Tonantzitlolone cytotoxicity toward renal cancer cells is PKCθ- and HSF1-dependent.

Elucidating the targets and mechanism of action of natural products is strategically important prior to drug development and assessment of potential clinical applications. In this report, we elucidated the main targets and mechanism of action of the natural product tonantzitlolone (TZL) in clear cell renal cell carcinoma (CCRCC). We identified TZL as a dual PKCα and PKCθ activator in vitro, although in CCRCC cells its activity was mostly PKCθ-dependent. Through activation of PKCθ, TZL induced an insulin resistant phenotype by inhibiting IRS1 and the PI3K/Akt pathway. Simultaneously, TZL activated the heat shock factor 1 (HSF1) transcription factor driving glucose dependency. Thus, similar to the selective PKCθ activator englerin A, TZL induces a metabolic catastrophe in CCRCC, starving cells of glucose while simultaneously increasing their glycolytic dependency.


INTRODUCTION
Multiple signaling pathways are involved in regulating cell metabolism. Several, including components of the insulin pathway, are regulated by members of the protein kinase C (PKC) family. The multiple isoforms of PKC directly or indirectly regulate diverse aspects of metabolism in an isoform specific manner [1,2]. For example, PKCθ and -ε induce insulin resistance by inhibiting insulin receptor substrate 1 (IRS1) and the insulin pathway, while PKCδ and -λ have the opposite effect and stimulate the insulin pathway by activating IRS1 [2]. PKC isoforms can also regulate additional aspects of cell metabolism, e.g. PKCζ has recently been shown to regulate glutamine metabolism [3]. Therefore, developing isoform-specific regulators of PKCs might have useful clinical applications for metabolically deregulated diseases, including most epithelial cancers. One example of this is the recent identification of the natural product englerin A (EA; C 26 H 34 O 6 ) as a potent and selective PKCθ activator [4]. This finding is furthering our understanding of the role played by PKCθ in tumor tissues and identifies PKCθ interaction with the heat shock factor 1 (HSF1) transcription factor as a critical link between cell metabolism and cell stress response.
Tonantzitlolone (C 26 H 40 O 7 ) is a diterpene ester identified in 1990 from the native Mexican plant Stillingia sanguinolenta Müll. Arg. (Euphorbiaceae). The initial interest towards investigating S. sanguinolutea biomedical properties comes from native Mexican, Navajo, and Creek traditional use as a medicinal plant [5]. Reports have since addressed TZL's structure, absolute configuration, and total synthesis [6,5]. Although preliminary data have suggested that TZL might have an anticancer effect, a thorough characterization of the biological effects of TZL is lacking and its targets as well as its mechanisms of action are yet to be discovered. In this report, our goal was to identify TZL primary targets and to characterize the molecular effect of TZL on tumor cells.

Tonantzitlolone is a PKC activator with weak isoform selectivity
Analysis of NCI 60 selectivity data indicated that TZL had a pattern of response very similar to englerin A, with a Pearson correlation of 0.91 at the GI-50 level of response; Table 1 [8]. All other pure compound correlations (n = 9) above 0.8 were to englerin analogues, with one exception (#741581, sodwanone W [7]). In addition, quantitative structure activity relationship (QSAR) analysis revealed that TZL is structurally similar to the protein kinase C (PKC) activator bryostatin 1 and analogues, suggesting that TZL is likely to have an effect on the protein kinase C (PKC) family (Tables 2 and 3; [8]). To validate these results, we used a pan-PKC kinase assay. Whole protein extracts of the clear cell renal cell carcinoma (CCRCC) cell line 786-0 were directly treated with TZL (5 μM; 1 h at 30°C). We used the anticancer agent EA, a selective PKCθ activator [4], as a positive control. This assay revealed that TZL is a potent PKC activator ( Figure 1A). Because of the variable functions of the different PKC isoforms, we assessed TZL isoform selectivity by in vitro kinase assay using purified proteins. As shown in Figure 1B, TZL efficiently activated PKCθ and PKCα, but not PKCδ.

Tonantzitlolone is a potent anti-tumor agent
To further assess the biologic effect of TZL on the viability of tumor cells, we used the NCI-60 screen, a platform containing 60 different cancer cell lines [9]. Interestingly, TZL displayed a preferred cytotoxicity towards renal cancer cells (Figure 2). In a similar screen, EA also displayed renal selectivity [10]; however this was not seen with the pan-PKC activator bryostatin 1 ( Figure 3; bryostatin's data were extracted from the public repository database CellMiner; NSC#339555) [11]. These data suggest that the difference in toxicity might be due to a variation in isoform selectivity, and that TZL's effect might be PKCθ-dependent.

Tonantzitlolone cytotoxicity in renal cancer cells is PKCθ-dependent
We then sought to assess whether TZL cytotoxicity was dependent on a particular PKC isoform. Because of the direct effect of TZL on PKCθ and PKCα purified proteins ( Figure 1B), we separately silenced these two kinases in 786-0 and A498 cells using small interference RNA prior to exposure to TZL. As shown in Figure 4A-B, silencing of PKCθ partially rescued the cells from TZL cytotoxicity while silencing of PKCα did not. This suggests that PKCθ expression is necessary to mediate TZL cytotoxicity. HEK293 cells are not sensitive to TZL ( Figure 4A) and express low levels of PKCθ compared to 786-0 and A498, but express robust levels of PKCα ( Figure 4B; [6]), suggesting that TZL's effect in CCRCC cells is likely mediated by PKCθ. To further evaluate the effect of TZL on PKCθ signaling pathway in CCRCC cells, we treated 786-0 and A498 CCRCC cell lines with TZL and assessed the effect of treatment on the insulin pathway. As shown in Figure 4C, TZL increased phosphorylation of IRS1 on serine 1101 (a PKCθ-dependent phosphorylation site that inhibits IRS1 activity) and decreased Akt phosphorylation. Moreover, glucose uptake was also inhibited after TZL treatment in a PKCθ-dependent manner (5 μM, 16 h; Figure 4D), suggesting that TZL limits tumor cells access to glucose by inducing an insulin resistant phenotype via activation of PKCθ.
We previously established a link between PKCθ and HSF1, a transcription factor known to induce tumor glucose dependency [12,13], namely direct phosphorylation and heat-shock-independent activation of HSF1 by PKCθ [4]. We thus investigated whether TZL also induced a direct heat-shock-independent activation  Figure 5A). Furthermore treatment of 786-0 cells with TZL (5 μM, 24 h) increased HSF1 transcriptional activity as measured with a luminescence-reporter plasmid assay ( Figure 5B) and similarly to effects observed with EA, TZL cytotoxicity was observed only when HSF1 and PKCθ were simultaneously overexpressed in HEK293 cells ( Figure 5C-5D). Also, TZL's cytotoxicity was dependent on the extracellular glucose concentration ( Figure 5E). Indeed, in low glucose media (1 g/L), TZL was significantly more cytotoxic than in high glucose media (4.5 g/L).Together these data demonstrate the ability of TZL to induce glucose dependency in CCRCC cell lines.

DISCUSSION
Understanding the intertwined role that members of the PKC family play in signaling pathways and cancer has the potential to lay the foundation for development of novel targeted therapies. The resurgent interest of the scientific community in understanding the molecular     Clear cell renal cell carcinoma is the most common type of both sporadic and hereditary kidney cancer. It is primarily induced by mutations occurring in the tumor suppressor gene VHL [14]. Defective VHL protein expression leads to stabilization of hypoxia-inducible factors 1and 2 [15,16], increased DNA damage [17], and to activation of the PI3K/Akt pathway stimulating cell proliferation and tumor growth [18]. Thus VHL-deficient cells display a pseudo-hypoxic phenotype with a restructured metabolism oriented toward aerobic glycolysis and favoring anabolic growth [12]. Interestingly, PKC members have been shown to be induced by hypoxia and novel PKC isoforms are known to interact with VHL, suggesting that VHL might directly regulate the expression of some of the PKC isoforms. Also, besides a well-characterized role in regulating cell cycle and survival [19,20,21], PKC isoforms are also critical in regulating metabolic processes [2,3]. Hence, CCRCC cells are useful tools for studying PKC isoform functions. Although structurally different from the natural product EA [5,10], TZL has a similar profile in the NCI-60 screen suggesting that the two compounds might share a similar mechanism of action. Using in vitro kinase assays, we showed that TZL activated PKCθ and not PKCδ and that its cytotoxicity was PKCθ dependent in CCRCC cells. Because of the limited expression profile of PKCθ to some solid tumors and immune cells, and the fact that HSF1 is often overexpressed in tumors, TZL might only be lethal to PKCθ-expressing, HSF1-addicted glycolytic tumor cells.
Despite the well-characterized role of HSF1 in cancer development and progression [13,12,22,23], in the context of PKCθ activation and the inhibition of the insulin pathway, non-heat-shock HSF1 activation appears to provide a therapeutic advantage. HSF1 is a tumor growth facilitator that enhances tumor glucose dependence via a heat-shock-independent transcriptional program [12]. Since TZL activates HSF1 and its cytotoxicity is inversely related to extracellular glucose concentration ( Figure 5E), the fact that HSF1 increases glucose dependency might provide a rationale for why activation of HSF1 underlies the cytotoxicity of PKCθ activators such as TZL.
In conclusion, our data suggest that TZL and the PKCθ activator EA share a similar mechanism of action in CCRCC cell lines-that is PKCθ activation that leads to a metabolic catastrophe by simultaneously inhibiting the insulin pathway via inhibitory phosphorylation of IRS1 while inducing glucose addiction via HSF1 activation. These data also further support PKCθ as a strategic therapeutic target for patients with highly glycolytic tumors, such as CCRCC.

Prediction of tonantzitlolone targets
Prediction of TZL target was performed using the software Metadrug (Genego Inc, Carlsbad, CA, USA), a systems pharmacology platform using QSAR modeling to analyze and compare biological effects of small molecule, as previously described [4]. The COMPARE algorithm was essentially as described in [8].

Non-radioactive PKC kinase assay
PKC kinase activity of cell lysates was measured using the pan-PKC activity assay from EnzoLife Sciences. Cells were lysed in TNESV lysis buffer (50 mM Tris, 1% Nonidet P-40, 2 mM EDTA, 100 mM NaCl and 2 mM Na 3 VO 4 ). After centrifugation (15 min; 15,000 rpm; 4ºC), clarified supernatant was incubated with 10 μM TZL in the kinase buffer provided by the manufacturer (1 h at 30°C with 10 μM ATP). The reaction was stopped and the phosphorylation of a PKC substrate was measured by spectrophotometry as indicated in the manufacturer protocol. PKCα, -δ, and θ kinase assays were performed in a similar manner using 5 ng of purified PKC proteins (incubation for 1 h at 30°C with 10 μM ATP).

Radioactive in vitro kinase assay
Purified HSF1 (50 ng) was incubated with purified PKCθ (50 ng) in presence or absence of TZL (1 μM). Reactions were initiated by the addition of 10 μM nonradioactive ATP and 6 μCi (0.2 μM) of [ 32 P]-ATP and incubated at 30°C for 30 min with periodic mixing. Proteins in the kinase reactions were separated by SDS-PAGE and transferred to PVDF membrane. Phosphorylation of HSF1 was assessed by radiography of PVDF membranes. Total HSF1 was then immunoblotted to ensure equal loading.

Cell lines and cell culture
The sporadic VHL-deficient kidney tumor cell lines 786-0 and A498, and the embryonic kidney epithelial cell line HEK293 were all purchased from American Type Culture Collection (ATCC; Manassas, VA, USA). Cells were cultured in Dulbecco's modified Eagle's medium High Glucose, pyruvate-free (DMEM; Cellgro) supplemented with 10% fetal bovine serum (Invitrogen, Grand Island, NY, USA). Viability experiments were performed in serum-free media.

Analysis of cell viability in vitro
Cell viability was assessed by Thiazolyl Blue Tetrazolium Bromide (MTT) assay as previously described [24]. Briefly 5, 000 cells/well were plated in 96-well plates and treated as indicated in the figure legends prior to assessing cell viability by MTT assay.

Immunoblot analysis
Analysis of protein expression and/or phosphorylation was made by immunoblotting as previously described [25]. Briefly, 20 μg of protein were separated in 4-20% SDS-PAGE gels (Bio-Rad, Hercules, CA, USA) and transferred onto PVDF membranes. Membranes were blocked for 1 h at room temperature in 5% fat-free milk diluted in TBST (10 mM Tris-HCl, 100 mM NaCl, 0.5 M EDTA, 0.1% Tween 20). Primary antibodies diluted in 5% fat-free milk were incubated overnight at 4ºC. After 3 washes, horseradish peroxidase-linked secondary antibodies (Sigma-Aldrich, St Louis, MO, USA) were incubated for 1 h at room temperature. Signal was detected using the ECL protein detection system (Pierce, Thermo Scientific, Rockford, IL, USA). PKCθ antibody was from Abcam (Cambridge, MA); antibodies for IRS1, IRS1-ser1101, HSF1, tubulin, Akt, and p-Akt were from Cell Signaling Technology (Danvers, MA, USA); HSP70 antibody was from Thermo Scientific.

Glucose uptake assay
Glucose uptake was measured using a fluorescent non-metabolizable D-glucose analog 2-[N-(7-nitrobenz-2-oxa-1, 3-diazol-4-yl) amino]-2-deoxy-D-glucose (2-NBDG, Cayman Chemicals, Ann Arbor, MI) as previously described [26]. Briefly 3, 000 cells were plated in black-well 96-well plates and were transfected the day after with small interference RNA against PKCθ or PKCα (10 nM; Santa Cruz Biotechnology, Dallas, TX, USA) using Dharmafect 3 transfection reagent as recommended in the manufacturer instructions (Dharmacon, Thermo Scientific, Pittsburgh, PA, USA). Eight hours post-transfection, cells were treated as indicated (16 h). At the end of the treatment, cells were incubated for 45 minutes in KREB buffer containing 1g/L glucose in presence or absence of 20 μM 2-NBDG. Cells were then washed 2 times for 10 minutes with PBS to remove all residual extracellular 2-NBDG. The amount of 2-NBDG imported into the cells was measured by assessing fluorescence at 488 nm.

HSF1 transcriptional activity reporter assay
HSF1 transcriptional activity reporter assay was performed as previously described [4].