Characterization of WAC interactions with R2TP and TTT chaperone complexes linking glucose and glutamine availability to mTORC1 activity

Abstract

TELO2-TTI1-TTI2 (TTT) and R2TP are multi-subunit chaperones that cooperate with HSP90 to assemble matured complexes of the PIKK family of kinases, including mTOR complex 1 (mTORC1). WAC, a protein previously implicated in transcription, H2B ubiquitination, and autophagy, was recently identified as a regulator of mTORC1 in response to glucose and glutamine availability, acting in concert with R2TP and TTT. However, the molecular basis of the interactions of WAC with R2TP and TTT and their role in mTORC1 regulation remains poorly defined. Here, we characterized the interactions of WAC with mTOR, R2TP, and TTT and how these are affected by nutrient conditions. Using purified proteins, we establish that WAC directly binds to mTOR-mLST8, R2TP, and TELO2, but not TTI1 and TTI2. In cells, WAC is part of complexes containing components of mTORC1, R2TP, and TTT, and these associations are modulated by nutrient availability. Notably, WAC and TELO2 strongly associate with mTOR under glucose and glutamine deprivation, and these interactions are weakened minutes after nutrient refeeding. These dynamics correlate with changes in mTORC1 activity. Transcriptomic and proteomic analysis shows that WAC, mTOR, R2TP, and TTT are co-expressed across several human cancers, supporting that WAC is part of a functional pathway with mTOR, R2TP, and TTT. Together, our findings reveal the formation and disassembly of a WAC complex with mTOR and TELO2 that contributes to regulate mTORC1 in response to glucose and glutamine availability.

Keywords: R2TP; RUVBL1‐RUVBL2; TELO2; WAC; mTORC1.

Fig. 1

WAC directly interacts with mTOR and TELO2. (A) Pull‐down experiment testing the interaction between WAC‐3xFlag and purified mTOR‐mLST8. Western blot of the co‐eluted proteins was only performed against the mTOR subunit. (B) WAC interacts directly with mTOR‐mLST8. Immunoblot analysis of the interaction of WAC with mTOR and mLST8 after elution of WAC from an affinity chromatography from lysates of HEK293 cells co‐expressing WAC‐3xFlag and mTOR‐mLST8. Two consecutive elutions are shown for the WAC + mTOR‐mLST8 sample. (C) Same experiment as in (A) but testing the interaction of WAC with TELO2. (D) Interaction of WAC and TTI‐TTI2 as in (C). Inputs and elutions were analyzed by SDS/PAGE and western blot. Inputs show the soluble fraction of Expi293 cell lysates expressing either 3xFlag as a control (labeled as Expi293 lysate) or WAC‐3xFlag (labeled as WAC lysate).

Fig. 2

WAC interacts with the R2TP complex. (A) Pull‐down experiment testing the interaction between WAC‐3xFlag and the R2TP complex. WAC‐3xFlag, used as bait, was incubated with a purified, already assembled R2TP complex. (B) To map the interaction of WAC with the subunits of the R2TP complex, a similar experiment as in (A) was performed to test the interaction with the RUVBL1‐RUVBL2 complex. Similar experiments are shown but using RPAP3 (C) and RPAP3‐PIH1D1 (D) as prey. Assays were performed as explained in the Methods section, similarly to the experiments shown in Fig. 3C–E.

Fig. 3

Analysis of gene and protein levels for WAC, mTORC1, R2TP, and TTT across cancer types. Gene (A) and protein (B) levels of the subunits of the mTOR‐WAC‐RUVBL1‐RUVBL2‐TTT complex in 10 cancer types with CPTAC tumor samples from LinkedOmics. Z‐scores are shown to represent normalized expression levels, with red indicating relatively high expression and blue indicating relatively low expression. The number of cancer types showing a positive correlation in gene (C) and protein (D) levels between subunits of the mTOR‐WAC‐RUVBL1‐RUVBL2‐TTT complex across 10 cancer types. The number within each box indicates the number of cancer types with a positive correlation out of the 10 tested. A positive correlation is defined as a Pearson correlation coefficient > 0.2 with a P‐value < 0.01. (E) Distribution of WAC protein expression in tumor samples versus matched normal samples across cancer types (where normal samples are available). Statistically significant differences between tumor and normal conditions were assessed using the two‐sided Wilcoxon rank sum test. Units in the Y axis correspond to protein expression levels in LinkedOmics, provided as log₂‐transformed MS1 intensities, which represent relative protein abundances measured via mass spectrometry and normalized for comparative analyses. (F) Differences in protein levels of the components of the WAC‐RUVBL1‐RUVBL2‐TTT complex between tumor and adjacent normal tissue (NAT) across cancer types. Red indicates higher expression in tumors compared to normal samples, while blue indicates higher expression in normal samples compared to tumors. Statistically significant differences were assessed using the two‐sided Wilcoxon rank sum test. The meta P‐value represents pan‐cancer significance when all samples are analyzed together using the two‐sided Wilcoxon rank sum test. *P < 0.05, **P < 0.01, ***P < 0.001. BRCA, breast invasive carcinoma; CCRCC, clear cell renal cell carcinoma; COAD, colon adenocarcinoma; GBM, glioblastoma; HNSCC, head and neck squamous cell carcinoma; LSCC, lung squamous cell carcinoma; LUAD, lung adenocarcinoma; OV, ovarian serous cystadenocarcinoma; PDAC, pancreatic ductal adenocarcinoma; UCEC, uterine corpus endometrial carcinoma.

Fig. 4

WAC preferentially interacts with mTOR and the TTT complex in the absence of glucose and glutamine. (A) Schematic illustration of nutrient stress protocol used. Color codes for each condition are used for the rest of the panels in the fig. (B) Immunoblot analysis of endogenous mTOR immunoprecipitation (IP) assay performed in HEK293T under Glc/Gln modulation. Representative blot of 5 independent experiments is shown. Each lane is labeled with a color code corresponding to the conditions indicated in (A). A control experiment, indicated as a B within a green circle, was performed using only beads without antibody and using complete media. (C) Immunoprecipitation data (mean ± SEM) from (B) were normalized by total mTOR immunoprecipitated and expressed as fold change of association with respect to the basal condition. Statistical significance was analyzed using ordinary one‐way ANOVA and Tukey's multiple comparisons test. P‐values, *P < 0.05, **P < 0.01, ***P < 0.001. (D) Immunoblot analysis of phosphorylation status of S6 is shown as a control of mTORC1 modulation in (B) and (C). (E) Immunoblot analysis of endogenous mTOR performed as in (B), treating cells with rapamycin for 30 min prior to Glc/Gln recovery. Representative blot of three independent experiments is shown. Color used for conditions using rapamycin is shown at the bottom and used in other panels. (F) Immunoprecipitation data (mean ± SEM) from (E) were normalized by total mTOR immunoprecipitated and expressed as fold change of association with respect to the starving condition. Statistical significance was analyzed using ordinary one‐way ANOVA and Tukey's multiple comparisons test. P‐values, *P < 0.05. (G) Immunoblot analysis of phosphorylation status of S6 is shown as a control of mTORC1 modulation in (E) and (F).

Fig. 5

WAC does not localize at lysosomes. (A) WAC KO‐TMEM192 HA Lyso cells transiently overexpressing 3xFlag‐tagged WAC were stimulated with Glc/Gln for 30 min after starvation. Representative confocal micrographs (maximum intensity projections), before (top panels) and after (bottom panels) addition of Glc/Gln, showing signals of Flag (green), TMEM192 (red) and DAPI (blue). Scale bars, 10 μm. (B) Immunoblots of Lyso IP samples from control and TMEM192 HA Lyso WT HEK293T cells cultured under fed conditions. HA‐tagged TMEM192 was immunoprecipitated with HA beads, and IP and P post nuclear supernatant (PNS) were probed for the indicated proteins. Representative blot of two independent experiments. (C) Quantification of the experiment shown in (B), measuring the fold change (FC) in protein signal over the control. Statistical significance was analyzed using ordinary one‐way ANOVA and Tukey's multiple comparisons test. P‐values, **P < 0.005.