p53 deficiency resulted in enhanced IGF-1R signaling pathway inhibition by metformin in lung adenocarcinoma cells

Research in recent years revealed that the antidiabetic drug metformin has some anticancer properties, and p53 plays an important role in the metformin-induced tumor inhibition. But less clear is the mechanism by which p53 status affects metformin mediated cancer cell responses. In this study, the effect of metformin on cancer cells with different p53 genotypes was investigated, and we found metformin could inhibit cancer cell proliferation especially in p53-deficient cells. So a paired isogenic nonsmall cell lung cancer cell lines H1299 p53+/+ and H1299 p53-/was involved to explore the possible mechanisms. We detected the differences in cell proliferation and solid tumor formation induced by metformin, furthermore, the signaling transduction of insulin like growth factor-1 receptor (IGF-1R) was evaluated, which was essential in cancer transformation and progression. Our results indicated that p53 deficiency lead to more efficient inhibition of IGF-1R signaling transduction, receptor degradation and ubiquitination induced by metformin, which resulted in enhanced cell apoptosis and proliferation inhibition in vitro, and more significant solid tumor growth retardation and cancer cell death in vivo. Although more cell lines and multiple experiments were required to make a general conclusion, our study suggested the enhanced inhibition of IGF-1R signaling pathway contribute to the metformin’s selective toxicity in p53deficient lung adenocarcinoma cells.


INTRODUCTION
The biguanide metformin is the most widely used treatment drug for type 2 diabetes. It is believed to exert its effect by reducing hepatic glucose production and increasing insulin sensitivity as well as glucose use by peripheral tissues [1]. Recent studies showed encouraging findings that metformin could reduce cancer risk or improve cancer prognosis both in vitro and in vivo [2]. Two types of distinct but not exclusive anticancer effect were concluded: first, by decreasing insulinemia and glycaemia, metformin could block PI3K/MAPK signaling pathway implicated in cell growth; second, through the activation of the AMPK pathway, metformin could directly act on cancer cells by targeting various pathways including tumor metabolism, inflammation, angiogenesis and cancer stem cells [3]. The AMPK pathway was agreed to play the major part in the anticancer mechanisms of metformin.
The TP53 tumor suppressor gene is one of the most frequently (>50%) mutated genes in lung cancer [4], which has also been associated with poor progress and therapy resistance [5]. Generally, functional p53 was crucial in the regulation of cell cycle, apoptosis and senescence in response to DNA damage, hypoxia, and oncogenic activation [6]. Also as an important regulator of cancer cell metabolism [7], the loss of p53 in cancer cells might be a main force behind the acquisition of glycolytic phenotype [8], lead to the abnormality in cancer cell glucose metabolism. Recent reports have revealed www.impactjournals.com/oncotarget/ Oncotarget, Advance Publications 2017 that p53 plays an important role in the metformininduced inhibition of proliferation and AMPK activation [3,9,10]. And data from present study demonstrated p53 knockdown or mutation has a negative effect on metformin-induced growth inhibition, senescence and apoptosis [11]. However, the possible mechanism by which p53 status influenced metformin mediated cellular responses is still an important question to be addressed.
The insulin like growth factor-1 receptor (IGF-1R) is an important upstream regulator of AMPK pathway [12], and IGF-1R signaling plays a critical role in cancer transformation and progression [13]. Overexpression of IGF-1R is observed in wide range of human cancers including lung cancer, and it is becoming one of the most intensively investigated molecular targets in oncology [14]. Recently, there are researches on the emerging effects of metformin on IGF-1R system. In our previous studies, we found metformin could inhibit IGF-1R signaling pathway and have synergistic effects with IGF-1R antibody in both small cell and non-small cell lung cancer cell lines [15,16]. And similar regulation effect of metformin on IGF signaling was found in endometrial cancer cells [17].
In this study, the proliferation inhibition of metformin on cancer cells with different p53 genotypes was investigated, and a paired isogenic cell lines H1299 p53 +/+ and H1299 p53 -/was involved to explore the difference caused by p53 deficiency in vitro and in vivo. Moreover, we detected the changes in IGF-1R signaling pathway to explore the possible mechanisms of p53 status in metformin-induced antitumor effects.

Metformin-induced proliferation inhibition in cancer cells with different p53 status
The p53 expression and transcription level in five different non-small cell lung cancer cell lines were checked by WB and RT-PCR. In accordance with previous researches, p53 protein and mRNA was detected in A549 and H460 cells (p53 wild type), but not in H520, H1975 (mutant type) and H1299 (deleted type). The cell viability and IC50 of metformin were detected by PrestoBlue Cell Viability Assay, and we found that the p53 mutant and deleted type cancer cells (H520, H1975 and H1299) were less resistance to metformin induced antitumor effects, with statistically significant lower metformin half maximal inhibitory concentration (IC50) and cell viability curves ( Figure 1A&1B). Then flow cytometry was performed to evaluate the metformin induced apoptosis of cancer cells with different p53 status, and we found both the early and late cell apoptosis of p53 mutant and deleted type cancer cells were significantly increased than p53 wild type cancer cells ( Figure 1C). Cancer cell proliferation and apoptosis of H1299 p53 +/+ and H1299 p53 -/-Since cancer cells with different p53 status showed significant difference in the treatment reaction to metformin, a paired isogenic H1299 p53 +/+ and H1299 p53 -/was enlisted in further experiment to confirm the difference caused by p53 deficiency and to explore the possible mechanism.
The cell viability and apoptosis experiments were repeated with new cell lines, and once again, metformin significantly inhibit cancer cells proliferation especially in p53-deficient cells, more significant reduction in cell growth and increased cell death was detected, and dosedependent inhibition was shown in Figure 2.

Solid tumor formation and cancer cell apoptosis
In order to investigate the possible changes in tumor formation ability of cancer cells with different p53 status, we subcutaneously transplanted H1299 p53 +/+ and p53 -/cells in BALB/c-nu mice, and the inhibition of solid tumor formation by metformin was detected. The tumor growth in size and weight was recorded and cancer cell apoptosis was evaluated. We found that metformin could significantly inhibit solid tumor growth and induce more cancer cell apoptosis in both H1299 p53 +/+ and p53 -/cells compared to control groups, however, more significant retardation of tumor development and cancer cell death was found in the tumor model of p53-deficient cells ( Figure 3).

IGF-1R tyrosine phosphorylation and signaling transduction
To explore the possible mechanism by which p53 status affect the antitumor reactions induced by metformin, we investigated the difference of H1299 p53 +/+ and p53 -/cells in IGF-1R signaling pathway. Following serum starvation for 12 h, H1299 p53 +/+ and p53 -/cell lines were treated with 50 ng/ml IGF-1 or first treated with 3 mM metformin for 30 min and then stimulated with 50 ng/ml IGF-1 for 2, 5, 10, 30 and 60 min. As seen in Figure 4, metformin could significantly decrease the IGF-1 (the IGF-1R ligand) induced IGF-1R tyrosine phosphorylation in both the two cell lines. And also the IGF-1 induced Akt and ERK activation was significantly inhibit, which are the two main proteins of IGF-1R signaling transduction. Moreover, there was significant difference in both IGF-1R phosphorylation and signaling transduction between the two cell lines with or without p53 deficiency ( Figure 4C). It could be concluded that metformin could more efficiently inhibit IGF-1R tyrosine phosphorylation and signaling transduction in p53deficient cancer cells. www.impactjournals.com/oncotarget

IGF-1R degradation and ubiquitination
Receptor degradation and ubiquitination both were important biological behaviors, and could determine the receptor activity to some extent. To detect the IGF-1R degradation, H1299 p53 +/+ and p53 -/cells were treated with 100 ng/ml IGF-1 or 5 mM metformin in serum free medium for 0, 6, 12, and 24 h. We found that metformin significantly down-regulated the IGF-1R expression as the ligand IGF-1 ( Figure 5). For the ubiquitination, cells were treated with 100 ng/ml IGF-1 or 5 mM metformin for 10 min after 24 h serum starvation, and then immunoprecipitation and immunoblotting as above. The results indicated that metformin could induce the IGF-1R ubiquitination after 10 min stimulation in both the two cell lines. Moreover, ubiquitination induced by metformin was much stronger in H1299 p53 -/cells, which indicated metformin-mediate IGF-1R ubiquitination was enhanced in the absent of p53 protein.

DISCUSSION
Metformin has recently received increased attention for its potential antitumor effects in clinical and experimental studies. The anticarcinogenic effects of metformin have been attributed including activation of AMPK pathway, induction of cell cycle arrest and/ or apoptosis, inhibition of protein synthesis, reduction in circulating insulin levels, inhibition of the unfolded protein response (UPR), activation of the immune system, and eradication of cancer stem cell (well reviewed in [1,3,18]). Besides these, in our study we found metformin could inhibit phosphorylation of IGF-1R and downstream Akt and ERK signaling, induce significant IGF-1R degradation and ubiquitination, which lead to cancer cell proliferation inhibition in vitro and solid tumor growth retardation in vivo. Moreover, we found p53 played an important role in these metformin induced IGF-1R dependent tumor suppression.
Recent studies have revealed the undeniable fact that most human cancers display enhanced IGF-1R concentrations and express high IGF-1R mRNA levels [19,20]. IGF-1R activation initiates downstream cascades which are essential in cancer cell metabolism, mainly including phosphatidylinsitol-3-Kinase (PI3K), Akt and mTOR; and Ras, Raf and MAPK/ERK [21]. Activation of these pathways led to oncogenic transformation, proliferation and survival of cancer cells [22]. In our study, we found metformin could significantly inhibit IGF1 (the ligand) induced IGF-1R activation and signaling transduction, and the downstream Akt and ERK activation was believed to mediate AMPK inhibition. In another word, metformin induced IGF-1R signaling inhibition also paved the way to metformin induced AMPK activation, this feedback loop might be a good supplement for the anticarcinogenic effects of metformin as listed above.
In addition to its role in cell growth-inducing activities, IGF-1R was reported to have important function in metabolic regulation [23]. The PI3K pathway, which is activated following IGF-1R activation, not only provides antiapoptotic and mitogenic signals but also has significant effect on cancer cell metabolism. The downstream proteins of IGF-1R, Akt/protein kinase, mTOR induce energy production in cancer cells and stimulate lipid and protein synthesis, which is responsible for a number of metabolic adaptations [24,25]. So inhibiting the IGF-1R pathway may induce significant changes in cancer cell metabolism and lead to retardation in proliferation, which is consist with our founding in this study.
p53, which was reported as tumor suppressor, has also an important role in the control of cell metabolism [26]. Activation of p53 allows cells to respond to glucose deprivation by arresting their proliferation until glucose is restored. And it is conceivable that loss-offunction mutation of p53 in cancer cells may lead to stimulation of the glycolytic pathway [23]. Moreover, recent reports have revealed that p53 acts as a major energy conserving component following metformininduced AMPK activation [9,10]. Tumor cells with loss of function of p53 are less able to compensate the metabolic stress induced by metformin and are more likely to undergo an energetic crisis because of their inability to stop proliferation [3], so p53-deficient cells showed an impaired ability to survive when treated with metformin compared with p53 +/+ cells, which was highly in accordance with our data here.
As an upstream regulator of the IGF-1R gene, p53 effects converge on decreased IGF/IGF1R signaling by reducing both IGF bioavailability and IGF-1R density [27]. Cross-talk and coordination between the p53 and IGF-1R/Akt/TOR pathways have been proposed as means  of integrating growth factor signaling, changes in nutrient levels and stress signals into regulation of cell growth, mitogenesis and apoptosis [28]. The IGF-1R signaling tends to be more malignant due to p53 deficiency, but seemed to be more susceptible to metformin treatment. Our data here indicated that more efficient retardation of IGF-1R signaling pathway and receptor degradation and ubiquitination were induced by metformin in the absent of p53 protein, resulting in enhanced proliferation inhibition, cell apoptosis in vitro and more significant solid tumor growth retardation and cancer cell death in vivo.
Although more cell lines and multiple animal experiments were required to make a general conclusion, our study suggested that the change in sensitivity of IGF- Intensity of bands for IGF-1R at different time points were quantified by densitometry, normalized to GAPDH, data were presented as mean ± S.E.M in the curves. There was significant difference between H1299 p53 +/+ and H1299 p53 -/cancer cells in both IGF1 and metformin induced receptor degradation ( * P<0.05, ** P<0.01). (C) Cells were simulated with 3 mM metformin for 10 min after serum starvation for 24 h. Cells were harvested with IP buffer. The IGF-1R antibody was used for immunoprecipitation, and immunobloting with ubiquitin antibody. Ubiquitination induced by metformin was significantly stronger in H1299 p53 -/cancer cells (P<0.05). www.impactjournals.com/oncotarget 1R signaling pathway to metformin may be the possible mechanism of metformin's selective toxicity in p53deficient cancer cells.
However, the clinical use of metformin as an antitumor agent is still awaited, although there were ongoing trails amassing evidence to ascertain if a survival benefits is associated with the use of metformin in various malignancies, the potential use of metformin as an immunotherapy agent needs to be substantiated with further evidence [29].

Cell viability assay and flow cytometry
Cells were cultured in 96 well plates overnight in the presence or absence of serum. Appropriate substrates were added to the wells for 12, 24, 48h, and then the cell viability was measured by PrestoBlue Cell Viability Assay (Invitrogen, CA, USA). Fluorescence was measured with excitation at 560 nm and emission at 590 nm wavelengths. Cell number was calculated from standard curve of fluorescence measurement.
For flow cytometry assay, cells were cultured in 6 well plates overnight and metformin was added to the wells for 24 h. After that cells were collected in tubes and treated in accordance with Annexin V-FITC/PI apoptosis detection kit (Invitrogen, CA, USA). The cells apoptosis were imaged under a fluorescent microscope using 488 nm excitation, 515 nm emission for FITC and 560 nm emission for PI. Dead cells were labeled with Annexin V (−) propidium iodide (PI) (+) and are shown in the Q1 area; late apoptotic cells were labeled with Annexin V (+) PI (+) and are shown in the Q2 area; live cells were labeled with Annexin V (−) PI (−) and are shown in the Q3 area; early apoptotic cells were labeled with Annexin V (+) PI (−) and are shown in the Q4 area (as shown in Figures 1C&2B).

SDS-PAGE and western blotting
Protein samples were dissolved in LDS sample buffer (Invitrogen, CA, USA) containing 1 mM β-mercaptoethnol, and analyzed by SDS-PAGE with 4-12% Bis-Tris gel (Invitrogen, CA, USA). After separation, the proteins were transferred to nitrocellulose membranes (GE Healthcare) at 4°C for 1 hour and then blocked for 1 hour at room temperature in a solution of 5% (w/v) skimmed milk powder and 0.1% (w/v) Tween 20 in Tris-buffered saline (TBS), pH 7.5 (TBS-T). After Incubated with appropriate primary antibodies overnight at 4°C, membranes were washed three times in TBS-T, and then incubated with a horseradish peroxidase-labeled secondary antibody (Pierce, IL, USA) for 1 hour at room temperature. The detection was made with ECL substrate (Pierce, IL, USA) and exposure to x-ray film.

Real-time PCR and immunoprecipitation (IP)
Total RNA of different cell lines was obtained using TRIzol reagent (Invitrogen), according to manufacturer's instructions. The RNA concentration was determined spectrophotometrically and intensity was detected by gel electrophoresis, and quality was confirmed in an Agilent 2100 Bioanalyzer (Agilent Technologies). The P53 mRNA was quantified by SYBR Green-quantitative real-time PCR and normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as in our previous study [30]. The PCR primers used were as follows: P53 forward 5'-CCTCAGCATCTTATCCGAGTGG-3' and reverse 5'-TGGATGGTGGTACAGTCAGAGC-3', GAPDH forward 5'-TGACCACATCCTATGCCACT-3' and reverse 5'-CCACCTTCTTGCTGTAGCT-3'.
Cells were lysed with IP lysis buffer supplied by Invitrogen, using 500 μl lysis buffer with 10 mM N-ethylmaleimide, 50 μM MG132, and protease inhibitor cocktail tablet (Roche) per well in the 6-well plates. After incubated with 1 mg antibody for 3 h at 4°C, the samples were added with 15 μl Dynabeads Protein G (Invitrogen, CA, USA), and incubated on a rocker platform at 4°C overnight. Then the immunoprecipitated complexes were collected by magnet, the supernatant discarded, and the beads were washed three times with lysis buffer and dissolved in sample buffer for western blot.

Solid tumor formation and TUNEL
BALB/c-nu mice (female, 5-6 weeks old) were purchased from vital River Laboratory Animal Technology (Beijing, China) and maintained under specific pathogenfree conditions. H1299 p53 +/+ and p53 -/cells (4×10 6 ) were subcutaneously injected. Tumor sizes were measured by longitudinal (L) and transverse (W) diameters every other day and tumor volumes were calculated as V=L×W 2 /2 [31]. The endpoint for this study was set to be three weeks after tumor cells administration, and then all mice were sacrificed under anesthesia and tumors were exercised and weighed. Tissue samples were embedded in paraffin wax and serially sectioned. All animal experimental protocols were approved by the Ethics Committee of Shandong University.
Tumor sections were deparaffinized and rehydrated, incubated with protease K for 25 min at room temperature for antigen retrieval, and then operated according to the protocol of the TUNEL apoptosis detection kit (Zhong shan jin qiao, China). The red-labeled TUNEL positive cells were imaged under a fluorescent microscope using 488 nm excitation and 530 nm emission. An image analyser (Media Cybernetics) was used to score the staining sections of TUNEL, and results were present in mean optical density (MOD = integrated optical density/ area) as in our previous study [32].

Statistical analysis
All experiments were repeated at least three times and the values were expressed as the mean ± standard error. Statistical significance was determined by Student's t-test using SPSS 17.0 statistical software (SPSS Inc, Chicago, IL). For all tests, a p value less than 0.05 was considered to be statistically significant.