcriptomic and proteomic profiling, along with genome-wide suppressor screening approaches, we found that V-ATPase-dependent autophagy is regulated through a unique mechanism distinct from classical nitrogen starvation-induced autophagy. Tryptophan metabolism negatively regulates V-ATPase-dependent autophagy via two parallel effectors. On the one hand, it activates ribosome biogenesis, thus repressing the translation of the transcription factor Gcn4/ATF4. On the other hand, tryptophan fuels NAD+ de novo biosynthesis to inhibit autophagy. These results provide an explanation for the mutational activation of autophagy seen in follicular lymphoma patients.
Similar content being viewed by others
Introduction
The lysosome/vacuole is the catabolic center of the cell. Besides its primary role in protein degradation, it participates in other important cellular processes such as metabolic signaling and gene regulation in response to environmental cues1,2. A functional lysosome/vacuole requires an acidic lumenal environment3, which is achieved by the vacuolar-type H+-translocating ATPase (V-ATPase). The V-ATPase is composed of two multisubunit domains: the peripheral V1 subcomplex and the integral V0 subcomplex. The V1 domain hydrolyzes ATP to promote H+ pumping through the V0 domain4,5.
Pathologically, altered V-ATPase activity has been implicated in cellular aging, neurodegenerative diseases, and cancer6,7,8,9. For example, in follicular lymphoma (FL) patients, recurrent loss-of-function mutations are identified in the ATP6V1B2, VMA21, and ATP6AP1 genes encoding a V-ATPase V1 domain component and two assembly factors10,11,12,13,14. Yet it is unclear how tumor cells could benefit from a dysfunctional V-ATPase.
Macroautophagy, hereafter autophagy, is a highly regulated cellular process in which cytoplasmic contents are degraded within the lysosome/vacuole15. The essentiality of V-ATPase genes in mammalian cells hinders the investigations of the V-ATPase’s function with regard to autophagy regulation. The current paradigm is that compromised V-ATPase function impairs starvation-induced autophagy16, because most lysosomal/vacuolar hydrolytic enzymes have optimal activity at acidic pH17. However, our recent characterizations of FL-associated mutations in ATP6V1B2/VMA2, VMA21, and ATP6AP1/BIG1, in human cell lines and a yeast model, reveal that basal autophagy flux is enhanced in nutrient-replete medium even though the lysosome/vacuole is indeed deacidified14,18,19. In contrast to the paradigm for autophagy regulation, this induction occurs even though MTOR/TOR, the major negative regulator of autophagy20, is still active. Abnormal activation of basal autophagy can promote tumorigenesis21,22. Therefore, one hypothesis is that mutations such as those affecting the V-ATPase in the FL tumor cells may benefit tumor survival by allowing autophagy activation (i.e., supplying nutrients) while MTOR is active (i.e., allowing cell growth) at the expense of less efficient lysosomal/vacuolar hydrolytic activity. Nonetheless, the underlying mechanisms of this unexpected mutational activation of autophagic flux by recurrent hotspot mutations in genes associated with V-ATPase activity remain unclear.
In this article, we discovered that the V-ATPase dysfunction can induce a selective form of autophagy by mechanisms that are distinct from classical nitrogen starvation-induced autophagy. We found that tryptophan metabolism plays a key role in this novel autophagy pathway. We characterized two parallel downstream effectors of tryptophan metabolism: ribosome biogenesis and NAD+ biosynthesis pathways. Our findings uncovered a novel function of V-ATPase in autophagy regulation and may explain the autophagy-activating effect of dysfunctional V-ATPase observed in FL patients.
Results
Vacuolar deacidification activates autophagy flux in nutrient-rich medium
Dysfunctional V-ATPase and lysosomal/vacuolar deacidification are thought to impair autophagy activity by hindering the degradative efficiency of lysosomal/vacuolar hydrolytic enzymes16,23. In addition, basal autophagy flux is maintained at a very low level in nutrient-rich medium (Fig. 1a). Therefore, for loss-of-function FL-associated V-ATPase mutants that cause lysosomal/vacuolar deacidification, it is counterintuitive to observe an autophagy activation phenotype14,18,19 (Fig. 1a); although there is only a partial induction of autophagy (less than seen under starvation conditions), it appears to be of significant value to cancer cells18.
a Summary of previously published data about the effects of FL-associated ATP6V1B2/VMA2 mutants on autophagy flux in human cells and in yeast. In both systems, an FL-associated R to Q mutation on a conserved residue in ATP6V1B2/VMA2 causes deacidification of lysosomal/vacuolar compartments and an increase of basal autophagy flux in nutrient-rich medium. This autophagy flux triggered by dysfunctional V-ATPase under nutrient-rich conditions is less robust than the one induced by starvation. b Schematic of inducible degradation of the target protein (Vma2) via the auxin-inducible degron system. Addition of the auxin molecule 3-IAA recruited the adapter protein OsTIR1 and the Cul1 E3 ubiquitin ligase complex to Vma2-AID*−9MYC. After poly-ubiquitination, the majority of Vma2-AID*-9MYC was degraded by the 26S proteasome. c–e SEY6210 VMA2-AID*-9MYC cells were grown to mid-log phase in YPD (0 h), centrifuged and resuspended in YPD + 0.1% DMSO, YPD + 300 μM 3-IAA, YPD + 500 nM conc. A, or SD-N medium for the indicated times. c Cell lysates were prepared, subjected to 10% SDS-PAGE, and analyzed by western blot. The ratio of free GFP to total GFP (free GFP plus GFP-Atg8) was quantified to indicate autophagy flux. Pgk1 was used as a loading control. The MYC blot detected Vma2-AID*-9MYC and OsTIR-9MYC. Three biological replicates were repeated with similar results. d Pho8-SEP cells were harvested and analyzed by flow cytometry to monitor vacuolar deacidification. A.U., arbitrary units. e WT, atg1∆ and pep4∆ cell lysates were processed and analyzed using the GFP-Atg8 processing western blot assay. *, non-specific band. Three biological replicates were repeated with similar results. f Hypothetical model of V-ATPase-dependent autophagy. V-ATPase-dependent autophagy describes the autophagy-activating effect of V-ATPase impairment in nutrient-rich medium. Under nutrient-rich conditions, the basal autophagy flux is very low, and the activating effect of V-ATPase dysfunction outweighs the defects in vacuolar hydrolytic activity; thus, the net outcome is enhancement of autophagy flux. During nitrogen starvation, autophagy activation induced by the starvation signal is much stronger than the one caused by V-ATPase dysfunction; the net result of V-ATPase dysfunction (i.e., the decrease in vacuolar hydrolytic activity) on starvation-induced autophagy is inhibitory.
To resolve this puzzle, we selected the budding yeast S. cerevisiae as the system for investigation, because the FL-associated mutations in ATP6V1B2/VMA2 and the autophagy activation effect are conserved between yeast and human18. Given the deleterious effect of the FL-associated mutations in ATP6V1B2/VMA2 on cell growth, it is challenging to directly perform substantial mechanistic characterization in the FL-associated mutant strain. To circumvent this difficulty and to reduce the possibility of suppressor mutations that compensate for chronic loss of V-ATPase structural or assembly genes, we fused an AID* degron domain (a 71-114 truncated fragment of the IAA17 protein that proved to work efficiently in budding yeast)24 to the endogenous Vma2 protein. We then asked if transient degradation of the Vma2 protein, induced by the addition of the auxin molecule indole-3-acetic acid (3-IAA), could mimic the effect of FL-associated mutations and activate autophagy flux in nutrient-rich yeast extract, peptone, dextrose (YPD) medium (Fig. 1b). As expected, adding 3-IAA rapidly resulted in the degradation of the majority ( > 95%) of the Vma2-AID*-9MYC protein (referred to as Vma2 hereafter) within 2 h (Fig. 1c). To quantitively measure autophagy flux, we transformed yeast cells with a GFP-Atg8 reporter construct. GFP-Atg8 is covalently attached to the phagophore and remains associated with the autophagosome inner membrane; after fusion with the vacuole, Atg8 is degraded, whereas the green fluorescent protein (GFP) is relatively stable and accumulates in the free form25. In the control (DMSO-treated) cells, free GFP was minimal over the 8-h time course (Fig. 1c). In contrast, after shifting to nitrogen starvation (SD-N) medium in the absence of 3-IAA, Vma2 was still abundant, and the majority of the chimeras were processed to generate free GFP (Fig. 1c). Following 3-IAA addition to cells in nutrient-rich medium, the autophagy flux became detectable at approximately 4 h (Fig. 1c) and was saturated and showed no further increase after 8 h (Fig. 1C and Supplementary Fig. 1a). To ensure that 3-IAA treatment was not simply affecting the amount of GFP-Atg8 present in the cells through regulation of the ATG8 promoter, we also examined autophagy flux in cells expressing the chimera under the control of the constitutive CUP1 promoter (CUP1p). Indeed, abundant GFP-Atg8 processing was also observed following the addition of 3-IAA in the CUP1p-GFP-ATG8 cells (Supplementary Fig. 1b). Furthermore, expression of additional copies of non-AID*-tagged Vma2 prevented the increase in autophagy flux triggered by Vma2-AID*-9MYC degradation (Supplementary Fig. 1c), supporting the conclusion that the autophagy flux observed here resulted from Vma2 temporal degradation.
Next, to confirm the results obtained with the GFP-Atg8 processing assay, we used additional methods to monitor autophagy under the same conditions. First, we transformed the VMA2-AID* cells with an RFP-Atg8 plasmid and examined if there was vacuolar accumulation of the RFP fluorescent signal. Consistent with the western blot-based method, we found that this microscope-based autophagy assay also detected strong autophagy activity (Supplementary Fig. 1d). Second, during autophagy, the Atg8 protein is covalently conjugated to phosphatidylethanolamine (PE; referred to as lipidation) and localizes to the phagophore assembly site/PAS, the location where phagophores are initiated and autophagosomes are formed26. Thus, the appearance of GFP-Atg8 puncta correlates with autophagy induction. Accordingly, we quantified GFP-Atg8 puncta formation upon Vma2 temporal degradation in nutrient-rich medium by fluorescence microscopy. Consistent with the western blot data, abundant GFP-Atg8 puncta were detected following the addition of 3-IAA at a level similar to that seen following a 1-h shift to nitrogen-starvation conditions (Supplementary Fig. 1e). Furthermore, we examined the generation of Atg8–PE as an additional means of monitoring autophagy induction. Due to the presence of the PE moiety, Atg8–PE migrates more rapidly than Atg8 during urea-SDS-polyacrylamide gel electrophoresis26. We found that the addition of 3-IAA for 4 h to cells growing in YPD medium resulted in a substantial increase in total Atg8 protein, with the majority of the protein being in the lipidated form, indicating robust autophagy induction (Supplementary Fig. 1f).
To test if the inducible degradation of Vma2 protein affects vacuolar acidification, we fused the C terminus of Pho8, a vacuolar-resident protein with the C terminus facing the lumen, with a pH-sensitive super ecliptic pHluorin (SEP) protein that displays increased fluorescence with higher pH levels27. A flow cytometry analysis suggested that 3-IAA treatment caused robust vacuolar deacidification to a similar extent as the V-ATPase inhibitor concanamycin A (conc. A)28 (Fig. 1d). Next, we asked if Vma2 temporal degradation triggers autophagy flux through vacuolar deacidification. To this end, we treated cells with the V-ATPase inhibitor conc. A in a nutrient-rich medium and observed strong GFP-Atg8 processing similar to that seen with 3-IAA treatment (Supplementary Fig. 2a). Furthermore, we used the AID strategy to conditionally deplete several subunits of the cytosolic V1 subcomplex of the V-ATPase (Vma1, Vma4 and Vma13) as well as those of the integral membrane V0 subcomplex (Vma6 and Vma9). We found that temporal degradation of different V-ATPase subunits recapitulated the autophagy-activating effect of the Vma2-AID*-9MYC protein; that is, we detected an increase in total GFP-Atg8 as well as an increase in free GFP (Supplementary Fig. 2b). Furthermore, we constructed a vacuolar gene AID strain library that tagged 208 yeast vacuolar proteins with an AID* degron domain. We induced protein degradation in nutrient-rich YPD medium and screened for vacuolar genes whose temporal degradation could recapitulate the autophagy activation effect of the V-ATPase subunits (Supplementary Fig. 2c). As expected, genes (EGO1 and EGO3) involved in TORC1 signaling were identified from this pooled screen (Supplementary Fig. 2d). In addition, Ypq1, a vacuolar lysine transporter, when temporally degraded, was sufficient to activate autophagy (Supplementary Fig. 2d). The autophagy activity triggered by Ypq1 degradation was milder than that caused by Vma2 degradation, suggesting that Ypq1 may act as one of the downstream effector molecules of V-ATPase dysfunction.
To rule out the possibility that the cleavage of GFP-Atg8, the formation of GFP-Atg8 puncta, and the increase of the Atg8–PE protein are caused by a non-autophagic process, we deleted the gene encoding the autophagy core machinery protein Atg1. We found that there was a complete block in the generation of free GFP from the cleavage of GFP-Atg8 in the atg1∆ cells (Fig. 1e). Similarly, GFP-Atg8 puncta no longer formed in the atg1∆ strain (Supplementary Fig. 1e). Deletion of the gene encoding the vacuolar protease Pep4, a key protease required for the activation of most vacuolar enzymes, also abolished autophagy flux, indicating that the observed cleavage of free GFP was generated from vacuolar hydrolysis (Fig. 1e). The deletion of genes encoding other autophagy core machinery proteins (Atg5, Atg9, Atg13, and Atg15) also completely prevented GFP-Atg8 processing (Supplementary Fig. 3).
According to a previous study, autophagic degradation during starvation is impaired in the V-ATPase null mutants16. To reconcile the opposite effects of V-ATPase dysfunction on autophagy flux in nutrient-rich versus starvation medium, we reasoned that dysfunction of the V-ATPase may have dual roles in autophagy regulation: On the one hand, it hampers vacuolar hydrolytic activities (but vacuolar proteases such as Pep4 are still partially active); on the other hand, it also activates autophagy induction (the final degradation step of this V-ATPase dysfunction-induced autophagy still requires the remaining hydrolytic activities). During nitrogen starvation, the starvation signal triggers robust autophagy induction, thus masking the milder autophagy activation effect of V-ATPase dysfunction. In a nutrient-rich medium, however, because basal autophagy flux is very low, the autophagy activation effect can be detected even though the autophagic degradation activity is partially compromised due to the altered pH. To test this, we performed a time-course analysis of GFP-Atg8 processing in SD-N + DMSO and SD-N + 3-IAA. We found that the GFP-Atg8 processing was minimal after 1 h in SD-N + 3-IAA (Supplementary Fig. 4a). However, after 4 h and 24 h, GFP-Atg8 processing essentially reached saturation in SD-N + 3-IAA and there was little difference between SD-N + DMSO and SD-N + 3-IAA (Supplementary Fig. 4a). In agreement with this, Atg8 lipidation was also not inhibited in the SD-N + 3-IAA condition (Supplementary Fig. 4b).
Based on these results, we developed the following hypothesis: The net effects of V-ATPase dysfunction and vacuolar deacidification on autophagy flux depend on the nutrient conditions. The net results of these defects promoted autophagy flux when cells were growing in nutrient-rich medium, while they caused a relative inhibition of autophagy flux during nitrogen starvation (Fig. 1f). Because this autophagy flux is induced by V-ATPase dysfunction in nutrient-rich medium, we named it “V-ATPase-dependent autophagyâ€.
TORC1 remains active during V-ATPase-dependent autophagy
In our previous reports, one unique feature of autophagy induced by FL-associated mutations in V-ATPase subunits is that TORC1 remains active18,19. To examine whether this property is also shared by the AID model, we measured the phosphorylation level of Atg13 and Npr1, two well-characterized TORC1 substrates29,30. Phosphorylation of Atg13 or Npr1 increases their molecular weight, leading to a slower mobility in an SDS-polyacrylamide gel. Treatment with the TORC1 inhibitor rapamycin caused robust dephosphorylation of Atg13 and Npr1. In contrast, both Atg13 and Npr1 remained hyperphosphorylated upon Vma2 transient degradation (Fig. 2a and Supplementary Fig. 5a). These results suggested that TORC1 remains active in the V-ATPase-dependent autophagy pathway.
a–e SEY6210 VMA2-AID*-9MYC cells were grown to mid-log phase in YPD (0 h), centrifuged and resuspended in YPD + 0.1% DMSO, YPD + 300 μM 3-IAA, YPD + 200 nM rapamycin, or SD-N medium for the indicated times. a ATG13-TAP cell lysates were analyzed by western blot. A 0.5-h treatment with the TORC1 inhibitor rapamycin triggers robust dephosphorylation of Atg13-TAP. Three biological replicates were repeated with similar results. b WLY176 WT and atg13∆ cell lysates were collected for the Pho8∆60 assay. Pho8∆60 activity was normalized to WT cells cultured in SD-N for 4 h (set to 100%). Statistical analysis was carried out by a two-tailed unpaired t-test with different variance. The summary data are presented as the mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, and NS, not significant (n = 3). c SEY6210 WT and atg11∆ cell lysates were collected and analyzed by western blot. Three biological replicates were repeated with similar results. d RPS6A-GFP WT, atg1∆, doa1∆, and ubp3∆ cell lysates were collected for western blot analysis. DMSO (0.1%) or 3-IAA (300 μM) were re-added every 12 h to maintain low levels of Vma2-AID*-9MYC. Statistical analysis was carried out by a one-tailed unpaired t-test with different variances. The summary data are presented as the mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, and NS, not significant (n = 3). e Total RNA was extracted from SEY6210 WT cells. The mRNA level of individual genes was normalized to the mRNA of the corresponding genes in cells treated with YPD + 0.1% DMSO for 4 h, which was set to 1. The data represent the average of 3 independent biological replicates. Statistical analysis was carried out by a one-tailed unpaired t-test with different variance. The summary data are presented as the mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, and NS, not significant (n = 3). f Hypothetical model of V-ATPase-dependent autophagy. The autophagy triggered by V-ATPase dysfunction features active TORC1, cargo selectivity, and less transcriptional activation of ATG genes.
V-ATPase-dependent autophagy is cargo selective
To further characterize the unique features of V-ATPase-dependent autophagy, we asked if it has cargo selectivity. Pho8∆60 is an engineered form of the Pho8 phosphatase where the N-terminal 60 amino acids are truncated, removing an internal, uncleaved signal sequence; thus, the altered protein can only enter the vacuole via the non-selective autophagy pathway31. Vacuolar delivery can be monitored by phosphatase activity resulting from the removal of a C-terminal propeptide. Pho8∆60 vacuolar delivery was strongly induced by nitrogen starvation, but not by Vma2 temporal degradation in nutrient-rich medium (Fig. 2b). This finding implied that, unlike nitrogen-starvation-induced autophagy, the V-ATPase-dependent autophagy has cargo selectivity.
Next, we examined the requirement for Atg11, the scaffold and adapter protein important for selective autophagy but not bulk autophagy32, in V-ATPase-dependent autophagy. Because the cargos of nitrogen-starvation-induced autophagy are largely non-selective, deletion of the gene encoding Atg11 only modestly affected the autophagy flux induced by nitrogen starvation (Fig. 2c). In contrast, V-ATPase-dependent autophagy was dramatically inhibited in the atg11∆ strain. Consistent with the GFP-Atg8 processing results, GFP-Atg8 puncta formation was also prevented in the atg11∆ strain in 3-IAA-treated cells growing in YPD medium (Supplementary Fig. 5b), indicating that V-ATPase-dependent autophagy was Atg11 dependent.
To identify the selective cargos of V-ATPase-dependent autophagy, we tagged different organellar proteins with GFP at the C-terminal end and performed GFP-processing assays: We found a clear increase in the vacuolar degradation of ribosomal proteins Rps6A and Rpl10, indicating the induction of ribophagy (Fig. 2d and Supplementary Fig. 5c). In contrast, although vacuolar deacidification caused abnormal mitochondria function33, the vacuolar degradation of the GFP-tagged mitochondrial matrix protein Idh1 was not increased in the context of V-ATPase-dependent autophagy, suggesting that minimal mitophagy occurred (Supplementary Fig. 5d). Previous studies reported that the ubiquitin protease Ubp3 and WD-repeat protein Doa1 regulate ribophagy during nitrogen starvation34,35. In the context of V-ATPase-dependent autophagy, individual deletion of ATG1, UBP3, or DOA1 significantly impaired ribophagy flux (Fig. 2d).
Moderate transcriptional activation of ATG genes in V-ATPase-dependent autophagy
Under nitrogen-starvation conditions, a broad spectrum of ATG genes is strongly transcriptionally activated36,37,38. In V-ATPase-dependent autophagy, however, only ATG1 and ATG9 mRNA levels were elevated, and to a much lesser extent than the starvation-induced transcriptional activation (Fig. 2e). Of note, although the ATG8 mRNA level was not changed upon Vma2 transient degradation (Fig. 2e), the Atg8 protein level was strongly induced (Fig. 1c and Supplementary Fig. 1f), suggesting the existence of post-transcriptional regulatory mechanisms in V-ATPase-dependent autophagy.
In summary, unlike nitrogen-starvation-induced autophagy, V-ATPase-dependent autophagy displayed three unique features: 1) TORC1 remained active; 2) a minimal level of non-selective autophagy, but instead the activation of selective forms of autophagy such as ribophagy; and 3) less transcriptional activation of ATG genes (Fig. 2f).
The Gcn2-Gcn4 pathway positively regulates V-ATPase-dependent autophagy
To further explore the unique molecular signature of V-ATPase-dependent autophagy, we aimed to compare differences in the transcriptome and proteome between cells treated with or without 3-IAA in nutrient-rich medium. To this end, we performed RNA-seq and stable isotope labeling by amino acids in cell culture (SILAC) experiments. For the RNA-seq experiments, we harvested cells to extract RNA after 4 h with or without 3-IAA treatment in nutrient-rich medium. For the SILAC experiments, we constructed an arg4∆ lys2Δ strain, and seeded cells from the same colony in synthetic medium with cold (R0K0) or hot (R6K4) amino acid isotopes. After cells reached early log phase, they were harvested and resuspended in cold/hot synthetic medium with DMSO (R0K0) or with 3-IAA (R6K4). After another 8-h culturing, cell lysates were collected from two culture tubes and mixed in a 1:1 ratio for two-way SILAC analysis. We found that Gcn4 targets were strongly induced at both mRNA and protein levels (Fig. 3a, b). This finding was further validated by RT-qPCR experiments: the transcription of Gcn4-target genes (CPA1, MET28, GRE1, MET16, and BNA1) was strongly activated in the context of the V-ATPase-autophagy axis (Supplementary Fig. 6a). Because Gcn4 activity is mainly regulated at the translational level39, we next used a high-affinity anti-Gcn4 antibody40 to monitor Gcn4 protein dynamics. The Gcn4 protein level was very low in nutrient-rich medium and was strongly increased when the Vma2 protein was degraded (Fig. 3c). This induction was less than the Gcn4 protein seen when cells were nitrogen starved, consistent with the fact that autophagy flux induced by V-ATPase-dependent autophagy was lower than that observed during nitrogen starvation. We further examined the Gcn4 protein level in FL-associated Vma2R381Q mutant strains, where autophagy activity increases in response to a defective V-ATPase. Consistent with the autophagy flux results, the basal protein level of Gcn4 in the Vma2R381Q mutant strain was enhanced in nutrient-rich medium, but to a lesser extent than seen in the WT strain under nitrogen starvation conditions (Supplementary Fig. 6b).
a Volcano plot of transcriptional changes in SEY6210 VMA2-AID*-9MYC cells treated with YPD + 300 μM 3-IAA for 4 h compared to the one treated with YPD + 0.1% DMSO for 4 h. DESeq2 was used with the Wald test (two-sided) to assess differential expression. Adjusted p-values were calculated using the Benjamini–Hochberg procedure to control for the FDR. Differentially expressed genes, padj<0.05, log2 (fold change)>1.5 were colored in blue (downregulated in cells treated with YPD + 300 μM 3-IAA for 4 h) and red (upregulated in cells treated with YPD + 300 μM 3-IAA for 4 h). Differentially expressed Gcn4 target genes were labeled in blue with black circles. Differentially expressed RPGs were labeled in red with black circles. b SEY6210 VMA2-AID*-9MYC arg4∆ lys2∆ cells were grown in SILAC medium supplemented with non-labeled lysine and arginine (R0K0) or stable isotype-labeled lysine and arginine (R6K4) to early log phase (OD600 = 0.02 ~ 0.04), centrifuged and resuspended in SILAC medium (R0K0) + 0.1% DMSO or SILAC medium (R6K4) + 300 μM 3-IAA for 8 h. Cell pellets were harvested and analyzed by two-way SILAC. Selected upregulated and downregulated pathways adapted from G-profiler are presented. c, d SEY6210 VMA2-AID*-9MYC cells were grown to mid-log phase in YPD (0 h), centrifuged and resuspended in YPD + 0.1% DMSO, YPD + 300 μM 3-IAA, or SD-N medium for the indicated times. Three biological replicates were repeated with similar results. c WT and gcn4∆ cell lysates were prepared and analyzed by western blot. Gcn4 protein was detected with an anti-Gcn4 antibody. *, non-specific band. d WT, gcn2∆, and gcn4∆ cell lysates were prepared and analyzed by western blot. e Schematic of V-ATPase-dependent autophagy. V-ATPase dysfunction activates Gcn2 kinase activity and induces Gcn4 translation to promote autophagy.
To examine the function of Gcn4 in V-ATPase-dependent autophagy, we analyzed autophagy flux using the GFP-Atg8 processing assay in VMA2-AID*-9MYC gcn4∆ cells. The deletion of GCN4 only partially impaired autophagy flux during nitrogen starvation but completely blocked GFP-Atg8 cleavage in the V-ATPase-dependent autophagy protocol (Fig. 3d). Because Gcn2 positively regulates Gcn4 induction and is also required for classical nitrogen-starvation-induced autophagy39,41, we next constructed VMA2-AID*-9MYC gcn2∆ cells. Similar to gcn4 deletion, deletion of GCN2 also completely abolished the autophagy flux induced by Vma2 temporal degradation while having only a partial effect on autophagy induced by nitrogen starvation (Fig. 3d). Furthermore, because Gcn2 enhances Gcn4 translation by phosphorylating the translational factor Sui2/eIF2α42, we asked if Gcn2 kinase activity is changed in V-ATPase-dependent autophagy. To this end, we used a phospho-specific antibody to monitor the phosphorylation of Sui2/eIF2α at the Ser51 residue. Indeed, the p-Ser51-Sui2/eIF2α level was strongly increased upon Vma2 temporal degradation in the WT but not in the gcn2∆ cells (Supplementary Fig. 6c). Because Gcn2 contains a regulatory domain that can sense uncharged tRNA43, we hypothesized that V-ATPase dysfunction and vacuolar pH dysregulation alter amino acid compartmentalization, leading to the activation of Gcn2, which in turn triggers Gcn4 induction. In summary, the Gcn2-Gcn4 pathway positively regulates V-ATPase-dependent autophagy (Fig. 3e).
Tryptophan prototrophy inhibits V-ATPase-dependent autophagy
Because nitrogen starvation and V-ATPase dysfunction represent two different types of stress, we reasoned that in addition to the Gcn2-Gcn4 pathway shared by both scenarios, V-ATPase-dependent autophagy must have some unique aspects to its regulatory mechanisms. To this end, we used a high-copy plasmid genomic DNA library to perform an unbiased genome-wide suppressor screen44: We transformed the VMA2-AID* cells with 1588 overexpression plasmids from the library and examined the induction of V-ATPase-dependent autophagy in each transformant by using the western blot-based GFP-Atg8 processing as the readout assay. In summary, we isolated 45 suppressor plasmids out of the total 1588 plasmids whose overexpression inhibited autophagy flux in the context of V-ATPase-dependent autophagy (Fig. 4a). Because each suppressor plasmid from the library includes multiple genes, these 45 suppressor plasmids generated a candidate gene list with 145 genes. Next, we individually overexpressed these 145 candidate genes and finally obtained 32 suppressor genes whose overexpression could reproducibly suppress V-ATPase-dependent autophagy (Fig. 4a, b). Three major groups that were revealed from the 32 suppressor genes included ribosome biogenesis factors (8/32, 25%), metabolic enzymes (7/32, 21.9%), and transcriptional regulators (6/32, 18.9%) (Fig. 4b).
a Schematic of suppressor screen design. For the first round of the suppressor screen, one biological replicate was performed for each transformant from the 1588 genome tiling library. A ratio of free GFP versus total GFP below 0.2 was considered as a suppressive effect on V-ATPase-dependent autophagy. For the second round of individual gene overexpression validation, 3 different colonies were tested, and candidates were considered suppressor genes if more than two colonies showed a suppressive effect. b Summary of 32 suppressor genes. c SEY6210 VMA2-AID*-9MYC cells (pRS307-CUP1p-GFP-ATG8) transformed with empty pRS405 plasmid and pRS405-CUP1p-TRP1-TAP plasmid were grown to mid-log phase in YPD (0 h), centrifuged and resuspended in YPD + 0.1% DMSO or YPD + 300 μM 3-IAA for 8 h. Cell lysates were prepared and analyzed by western blot. Three biological replicates were repeated with similar results. d SEY6210 VMA2-AID*-9MYC cells (pRS307-CUP1p-GFP-ATG8) with overexpressed Trp permease Tat2 protein (pRS405-CUP1p-TAT2-3xHA) were grown to mid-log phase in YPD (0 h), centrifuged and resuspended in YPD + 0.1% DMSO or YPD + 300 μM 3-IAA with or without 1 mM Trp for 8 h. Cell lysates were processed and analyzed by western blot. Three biological replicates were repeated with similar results. e Volcano plot of transcriptional changes in cells transformed with pRS405-CUP1p-TRP1-3xHA plasmid compared to empty pRS405 plasmid after 4 h in YPD + 300 µM 3-IAA medium. DESeq2 was used with the Wald test (two-sided) to assess differential expression. Adjusted p-values were calculated using the Benjamini–Hochberg procedure to control for the FDR. Differentially expressed genes, padj<0.05, log2 (fold change)>1.5 are colored in blue (downregulated in cells transformed with pRS405-CUP1p-TRP1-3xHA plasmid) and red (upregulated in cells transformed with pRS405-CUP1p-TRP1-3xHA plasmid). Gcn4 target genes and RPGs were highlighted with black circles. f SEY6210 VMA2-AID*-9MYC gcn4∆ cells and WT cells transformed with empty pRS405 plasmid and pRS405-CUP1p-TRP1-3xHA plasmid were grown to mid-log phase in YPD medium (0 h), centrifuged and resuspended in YPD + 0.1% DMSO or YPD + 300 μM 3-IAA medium for 8 h. Cell lysates were prepared and analyzed by western blot. Gcn4 protein was detected with anti-Gcn4 antibody. *, non-specific band. Three biological replicates were repeated with similar results. g BY4742 WT, trp1∆, and gcn2∆ trp1∆ cells were grown to mid-log phase in YPD (0 h), centrifuged and resuspended in YPD + 0.1% DMSO, YPD + 300 μM 3-IAA, or amino acid starvation medium for the indicated times. Cell lysates were processed and analyzed by western blot. p-Ser51-Sui2/eIF2α was detected with a phospho-specific antibody. Three biological replicates were repeated with similar results. h Schematic of V-ATPase-dependent autophagy. Trp prototrophy inhibits Gcn4 induction and suppresses autophagy flux upon Vma2 degradation.
Among all these suppressor genes, we first focused on TRP1, the gene that encodes an essential enzyme in the tryptophan (Trp) biosynthetic pathway45. In mammals, Trp is an essential amino acid and is a precursor for several important metabolites such as serotonin, melatonin, and nicotinamide adenine dinucleotide (NAD+)46. The levels of Trp metabolites are intimately associated with many clinical features, and several recent studies found that modulating Trp metabolism can control disease progression47. Unlike mammals, which must obtain Trp from their diet, wild-type yeast cells can synthesize Trp through a de novo pathway48. However, all of the above characterizations of V-ATPase-dependent autophagy were performed in strains constructed in the SEY6210 genetic background; SEY6210 contains a truncation mutation within the TRP1 gene and is accordingly Trp auxotrophic49, leading to the possibility that the Trp biosynthetic pathway regulates V-ATPase-dependent autophagy.
Accordingly, we transformed SEY6210 VMA2-AID*-9MYC cells with integrative plasmids expressing the Trp1-TAP protein. Expression of Trp1-TAP completely prevented autophagy flux upon Vma2 temporal degradation (Fig. 4c). In addition, Trp1-TAP expression also partially rescued the cell growth inhibition that occurred due to vacuolar deacidification (Supplementary Fig. 7a). In parallel, deletion of the TRP1 gene in BY4742 slightly impaired cell growth in the SMD + 3-IAA medium (Supplementary Fig. 7b). Although deletion of the ATG1 gene did not affect cell growth, simultaneous disruption of the autophagy pathway and the Trp biosynthetic pathway severely inhibited cell growth. This finding suggests that when intracellular Trp is scarce due to auxotrophy, and vacuoles are de-acidified, the V-ATPase-dependent autophagy pathway is activated and is required for optimal cell growth. To examine if an intact endogenous Trp biosynthetic pathway is sufficient to block V-ATPase-dependent autophagy, we constructed the VMA2-AID*-9MYC strain in a Trp prototrophic background. In this case, Vma2 could be efficiently degraded under the same YPD + 3-IAA protocol (Supplementary Fig. 7c). The autophagy flux, however, was substantially blocked. Individual disruptions of different essential genes in the Trp biosynthetic pathway reactivated V-ATPase-dependent autophagy (Supplementary Fig. 7c). The inhibitory effect of Trp prototrophy on autophagy flux was specific to V-ATPase-dependent autophagy, because nitrogen-starvation-induced autophagy was not affected by Trp prototrophy (Supplementary Fig. 7c). Furthermore, we asked if V-ATPase-dependent autophagy is regulated by Trp and its derivative metabolites or any upstream intermediate molecules in the Trp biosynthetic pathway. To test this, we supplemented nutrient-rich YPD medium with additional exogenous Trp and found that autophagy flux was largely prevented in the Trp auxotrophic SEY6210 strain (Fig. 4d). This finding indicated that the Trp prototrophy inhibited V-ATPase-dependent autophagy via Trp and/or its derivative metabolites.
To investigate how Trp metabolism regulates V-ATPase-dependent autophagy, we performed RNA-seq profiling experiments in a Trp auxotrophic SEY6210 background strain transformed with an empty vector or a vector that encodes the TRP1 gene under 4 h YPD + 3-IAA treatment. We found that compared to Trp auxotrophic cells, induction of Gcn4 target genes was dramatically inhibited (Fig. 4e). Additional RT-qPCR validation confirmed that complementation of the Trp biosynthetic pathway partially inhibited Gcn4 target gene (CPA1, MET16, MET28) activation (Supplementary Fig. 7d). Consistent with this result, induction of the Gcn4 protein was also partially inhibited in the TRP1-overexpressing cells (Fig. 4f). In contrast, the increase of Gcn2 activity (as assessed by p-Ser51 phosphorylation of Sui2) in the context of V-ATPase-dependent autophagy was not affected by Trp prototrophy (Fig. 4g and Supplementary Fig. 7e). This finding suggested that Trp metabolism did not affect general amino acid sensing by Gcn2. Considering the fact that Trp is the least abundant amino acid in the cell48, we concluded that Trp metabolism involves a unique mechanism that negatively regulates V-ATPase-dependent autophagy via suppressing Gcn4 induction while Gcn2 remains active (Fig. 4h).
NAD+ acts downstream of the Trp biosynthetic pathway
Because Trp derivatives play important roles in multiple biological processes, we hypothesized that Trp metabolism regulates V-ATPase-dependent autophagy via its secondary metabolites. For example, Trp is the necessary precursor for the de novo NAD+ biosynthetic pathway (also known as the kynurenine pathway). Therefore, we next asked if the NAD+ biosynthetic pathway acts downstream of the Trp biosynthetic pathway. To this end, we deleted BNA4, the gene encoding the essential enzyme for the conversion from Trp to kynurenine in a Trp prototrophic genetic background (Fig. 5a). Similar to the result seen in the trp1∆ cells, disruption of the kynurenine pathway in the bna4∆ cells reactivated V-ATPase-dependent autophagy. In addition, metabolomic profiling showed that intermediates of the kynurenine pathway, such as kynurenine and nicotinic acid mononucleotide/NaMN, were dramatically downregulated in the context of V-ATPase-dependent autophagy (Fig. 5b). Refueling the Trp biosynthetic pathway by expressing the TRP1 gene could rescue the depletion of the kynurenine pathway intermediates.
a–f BY4742 or SEY6210 VMA2-AID*-9MYC cells (pRS307-CUP1p-GFP-ATG8) were grown to mid-log phase in YPD (0 h), centrifuged and resuspended in YPD + 0.1% DMSO, YPD + 300 μM 3-IAA, or SD-N medium with or without 1 mM niacin addition for the indicated times. a BY4742 WT, trp1∆, and bna4∆ cell lysates were prepared and analyzed by western blot. Three biological replicates were repeated with similar results. b SEY6210 VMA2-AID*-9MYC whole cell lysates were collected, and intermediate metabolites involved in the kynurenine pathway were analyzed. Statistical analysis was carried out using a one-tailed unpaired t-test. The summary data are presented as the mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, and NS, not significant (n = 3). c SEY6210 WT cell lysates were processed and analyzed by western blot. Three biological replicates were repeated with similar results. d SEY6210 WT and gcn4∆ cell lysates were prepared and analyzed by western blot. Gcn4 protein was detected with an anti-Gcn4 antibody. *, non-specific band. Three biological replicates were repeated with similar results. e Total RNA was extracted from SEY6210 WT cells. The mRNA level of individual genes was normalized to the mRNA of the corresponding genes in cells treated with YPD + 0.1% DMSO without niacin addition for 4 h, which was set to 1. The data represent the average of 3 independent biological replicates. Statistical analysis was carried out by a one-tailed unpaired t-test. The summary data are presented as the mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, and NS, not significant (n = 3). f SEY6210 WT and gcn2∆ cell lysates were processed and analyzed by western blot. p-Ser51-Sui2/eIF2α was detected with a phospho-specific antibody. Three biological replicates were repeated with similar results. g Schematic of V-ATPase-dependent autophagy. NAD+ biosynthesis acts downstream of Trp metabolism but is not responsible for inhibition of Gcn4 induction.
In addition to the kynurenine pathway, NAD+ can also be generated from a salvage pathway50. To test if enhancing the NAD+ salvage pathway can inhibit the V-ATPase-autophagy axis, we supplemented the Trp auxotrophic SEY6210 VMA2-AID*-9MYC cells with exogenous niacin, the precursor that fuels the NAD+ salvage pathway. We found that the addition of niacin partially blocked V-ATPase-dependent autophagy (Fig. 5c). In summary, NAD+ acts downstream of the Trp biosynthetic pathway to negatively regulate V-ATPase-dependent autophagy.
To examine if NAD+ mediates the repressive effect on Gcn4 induction by Trp prototrophy, we monitored the Gcn4 protein level in cells treated with YPD + 3-IAA with or without niacin supplementation. Addition of 1 mM niacin did not inhibit Gcn4 protein induction (Fig. 5d). Furthermore, we measured the mRNA level of Gcn4 target genes (ATG1, BNA1, CPA1, MET16). Similar to the result with Gcn4 protein, the induction of these Gcn4 target genes was not affected (Fig. 5e). Gcn2 activation was also unaltered by niacin (Fig. 5f). These findings suggested that although acting downstream of Trp prototrophy, NAD+ regulated V-ATPase-dependent autophagy through a parallel mechanism without inhibiting Gcn4 induction (Fig. 5g).
Ribosome biogenesis acts downstream of Trp prototrophy and inhibits Gcn4 induction
Three parallel lines of evidence converged on the hypothesis that the ribosome biogenesis pathway mediates the repressive effect on Gcn4 induction by Trp metabolism: First, under nutrient-rich conditions, ribosomal proteins can repress the translation of GCN4 mRNA as a regulatory mechanism to maintain the Gcn4 protein at a relatively low basal level51. Second, our RNA-seq profiling found that ribosomal protein genes (RPGs) were regulated in an opposite direction to Gcn4 target genes in the context of V-ATPase-dependent autophagy. Upon Vma2 degradation, RPGs were downregulated (Fig. 3a), which could be reversed by Trp prototrophy (Fig. 4e). Third, genes involved in the ribosome biogenesis pathway constituted the largest category of the 32 suppressor genes from the unbiased genome-wide screen (Fig. 4b), suggesting that the ribosome biogenesis pathway is functionally implicated in V-ATPase-dependent autophagy regulation.
To test this hypothesis, we first validated the RNA-seq results by RT-qPCR experiments: cytosolic RPGs (RPS6A/RPS6B, RPS22A, RPL38, RPL31A), as well as a mitochondrial RPG (MRPL50), were significantly downregulated in the context of V-ATPase-dependent autophagy (Fig. 6a). This downregulation of cytosolic RPGs was reversed when TRP1 was expressed in the Trp auxotrophic cells (Fig. 6a). Supplementation of exogenous niacin, however, did not prevent an RPG decrease (Supplementary Fig. 8a), indicating that ribosome biogenesis was acting downstream of Trp prototrophy but not of the NAD+ pathway. In contrast, neither TRP1 expression nor niacin supplementation reversed the downregulation of mitochondrial RPG MRPL50, suggesting that cytosolic RPGs, but not mitochondrial RPGs act downstream of Trp metabolism.
a–c SEY6210 VMA2-AID*-9MYC cells were grown to mid-log phase in YPD (0 h), centrifuged and resuspended in YPD + 0.1% DMSO, YPD + 300 μM 3-IAA, or SD-N medium for the indicated times. a Total RNA was extracted from SEY6210 VMA2-AID*-9MYC cells transformed with empty pRS405 plasmid and pRS405-CUP1p-TRP1-TAP plasmid or SEY6210 VMA2-AID*-9MYC cells treated with or without 1 mM niacin. The mRNA level of individual genes was normalized to the mRNA of the corresponding genes in cells transformed with an empty vector treated with YPD + 0.1% DMSO for 4 h, which was set to 1. The data represent the average of 3 independent biological replicates. Statistical analysis was carried out by a one-tailed unpaired t-test. The summary data are presented as the mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, and NS, not significant (n = 3). b SEY6210 VMA2-AID*-9MYC WT and gcn4Δ cells transformed with empty pRS405 plasmid and pRS405-CUP1p-RPS6A-3xHA plasmid were harvested, processed and analyzed by western blot. Gcn4 protein was detected with an anti-Gcn4 antibody. *, non-specific band. Three biological replicates were repeated with similar results. c SEY6210 VMA2-AID*-9MYC cells transformed with empty pRS405 plasmid and pRS405-CUP1p-RPS6A-TAP plasmid were harvested, processed and analyzed by western blot. Three biological replicates were repeated with similar results. d Schematic of V-ATPase-dependent autophagy. Trp prototrophy prevents downregulation of ribosome biogenesis upon Vma2 temporal degradation and therefore inhibits Gcn4 induction to block autophagy flux. NAD+ homeostasis does not affect ribosome biogenesis and Gcn4 induction. Three biological replicates were repeated with similar results.
Next, we selected RPS6A as an RPG for further functional characterization. RPS6A encodes a component of the small (40S) ribosomal subunit and was revealed as a suppressor gene from the suppressor screen (Fig. 4b). If the ribosome biogenesis pathway mediates Trp prototrophy’s repressive effect on Gcn4 induction, overexpression of RPS6A will block Gcn4 induction but will not affect Gcn2 activation. Indeed, similar to Trp prototrophy, overexpression of RPS6A led to partial inhibition of Gcn4 induction (Fig. 6b). In contrast, the phosphorylation of Sui2/eIF2α at the Ser51 residue was unaffected by RPS6A overexpression (Supplementary Fig. 8b), indicating that the ribosome biogenesis pathway did not alter Gcn2 activation. Furthermore, when overexpressed, RPS6A significantly inhibited autophagy flux in the context of the V-ATPase-autophagy axis but only modestly inhibited nitrogen-starvation-induced autophagy (Fig. 6c). Therefore, in parallel to the NAD+ pathway, ribosome biogenesis acts as the second downstream effector of Trp prototrophy and is responsible for the repressive effect on Gcn4 induction (Fig. 6d).
Rpc53 phosphorylation regulates V-ATPase-dependent autophagy by modulating the ribosome biogenesis pathway
In the context of V-ATPase-dependent autophagy, transcription of RPGs was only partially decreased (Fig. 6a), and it is questionable whether this extent of inhibition is sufficient to trigger strong downregulation of ribosome biogenesis. We hypothesized that, besides transcriptional control of RPGs, there are alternative regulatory mechanisms contributing to the downregulation of the ribosome biogenesis pathway in V-ATPase-dependent autophagy. To uncover such mechanisms, we next focused on the RNA polymerase III subunit Rpc53. During nitrogen starvation, Rpc53 is hyperphosphorylated and inactivated, leading to a decrease in ribosome biogenesis (Fig. 7a)52. To examine if Rpc53 is phosphorylated upon Vma2 temporal degradation, we constructed an RPC53-TAP strain. Hyperphosphorylated Rpc53 migrates slower in the SDS-polyacrylamide gel due to a higher molecular weight52. In the context of V-ATPase-dependent autophagy, approximately half of the total Rpc53 was phosphorylated compared to the effect of treating cells with rapamycin (Fig. 7b). To ask if phosphorylation of Rpc53 is functionally important for V-ATPase-dependent autophagy, we generated a non-phosphorylatable Rpc53T228A strain. Rpc53 hyperphosphorylation was completely prevented in the Rpc53T228A strain, and we found that Gcn4 induction was partially inhibited in this strain (Fig. 7c), suggesting that Rpc53 phosphorylation mediated the regulation of Gcn4 induction in V-ATPase-dependent autophagy. Because full phosphorylation and inactivation of Rpc53 requires two kinases, Kns1 and Mck152, we next constructed a kns1∆ mck1∆ double-knockout strain and examined its functional importance in connection to V-ATPase-dependent autophagy. We found that double-knockout of KNS1 and MCK1 had additive effects, leading to a more dramatic autophagy defect than either kns1∆ or the mck1∆ single knockout (Supplementary Fig. 9a). Furthermore, similar to the Rpc53T228A strain, the kns1∆ mck1∆ double-knockout strain also resulted in a partial inhibition of Gcn4 induction (Supplementary Fig. 9b).
a Schematic showing the phosphosites in Rpc53. The zoomed-in sequence shown with a proposed Kns1 motif (RXXS/TP) underlined and the phosphosite boxed in gray. Two overlapping GSK3 motifs (S/TXXXS/T) direct Mck1 phosphorylation of primed substrates (blue arrows) at phosphosites boxed in gray. Adapted from Lee et al., (2012). During nitrogen starvation or rapamycin treatment, Rpc53 is hyperphosphorylated and inactivated. As a result, ribosome biogenesis is repressed. b–d SEY6210 VMA2-AID*-9MYC cells (pRS307-CUP1p-GFP-ATG8) were grown to mid-log phase in YPD (0 h), centrifuged and resuspended in YPD + 0.1% DMSO, YPD + 300 μM 3-IAA, YPD + 200 nM rapamycin, or SD-N medium for the indicated times. Three biological replicates were repeated with similar results. b RPC53-TAP cell lysates were prepared and analyzed by western blot. Rapamycin treatment was the positive control that was known to induce complete Rpc53 phosphorylation. Rpc53-TAP protein was detected with anti-PAP antibody. The upper band represents phosphorylated Rpc53 (p-Rpc53-TAP) with lower mobility on an SDS-PAGE gel. c SEY6210 RPC53-TAP and RPC53T228A-TAP cell lysates were processed and analyzed by western blot. Gcn4 protein was detected with an anti-Gcn4 antibody. *, non-specific band. d RPC53-TAP and RPC53T228A-TAP cells transformed with empty pRS405 plasmid and pRS405-CUP1p-TRP1-3xHA plasmid were harvested, processed, and analyzed by western blot. e Graphical model of V-ATPase-dependent autophagy. 1 A, FL-associated mutations of the V-ATPase subunits cause vacuolar deacidification. 1 B, unlike nitrogen starvation-induced autophagy, the TORC1 complex remains active in this context. 2 A, The Gcn2-Gcn4 pathway is activated by the vacuolar deacidification signal. 2B, Rpc53 is partially phosphorylated by Kns1 and Mck1 and thus inactivated. This induces a downregulation of ribosome biogenesis. 2 C, the downregulation of ribosome biogenesis releases the inhibitory effect on Gcn4 translational activation, thus further stimulating the Gcn2-Gcn4 pathway. 3 A, Trp prototrophy partially inhibits Rpc53 phosphorylation, and therefore maintains ribosome biogenesis in an active state. 3B, in addition, the Trp biosynthetic pathway fuels the intracellular NAD+ pool, which inhibits V-ATPase-dependent autophagy without affecting Gcn4 induction.
Finally, to test if Trp prototrophy regulates ribosome biogenesis by modulating Rpc53 phosphorylation, we transformed SEY6210 RPC53-TAP cells with a vector encoding the TRP1 gene (Fig. 7d). In comparison to the empty vector control cells, phosphorylation of Rpc53 was partially inhibited in TRP1-expressing cells (Fig. 7d), suggesting that besides direct activation of RPG transcription, Trp metabolism also modulated ribosome biogenesis by controlling Rpc53 phosphorylation. In addition, supplementation with niacin did not inhibit Rpc53 phosphorylation (Supplementary Fig. 9c), consistent with the fact that ribosome biogenesis was positively regulated by Trp prototrophy but not NAD+ (Fig. 6a and Supplementary Fig. 9a). In summary, Vma2 degradation promoted Rpc53 phosphorylation by Kns1 and Mck1, resulting in downregulation of ribosome biogenesis, thus derepressing Gcn4 induction. Repletion of Trp inhibited Rpc53 phosphorylation to prevent downregulation of the ribosome biogenesis pathway, thus repressing Gcn4 induction (Fig. 7e).
Discussion
The V-ATPase plays important roles in vacuolar/lysosomal homeostasis. The inessentiality of V-ATPase genes in yeast makes it a convenient system to dissect the regulatory functions of the V-ATPase on autophagy activity. In this article, we demonstrated that V-ATPase dysfunction has a dual effect on autophagy: on the one hand, it impairs lysosomal/vacuolar hydrolytic activity to inhibit the final step of autophagy; on the other hand, it can activate earlier steps of autophagy (e.g., phagophore formation) in an ATG gene-dependent manner. Unlike starvation-induced autophagy that responds to external nutrient availability, V-ATPase-dependent autophagy is triggered by intrinsic change (i.e., by intracellular signals based on the lysosomal/vacuolar status). We have also performed a pooled genetic screen on vacuolar proteins and found that conditional depletion of several vacuolar transporters can partially phenocopy the autophagy-activating effect of the V-ATPase dysfunction (Supplementary Fig. 2d). This result suggests that the solute gradients between the vacuolar lumen and cytoplasm directly regulate autophagy activity.
Given that cancer cell evolution involves the accumulation of genomic mutations, it is plausible to hypothesize that mutations in V-ATPase genes (and other genes such as KRAS) benefit cancer cells by activating autophagy flux even in nutrient-rich environments to promote cell survival as a universal strategy53. Reduced systemic Trp levels have been measured in patients with various types of cancer54,55,56,57,58,59, raising the possibility that Trp metabolism may regulate autophagy flux in cancer cells.
Our work prompts a reconsideration of the widespread use of V-ATPase inhibitors in the autophagy assays in mammalian cells. The V-ATPase inhibitor bafilomycin A1 is the most commonly used compound to collapse the lysosomal pH gradient and to block autophagosome-lysosome fusion, thus enabling quantitative measurement of autophagy flux based on accumulated lipidated LC323,60. Our study suggests that the accumulation of lipidated LC3 may also result from activation of V-ATPase-dependent autophagy.
A recent study reports that in nutrient-replete conditions, V-ATPase activity and lysosomal/vacuolar pH dynamically fluctuate within a relatively wide range, which is associated with cell cycle progression61. Another recent analysis in nematodes finds that knockdown of specific V-ATPase subunits can promote autophagy activity and exert a beneficial effect on longevity62. These studies suggest that, as a more fundamental mechanism to maintain cellular homeostasis, the function of V-ATPase-dependent autophagy may expand from pathological FL tumor cells to a broader physiological scenario.
One major limitation of the current study is not examining the functions of components of the V-ATPase-dependent autophagy pathway in human cells. Consistent with the findings from yeast in the current work, our previous characterizations of FL-associated V-ATPase mutants in human cells also found that when an FL-associated V-ATPase mutant is expressed, ATF4/Gcn4 is strongly induced18, suggesting that the V-ATPase-dependent autophagy pathway shares conserved regulatory mechanisms in eukaryotes. Establishing an AID system for V-ATPase subunits in mammalian cell lines is challenging but will be of significant importance to expand findings from the yeast system to human cells in future studies.
Methods
Yeast strains, media, and growth conditions
Yeast strains used in this study are listed in Table S1. Gene deletions were generated according to the standard method63. Strain JH416 (Rpc53T228A) was made by mutating RPC53 at the corresponding sites in the genome64. Yeast cells were cultured at 30 °C in rich medium (YPD; 1% yeast extract, 2% peptone, and 2% glucose). To induce autophagy, cells in the mid-log phase (OD600 = 0.8–1.0) were shifted to nitrogen-starvation medium with glucose (SD-N; 0.17% yeast nitrogen base without ammonium sulfate or amino acids, and 2% glucose) for the indicated times. For the SILAC assay, -Lys -Arg synthetic medium (0.17% yeast nitrogen base without amino acids, 2% glucose, 1 mM alanine, 1 mM asparagine, 1 mM aspartic acid, 1 mM cysteine, 1 mM glutamic acid, 1 mM glutamine, 1 mM glycine, 1 mM methionine, 1 mM proline, 2 mM leucine, 2 mM serine, 2 mM threonine, 0.75 mM isoleucine, 0.75 mM phenylalanine, 0.5 mM tyrosine, 0.3 mM histidine, 0.4 mM tryptophan, 2.5 mM valine, 0.15 mM adenine, 0.2 mM uracil) is supplemented with 30 mg/L L-lysine-2HCl (Thermo Scientific, 89987) or L-lysine-2HCl, 4,4,5,5-D4 (Thermo Scientific, 88437) and 5 mg/L L-arginine-HCl (Thermo Scientific, 89989) or L-arginine-HCl, 13C6 (Thermo Scientific, 88210). For the flow cytometry assay, cells were resuspended in minimal synthetic medium (0.17% yeast nitrogen base without amino acids, 2% glucose).
Auxin-inducible degradation
To set up the AID system, SEY6210 and BY4742 (WT) cells were first transformed with the plasmid pNHK53 (ADH1p-OsTIR1-9MYC). Vma2 and other V-ATPase genes were then tagged with AID*-9MYC by homologous recombination. The DNA fragments used for transformation were amplified with pHIS3-AID*-9MYC (Addgene, 99524; deposited by Dr. Helle Ulrich) and pCloNAT-AID*-9MYC (this study) as the template DNA. The “AID*“ refers to the 71–114 amino acids of the IAA17 protein in plants24. To conditionally knock down VMA2, the cells were first cultured in YPD overnight until mid-log phase and then diluted and treated with 300 μM IAA (Sigma, I2886) for the indicated times to induce the degradation of Vma2. For starvation, cells were shifted to SD-N medium, and 300 μM IAA was added to the medium to maintain degradation of Vma2. After an appropriate time period of treatment or starvation, samples were collected for western blot, RT-qPCR analysis or vacuole isolation.
Autophagic flux assays and western blotting
For the GFP-Atg8 and Pgk1-GFP processing assays, 1 OD600 units of yeast cells were harvested, resuspended in 500 μL of 10% TCA, and kept at 4 °C for 30 min. The samples were centrifuged at 16,000 g for 5 min at 4 °C, the supernatant was removed, and the pellets were resuspended in 200 μL acetone. The samples were subject to centrifugation under the same conditions; the supernatant fraction was removed, and the pellet fractions were dried in a fume hood at room temperature for 15 min. Next, 50 μL MURB buffer (50 mM NaH2PO4, pH 7.0, 25 mM MES, pH 7.0, 1% SDS, 3 M urea, 0.5% BME, 1 mM NaN3, 0.2 μg/μl bromophenol blue) was added to the pellets, followed by 5 min vigorous mixing by vortex at 4 °C. Before loading onto the SDS-PAGE gel, protein samples were incubated in a 55 °C water bath for 10 min31,65,66.
For the Pho8∆60 assay, 1 OD600 units of yeast cells were harvested, resuspended in 300 μL Pho8∆60 lysis buffer (20 mM PIPES-KOH, pH 6.8, 50 mM KCl, 100 mM KOAc, 10 mM MgSO4, 10 μM ZnSO4, 0.5% Triton X-100), followed by 5 min of vigorous mixing by vortex at 4 °C. After 15 min, the samples were subjected to centrifugation at 800 g for 15 min at 4 °C; 50 μL of the supernatant fraction was used for a BCA assay (Thermo Scientific Pierce BCA protein assay kit, 23225) and 150 μL of the supernatant freaction was mixed with 600 μL Pho8∆60 reaction buffer (250 mM Tris-HCl, pH 8.5, 10 mM MgSO4, 10 μM ZnSO4, 0.4% Triton X-100, 1.25 mM p-nitrophenyl phosphate [pNPP; Sigma N9389]). For the BCA assay, 50 μL of the supernatant were mixed with 950 μL of the BCA solution, incubated at 37 °C for 15 min, and analyzed using a spectrophotometer (562 nm). For the Pho8∆60 reaction assay, the mixed solution was incubated at 30 °C for 30 min and again analyzed using a spectrophotometer (405 nm).
Antisera were from the following sources: Atg8 (1:1000)67, Pgk1 (1:100,000; a generous gift from Dr. Jeremy Thorner, University of California, Berkeley), monoclonal YFP (1:5000; Clontech, 632381), antibody to PAP (1:5000; Jackson ImmunoResearch, 323–005-024), antibody to Gcn4 (1:1000; C11L34 clone, Novus, NBP2-81274), antibody to p-Ser51-Sui2 (1:1000; Cell Signaling Technology, 9721) and anti-MYC antibody (1:5000; Sigma, M4439). The blots were imaged using a ChemiDoc Touch imaging system (Bio-Rad) and quantified using Bio-Rad Image Lab software.
Flow cytometry analysis for vacuolar deacidification
Pho8-SEP cells were harvested under the indicated conditions, washed with 0.85% NaCl, resuspended in minimal synthetic medium and analyzed on an Attune NXT 4 Laser Flow Cytometer (Attune NXT4). Cells were differentiated from debris by forward scatter and side scatter. Pho8-SEP was detected via the BL-1 channel at 325 V. Unless otherwise indicated, the fluorescence of 2 × 104 cells was analyzed per sample. Data analysis was carried out in FlowJo.
RNA extraction and RT-qPCR
To eliminate genomic DNA contamination after total RNA extraction, an additional DNase treatment was performed according to the RNeasy kit instructions with the RNase-free DNase set (Qiagen). One microgram of total RNA was reverse-transcribed into cDNA in a 20-μl reaction mixture. The cDNA levels were then analyzed using the Eppendorf Realplex4 with the gene-specific primers listed in Table S2. Each sample was tested in a 384-well plate (Applied Biosystems). The reaction mix (15-μl final volume) consisted of 7.5 μl of Power SYBR Green master mix (Applied Biosystems), 0.75 μl of each primer (400 nM final concentration), and 6 μl of a 1:5 dilution of the cDNA preparation. The thermocycling program consisted of one hold at 95 °C for 10 min, followed by 40 cycles of 15 s at 95 °C and 1 min at 60 °C. After completion of these cycles, melting-curve data were then collected to verify PCR specificity and the absence of primer dimers, and to examine potential contamination. The transcript abundance in samples was determined using a comparative threshold cycle method. The relative abundance of the reference mRNA of TAF10 in each sample was determined and used to normalize for differences in total RNA amount according to the method described previously68. Unless specified, the mRNA level of individual ATG genes was normalized to the mRNA level of the corresponding gene in wild-type cells grown in rich conditions, which was set to 1.
RNA-seq profiling
cDNA library preparation and Illumina high-throughput sequencing were performed by the Advanced Genomics Core at the University of Michigan. Reads from multiplexed libraries were scanned for a perfect match to each unique 6-nucleotide barcode. If more than one barcode was matched perfectly, the 30-most barcode was chosen. Remaining reads were scanned for one mismatch to each 6-nucleotide barcode. If more than one barcode was matched with one mismatch, the 30-most barcode was chosen. The mean sequencing error rate for each library was estimated from the mean quality score. Reads were mapped using Bowtie269 (parameters: -f -v 3 -k 500 --best --strata) for the yeast transcriptome derived from yeast genome version S288C. Alignments with a quality score less than 88 were discarded. A majority of reads mapped to a unique locus in the transcriptome; these reads were used to calculate reads per kilobase per million mapped reads/RPKM for each yeast gene. The raw read counts without normalization of the two batches of RNA-seq experiments are included in Table S3. The accession code for the source data is: GSE290977.
SILAC sample preparation and LC-MS/MS analysis
Five OD cell lysates were harvested in SILAC medium after 8 h in 0.1% DMSO (R0K0) or 300 µM 3-IAA (R6K4).
Samples were sent to the University of Michigan Proteomics & Peptide Synthesis Core and subjected to a two-way SILAC analysis service, according to their standardized procedure as described below.
The beads were resuspended in 50 μl of 0.1 m ammonium bicarbonate buffer (pH ∼8). Cysteines were reduced by adding 50 μl of 10 mm DTT and incubating at 45 °C for 30 min. The samples were cooled to room temperature, and alkylation of cysteines was achieved by incubating with 65 mm 2-chloroacetamide, in the dark, for 30 min at room temperature. An overnight digestion with 1 μg of sequencing-grade, modified trypsin was carried out at 37 °C with constant shaking in a Thermomixer. Digestion was stopped by acidification, and peptides were desalted using SepPak C18 cartridges, following the manufacturer’s protocol (Waters). The samples were completely dried using a vacuum concentrator (Eppendorf). The resulting peptides were dissolved in 8 μl of 0.1% formic acid, 2% acetonitrile solution, and 2 μl of the peptide solution were resolved on a nano-capillary reverse phase column (Acclaim PepMap C18, 2 µm, 50 cm, Thermo Scientific), using a 0.1% formic acid, 2% acetonitrile (buffer A) and 0.1% formic acid, 95% acetonitrile (buffer B) gradient at 300 nl/min over a period of 180 min (2–22% buffer B in 110 min, 22–40% in 25 min, 40–90% in 5 min, followed by holding at 90% buffer B for 5 min and re-equilibration with buffer A for 25 min). Eluent was directly introduced into an Orbitrap Fusion tribrid mass spectrometer (Thermo Scientific), using an EasySpray source. MS1 scans were acquired at 120 K resolution (automatic gain control target = 1 × 106; max ionization time = 50 ms). Data-dependent collision-induced dissociation MS/MS spectra were acquired using the Top speed method (3 s), following each MS1 scan (normalized collision energy ∼32%; automatic gain control target 1 × 105; max ionization time 45 ms).
LC-MS/MS measurements were performed on a QExactive (QE) Plus and HF-X mass spectrometer coupled to an EasyLC 1000 and EasyLC 1200 nanoflow-HPLC, respectively (all Thermo Scientific). Peptides were fractionated on a fused silica HPLC-column tip (I.D. 75 μm, New Objective, selfpacked with ReproSil-Pur 120 C18-AQ, 1.9 μm (Dr. Maisch, r119.ag.0001) to a length of 20 cm) using a gradient of A (0.1% formic acid in water) and B (0.1% formic acid in 80% acetonitrile in water): samples were loaded with 0% B with a flow rate of 600 nL/min; peptides were separated by 5%–30% B within 85 min with a flow rate of 250 nL/min. Spray voltage was set to 2.3 kV and the ion-transfer tube temperature to 250 °C; no sheath or auxiliary gas were used. Mass spectrometers were operated in the data-dependent mode; after each MS scan (mass range m/z = 370–1750; resolution: 70,000 for QE Plus and 120,000 for HF-X) a maximum of ten, or twelve MS/MS scans were performed using a normalized collision energy of 25%, a target value of 1000 (QE Plus)/5000 (HF-X) and a resolution of 17,500 for QE Plus and 30,000 for HF-X. MS raw files were analyzed using MaxQuant (version 1.6.2.10)70 using a UniProt full-length S. cerevisiae database (March 2016) and common contaminants such as keratins and enzymes used for in-gel digestion as reference. Carbamidomethylcysteine was set as a fixed modification, and protein amino-terminal acetylation, serine, threonine and tyrosine phosphorylation, and oxidation of methionine were set as variable modifications. The MS/MS tolerance was set to 20 ppm, and three missed cleavages were allowed using trypsin/P as enzyme specificity. Peptide, site, and protein false discovery rate (FDR) based on a forward-reverse database were set to 0.01, the minimum peptide length was set to 7, the minimum score for modified peptides was 40, and the minimum number of peptides for identification of proteins was set to one, which must be unique. The “match-between-run†option was used with a time window of 0.7 min. MaxQuant results were analyzed using Perseus71. The raw data of SILAC experiments are included in Table S4.
Metabolomic profiling
Yeast samples were prepared for LC-MS analysis by solvent extraction. The extraction solvent was a 1:1:1:1 mixture of methanol:acetonitrile:acetone:water containing 4 µM D5-tryptophan to serve as an internal standard. To each yeast cell pellet, 300 µL of ice-cold extraction solvent with internal standard were added to each sample, followed by immediate probe sonication for 20 s at power level 2, 20% duty cycle using a Branson 450 probe sonicator. Following sonication, all sample tubes were incubated on ice for 10 min, then centrifuged for 10 min at 15,000 xg at 4 °C. An aliquot (225 µL) of the supernatant was transferred to autosampler vials containing a 450-µL insert. The supernatant was dried completely under a stream of nitrogen gas and was then reconstituted in 75 µL of an 8:2 mixture of methanol:water. LC-MS analysis was performed on an Aglent 1290 Infinity II LC system coupled to an Agilent 6530 Q-TOF mass spectrometer (Agilent Technologies, Santa Clara, CA) equipped with a JetStream ESI source. The injection volume was 5 µL. The UPLC was equipped with a 10-port valve configured to allow the column to be either eluted to the mass spectrometer or back-flushed to waste. The chromatographic separation was performed on an Agilent ZORBAX RRHD Extend 80 Å C18, 2.1 × 150 mm column (particle diameter 1.8 µm) equipped with an Agilent ZORBAX SB-C8, 2.1 mm × 30 mm guard column. The column temperature was held at 35 °C. Mobile phase A consisted of 97:3 water: methanol, and mobile phase B was 100% methanol; both A and B contained tributylamine and glacial acetic acid at concentrations of 10 mM and 15 mM, respectively. The column was back-flushed with mobile phase C (100% acetonitrile, no additives) between injections for column cleaning. The LC gradient was as follows: 0-2 min, 0% B; 2-12 min, linear ramp to 99% B; 12–17.5 min, 99% B. At 17.5 min, the 10-port valve was switched to reverse flow (back-flush) through the column, and the solvent composition changed to 99% C. From 20.5–21 min the flow rate was ramped to 0.8 mL/min, held until 22.5 min, then reduced to 0.6 mL/min. From 22.7–23.5 min the solvent was ramped from 99% to 0% C while flow was simultaneously ramped down from 0.6–0.4 mL/min and held until 29.4 min, at which point flow rate was returned to starting conditions at 0.25 mL/min. The 10-port valve was returned to restore forward flow through the column at 28.5 min. An isocratic pump was used to introduce the reference mass solution through the reference nebulizer for dynamic mass correction. The total run time was 30 min. The mass spectrometer was operated in negative ionization mode, using full-scan MS1 data acquisition from m/z 50 to 1200. The following source parameters were used: Gas temp: 250 °C; gas flow: 13 L/min; nebulizer: 35 psi; sheath gas temp: 325 °C; sheath gas flow: 12 L/min; capillary: 3500 V; nozzle voltage: 1500 V. Data were analyzed using Agilent Quantitative Analysis software, with peaks identified by matching accurate mass and RT for targeted compounds to those of an authentic standard. Data were analyzed for relative quantification using Agilent Quantitative Analysis software (v10.0), and compound levels were reported as peak area values.
Fluorescence microscopy
For imaging, yeast strains tagged with fluorescent markers were cultured in nutrient-rich medium (YPD) and were treated with 300 μM IAA for 8 h for conditional Vma2 degradation or shifted to nitrogen starvation (SD-N) medium for 2 h. Cells were collected at an optical density (OD) of 0.6–1.0 and were observed using a Leica DMi8 microscope with a 100× objective and Leica Thunder imager. Image processing was performed using Fiji ImageJ software and its plugins. Stacks of 5 image planes were collected with a spacing of 0.88 μm. The proportion of cells with GFP-tagged Atg8 protein forming puncta and RFP-tagged Atg8 protein entering the vacuole was determined by counting these cells among at least 200 yeast cells per condition.
Declaration of generative AI and AI-assisted technologies in the writing process
No AI technology was applied in this work.
Statistical analysis
The two-tailed Student’s t-test was used to determine statistical significance in all western blots for at least 3 independent biological replicates. Differences were considered significant at p < 0.05. Unless indicated otherwise, all summary data are presented as the mean and SD. For the microscopy quantification, ordinary one-way ANOVA was used.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data availability
All raw sequencing data generated in this study are available at the GEO with the Accession number GSE290977. Data supporting the findings of this study are also provided in the Supplementary data. Source data are provided with this paper.
References
Ballabio, A. & Bonifacino, J. S. Lysosomes as dynamic regulators of cell and organismal homeostasis. Nat. Rev. Mol. Cell Biol. 21, 101–118 (2020).
Zoncu, R. et al. mTORC1 senses lysosomal amino acids through an inside-out mechanism that requires the vacuolar H(+)-ATPase. Science 334, 678–683 (2011).
Mindell, J. A. Lysosomal acidification mechanisms. Annu Rev. Physiol. 74, 69–86 (2012).
Kane, P. M. The where, when, and how of organelle acidification by the yeast vacuolar H+-ATPase. Microbiol Mol. Biol. Rev. 70, 177–191 (2006).
Forgac, M. Vacuolar ATPases: rotary proton pumps in physiology and pathophysiology. Nat. Rev. Mol. Cell Biol. 8, 917–929 (2007).
Chen, F., Kang, R., Liu, J. & Tang, D. The V-ATPases in cancer and cell death. Cancer Gene Ther. 29, 1529–1541 (2022).
Colacurcio, D. J. & Nixon, R. A. Disorders of lysosomal acidification-The emerging role of v-ATPase in aging and neurodegenerative disease. Ageing Res Rev. 32, 75–88 (2016).
Hughes, A. L. & Gottschling, D. E. An early age increase in vacuolar pH limits mitochondrial function and lifespan in yeast. Nature 492, 261–265 (2012).
Molin, M. & Demir, A. B. Linking peroxiredoxin and vacuolar-atpase functions in calorie restriction-mediated life span extension. Int J. Cell Biol. 2014, 913071 (2014).
Green, M. R. et al. Mutations in early follicular lymphoma progenitors are associated with suppressed antigen presentation. Proc. Natl Acad. Sci. USA 112, E1116–E1125 (2015).
Krysiak, K. et al. Recurrent somatic mutations affecting B-cell receptor signaling pathway genes in follicular lymphoma. Blood 129, 473–483 (2017).
Okosun, J. et al. Recurrent mTORC1-activating RRAGC mutations in follicular lymphoma. Nat. Genet 48, 183–188 (2016).
Wang, F. et al. Follicular lymphoma-associated mutations in the V-ATPase chaperone VMA21 activate autophagy, creating a targetable dependency. Autophagy 18, 1982–2000 (2022).
Yang, Y., Zhang, Z. & Klionsky, D. J. Follicular lymphoma-associated mutations in the V-ATPase chaperone Vma21 activate autophagy by dysfunctional V-ATPase assembly. Autophagy Rep. 1, 226–233 (2022).
Feng, Y., He, D., Yao, Z. & Klionsky, D. J. The machinery of macroautophagy. Cell Res. 24, 24–41 (2014).
Nakamura, N., Matsuura, A., Wada, Y. & Ohsumi, Y. Acidification of vacuoles is required for autophagic degradation in the yeast, Saccharomyces cerevisiae. J. Biochem 121, 338–344 (1997).
Coffey, J. W. & De Duve, C. Digestive activity of lysosomes. I. The digestion of proteins by extracts of rat liver lysosomes. J. Biol. Chem. 243, 3255–3263 (1968).
Wang, F. et al. Follicular lymphoma-associated mutations in vacuolar ATPase ATP6V1B2 activate autophagic flux and mTOR. J. Clin. Invest 129, 1626–1640 (2019).
Lei, Y. et al. Big1 is a newly identified autophagy regulator that is critical for a fully functional V-ATPase. Mol. Biol. Cell 35, br20 (2024).
Deleyto-Seldas, N. & Efeyan, A. The mTOR-autophagy axis and the control of metabolism. Front Cell Dev. Biol. 9, 655731 (2021).
Santana-Codina, N., Mancias, J. D. & Kimmelman, A. C. The role of autophagy in cancer. Annu Rev. Cancer Biol. 1, 19–39 (2017).
Russell, R. C. & Guan, K. L. The multifaceted role of autophagy in cancer. Embo j. 41, e110031 (2022).
Mauvezin, C., Nagy, P., Juhász, G. & Neufeld, T. P. Autophagosome-lysosome fusion is independent of V-ATPase-mediated acidification. Nat. Commun. 6, 7007 (2015).
Morawska, M. & Ulrich, H. D. An expanded tool kit for the auxin-inducible degron system in budding yeast. Yeast 30, 341–351 (2013).
Shintani, T. & Klionsky, D. J. Cargo proteins facilitate the formation of transport vesicles in the cytoplasm to vacuole targeting pathway. J. Biol. Chem. 279, 29889–29894 (2004).
Xie, Z., Nair, U. & Klionsky, D. J. Atg8 controls phagophore expansion during autophagosome formation. Mol. Biol. Cell 19, 3290–3298 (2008).
Sankaranarayanan, S., De Angelis, D., Rothman, J. E. & Ryan, T. A. The use of pHluorins for optical measurements of presynaptic activity. Biophys. J. 79, 2199–2208 (2000).
Dröse, S. et al. Inhibitory effect of modified bafilomycins and concanamycins on P- and V-type adenosinetriphosphatases. Biochemistry 32, 3902–3906 (1993).
Stephan, J. S., Yeh, Y. Y., Ramachandran, V., Deminoff, S. J. & Herman, P. K. The Tor and PKA signaling pathways independently target the Atg1/Atg13 protein kinase complex to control autophagy. Proc. Natl Acad. Sci. USA 106, 17049–17054 (2009).
Schmidt, A., Beck, T., Koller, A., Kunz, J. & Hall, M. N. The TOR nutrient signalling pathway phosphorylates NPR1 and inhibits turnover of the tryptophan permease. Embo j. 17, 6924–6931 (1998).
Noda, T. & Klionsky, D. J. The quantitative Pho8Delta60 assay of nonspecific autophagy. Methods Enzymol. 451, 33–42 (2008).
Zientara-Rytter, K. & Subramani, S. Mechanistic insights into the role of Atg11 in selective autophagy. J. Mol. Biol. 432, 104–122 (2020).
Hughes, C. E. et al. Cysteine toxicity drives age-related mitochondrial decline by altering iron homeostasis. Cell 180, 296–310.e218 (2020).
Kraft, C., Deplazes, A., Sohrmann, M. & Peter, M. Mature ribosomes are selectively degraded upon starvation by an autophagy pathway requiring the Ubp3p/Bre5p ubiquitin protease. Nat. Cell Biol. 10, 602–610 (2008).
Ossareh-Nazari, B. et al. Cdc48 and Ufd3, new partners of the ubiquitin protease Ubp3, are required for ribophagy. EMBO Rep. 11, 548–554 (2010).
Bernard, A., Jin, M., Xu, Z. & Klionsky, D. J. A large-scale analysis of autophagy-related gene expression identifies new regulators of autophagy. Autophagy 11, 2114–2122 (2015).
Feng, Y., Yao, Z. & Klionsky, D. J. How to control self-digestion: transcriptional, post-transcriptional, and post-translational regulation of autophagy. Trends Cell Biol. 25, 354–363 (2015).
Lei, Y. & Klionsky, D. J. Transcriptional regulation of autophagy and its implications in human disease. Cell Death Differ. 30, 1416–1429 (2023).
Hinnebusch, A. G. Translational regulation of GCN4 and the general amino acid control of yeast. Annu Rev. Microbiol 59, 407–450 (2005).
Hanes, J., Jermutus, L., Weber-Bornhauser, S., Bosshard, H. R. & Plückthun, A. Ribosome display efficiently selects and evolves high-affinity antibodies in vitro from immune libraries. Proc. Natl Acad. Sci. USA 95, 14130–14135 (1998).
Tallóczy, Z. et al. Regulation of starvation- and virus-induced autophagy by the eIF2alpha kinase signaling pathway. Proc. Natl Acad. Sci. USA 99, 190–195 (2002).
Vlahakis, A., Graef, M., Nunnari, J. & Powers, T. TOR complex 2-Ypk1 signaling is an essential positive regulator of the general amino acid control response and autophagy. Proc. Natl Acad. Sci. USA 111, 10586–10591 (2014).
Dong, J., Qiu, H., Garcia-Barrio, M., Anderson, J. & Hinnebusch, A. G. Uncharged tRNA activates GCN2 by displacing the protein kinase moiety from a bipartite tRNA-binding domain. Mol. Cell 6, 269–279 (2000).
Jones, G. M. et al. A systematic library for comprehensive overexpression screens in Saccharomyces cerevisiae. Nat. Methods 5, 239–241 (2008).
Braus, G. H. Aromatic amino acid biosynthesis in the yeast Saccharomyces cerevisiae: a model system for the regulation of a eukaryotic biosynthetic pathway. Microbiol Rev. 55, 349–370 (1991).
Xue, C. et al. Tryptophan metabolism in health and disease. Cell Metab. 35, 1304–1326 (2023).
Platten, M., Nollen, E. A. A., Röhrig, U. F., Fallarino, F. & Opitz, C. A. Tryptophan metabolism as a common therapeutic target in cancer, neurodegeneration and beyond. Nat. Rev. Drug Discov. 18, 379–401 (2019).
Barik, S. The uniqueness of tryptophan in biology: properties, metabolism, interactions and localization in proteins. Int. J. Mol. Sci. 21, https://doi.org/10.3390/ijms21228776 (2020).
Robinson, J. S., Klionsky, D. J., Banta, L. M. & Emr, S. D. Protein sorting in Saccharomyces cerevisiae: isolation of mutants defective in the delivery and processing of multiple vacuolar hydrolases. Mol. Cell Biol. 8, 4936–4948 (1988).
Croft, T., Venkatakrishnan, P. & Lin, S. J. NAD(+) Metabolism and regulation: lessons from yeast. Biomolecules 10, https://doi.org/10.3390/biom10020330 (2020).
MartÃn-Marcos, P., Hinnebusch, A. G. & Tamame, M. Ribosomal protein L33 is required for ribosome biogenesis, subunit joining, and repression of GCN4 translation. Mol. Cell Biol. 27, 5968–5985 (2007).
Lee, J., Moir, R. D., McIntosh, K. B. & Willis, I. M. TOR signaling regulates ribosome and tRNA synthesis via LAMMER/Clk and GSK-3 family kinases. Mol. Cell 45, 836–843 (2012).
Wang, X. et al. Oncogenic RAS induces a distinctive form of non-canonical autophagy mediated by the P38-ULK1-PI4KB axis. Cell Res. 35, 399–422 (2025).
Giusti, R. M. et al. Differential patterns of serum biomarkers of immune activation in human T-cell lymphotropic virus type I-associated myelopathy/tropical spastic paraparesis, and adult T-cell leukemia/lymphoma. Cancer Epidemiol. Biomark. Prev. 5, 699–704 (1996).
Huang, A. et al. Serum tryptophan decrease correlates with immune activation and impaired quality of life in colorectal cancer. Br. J. Cancer 86, 1691–1696 (2002).
Schroecksnadel, K., Winkler, C., Fuith, L. C. & Fuchs, D. Tryptophan degradation in patients with gynecological cancer correlates with immune activation. Cancer Lett. 223, 323–329 (2005).
Weinlich, G., Murr, C., Richardsen, L., Winkler, C. & Fuchs, D. Decreased serum tryptophan concentration predicts poor prognosis in malignant melanoma patients. Dermatology 214, 8–14 (2007).
Suzuki, Y. et al. Increased serum kynurenine/tryptophan ratio correlates with disease progression in lung cancer. Lung Cancer 67, 361–365 (2010).
Zhai, L. et al. The kynurenine to tryptophan ratio as a prognostic tool for glioblastoma patients enrolling in immunotherapy. J. Clin. Neurosci. 22, 1964–1968 (2015).
Klionsky, D. J. et al. Guidelines for the use and interpretation of assays for monitoring autophagy (4th edition). Autophagy 17, 1-382, (2021).
Okreglak, V. et al. Cell cycle-linked vacuolar pH dynamics regulate amino acid homeostasis and cell growth. Nat. Metab. 5, 1803–1819 (2023).
Li, T. Y. et al. A lysosomal surveillance response to stress extends healthspan. Nat. Cell. Biol. https://doi.org/10.1038/s41556-025-01693-y (2025).
Gueldener, U., Heinisch, J., Koehler, G. J., Voss, D. & Hegemann, J. H. A second set of loxP marker cassettes for Cre-mediated multiple gene knockouts in budding yeast. Nucleic Acids Res. 30, e23 (2002).
Gardner, J. M. & Jaspersen, S. L. Manipulating the yeast genome: deletion, mutation, and tagging by PCR. Methods Mol. Biol. 1205, 45–78 (2014).
Cheong, H. & Klionsky, D. J. Biochemical methods to monitor autophagy-related processes in yeast. Methods Enzymol. 451, 1–26 (2008).
Welter, E., Thumm, M. & Krick, R. Quantification of nonselective bulk autophagy in S. cerevisiae using Pgk1-GFP. Autophagy 6, 794–797 (2010).
Huang, W. P., Scott, S. V., Kim, J. & Klionsky, D. J. The itinerary of a vesicle component, Aut7p/Cvt5p, terminates in the yeast vacuole via the autophagy/Cvt pathways. J. Biol. Chem. 275, 5845–5851 (2000).
Vandesompele, J. et al. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol. 3, Research0034 (2002).
Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat. Methods 9, 357–359 (2012).
Cox, J. & Mann, M. MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification. Nat. Biotechnol. 26, 1367–1372 (2008).
Tyanova, S., Temu, T. & Cox, J. The MaxQuant computational platform for mass spectrometry-based shotgun proteomics. Nat. Protoc. 11, 2301–2319 (2016).
Acknowledgments
We thank the members of the Klionsky lab for their helpful suggestions and comments. We thank Drs. Longhua Guo, Feng Shao, Haoxing Xu, Ming Li, Yue Xu, and Xiao-Wei Chen for constructive suggestions on the manuscript writing. We thank Dr. Laura Buttitta’s lab for sharing the use of the Attune NXT 4 Laser Flow Cytometer (Attune NXT4). Metabolomics measurements were performed by the University of Michigan Metabolomics Core (University of Michigan Medical School, Biomedical Research Core Facilities, Ann Arbor, MI, USA). This work was supported by the National Institute of General Medical Sciences [GM131919].
Author information
Authors and Affiliations
Contributions
D.J.K. supervised the project. Y.H. conceived the original concept, designed and performed most of the experiments in the study. D.D. performed the fluorescence microscopy experiments; Z.Z. and Y.L. helped to construct part of the validation plasmids for the second round of the suppressor screen. C.E. optimized and performed the metabolomic profiling.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Source data
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Huang, Y., Dialynaki, D., Lei, Y. et al. V-ATPase-dependent induction of selective autophagy. Nat Commun 16, 8508 (2025). https://doi.org/10.1038/s41467-025-63472-5
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-025-63472-5
