Renal cell carcinoma (RCC) mainly originates from renal proximal tubules. Intriguingly, disruption of genes frequently mutated in human RCC samples thus far has only generated RCC originated from other renal tubule parts in mouse models. This hampers our understanding of the pathogenesis of RCC. Here we show that mTOR signaling, often activated in RCC samples, initiates RCC development from renal proximal tubules. Ablation of Tsc1, encoding an mTOR suppressor, in proximal tubule cells led to multiple precancerous renal cysts. mTOR activation increased MEK1 expression and ERK activation, and Mek1 ablation or inhibition diminished cyst formation in Tsc1-deficient mice. mTOR activation also increased MKK6 expression and p38MAPK activation, and ablation of the p38α-encoding gene further enhanced cyst formation and led to RCC with clear cell RCC features. Mechanistically, Tsc1 deletion induced p53 and p16 expression in a p38MAPK-dependent manner, and deleting Tsc1 and Trp53 or Cdkn2a (encoding p16) enhanced renal cell carcinogenesis. Thus, mTOR activation in combination with inactivation of the p38MAPK–p53/p16 pathway drives RCC development from renal proximal tubules. Moreover, this study uncovers previously unidentified mechanisms by which mTOR controls cell proliferation and suggests the MEK–ERK axis to be a potential target for treatment of RCC.

Significance:

Mouse modeling studies show that mTOR activation in combination with inactivation of the p38MAPK–p53/p16 axis initiates renal cell carcinoma that mimics human disease, identifying potential therapeutic targets for RCC treatment.

Renal cell carcinomas (RCC) account for 2% to 3% of malignancies and are a group of heterogeneous tumors classified into clear cell, papillary, and chromophobe subtypes, with clear cell RCC being the major subtype (1, 2). Clear cell and papillary RCCs are mainly derived from renal proximal tubules, sites for resorption of ∼80% of the filtrate nutrients and electrolytes (3). RCCs are usually refractory to radiotherapy and chemotherapy but can be treated with targeted therapies including inhibitors for VEGF, mTOR, or immune checkpoints or combination of these drugs (4, 5). However, only a portion of patients are responsive to these treatments and many patients develop drug resistance (1, 6). Improved therapeutic regimens are thus needed, this requires thorough understanding of the pathogenesis of RCCs (1).

Molecular and genomic analyses have identified somatic mutations in VHL/HIF/VEGF, PI3K/Akt/mTOR, and other mitogenic pathways, although mutation in tumor suppressor gene Trp53 is uncommon compared with other cancer types (7–10). Mutations are most common in chromosome 3p that contains Vhl, BAP1, and PBRM1 (5, 10), which are believed to activate mTOR signaling and HIF1α-mediated expression of VEGF and PDGF to induce RCC development (11, 12). Intriguingly, deletion of both Vhl and Bap1 or Pbrm1 in proximal tubular cells (with Villin-Cre or Sglt2-Cre mice) fails to induce RCC (9), although deletion of Vhl and Bap1 or Pbrm1 or deletion of Vhl, Rb, and Trp53, in Bowman capsule, distal tubules, and collecting tubule (with Pax8-Cre or Ksp-Cre mice) induces RCCs (9, 13, 14). In addition, deletion of Vhl, Trp53, and Kif3 or deletion of Vhl and Cdkn2a combined with Myc overexpression leads to cyst formation and RCC in distal and collecting tubules (15–17). Thus, there is a lack of RCC mouse models with proximal tubule origins (9, 18).

mTOR signaling is often hyperactivated in RCC samples and mTOR inhibitor Sirolimus has been approved for RCC treatment (5). mTOR is the sensor of nutrients and growth factors (19, 20). It forms two complexes with mTORC1 promoting cell proliferation and mTORC2 regulating cell survival and cytoskeleton (21). It is believed that mTORC1 signaling promotes cell proliferation mainly by increasing global protein synthesis (6, 22, 23), lipid synthesis (24), and nucleotide synthesis (25). A number of studies have shown that 25% to 76% RCC samples show hyperactive mTORC1 signaling due to mutations in genes including TSC1, TSC2, PI3KCA, and PTEN (10, 26). Patients with TSC (loss of TSC1 or TSC2) often develop angiomyolipoma and renal cysts and 2% to 4% of the patients also develop RCCs (27). Whole body deletion of one allele of Tsc1 or Tsc2 leads to development of cyst, cystadenoma, and carcinoma in mice (8, 28–30). Although whole body deletion of Tsc1, Fox1, Fox3, and Fox4 causes renal tumors in 44% to 72% of mice, the origin and identity of these tumors are unclear (31). In general, the role of mTOR signaling in RCC pathogenesis is believed to promote proliferation via increasing synthesis of proteins and other molecules.

Here, we delete Tsc1 in Villin+ proximal tubular cells and find that mTOR activation induces renal cyst formation by enhancing MEK1–ERK activation and Myc and cyclins expression. Mek1 deletion or inhibition diminishes Tsc1 deletion-induced cyst formation. The lack of RCC development in these mice is likely due to mTOR-induced activation of the MKK6–p38MAPK–p53/p16 pathway as inactivation of Mapk14 (encoding p38α), Trp53, or Cdkn2a (encoding p16) enhances renal cyst formation and causes RCC in Tsc1−/− proximal tubular cells, which show some features of clear cell RCC. The MKK6–p38MAPK–p53/p16 pathway appears to act as a safeguard mechanism to contain mTOR-driven cell overproliferation. Thus, cooperation of the mTOR and p53/p16 pathways, which are often activated and silenced (through mechanisms other than mutations) respectively, in RCC samples (10), in the presence of mutations in Vhl, Bap1, or Pbrm1, initiates RCC development from proximal tubules. This study expands our understanding of RCC pathogenesis, identifies MEK–ERK as a potential target for RCC treatment, and uncovers novel mechanisms whereby mTOR controls cell proliferation by balancing two MAPKs.

Mouse strains and maintenance

Tsc1f/f, Villin-Cre (Tg(Vil1-cre)20Syr/J), and Trp53−/− mouse lines were purchased from The Jackson Laboratories. Mek1f/f mouse was generated in Charron's lab. Mapk14f/f mouse was generated in Yibin Wang's lab. The Cdk2n1−/− mouse line was generated in Ron Depinho's lab. These mice were maintained in SPF mouse facility. Animal experiments were carried out in accordance with recommendations in the National Research Council Guide for Care and Use of Laboratory Animals, with the protocols approved by the Institutional Animal Care and Use Committee.

TSC patient samples

Control and TSC patient kidney and intestine tissues were obtained from the Brain and Tissue Bank for Development Disorders (University of Maryland). The tissue bank has obtained the written informed consent from all patients. The study has been approved by the Institutional Review Board (HSC-E-034–061) of the University of Texas Health Science Center, and experiments were conducted in accordance with recognized ethical guidelines (U.S. Common Rule).

Mouse drug administration

To inhibit MEK1, mice were injected with 25 mg/kg U0126 (Selleck, s1102) daily for 1 month in the study. To inhibit mTOR, mice were injected with Rapamycin (3 mg/kg; Selleck, s1039) daily for the period indicated in the study. Solvent was used as a control in these studies.

Histology and IHC

The kidneys were harvested immediately after sacrifice, fixed with 4% paraformaldehyde, and prepared for paraffin embedding. The samples were sectioned at 4 μm. Hematoxylin and eosin staining was performed following standard protocols. For immunostaining, antigen retrieval was carried out by boiling the sections in citrate buffer, pH 6, for 10 minutes, followed by cooling for 60 minutes at room temperature. To eliminate endogenous peroxidases, tissues were treated in methanol containing 3% H2O2 for 30 minutes. Tissues were permeabilized with 0.1% Triton X-100 for 30 minutes. The sections were blocked in 10% goat serum for 30 minutes followed by primary antibodies incubation at 4°C overnight. Tissues were washed by PBS and subsequently incubated with secondary antibody at 37°C for 60 minutes. Antibodies against Ki67 (ab15580) were purchased from Abcam. Antibodies against p-S6 (2211), p-p38 (9211), TSC1 (4906s), TSC2 (3612), and pERKs (9106) were purchased from Cell Signaling Technology, PAX8 (sc-81353), PKD (sc-130554), SDHB (sc-271548), AQP1 (25287), and CAIX (sc-365900) antibodies were purchased from Santa Cruz Biotechnology, whereas vimentin (ab92547) antibodies were purchased from Abcam, DBA(RL-1032) was purchased from VectorLabs.

Urine analysis

For kidney function assays, we collected mouse serum and determined the levels of urea, creatine, and uric acid using detection kits purchased from FOSUNPHARMA (1.02.0303, 1.02.4102, and 1.02.2006, respectively), following the protocols provided by the manufacturer.

Isolation and culture of primary MEFs

Primary MEFs were derived from Tsc1f/f mice following standard protocols. To delete Tsc1 in these MEFs, cells were seeded in 10-cm dishes and cultured overnight. Five milliliters of medium containing control retrovirus or Cre-expressing retrovirus was used to infect the MEFs with polybrene (8 mg/mL). Deletion of Tsc1 was verified by Western blot analysis. Cells were cultured overnight and directly used for further experiments.

Tsc1 knockdown in kidney cell lines

HEK293 and HK2 were purchased from ATCC in March 2016 and October 2017, respectively, and were frozen at early passages. Experiments were conducted within five passages after thawing of the initial frozen cells. Cell authentication was done using short tandem repeat DNA profiling and Mycoplasma was determined using Mycoplasma Detection Kit (done by Mingzhoubio Inc.) before experiments. HEK293 or HK2 cells were cultured in DMEM supplemented with 10% FCS. For knockdown experiments, Tsc1 (ON-TARGETplus SMART pool) siRNA were used.

S35-methionine incorporation assay

HEK293 cells were pre-incubated with DMEM without methionine for 30 minutes and then labeled with S35-methionine for 1 hour. Cell extracts were divided into three portions, which were subjected to immunoprecipitation with antibodies against MKK6 (9264), MEK1 (9122; Cell Signaling Technology), or GAPDH (sc32233; Santa Cruz Biotechnology). GAPDH was used for both MEK1 and MKK6. The immunoprecipitates were eluted and applied to SDS-PAGE, and the gel was dried and visualized with phospho-imaging.

Western blot analysis

Tissues were lysed in T-PER Tissue protein extraction reagent (Thermo Fisher Scientific) and cells were lysed in TNEN buffer. The same amounts of samples were prepared and resolved on multiple SDS-PAGE gels, which were transferred to polyvinylidenedifluoride membranes (Millipore). Antibodies against MKK6 (9264), MEK1 (9122), p-p38 (9211), p38 (9212), pERKs (9106), ERKs (9102), p53 (2524), p-S6 (1211) were purchased from Cell Signaling Technology. Antibodies against β-actin (sc81178), and p16 (sc1207) were purchased from Santa Cruz Biotechnology. All Western blot-related experiments were repeated three times and a representative result was presented. For quantitation, the protein bands were quantitated using the software provided by FluorChem M system (ProteinSimple). To quantify Western blot results, the signal of a given protein was normalized to that of actin. This ratio in control samples was set at 1.0 and the ratio of mutant samples was normalized to the control.

qPCR

Total RNA was extracted from cells or tissues with Trizol reagent (Invitrogen). Complementary DNA was synthesized using Transcriptor First Strand cDNA Synthesis Kit (Roche). The detection and quantification of target mRNA were performed with qPCR, which were normalized to the levels of β-actin. Dissociation curves were run to validate the specificity of the primers. qPCR was carried out using the Applied Biosystems 7500 system. The primers are listed in Supplementary Table S1.

RNA sequencing and data analysis

RNA sequencing (RNA-seq) was done at BGI Group using three Tsc1 deficient and control kidney samples. For data analysis, SOAPnuke (v1.5.2) was used to filter reads and generate FASTQ format. Clean reads were then aligned to the reference mouse genome using the Bowtie2 (v2.2.5) with default parameters and gene expression level was calculated with RSEM (v1.2.12). DEGs were detected with DEGseq as requested and statistically significant (adjusted P value ≤0.001) genes with large expression changes (fold change ≥2) were defined as differentially expressed genes. In GO analysis and pathway analysis of DEG, DEGs were classified according to official classification with the GO or KEGG annotation result and phyper (a function of R) was performed in GO and pathway functional enrichment. The P value calculating formula was displayed on wiki (https://en.wikipedia.org/wiki/Hypergeometric_distribution) and FD) was calculated for each P value with FDR ≤ 0.01 was defined as significant enriched.

Statistical analysis

Data are given as mean ± SEM when each value (mouse) was an average of several measurements whereas data are given as mean ± SD when each value represents one measurement or replicates. Differences between two groups were measured by the Student t test. A P value less than 0.05 is defined as statistically significant difference.

Data availability

The RNA-seq data of three kidney samples of Villin-Cre; Tsc1f/f and age-matched normal mice have been deposited in the SRA database under accession code PRJNA PRJNA592055 with submission ID: SUB6617625.

Deletion of Tsc1 in proximal tubules leads to renal cysts

To further understand the pathogenesis of RCC, we ablated Tsc1 using the Villin-Cre mouse line, which labels renal proximal tubular cells and intestinal villus epithelial cells (32, 33). We found that in normal adult mice, mTOR was activated mainly in renal tubular structures as manifested by immunostaining of p-S6 and p-4EBP1, indicators of mTORC1 activation and drivers of protein synthesis (Supplementary Fig. S1A), consistent with the findings that mTOR plays critical roles in renal homeostasis and diseases (27, 34, 35). Tsc1 deletion resulted in a drastic increase in p-S6 and p-4E-BP1 signals in kidney samples (Fig. 1A). Immunostaining revealed that enhanced mTOR activation mainly occurred in renal tubular structures positive for proximal tubular marker AQP1, but not for distal renal tubular marker DBA (Supplementary Fig. S1B). These mice had an average lifespan of 7 to 8 months (Supplementary Fig. S1C), which could be caused by defects in renal tubular cells or other cells marked by Villin. The mutant mice showed progressive kidney enlargement and cyst formation (Fig. 1B), which were quantitated using the method described by Boletta and colleagues (30). The mutant mice also showed tubular cell overproliferation, yet, they did not develop obvious renal tumors (Fig. 1C), in contrast to Tsc1+/− or Tsc2+/− mice, which developed cystadenoma and carcinomas at advanced age besides cysts (28, 29). Moreover, we found that levels of urea and creatine in the urine were increased whereas the uric acid level was reduced (Fig. 1D), suggesting that kidney function was impaired. These phenotypes were corrected by RAP treatment (Supplementary Figs. S1D and S1E), which mainly targets mTORC1, suggesting that mTORC1 plays a pro-growth role in renal proximal tubular cells.

Tsc1 deletion increases expression of both pro- and anti-proliferation genes including Trp53 and Cdkn2a

To understand the molecular basis of mTOR-induced renal cyst formation, we carried out RNA sequencing of kidney samples of 2-month-old Tsc1 deficient and control mice. A large number of genes showed altered expression (Supplementary Fig. S2A). Tsc1 deletion did not affect expression of the RCC disease genes including BAP1, PBRM1, SETD2, KDM5C, MET, NRF2, NF2, or Hif2α (Supplementary Fig. S2B), yet it led to increased expression of Vhl, Hif1α, VEGFB, and VEGFC (Supplementary Fig. S2B), suggesting that mTOR activation may regulate angiogenesis in addition to promoting cyst formation.

GO and KEGG analyses revealed that catabolic pathway genes, e.g., protein degradation, amino acid degradation, and lipolysis, were suppressed whereas expression of anabolic genes including Tif-1α (protein synthesis), Srebp1 and Pparγ (lipid synthesis), Atf4, Cad, and Mthfd2 (nucleotide synthesis), and Creb5 and Hif1α (glucose metabolism; thereafter referred to as anabolic genes) were increased (Supplementary Figs. S2C–S2E), supporting that mTORC1 activation in general augments cell metabolism (20). Interestingly, expression of cell-cycle signature proteins was elevated including Myc, cyclins, E2F1, E2F3, and CDK4, as well as anti-proliferation proteins including p15, p16, Wee1, Bax1, p21, many of which are p53 target genes (Fig. 1E; Supplementary Fig. S3). Western blot analysis confirmed the increases in pro-proliferation proteins Myc, cyclin E, and PCNA and anti-proliferation proteins p53, p21, and p16 (Fig. 1F). Thus, Tsc1 deletion leads to concurrent increases in pro- and anti-proliferation proteins in addition to metabolic proteins, with the pro-proliferation signals apparently dominating over the anti-proliferation signals, resulting in increased cell proliferation and cyst formation.

Tsc1−/− mouse and TSC patient samples show increased MEK1 and MKK6 expression and ERK and p38 activation

What causes elevated expression of these pro- and anti-proliferation genes at the mRNA levels? KEGG analysis revealed activation of multiple signaling pathways including MAPKs in Tsc1 deficient samples (Fig. 1E; Supplementary Figs. S2C and S3). Western blot analysis confirmed that Tsc1 deletion led to activation of the ERK and p38MAPK pathways but not JNK, β-catenin, Smad1/5/8, Smad2/3, or Akt1 (Fig. 2A; Supplementary Fig. S4A). Immunostaining confirmed strong ERK and p38 activation but not Akt activation (Fig. 2B; Supplementary Fig. S4B). Interestingly, Tsc1 deficient kidney samples also showed increased protein levels of MEK1, an upstream activator of ERKs, and MKK6, an upstream kinase of p38MAPKs, which required mTORC1 as RAP administration could dampen the increases (Fig. 2C).

The link to MEK1-ERKs and MKK6-p38MAPKs were also observed in TSC patient samples. We collected tissue samples from 6 patients with TSC, which had been used in our previous studies (36). The kidney samples showed angiomyolipomas (AML) and some cysts whereas the small intestine samples showed disrupted structures of the villi (Supplementary Figs. S5A and S5B). All the samples showed enhanced activation of mTOR, manifested by an increase in p-S6, decreased levels of TSC1, but normal levels of TSC2 and PDK1 (Fig. 2D; Supplementary Figs. S5C and S5D). Importantly, all samples showed 3- to 4-fold increases in MEK1 and MKK6 proteins and ERK and p38MAPK activation, revealed by Western blot analysis and confirmed by immunostaining of the patient samples (Fig. 2D; Supplementary Fig. S5E). In addition, transient knockdown of Tsc1 mRNA in human kidney cell lines HEK293 or HK2 with siRNA led to increases in MEK1 expression and ERK activation as well as increases in MKK6 expression and p38MAPK activation (Supplementary Figs. S6A and S6B). Moreover, transient deletion of Tsc1 in mouse embryonic fibroblasts (MEFs) quickly resulted in increased MEK1 and MKK6 expression and ERK and p38MAPK activation (Supplementary Fig. S6C). These results suggest that mTOR activation directly induces MEK1 and MKK6 expression in multiple cell types.

How does mTOR activation increase MEK1 and MKK6 expression? We found that Tsc1 deficiency did not significantly affect the mRNA levels of Mek1 or Mkk6 in HEK293 cells, HK29 cells, or MEFs (Supplementary Figs. S6D–S6F), We analyzed the mRNA levels of Mek1, Erk1, Erk2, MKK6, p38α in Villin-Cre;Tsc1f/f mouse kidney sample treated with rapamycin and found that the mRNA levels of these genes were not altered (Fig. 2E; Supplementary Fig. S6G). These results suggest the regulation occurs at posttranscriptional levels. mTOR is known to selectively increase protein synthesis of certain mRNAs species (37, 38). We then used the kidney cell line HEK293 to dissect the underlying mechanisms and found that Tsc1 knockdown significantly increased incorporation of radio-labeled Methionine into MEK1 and MKK6, which was blocked by RAP treatment (Fig. 2F). Moreover, we found that mRNA species for Mek1 and Mkk6 were enriched in polysomes in Tsc1-deficient cells compared with control cells (Supplementary Fig. S6H). These results suggest that mTORC1 activation promotes protein synthesis of MEK1 and MKK6, which in turn potentiate activation of the ERK and p38MAPK signaling cascades, respectively.

Mek1 deletion diminishes renal cyst formation in Tsc1−/− mice

ERK signaling is a primary pro-proliferation pathway (39). To test whether enhanced ERK activation contributes to renal cyst formation caused by Tsc1 deletion, we crossed Villin-Cre and Villin-Cre;Tsc1f/f mice to Mek1f/f mice (40, 41). Western blot analysis revealed that the increases in Myc, cyclin E, and PCNA were dampened by Mek1 deletion in Tsc1−/− kidney samples but not in Tsc1+/+ samples (Fig. 3A). However, Mek1 deletion did not affect the increases in p53, p16, or p21 proteins or mRNA levels of anabolic genes (Supplementary Figs. S7A and S7B), or mTORC1 activation, judged by unaltered levels of p-S6 and p-4E-BP1 (Fig. 3A; Supplementary Fig. S7C). Overall, these results indicate that mTOR-activation-induced Myc and cyclin E expression in tubular cell requires MEK1–ERK signaling and that the ERK pathway is downstream of mTOR signaling.

Mek1 ablation alone modestly inhibited tubular cell proliferation without significantly affecting kidney growth (Fig. 3B and C), likely due to existence of redundant MEK molecules. However, Mek1 deficiency restored ERK activation to the levels of normal cells and largely rescued renal tubular cell overproliferation, cyst formation, and renal secretion defects of Villin-Cre, Tsc1f/f mice (Fig. 3A–D). Overall, these data suggest that elevated MEK1-ERK activation mediates mTOR-induced cyst formation, yet its effects on tubular cell proliferation are minimal in normal mice.

Ablation of Tsc1 and Mapk14 in proximal tubules leads to RCC

p38MAPKs have anti-proliferation or pro-proliferation activities depending on the cell contexts (39, 42). We then ablated Mapk14, which encodes the dominant isoform p38α MAPK, to determine its function in Tsc1-deficient renal tubular cells (Fig. 4A). Villin-Cre;Mapk14f/f mice showed no cyst formation or significant alteration in kidney size or weight (Fig. 4B and C). However, ablation of Mapk14 led to further increases in kidney size and cyst formation in Villin-Cre; Tsc1f/f mice and aggravated the renal secretion defects (Fig. 4B–D). Most of these double knockout mice survived for 3 to 6 months with small tumors detected in kidneys of all mice (Fig. 4C; Supplementary Fig. S8A). The tumors showed no obvious oncolytic features including granular cytoplasm and prominent vacuoles, expressed very low levels of succinate dehydrogenase B (SDHB; ref. 43), but expressed certain levels of clear cell RCC markers vimentin, CAIX, and Pax8 (Fig. 4C; Supplementary Fig. S8B). These features suggest that deletion of Tsc1 and Mapk14 leads to development of RCCs, with similarity to clear cell RCCs described in Vhl and Bap1 double knockout mice (9, 13).

Mapk14 ablation suppresses Tsc1 deletion-induced p53 and p16 expression

p38MAPKs have many substrates that regulate various cellular activities (39). How does elevated p38MAPK activation suppress RCC development in Tsc1−/− tubular cells? We found that the kidney samples of Villin-Cre; Tsc1f/f; Mapk14f/f mice showed no change in protein levels of Myc, cyclin E, or PCNA or mRNA levels of the anabolic genes compared with Villin-Cre; Tsc1f/f mice (Supplementary Figs. S8C and S8D). Instead, Mapk14 deficiency dampened Tsc1 deletion-induced increases in p53 and p16 (Fig. 5A), suggesting that mTOR activation upregulates p53 and p16 via p38MAPKs in renal tubular cells. p38MAPKs may increase p53 by direct phosphorylation, S6K-mediated Mdm2 phosphorylation, and/or other mechanisms (36, 44).

We found that the kidney samples of Villin-Cre; Tsc1f/f; Mapk14f/f mice showed no change in the levels of MEK1 protein or ERK activation or the levels of p-S6 or p-4EBP1 compared with those of Villin-Cre; Tsc1f/f mice (Fig. 5A; Supplementary Fig. S8C). In addition, MKK6 expression, p38MAPK activation, and mTOR activation were not affected by Mek1 deletion in Villin-Cre; Tsc1f/f mice (Supplementary Fig. S7A; Fig. 3A). These results suggest that Tsc1 ablation-activated MEK1–ERK and MKK6–p38MAPK–p53/p16 pathways are independent of each other.

Inactivation of both Tsc1 and Trp53 in proximal tubule induces RCC

We then generated Villin-Cre; Tsc1f/f; Trp53−/− mice to test the effect of upregulated p53 in Tsc1 deficiency-induced cell proliferation (Fig. 5B). Some of Trp53−/− mice could survive up to 7 months. We observed no cyst formation or alteration in kidney structures in 2- or 7-month-old Trp53−/− mice compared with control littermates (Fig. 5C and D). However, Trp53 deficiency enhanced cyst formation and led to renal tumor formation in Villin-Cre; Tsc1f/f; Trp53−/− mice (Fig. 5D). Immunostaining revealed that some of the tumor cells were positive for RCC markers vimentin, CAIX, and Pax8 but not SDHB (Fig. 5D; Supplementary Fig. S8B). Overall, these results suggest that deletion of Tsc1 and Trp53 promotes RCC development.

Inactivation of both Tsc1 and Cdkn2a in proximal tubule induces RCC

We also generated Villin-Cre; Tsc1f/f; Cdkn2a−/− mice (Fig. 6A). Similar to Mek1, Cdkn2a ablation alone did not affect tubular cell proliferation (Fig. 6B and C), yet, it increased kidney size and cyst formation and induced renal tumor formation in Villin-Cre; Tsc1f/f mice (Fig. 6B and C). Immunostaining revealed that the tumor cells expressed vimentin, CAIX, and Pax8, although much lesser extents than Tsc1 and Trp53 double knockout mice (Fig. 6C). Thus, both p53 and p16, upregulated by mTOR signaling via p38MAPKs, help to suppress proliferation of Tsc1−/− tubular cells and thus act as a brake on cell proliferation whereas their effects on Tsc1+/+ cells are minimal.

MEK inhibitor diminishes cyst and RCC development

We then tested whether small molecular compound inhibitor of MEK1 could inhibit renal cyst formation. We injected U0126 or vehicle daily into 2-month-old Villin-Cre; Tsc1f/f mice for 3 months and found that ERK activation was inhibited whereas the levels of p-S6 and p-4EBP1 were unaltered (Fig. 7A), further supporting that the ERK pathway is downstream of mTOR signaling. U0126 did not significantly affect the size of kidney or the number of proliferating tubular cells in wild-type mice, yet, it dampened kidney overgrowth in Tsc1−/− mice (Fig. 7B and C). Tubular cell proliferation and cyst formation were largely suppressed as well (Fig. 7C).

Because renal cysts are precancerous lesions, we then repeated the experiments on Villin-Cre; Tsc1f/f; Trp53−/− mice, which developed RCCs at 2 to 3 months of age. We found that U0126 treatment (for 2 months starting at 1 month of age) diminished RCC development, kidney overgrowth, and expression of RCC markers (Fig. 7D; Supplementary Fig. S9). These data confirm that the MEK1–ERK pathway plays a critical role in mTOR-induced renal cyst and RCC development from proximal tubules and suggest that MEK1 inhibitors, which are developed as cancer therapeutic drugs (45), have the potential to treat RCC patients as well as patients with TSC.

The RCCs, derived mainly from renal proximal tubules, carry somatic mutations that lead to activation of HIF-VEGF/PDGF, Akt-mTOR, and other mitogenic pathways (10, 46). Yet, available mouse modeling studies show that disruption or combination of disruption of the commonly mutated genes fail to generate RCCs from proximal tubules (9). Here, we generate mouse models to show that RCC can originate from proximal tubules under conditions with activation of mTOR and inactivation of the p38MAPK–p53/p16 pathway. Deletion of Tsc1 alone in renal proximal tubules only induces cyst formation. The inability of mTOR activation to induce RCC is likely owing to mTOR-induced activation of the p38MAPK–p53/p16 pathway as deletion either of the 3 genes in Tsc1−/− proximal tubule cells results in RCC. Overall, our findings suggest that mTOR signaling may initially induce cyst formation in proximal tubules and then cooperate with mutations that compromise the p38MAPK–p53/p16 pathway to transform cyst into RCC.

We show that tumor cells in our mouse models are morphologically regular and express certain levels of CAIX, PAX8, and vimentin but very low levels of SDHB, similar to clear cell RCCs observed in Vhl and Bap1 double knockout mice (9, 13). Overall, our RCC mouse models simulate human RCC in cell origin, oncogenic drivers (mTOR and p53), and expression of some clear cell RCC markers, and may be useful in dissecting RCC pathogenesis and testing potential anti-RCC drugs.

RCCs are detected in only 2% to 4% of patients with TSC, which usually have acquired loss of the heterogeneity of TSC genes as well as somatic mutations in other tumorigenic genes (35). TSC-associated RCCs are also heterogeneous and have features such as female predisposition, onset at early age, and multifocality (35, 43, 47). We show here that Tsc1 and Mapk14 (or Trp53) double knockout mice developed renal tumors at early stages (2–4 months of age), which expressed some clear cell RCC markers. More importantly, mTOR hyperactivation is observed in 25% to 76% of RCC samples, which is often caused by mutations in other genes including VHL, BAP1, and PTEN. It is possible that activated mTOR may activate ERKs and p38MAPKs in both TSC-associated RCCs and common RCCs and activation of the p38MAPK–p53/p16 pathway may explain the low RCC incidence in patients with TSC. Examination of the p38MAPK-p53/p16 pathway in common RCCs and TSC-associated RCCs may help understand the pathogenesis of these diseases.

Our mouse models contrast those with deletion of both Vhl and Bap1 or Pbrm1, which mainly causes RCC in Bowman epithelial cells or distal tubules, suggesting that RCCs originated from different renal tubule segments may employ distinct carcinogenic pathways, likely due to differences in the nature of tubular cells and the environments (48). Interestingly, it has been reported that distal tubules have higher turnover rate than proximal tubules (18), this may explain why distal tubules cells are easier to transform by deletion of Vhl and Bap1 or Pbrm1. In renal proximal tubules, mutations in Vhl, Bap1, or Pbrm1 may cooperate with mTOR signaling to initiate RCC development. For example, these mutations may lead to increased blood vessel formation and thus promote tumor growth.

Although mTOR-induced cell proliferation and tumorigenesis are generally believed to be mediated by increased anabolic cellular processes (20), we show here that mTOR activation drives renal cyst formation and RCC development via MEK1–ERK signaling. mTOR activation leads to activation of the ERK pathway and Mek1 deficiency or inhibition dampens overgrowth of Tsc1 deficient renal tubular cells and eliminates renal cyst formation and RCC development (Fig. 7E). It is likely that enhanced ERK activation promotes proliferation while increased expression of anabolic genes provides building blocks in response to mTOR activation. Mek1 ablation or inhibition dampens mTOR-induced expression of Myc and cyclin E1 without affecting the protein levels of p53/p16, mRNA levels of anabolic genes, or levels of p-S6 or p-4EBP1 (translation regulators), supporting that the effect of MEK1–ERK on cell proliferation may go through Myc/cyclins.

Intriguingly, several studies have shown that ERKs phosphorylate TSC2 and subsequently activate mTOR signaling (49). However, we find that deletion of Mek1 or inhibition of Mek1 in Tsc1+/+ or Tsc1−/− cells did not affect mTOR activation in renal tubular cells. This discrepancy may be caused by different cell contexts (normal vs. cancer cells) and warrants further investigation. Nevertheless, our multiple lines of evidence supports that MEK–ERK signaling acts downstream not upstream of mTOR activation and it plays a critical role in Tsc1 deletion-induced renal cyst and RCC development.

We find that mTOR activation increases MKK6 expression and p38MAPK activation in renal tubular cells. Functionally, deletion of Mapk14 enhances renal cyst formation in Villin-Cre;Tsc1f/f mice and leads to RCC development whereas it has minimal effect on Tsc1+/+ cells. Mechanistically, although p38MAPKs have many substrates, p38MAPK inhibits renal cell proliferation and RCC development by increasing p53 and p16 expression in Villin-Cre;Tsc1f/f mice without affecting expression of Myc/cyclins or anabolic genes. We conceive that enhanced MKK6–p38MAPK–p53/p16 pathway may act as a safeguard mechanism to curtail mTOR-driven cell proliferation, especially in response to p38MAPK-activating stressors such as DNA damage, ROS, and proteostatic stress. In the absence of this axis, mTOR activation readily causes RCC as observed in young Villin-Cre; Tsc1f/f; Mapk14f/f, Villin-Cre; Tsc1f/f; Trp53−/−, and Villin-Cre; Tsc1f/f; Cdkn2a−/− mouse lines.

The findings of our study suggest that mTOR employs two independent MAPK pathways to regulate cell proliferation (Fig. 7E) and the links to MAPKs are verified in multiple cell or tissue types. Mechanistically, Tsc1-mTOR signaling regulates ERK and p38MAPK pathways by directly increasing MEK1 and MKK6 expression, respectively, unraveling previously unidentified mechanisms by which activated mTOR controls cell proliferation. Moreover, a study by Chen and colleagues reported that both mTOR and MAPK pathways were activated in subpopulations of RCC patient samples (50). The mTOR-MAPK link may exist in other tumor types that show enhanced mTOR activation.

In summary, the study establishes several RCC models with proximal tubular cell origins and identifies MAPKs as effector molecules in mTOR activation-driven cyst formation and RCC pathogenesis. Given that mTORC1 hyperactivation is common in RCC samples as well as tumors of many other organs, the findings of this study may have a broad clinical implication.

No disclosures were reported.

H. Wu: Performed research, analyzed data. D. He: Performed research. S. Biswas: Performed research. M. Shafiquzzaman: Performed research. X. Zhou: Performed research. J. Charron: Performed research. Y. Wang: Performed research. B.K. Nayak: Performed research. S.L. Habib: Designed research. H. Liu: Designed research, analyzed data. B. Li: Writing–original draft, writing–review and editing.

The work was supported by the National Key Research and Development Program of China (2018YFA0800803 to B. Li and 2017YFA0103602 to H. Liu), the National Natural Science Foundation of China (81520108012 and 91749201), and grants from the State Key Laboratory of Oncogenes and Related Genes (No. 90-17-03) to B. Li.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Bray
F
,
Ferlay
J
,
Soerjomataram
I
,
Siegel
RL
,
Torre
LA
,
Jemal
A
. 
Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries
.
CA Cancer J Clin
2018
;
68
:
394
424
.
2.
Capitanio
U
,
Montorsi
F
. 
Renal cancer
.
Lancet
2016
;
387
:
894
906
.
3.
Kotecha
RR
,
Motzer
RJ
,
Voss
MH
. 
Towards individualized therapy for metastatic renal cell carcinoma
.
Nat Rev Clin Oncol
2019
;
16
:
621
33
.
4.
Hu
SL
,
Chang
A
,
Perazella
MA
,
Okusa
MD
,
Jaimes
EA
,
Weiss
RH
, et al
The nephrologist's tumor: basic biology and management of renal cell carcinoma
.
J Am Soc Nephrol
2016
;
27
:
2227
37
.
5.
Choueiri
TK
,
Motzer
RJ
. 
Systemic therapy for metastatic renal-cell carcinoma
.
N Engl J Med
2017
;
376
:
354
66
.
6.
Hsieh
AC
,
Liu
Y
,
Edlind
MP
,
Ingolia
NT
,
Janes
MR
,
Sher
A
, et al
The translational landscape of mTOR signalling steers cancer initiation and metastasis
.
Nature
2012
;
485
:
55
61
.
7.
Cancer Genome Atlas Research N
. 
Comprehensive molecular characterization of clear cell renal cell carcinoma
.
Nature
2013
;
499
:
43
9
.
8.
Frew
IJ
,
Moch
H
. 
A clearer view of the molecular complexity of clear cell renal cell carcinoma
.
Annu Rev Pathol
2015
;
10
:
263
89
.
9.
Gu
YF
,
Cohn
S
,
Christie
A
,
McKenzie
T
,
Wolff
N
,
Do
QN
, et al
Modeling renal cell carcinoma in mice: bap1 and pbrm1 inactivation drive tumor grade
.
Cancer Discov
2017
;
7
:
900
17
.
10.
Sato
Y
,
Yoshizato
T
,
Shiraishi
Y
,
Maekawa
S
,
Okuno
Y
,
Kamura
T
, et al
Integrated molecular analysis of clear-cell renal cell carcinoma
.
Nat Genet
2013
;
45
:
860
7
.
11.
Fu
L
,
Wang
G
,
Shevchuk
MM
,
Nanus
DM
,
Gudas
LJ
. 
Generation of a mouse model of von hippel-lindau kidney disease leading to renal cancers by expression of a constitutively active mutant of HIF1alpha
.
Cancer Res
2011
;
71
:
6848
56
.
12.
Schonenberger
D
,
Harlander
S
,
Rajski
M
,
Jacobs
RA
,
Lundby
AK
,
Adlesic
M
, et al
Formation of renal cysts and tumors in Vhl/Trp53-deficient mice requires HIF1alpha and HIF2alpha
.
Cancer Res
2016
;
76
:
2025
36
.
13.
Harlander
S
,
Schonenberger
D
,
Toussaint
NC
,
Prummer
M
,
Catalano
A
,
Brandt
L
, et al
Combined mutation in Vhl, Trp53 and Rb1 causes clear cell renal cell carcinoma in mice
.
Nat Med
2017
;
23
:
869
77
.
14.
Leung
JY
,
Kim
WY
. 
Bap1 and Pbrm1: determinants of tumor grade and mTOR activation in VHL-deficient mouse models of renal cell carcinoma
.
Cancer Discov
2017
;
7
:
802
4
.
15.
Bailey
ST
,
Smith
AM
,
Kardos
J
,
Wobker
SE
,
Wilson
HL
,
Krishnan
B
, et al
MYC activation cooperates with Vhl and Ink4a/Arf loss to induce clear cell renal cell carcinoma
.
Nat Commun
2017
;
8
:
15770
.
16.
Guinot
A
,
Lehmann
H
,
Wild
PJ
,
Frew
IJ
. 
Combined deletion of Vhl, Trp53 and Kif3a causes cystic and neoplastic renal lesions
.
J Pathol
2016
;
239
:
365
73
.
17.
Frew
IJ
,
Thoma
CR
,
Georgiev
S
,
Minola
A
,
Hitz
M
,
Montani
M
, et al
pVHL and PTEN tumour suppressor proteins cooperatively suppress kidney cyst formation
.
EMBO J
2008
;
27
:
1747
57
.
18.
Mandriota
SJ
,
Turner
KJ
,
Davies
DR
,
Murray
PG
,
Morgan
NV
,
Sowter
HM
, et al
HIF activation identifies early lesions in VHL kidneys: evidence for site-specific tumor suppressor function in the nephron
.
Cancer Cell
2002
;
1
:
459
68
.
19.
Hall
MN
. 
An amazing turn of events
.
Cell
2017
;
171
:
18
22
.
20.
Saxton
RA
,
Sabatini
DM
. 
mTOR signaling in growth, metabolism, and disease
.
Cell
2017
;
169
:
361
71
.
21.
Guri
Y
,
Colombi
M
,
Dazert
E
,
Hindupur
SK
,
Roszik
J
,
Moes
S
, et al
mTORC2 promotes tumorigenesis via lipid synthesis
.
Cancer Cell
2017
;
32
:
807
23
e12.
22.
Faller
WJ
,
Jackson
TJ
,
Knight
JR
,
Ridgway
RA
,
Jamieson
T
,
Karim
SA
, et al
mTORC1-mediated translational elongation limits intestinal tumour initiation and growth
.
Nature
2015
;
517
:
497
500
.
23.
So
L
,
Lee
J
,
Palafox
M
,
Mallya
S
,
Woxland
CG
,
Arguello
M
, et al
The 4E-BP-eIF4E axis promotes rapamycin-sensitive growth and proliferation in lymphocytes
.
Sci Signal
2016
;
9
:
ra57
.
24.
Peterson
TR
,
Sengupta
SS
,
Harris
TE
,
Carmack
AE
,
Kang
SA
,
Balderas
E
, et al
mTOR complex 1 regulates lipin 1 localization to control the SREBP pathway
.
Cell
2011
;
146
:
408
20
.
25.
Robitaille
AM
,
Christen
S
,
Shimobayashi
M
,
Cornu
M
,
Fava
LL
,
Moes
S
, et al
Quantitative phosphoproteomics reveal mTORC1 activates de novo pyrimidine synthesis
.
Science
2013
;
339
:
1320
3
.
26.
Grabiner
BC
,
Nardi
V
,
Birsoy
K
,
Possemato
R
,
Shen
K
,
Sinha
S
, et al
A diverse array of cancer-associated MTOR mutations are hyperactivating and can predict rapamycin sensitivity
.
Cancer Discov
2014
;
4
:
554
63
.
27.
Salussolia
CL
,
Klonowska
K
,
Kwiatkowski
DJ
,
Sahin
M
. 
Genetic etiologies, diagnosis, and treatment of tuberous sclerosis complex
.
Annu Rev Genomics Hum Genet
2019
;
20
:
217
40
.
28.
Onda
H
,
Lueck
A
,
Marks
PW
,
Warren
HB
,
Kwiatkowski
DJ
. 
Tsc2(±) mice develop tumors in multiple sites that express gelsolin and are influenced by genetic background
.
J Clin Invest
1999
;
104
:
687
95
.
29.
Wilson
C
,
Idziaszczyk
S
,
Parry
L
,
Guy
C
,
Griffiths
DF
,
Lazda
E
, et al
A mouse model of tuberous sclerosis 1 showing background specific early post-natal mortality and metastatic renal cell carcinoma
.
Hum Mol Genet
2005
;
14
:
1839
50
.
30.
Drusian
L
,
Nigro
EA
,
Mannella
V
,
Pagliarini
R
,
Pema
M
,
Costa
ASH
, et al
mTORC1 upregulation leads to accumulation of the oncometabolite fumarate in a mouse model of renal cell carcinoma
.
Cell Rep
2018
;
24
:
1093
104
e6.
31.
Gan
B
,
Lim
C
,
Chu
G
,
Hua
S
,
Ding
Z
,
Collins
M
, et al
FoxOs enforce a progression checkpoint to constrain mTORC1-activated renal tumorigenesis
.
Cancer Cell
2010
;
18
:
472
84
.
32.
el Marjou
F
,
Janssen
KP
,
Chang
BH
,
Li
M
,
Hindie
V
,
Chan
L
, et al
Tissue-specific and inducible Cre-mediated recombination in the gut epithelium
.
Genesis
2004
;
39
:
186
93
.
33.
Kwiatkowski
DJ
,
Zhang
H
,
Bandura
JL
,
Heiberger
KM
,
Glogauer
M
,
el-Hashemite
N
, et al
A mouse model of TSC1 reveals sex-dependent lethality from liver hemangiomas, and up-regulation of p70S6 kinase activity in Tsc1 null cells
.
Hum Mol Genet
2002
;
11
:
525
34
.
34.
Fantus
D
,
Rogers
NM
,
Grahammer
F
,
Huber
TB
,
Thomson
AW
. 
Roles of mTOR complexes in the kidney: implications for renal disease and transplantation
.
Nat Rev Nephrol
2016
;
12
:
587
609
.
35.
Lam
HC
,
Siroky
BJ
,
Henske
EP
. 
Renal disease in tuberous sclerosis complex: pathogenesis and therapy
.
Nat Rev Nephrol
2018
;
14
:
704
16
.
36.
Lai
KP
,
Leong
WF
,
Chau
JF
,
Jia
D
,
Zeng
L
,
Liu
H
, et al
S6K1 is a multifaceted regulator of Mdm2 that connects nutrient status and DNA damage response
.
EMBO J
2010
;
29
:
2994
3006
.
37.
Dowling
RJ
,
Topisirovic
I
,
Alain
T
,
Bidinosti
M
,
Fonseca
BD
,
Petroulakis
E
, et al
mTORC1-mediated cell proliferation, but not cell growth, controlled by the 4E-BPs
.
Science
2010
;
328
:
1172
6
.
38.
Ben-Sahra
I
,
Manning
BD
. 
mTORC1 signaling and the metabolic control of cell growth
.
Curr Opin Cell Biol
2017
;
45
:
72
82
.
39.
Wagner
EF
,
Nebreda
AR
. 
Signal integration by JNK and p38 MAPK pathways in cancer development
.
Nat Rev Cancer
2009
;
9
:
537
49
.
40.
Bissonauth
V
,
Roy
S
,
Gravel
M
,
Guillemette
S
,
Charron
J
. 
Requirement for Map2k1 (Mek1) in extra-embryonic ectoderm during placentogenesis
.
Development
2006
;
133
:
3429
40
.
41.
Caunt
CJ
,
Sale
MJ
,
Smith
PD
,
Cook
SJ
. 
MEK1 and MEK2 inhibitors and cancer therapy: the long and winding road
.
Nat Rev Cancer
2015
;
15
:
577
92
.
42.
Gupta
J
,
del Barco Barrantes
I
,
Igea
A
,
Sakellariou
S
,
Pateras
IS
,
Gorgoulis
VG
, et al
Dual function of p38alpha MAPK in colon cancer: suppression of colitis-associated tumor initiation but requirement for cancer cell survival
.
Cancer Cell
2014
;
25
:
484
500
.
43.
Yang
P
,
Cornejo
KM
,
Sadow
PM
,
Cheng
L
,
Wang
M
,
Xiao
Y
, et al
Renal cell carcinoma in tuberous sclerosis complex
.
Am J Surg Pathol
2014
;
38
:
895
909
.
44.
Wong
ES
,
Le Guezennec
X
,
Demidov
ON
,
Marshall
NT
,
Wang
ST
,
Krishnamurthy
J
, et al
p38MAPK controls expression of multiple cell cycle inhibitors and islet proliferation with advancing age
.
Dev Cell
2009
;
17
:
142
9
.
45.
Samatar
AA
,
Poulikakos
PI
. 
Targeting RAS-ERK signalling in cancer: promises and challenges
.
Nat Rev Drug Discov
2014
;
13
:
928
42
.
46.
Ricketts
CJ
,
De Cubas
AA
,
Fan
H
,
Smith
CC
,
Lang
M
,
Reznik
E
, et al
The cancer genome atlas comprehensive molecular characterization of renal cell carcinoma
.
Cell Rep
2018
;
23
:
313
26
e5.
47.
Guo
J
,
Tretiakova
MS
,
Troxell
ML
,
Osunkoya
AO
,
Fadare
O
,
Sangoi
AR
, et al
Tuberous sclerosis-associated renal cell carcinoma: a clinicopathologic study of 57 separate carcinomas in 18 patients
.
Am J Surg Pathol
2014
;
38
:
1457
67
.
48.
Lee
JW
,
Chou
CL
,
Knepper
MA
. 
Deep sequencing in microdissected renal tubules identifies nephron segment-specific transcriptomes
.
J Am Soc Nephrol
2015
;
26
:
2669
77
.
49.
Mendoza
MC
,
Er
EE
,
Blenis
J
. 
The ras-ERK and PI3K-mTOR pathways: cross-talk and compensation
.
Trends Biochem Sci
2011
;
36
:
320
8
.
50.
Chen
F
,
Zhang
Y
,
Senbabaoglu
Y
,
Ciriello
G
,
Yang
L
,
Reznik
E
, et al
Multilevel genomics-based taxonomy of renal cell carcinoma
.
Cell Rep
2016
;
14
:
2476
89
.

Supplementary data