Gundelia Tournefortii Extracts Inhibit Progressions of Hepatocellular Carcinoma in Mice Model Through Decrease in p53/Akt/PI3K Signaling Pathway

Background: Gundelia (G.) tournefortii is one of the most famous plants used in Palestine for its traditional curative and nutritive properties. We evaluated effects of Gundelia (G.) tournefortii aqueous-extract in an ex-vivo and in in-vivo animal model of hepatocellular carcinoma (HCC). Methods: HCC cell-line (Hep3B) were injected in the back of male mice on NOD.CB17-Prkdc-scid/NCrHsD background. Tumor size, serum a-fetoprotein (αFP) and Glypican-3 (GPC3) were assessed at day 2 following HCC injections till day-12. G. tournifortii extracts were injected i.p at day 10 in HCC mice. Liver histology, hepatic-p53, p-Akt and p-PI3K expressions were evaluated. Primary-hepatocytes from HCC mice were ex-vivo treated with G. tournifortii and investigated for perturbation of the DNA cell cycle using propidium iodide (PI) staining assay, while apoptosis was estimated by staining with Annexin-V and PI by the ow cytometry. Results: Tumor mass increase in animals with HCC and was associated with elevated aFP and GPC3 (P<0.05). Mice receiving G. tournefortii extracts showed a signicant decrease in tumor at days 12 with decreased hepatic-p53 and phosphorylated Akt/PI3K signaling pathway (p=0.001). Histology H&E staining showed remarkable reduction in inammatory lesions in HCC livers receiving G. tournifortii extracts in line with delay in G2-M phase of HCC-primary-hepatocytes to 1.39-folds. Moreover, apoptotic activity was mostly enhanced by the 200mg/ml G. tournifortii extracts while a reduction of HCC necrosis to 2.4 fold (P<0.01). activity against human breast adenocarcinoma cell (MCF-7). In our current study, we choose to use G. tournifortii extracts in water solution to prevent any potential harm and hepatotoxicity with other organic solvent. We showed that the ower buds of G. tournifortii suggest an interesting anticancer agent that can potentially act on several levels of the cell cycle including inhibited molecular expressions of p53 and cell proliferations through Akt/PI3K, arresting the most attractive phase of mitosis (G2-M phase) and enhancing cell apoptotic activity. These anti-proliferative effects could be explain in part as suppression of serum tumor marker of αFP and GPC3.


Background
Herbal supplements and functional food products have gained increased interest recently due to their nutritional and health bene ts. [1][2][3]. Gundelia (G.) tournefortii is commonly known as Tumble thistle or tumbleweed, and a'kub in Arabic. This annual herbaceous plant belongs to Asteraceae (Compositae) family and native to the Irano-Turanian region and may be found in Mediterranean regions of the east and grow in semi-desert or sandy plains in Palestine, Jordan, Syria, Iraq, Iran, Azerbaijan, Anatolia, Armenia, Turkey and other areas [4][5][6].
G. tournefortii is a spiny thistle-like perennial plant about 60 cm height, and its stems branch from the base and almost hairless. Its leaves are leathery, rigid, and alternating with prominent yellow, red, or purple veins; pinnati d to pinnatisect; with spiny-dentate lobes [6]. While the ower can be white, yellow, burgundy, red or green in color [4]. It has been used to make traditional recipes and have been mentioned in local folkloric songs and proverbs [7]. Its use in food is very ancient, more than 2000 years ago, found in the Talmud of Babylonia and in the Biblical writings [8]. According to al-Muqaddasi, a Jerusalem geographer G. tournefortii was one of the plant species with which Palestine was privileged. Because of the excessive commercial gathering of G. tournefortii and thus the decline in its population, there were limitations over its gathering to become for domestic use only [9].
The stems, owers, leaves, and seeds of G. tournefortii can be used as food. In Palestine, Jordan, and Syria the tender plants ower buds are collected and cooked before owering [4,5]. In Turkey, dried plant is stacked for winter food, latex is used to make chewing gum and the seeds are used as coffee [10].
In folk medicine, the plant stalk is considered hepatoprotective and blood puri er in Iran and its latex is used for burning off the warts, drying up sores, and as an emetic and a cure for snakebite in Lebanon [4,5]. Moreover, it used in traditional medicine for the treatment of kidney diseases, anorexia, heart stroke, chest pain, gastric in ammation, diarrhea, bronchitis, gingivitis, vitiligo, and diabetes [4].
One of the fundamental features of cancer is tumor clonality, which means the development of tumors from single cells that begin to proliferate abnormally [12][13][14][15]. This process can be retarded by activities such as apoptosis, cytotoxicity, and anti-proliferative activity [16]. According to 2018 WHO statistics, liver cancer was among the most encountered causes of cancer mortality with 782,000 deaths by the end of that year [17 − 9]. For these reasons, the current study aims to determine the anti-hepatocellular cancer activity of G. tournefortii in an vivo and ex-vivo models, in speci c its' owering bud part.

Material And Methods
Plant material and extraction method The fresh owering buds of G. tournefortii were purchased from local Palestinian markets (Nablus), the plant was taxonomically characterized in the Pharmacognosy Laboratory at An-Najah National University, and the voucher specimen code is Pharm-PCT-1133. The fresh plant sample was washed several times using distilled water to remove any contaminates.
The cleaned owering buds of G. tournefortii (200 g) were cut in small pieces and boiled in a beaker with 1L of distilled water for 30 min. The boiled mixture was ltered using Buchner funnel. The extract of the sample was lyophilized into a powder form by using a freeze-drier. For cytotoxicity assessment, each G.
tournefortii extracts were diluted in sterile water in a concentration of 100 mg/ml. Cytotoxicity evaluation was carried out by using the primary isolated hepatocytes form HCC mice as compared to their counterparts from the mice received the vehicle treatments.

HCC mice models
Male mice on NOD.CB17-Prkdc-scid/NCrHsD background, 12 weeks of age, weighing 22 ± 0.5 g, received care according to ethic regulations of the An-Najah National University and NIH guidelines. Mice were purchased commercially from Harlan Laboratories, Jerusalem-Israel. All animal protocols were approved by the institutional animal care ethical committee at the An-Najah National University, and housed in a barrier facility. For the xenograft model 3.0 million of Hep3B cells/100ml (human liver hepatocellular carcinoma cell line) were injected subcutaneously at the back (N=6 animals). Tumor masses and mice weight were daily monitored macroscopically for 12 days at day 2 following cell injections. At day 10 of injections, One group (N = 6) received G. tournefortii extracts made with water were i.p injected at a dose of 60 mg/kg body weight, while the other group (N = 6) received only the vehicle (100% water). For tumor serum marker measurements, tail blood samples were drawn every 2 day intervals from each animal starting from day 2 (following Hep3B-injections). At sacri ce (day 12), tumor masses were collected and liver were obtained for both molecular biology and histopathology analyses and tumor size (weight and volumes). Serum samples were obtained for evaluating tumor serum levels. The animals were terminated intramuscularly with 0.1 ml of ketamine: xylazine: acepromazine (4:1:1) per 30g of body-weight prior to cervical dislocation.

Primary hepatocytes isolations
Brie y, abdominal area fur were cleaned with 70% alcohol and a midline incision to expose abdominal site. Livers were perfused from catheter into portal vein with a pump at a rate of 4.5 ml/minute with perfusion medium (ThermoFisher, Cat # 17701-038) and later for another 6-7 minutes with liver digest medium (ThermoFisher, Cat # 17703-034). Livers were then placed in a 100 mm plate lled with cold 20ml washing medium, were cut into pieces and transferred through 100/70µm lters into 50ml centrifuge tubes. Cells were centrifuged at 50Xg for 3 minutes at 4°C. We decant the supernatant and added 20ml 40% cold percoll (Sigma P1644-500ML) to each tube. We then centrifuged at 150-200Xg for 7 minutes at 4°C and viable hepatocytes were collected from the bottom of the tubes. Washed hepatocytes were cultured in collagen-coated plate at a concentration of 2.2X10 5 cells/ 6-well. Cell were incubated at 37°C incubator with 5% CO2 for 2.5-3.5 hours prior to washing once to remove dead cells and a new medium were then placed for additional 24 hours. The next day, medium were changes and hepatocytes were used for our experiments on the next day. Hepatocytes condition was accomplished by using Williams E' medium (ThermoFisher, Cat number: A1217601) supplemented with 6% FBS, glutamine, dexamethasone, glucagon and insulin. For the in vitro assay, G. tournefortii extracts made in water were incubated with the obtained primary hepatocytes from the HCC mice model (10 6 /ml) in concentrations of 50, 100 and 200 µg/ml for 24 hrs/37 o C.

Viability assay
The viability of hepatocytes were determined by trypan blue staining. In brie y, 100 μl of cells was aseptically transferred to a 1.5 ml clear tube and incubated for 3 min at room temperature (25 °C) with an equal volume of 0.4% (w/v) trypan blue solution (Sigma, USA). Cells were counted using a dual-chamber hemocytometer and a light microscope. Nonviable cells were stained blue and viable cells were unstained. These two types of cells were recorded separately, and the means of six independent cell counts were pooled for analysis.
Histological assessments of liver injury The posterior one third of the liver was xed in 10% formalin for 24 hours and then para n-embedded in an automated tissue processor. Seven-millimeter liver sections were cut from each animal. Sections ( resolved on a 10% (w/v) SDS-polyacrylamide gel (Novex, Groningen, The Netherlands), under reducing conditions. For immunoblotting, proteins were transferred to a Protran membrane. Blots were then incubated for 1 hr at room temperature in a blocking buffer containing 5% skim milk (w/v). Next, the blots were incubated with rabbit anti-mouse p53/p-AKT/ PI3Ks, (R&D System, Minneapolis, MN) diluted 1:1000 overnight in 4 o C, and subsequently, with peroxidase-conjugated with Anti-rabbit (Abcam, Israel, diluted 1/5000), for 1.5 hr at room temperature. Immunoreactivity was detected using an ECL kit (Abcam, Israel).

Flow cytometry analysis
Following cultures, the adjustment of the harvested primary hepatocytes to 10 6 / ml in staining buffer (in saline consisting of 1% bovine albumin was achieved. For viability measurements and apoptosis, the staining of fragmented DNA by propidium-iodide (PI) and the staining of phosphatidylserine by using annexin V-conjugated to FITC was done according to the instructions of the manufacturer (R&D System, Minneapolis, MN). After that, the apoptosis was marked as annexin-V (+) but propidium-iodide (

Results
G. tournefortii extracts decreased tumor mass and in part maintained liver histological characteristics in animals with hepatocellular carcinoma (HCC) HCC model was performed to study tumorigenicity outcome. Anti-cancer effects of G. tournefortii was evaluated through injecting the extracts as mentioned in materials and methods. Hepatic tumor mass and size were obtained at the end of experiment post Hep3B (HCC)-injections. Figure 1a shows macroscopic examination of the tumor extracted from the back of a representative mouse with HCC and a HCC mouse receiving the G. tournefortii extract as compared with the vehicle mice. Mice with the HCC mouse receiving the G. tournefortii extract had tumor size of 0.9 ± 0.1 cm as compared to 2.0 ± 0.6 cm in the HCC implanted counterparts. Figure 1B shows the gradual increase in tumor size following HCC injections. At day 10, mice groups were split into two groups; one group received the vehicle and the second group received extracts of G. tournefortii. Tumor size continued to increase with the vehicle on day 12 post-HCC injections, while HCC implanted mice receiving extracts of G. tournefortii showed a signi cant decrease in tumor size at day 12 (p = 0.01). To characterize whether liver histology is altered following the i.p injections of the G. tournefortii extract, we stained for H&E. The vehicle treatment together with the G. tournefortii extract showed no histopathological effects on the liver tissue (Fig. 1c&d). While the HCC implanted mice group exhibited hemorrhage, necrosis, dysplasia and hepatic cell carcinoma (Fig. 1e); the HCC implanted mice receiving the G. tournefortii extract showed a partial reduction in necro-in ammatory regions and less hemorrhage was observed (Fig. 1f).
HCC progression is known to be driven by chronic in ammation. Macrophages play a crucial role in chronic liver in ammation [18]. The tumor microenvironment plays a key role in the progression of HCC. Tumor-associated macrophages are a well-known component of the tumor microenvironment and abundantly in ltrate HCC microenvironment [19]. The roles of macrophages in the development and progression of HCC have been recognized [20]. For this purpose, immuno uorescence staining of liver macrophages from our mice model groups were examined by confocal microscopy. Images of confocal microscopy show presence of macrophage marker F480 (green) on liver sections obtained from HCC ( Fig. 1i) and HCC with the G. tournefortii extract (Fig. 1j). No F480 in ltrates were seen in the vehicle mice treated (Fig. 1h) or untreated (Fig. 1g) with the G. tournefortii extract. To quantify F480 in liver sections, we enumerated cells expressing F480 + / eld as indicated in Fig. 1K. Low numbers of F480 + were observed in liver biopsies of HCC mice treated with the G. tournefortii extract as compared to their untreated counterparts, thus indicating fewer in ltrated macrophages and less in ammatory response. (P = 0.0001)

Serum tumor markers and molecular characteristics of HCC mice model receiving G. tournefortii extracts
To further characterize our HCC mice model receiving G. tournefortii extracts we evaluated for αFP in order to assess the extent of malignant cells. Moreover, we evaluated for serum levels of GPC3, a heparan sulfate proteoglycan anchored to the plasma membrane. It has been demonstrated to interact with growth factors and modulate their activities. It has been shown that soluble GPC3 is sensitive in detecting well or moderately differentiated HCC [21]. Simultaneous determination of αFP and GPC3 markers improves overall sensitivity from 50-72% [22]. Figure 3 showed serum αFP (Fig. 3a) levels and GPC3 (Fig. 3b) levels were elevated during tumor progressions at a 2 day intervals following the HCC implantation. Injections with G. tournefortii extracts caused a remarkably reduction in both serum αFP and GPC3 levels in the HCC mice models to levels similar to the vehicle treated mice group (P < 0.01). To further extend our study on mechanisms behind the decrease in tumor marker following G. tournefortii extracts, we evaluated for molecular signaling in liver extracts though quantitation of phosphorylated proteins by western blot analysis. The tumor suppressor p53 is one of the most frequently mutated genes in liver cancer. p53 regulates expression of genes involved in cell cycle progression, cell death, and cellular metabolism to avert tumor development due to carcinogens [23]. Hepatic intracellular pathways of Protein kinase B (PKB), also known as Akt were evaluated. Akt plays a key role in multiple cellular processes such as glucose metabolism, apoptosis, cell proliferation, transcription, and cell migration [24]. Moreover, Phosphoinositide 3-kinases (PI3Ks) involved in cellular functions such as cell growth, proliferation, differentiation, motility, survival and intracellular tra cking, which in turn are involved in cancer [25] were also assessed. Figure 3c shows densitometry bands of western blot of from representative proteins of p53, p-Akt, Akt, p-PI3K, PI3K, and the house keep protein; β-actin. Average qutanitations of the measured protein are presented in Fig. 3d -f. Figure 3d show elevated p53 protein as well as phosphorylated Akt/ PI3K (Fig. 3e&f) in mice liver extracts receiving the HCC cells as mentioned in materials and methods. Quantitation of p53 protein were decreased and Akt/ PI3K signaling were dephosphorylated in mice liver extracts in HCC mice receiving G. tournefortii extracts (P < 0.05). No effects of G. tournefortii extracts were seen in naïve mice with or without the treatment. Overall data show that G. tournefortii extracts modulated tumor progressions through p53 inhibitions and delay in Akt/PI3K signaling pathway indicating its effects in decreasing proliferations of HCC.
G. tournefortii inhibit DNA cell cycle of HCC isolated primary hepatocytes In order to verify the ability of G. tournefortii extract to induce disturbances in the cell cycle of hepatocytes of HCC mice, ow cytometry analysis of propidium-iodide stained nuclei cells were used.
Perturbations in the cell cycle were investigated using the boiled extracts of G. tournefortii, as they exhibited a high e ciency. The extract was diluted in sterile water prior to incubation with hepatocytes at 37 °C for 24 hrs at a concentration of 100 µg/ml. The control cells were treated with water only. Doxorubicin (Dox) was used to a positive control to induce cell cycle progression [26]. The data in Fig. 3a shows a signi cant elevation in the proportion of cells in the G1 phase following treatment with G. tournefortii extract. Average values of 52.53 ± 3.32%, 56 ± 5.56% and 52.67 ± 4.04% were obtained with 50, 100 and 200 µ g/ml extracts, respectively, as compared to 41.6 ± 3% in untreated samples (P < 0.005). There was no signi cant differences between the three different concentrations of the extract. Moreover, a signi cant decrease in the S phase were seen following the treatments with the extract with the concentration of 200 µg/ml (Fig. 3b). On the other hand, extracts of the three different concentrations inhibited HCC hepatocytes cell cycle in the G2M phase to 3, 6.8 and 2.64-folds following the 50, 100 and 200 µg/ml of G. tournefortii extracts, respectively, as compared to untreated cells (percentages obtained were less than the Dox treatments; Fig. 3c). These data show signi cant disturbances in cell cycle parameters in the G2/M phase (mitosis state) along with a signi cant shifting to the G1 phase (naive state), indicating a marked delay in the mitotic phase following G. tournefortii extract suggesting a potentials of anti-cancer characteristics.

G. tournefortii increase the apoptotic and necrotic activity of HCC primary hepatocytes in an in vitro cultures
Phosphatidylserine (PS) one of the phospholipid components of membrane cell, normally PS appears facing the cytoplasm but in case of apoptosis, a ip op occurs, which means that PS now is facing external surface of the cell [27]. To verify that G. tournefortii induce an apoptotic activity in hepatocellular cancer cells isolated from the HCC mice groups, we used Annexin-V, which is a protein with a high a nity to phosphatidylserine, conjugated with FITC for detection of PS. We used another marker, propidium iodide (PI), to check DNA and to determine necrosis (last step of death) in the cells. Early apoptosis evaluated by Annexin-V + and PI − . This was distinguished from late apoptotic and necrotic cells tested (Annexin-V + and PI + ) and necrotic cell (Annexin-V − and PI + ). Three different concentrations of the G.
tournefortii extracts were used (50,100 and 200 µg/ml) in order to have a broad spectrum of its apoptotic effects in the in vitro settings. Figure 4 shows that 47.03 ± 6.2% of untreated HCC hepatocytes undergone early apoptosis as indicated by Annexin-V + / PI − . Following treatments with 50, 100 and 200 µ g/ml the G. tournefortii extracts elevated apoptotic activity was noticed up to 55.67 ± 4.4%, 53.67 ± 4.4% and 62.67 ± 7.3%, respectively (P < 0.02). Moreover, the G. tournefortii extracts decreased necrosis effects on the cells (Annexin-V − and PI + ) in favor to shifting the cells for late apoptosis and necrosis (Annexin-V + and PI + ) in a range of 1.4 to 1.8-fold increase (P < 0.004). Our data strongly suggest that G. tournefortii has an anticancer property through increasing the apoptotic and necrotic activity of hepatocellular cancer cells and thus shifting the cells to program cell death rather than causing death to the cells.

Discussion
Several studies were conducted to study anti-cancer properties of plants including Gundelia species. However, these papers only discussed one anticancer aspect and were focused in testing the extract in an in vitro setting. On the other hand, our study discussed the anticancer activity on liver cancer in both an in vivo animal model and ex-vivo hepatocellular carcinoma cells isolated from mice livers. Other studies on the anticancer activity of G. tournefortii were conducted by Saleh Abu-La et al. [28] and Betül Özaltun and Taner Dastan [29]. Abu-La et al. found that methanol and hexane extracts of G. tournifortii showed anti-cancer properties against the HCT-116 cancer cell line. However, the water extract showed no signi cant effects were achieved. On the other hand, Betül Özaltun and Taner Dastan found that aqueous plant extracts of different concentrations obtained from different parts exhibited potent cytotoxic activity against human breast adenocarcinoma cell (MCF-7). In our current study, we choose to use G. tournifortii extracts in water solution to prevent any potential harm and hepatotoxicity with other organic solvent. We showed that the ower buds of G. tournifortii suggest an interesting anticancer agent that can potentially act on several levels of the cell cycle including inhibited molecular expressions of p53 and cell proliferations through Akt/PI3K, arresting the most attractive phase of mitosis (G2-M phase) and enhancing cell apoptotic activity. These anti-proliferative effects could be explain in part as suppression of serum tumor marker of αFP and GPC3.
Akt, also known as protein kinase B, plays key roles in cell proliferation, survival and metabolism. Akt hyperactivation contributes to many pathophysiological conditions, including human cancers [30,31], and is closely associated with poor prognosis and chemo-or radiotherapeutic resistance [31]. Normally, Akt is activated by growth factors that activate PI3K; both Akt and PI3K were inhibited following treatments of G. tournifortii. Human hepatocellular carcinoma (HCC) is the most common primary malignancy of the liver [32,33], and represents a severe, worldwide threat to human health and quality of life. Patient survival after surgery remains relatively low, with 5-year survival rates after resection for earlystage disease ranging between 17 and 53%, and recurrence rates being as high as 70% [34,35]. Therefore, it is important to identify biomarkers that reliably distinguish patients at high risk of recurrence. G. tournifortii could contribute to the medicinal treatment of liver cancer and could be used a target for Akt activation and thus interfere with cell signaling responses and cycle checkpoints.