PI3K-Akt-mTOR inhibition by GNE-477 inhibits renal cell carcinoma cell growth in vitro and in vivo

Sustained activation of PI3K-Akt-mTOR cascade is important for renal cell carcinoma (RCC) cell progression. GNE-477 is a novel and efficacious PI3K-mTOR dual inhibitor. The current study tested its anti-RCC cell activity. In the primary cultured human RCC cells, GNE-477 potently inhibited cell growth, viability and proliferation, as well as cell cycle progression, migration and invasion. Furthermore, it induced robust apoptosis activation in primary RCC cells, but being non-cytotoxic to HK-2 epithelial cells and primary human renal epithelial cells. In the primary RCC cells GNE-477 inactivated PI3K-Akt-mTOR cascade by blocking phosphorylation of p85, Akt1, p70S6K1 and S6. Restoring Akt-mTOR activation by a constitutively-active Akt1 reversed GNE-477-induced anti-RCC cell activity. In nude mice intraperitoneal injection of GNE-477 potently suppressed RCC xenograft tumor growth. Collectively, targeting PI3K-Akt-mTOR cascade by GNE-477 inhibits RCC cell growth in vitro and in vivo.

Molecularly-targeted therapies are currently needed for better and advanced RCC treatments [7]. Our previous studies have shown that melanoma antigen A6 (MAGEA6), a cancer-specific ubiquitin ligase of AMPactivated protein kinase (AMPK), is uniquely expressed in human RCC tissues and cells. MAGEA6 silencing or knockout activated AMPK signaling to inhibit mammalian target of rapamycin (mTOR) cascade, thereby inhibiting RCC cell progression [8]. Furthermore, a long non-coding RNA (LncRNA) THOR is expressed in RCC tissues and cells. THOR silencing resulted in potent RCC cell growth inhibition in vitro and in vivo [9].
AGING A very recent study by Heffron et al., has identified GNE-477 as a potent and efficient PI3K and mTOR dual inhibitor [14]. By simultaneously targeting PI3K and mTOR, GNE-477 may have unique advantage over single-specific mTORC1 or PI3K inhibitors in inhibiting human cancer cells [14]. The results of this study will show that targeting PI3K-Akt-mTOR cascade by GNE-477 potently inhibits RCC cell growth in vitro and in vivo.

GNE-477 potently inhibits human RCC cell survival, proliferation, cell cycle progression, migration and invasion
First the primary human RCC cells ("RCC1" [8,9]) were cultured in FBS-containing complete medium, and treated with GNE-477 at 1-100 nM. After further culture for 24-96h, the cell viability was tested by CCK-8 assays. As demonstrated, GNE-477, in a dose-dependent manner, efficiently decreased CCK-8 viability in RCC1 cells ( Figure 1A). The dual PI3K-mTOR inhibitor also displayed a time-dependent response in inhibiting CCK-8 viability in RCC1 cells ( Figure 1A). The CCK-8 OD reduction was significant at 48h after GNE-477 treatment (10-100 nM), that lasted for at least 96h ( Figure 1A). The colony formation assay results, Figure 1B, show that the number of viable RCC1 colonies was significantly decreased following GNE-477 treatment (at 10-100 nM, for 10 days). Since in RCC1 cells 50 nM of GNE-477 resulted in potent cell viability reduction ( Figure 1A) and colony formation inhibition ( Figure 1B), this concentration was selected for further experiments.
In the primary human RCC cells-derived from two other RCC patients, RCC2 and RCC3, GNE-477 (50 nM) stimulation potently inhibited cell viability (CCK-8 OD, Figure 1G), proliferation (nuclear EdU incorporation, Figure 1H) and migration ( Figure 1I). In contrast, in HK-2 renal epithelial cells and primary human renal epithelial cells, the same GNE-477 (50 nM) treatment was completely ineffective and non-cytotoxic ( Figure 1G-1I). These results show that GNE-477 specially and potently inhibited RCC cell viability, proliferation, cell cycle progression, migration and invasion in vitro.

GNE-477 potently inhibits RCC xenograft tumor growth in mice
The potential anti-RCC activity of GNE-477 in vivo was tested. Using a previously described animal model [8,9], we subcutaneously (s.c.) injected RCC1 cells to the flanks of the nude mice. Within two weeks RCC1 xenograft tumors were established with the volume close to 100 mm 3 . Tumor growth curve results, Figure  4A, demonstrated that intraperitoneal (i.p.) injection of GNE-477, at 10 or 50 mg/kg (daily, for 3 weeks), potently inhibited RCC1 xenograft tumor growth in nude mice. Calculating the estimated daily tumor growth, using the formula (tumor volume at Day-35tumor volume at Day-0)/35, we again show that GNE-477 injection potently suppressed RCC1 xenograft tumor growth in vivo ( Figure 4B). At Day-35, tumors of all three groups were isolated and weighted. Tumors of GNE-477-treated mice were significantly lighter than those of the vehicle control mice ( Figure 4C).
Notably, at the dose of 10 mg/kg (i.p. daily for three weeks) the known mTOR kinase inhibitor AZD2014 [19][20][21] significantly inhibited RCC1 xenograft tumor growth in mice ( Figure 4A-4C). Importantly, GNE-477 was more potent in suppressing RCC1 xenograft growth than the same concentration of AZD2014 ( Figure 4A-4C). There was no significant difference in animal body weights among the four groups ( Figure 4D). Furthermore, no apparent toxicities were noticed in GNE-477-treated mice and AZD2014-treated mice. These results confirmed that GNE-477, at well-tolerated doses, potently inhibited RCC1 xenograft tumor growth in mice. AGING DISCUSSION mTOR protein lies in a central position in the PI3K-Akt-mTOR cascade [15,22,23]. There are at least two mTORC complexes identified thus far, mTORC1 and mTORC2 [15,22,23]. mTORC1 is composed of mTOR, Raptor, mLST8 and several others, required for p70S6K1 and 4EBP1 phosphorylation [15,22,23]. mTORC2 is composed of mTOR, Rictor, Sin1, and others, serving as the upstream kinase of Akt phosphorylation at Ser-473 [24,25]. Studies have shown that mTORC1 and mTORC2 are both essential for the progression of RCC. Both are important for cancer cell growth, proliferation and migration, angiogenesis, chemo-resistance and metastasis [10,11,26,27]. Importantly, the mTOR pharmacological inhibitors have displayed therapeutic values for the treatment of RCC [10,11,26,27]. mTORC1 inhibitors have been approved by the US FDA for the clinical treatment of advanced RCC patients after failure of either sunitinib or sorafenib [10,11]. Yet, mTORC1 inhibitors could still have several limitations and drawbacks, including incomplete mTOR inhibition and feedback activation of other oncogenic signalings [10,11].
In the present study, we show that GNE-477 blocked phosphorylation of p70S6K1-S6 and Akt (Ser-473 and Thr-308) in RCC1 cells. It thus inactivated both mTORC1 and mTORC2 cascades. Furthermore, phosphorylation of p85 was largely inhibited by GNE-477. These results show that GNE-477 blocked the whole PI3K-Akt-mTOR cascade in the RCC cells. This should explain the extremely high efficiency of this compound against RCC cells. Indeed, GNE-477 was significantly more potent than other PI3K-Akt-mTOR inhibitors (LY294002, AZD2014, perifosine) in inhibiting RCC cell survival and inducing cell apoptosis. In vivo, GNE-477 was more potent in suppressing RCC1 xenograft growth than AZD2014. These results support that this compound could have important therapeutic value for the treatment of RCC.
Our results imply that GNE-477-induced anti-RCC activity is due to PI3K-Akt-mTOR inhibition. Restoring Akt-mTOR activation, by a ca-Akt1 construct, completely reversed GNE-477-induced cytotoxicity against RCC1 cells. Furthermore, Akt1 depletion, by the CRISPR/Cas9 Akt1 KO construct, mimicked GNE-477's activity and potently inhibited RCC1 cell viability and proliferation. Importantly, adding GNE-477 in the Akt1-KO RCC1 cells failed to induce further cytotoxicity. These results suggest that PI3K-Akt-mTOR inactivation by GNE-477 led to cytotoxicity and growth inhibition in RCC cells. This should also explain why GNE-477 was completely ineffective in HK-2 cells and primary human kidney epithelial cells. Since previous studies have shown that the basal PI3K-Akt-mTOR activation is quite low in the normal epithelial cells [28][29][30]. We also show that GNE-477 did not induce apparent toxicities to the nude mice.

CONCLUSIONS
Together, we conclude that targeting PI3K-Akt-mTOR by GNE-477 inhibited human RCC cell growth in vitro and in vivo. It should be noted that a number of tested PI3K-mTOR kinase inhibitors failed to result in significant clinical improvement for RCC patients [10,11]. Certain PI3K-mTOR inhibitors are even more toxic and less efficacious than everolimus or temsirolimus [31]. Therefore, the current results of in vitro and animal studies could not be directly translated to humans, and thus the efficacy and safety of GNE-477 will definitely need further characterizations and research. Lower concentrations of GNE-477 could also be tested in mice.

Cell culture
As described previously [8,9] the primary human RCC cells (from Dr. Zheng at Nantong University [32]) were derived from three independent primary RCC patients with written-informed consents. They were named as "RCC1", "RCC2" and "RCC3". The primary cancer cells were cultured in the previously-described medium for primary cells [33]. Cultures of HK-2 cells and primary human renal epithelial cells were also described in our previous study [9]. Protocols of this study were approved by the Ethics Board of Wenzhou Medical University, according to the principles expressed in the Declaration of Helsinki.

Cell viability
The primary human RCC cells or the epithelial cells were seeded onto 96-well plates (at 4, 000 cells per well). After the indicated treatments, the cell viability was examined by a Cell Counting Kit-8 (CCK-8) (Dojindo Molecular Technologies, Japan) CCK-8 optical densities (ODs) were examined using a microplate reader at the test-wavelength of 550 nm.

EdU (5-ethynyl-20-deoxyuridine) incorporation
The primary human RCC cells or the epithelial cells were seeded onto the six-well tissue-culturing plates (1×10 5 cells per well). Following the indicated treatments an EdU Apollo-567 assay kit (RiboBio, Guangzhou, China) was utilized to test cell proliferation, with nuclear EdU and DAPI staining visualized under a fluorescent microscope (1×200 magnification, Leica, Shanghai, China.). In each treatment five random views with total 500 cells were included to calculate the nuclear EdU ratio (% vs. DAPI).

In vitro cell migration and invasion
The primary human RCC cells or the epithelial cells, with the applied treatments, were initially seeded on the upper chambers of "Transwell" (BD Biosciences, Shanghai, China) [34], at a density of 1×10 4 cells in 250 μL serum-free medium in each chamber. The complete medium (15% FBS) was added to the lower chambers. After incubation for 24h, on the lower surface the migrated cells were stained and counted manually. To test cell invasion, Matrigel (Sigma, Shanghai, China) was added on the upper chambers of "Transwell". For each condition, five repeated views were included to calculate the average number of migrated/invaded cells.

Cell cycle assay
The primary human RCC cells with or without GNC-477 treatment were stained with propidium iodide (PI, 5 μg/mL, Thermo-Fisher Invitrogen, Shanghai, China) and RNase (50 μg/mL, Thermo-Fisher Invitrogen). DNA contents were examined under a flow cytometer (BD Biosciences, Franklin Lakes, NJ). Cell cycle distributions were recorded.

Colony formation
RCC1 cells, in 0.25% agarose (Sigma)-containing complete medium, were initially seeded onto 10-cm tissue-culture dishes (at 1×10 4 cells per dish). The FBScontaining complete medium with or without GNE-477 (at tested concentrations) was renewed every 2 days (for a total of 10 days). Afterwards, viable cell colonies were counted manually.

Western blotting
As described previously [8,9], following the applied treatment, the quantified protein lysates (30-40 μg per sample) were separated by SDS-PAGE gels, that were transferred to the PVDF (polyvinylidene difluoride) blots (Merck-Millipore, Shanghai, China). After blocking (in milk-containing PBST), the blots were incubated with applied primary and secondary antibodies. The protein bands were visualized based on the molecular weights by using an enhanced chemiluminescence (ECL) kit (Pierce) [36]. Data quantification was through an ImageJ software (NIH, US).

Mitochondrial depolarization
As described [37], with mitochondrial depolarization in the stressed cells JC-1 red fluorescein shall aggregate into mitochondria to form green monomers [38]. Briefly, after the indicated treatments, RCC1 cells were stained with JC-1 dye at 10 μg/mL for 30 min under the dark. The JC-1 green intensity was examined via a fluorescence spectrofluorometer at 550 nm.

Constitutively-active mutant Akt1
The recombinant constitutively-active Akt1-GFP (caAkt1, S473D) adenovirus was provided by Dr. Zhang [39], that was transduced to RCC1 cells (cultured in the polybrene-containing complete medium). Cells were than subjected to FACS sorting of GFP to establish the monoclonal stable RCC1 cells, with caAkt1 expression verified by Western blotting analyses.

Akt1 knockout
A lenti-CRISPR/Cas9-GFP Akt1-KO construct was from Dr. Zhang's lab at Soochow University [40], transduced to primary RCC1 cells. Cells were then subjected to FACS to sort GFP-positive cells, which were further distributed to 96-well tissue culture plates, with Akt1 knockout (KO) screened. The monoclonal stable Akt1 KO RCC1 cells were established.

Mice xenografts
As described previously [9], the female nude mice (5-6 week of age, 18.2-19.1 grams in weights) were provided by the Animal Center of Wenzhou Medical University (Wenzhou, China), maintained under standard conditions. Six million RCC1 cells per mouse were inoculated s.c. to the right flanks. Within two weeks the xenograft tumors were established with the volume close to 0.1 cm 3 . The tumor-bearing mice were randomly assigned into three groups (n=10 per group), intraperitoneally injected with GNE-477 or the vehicle control [14]. Tumor recordings were described early [8,9]. The protocols were approved by the IACUC of Wenzhou Medical University, according to National Institutes of Health guide for the care and use of laboratory animals.

Statistical analyses
Data were presented as mean ± standard deviation (SD). Statistics were analyzed by one-way ANOVA using a SPSS software (21.0, SPSS Co., Chicago, CA). To test difference between two specific groups, a two tailed T Test was applied (Excel 2007, Microsoft). P values <0.05 were considered statistically different.