Abrus agglutinin targets cancer stem-like cells by eliminating self-renewal capacity accompanied with apoptosis in oral squamous cell carcinoma

Abstract

The accumulating evidences show that Abrus agglutinin, a plant lectin, displays a broad range of anticancer activity including cancer-specific induction of apoptosis; however, the underlying molecular mechanism of Abrus agglutinin–induced oral cancer stem cell elimination remains elusive. Our data documented that Abrus agglutinin effectively downregulated the CD44⁺ expression with the increased CD44⁻ population in different oral cancer cells. After 24-h Abrus agglutinin treatment, FaDu cells were quantified for orosphere formation in ultra-low attachment plates and data showed that Abrus agglutinin inhibited the number and size of orosphere in a dose-dependent manner in FaDu cells. Furthermore, Abrus agglutinin hindered the plasticity of FaDu orospheres as supported by reduced sphere formation and downregulated the self-renewal property via inhibition of Wnt-β-catenin signaling pathway. Introduction of LiCl, a glycogen synthase kinase 3β inhibitor, rescued the Abrus agglutinin–stimulated inhibition of β-catenin and phosphorylated glycogen synthase kinase 3β in FaDu cell–derived orospheres confirming importance of Wnt signaling in Abrus agglutinin–mediated inhibition of stemness. In this connection, our data showed that Abrus agglutinin restrained proliferation and induced apoptosis in FaDu-derived cancer stem cells in dose-dependent manner. Moreover, western blot data demonstrated that Abrus agglutinin increased the Bax/Bcl-2 ratio with activation of poly(adenosine diphosphate–ribose) polymerase and caspase-3 favoring apoptosis induction in orospheres. Abrus agglutinin induced reactive oxygen species accumulation in orospheres and pretreatment of N-acetyl cysteine, and a reactive oxygen species scavenger inhibited Abrus agglutinin–mediated caspase-3 activity and β-catenin expression indicating reactive oxygen species as a principal regulator of Wnt signaling and apoptosis. In conclusion, Abrus agglutinin has a potential role as an integrative therapeutic approach for combating oral cancer through targeting self-renewability of orospheres via reactive oxygen species–mediated apoptosis.

Introduction

Head and neck squamous cell carcinoma (HNSCC) comprising the cancer of several distinct subsites (e.g. nasopharynx, oropharynx, hypopharynx, larynx, lip, and tongue) ranks as the sixth most frequently occurring cancer globally. The Indian subcontinent is considered as the hub of oral cancer. ¹ World Health Organization (WHO) regards oral cancer as a major public health challenge in India.^2,3 It is the most common cancer of males and the fifth most common in females. ⁴ It ranks among the top three types of cancer in the country and accounts for 30%–40% of all the malignant tumors in India. Near about 83,000 new oral cancer cases are registered here and 46,000 deaths occur annually. ¹ Vital nutrient deficiency, unhygienic oral cavity, illiteracy, heavy smoking, alcohol, and tobacco chewing are the major reasons for such high oral cancer rise.^5,6 Despite the advancement in the modern treatment modalities and though the quality of life of oral cancer patients has improved, the 5-year survival rate has not improved in the past decades because of treatment resistance, loco-regional recurrences, and distant metastasis. ⁷ A recent report has ascertained the phenotypic heterogeneity of HNSCC tumors that supports the existence of a rare subset of intratumoral cells called as cancer stem cells (CSCs). ⁸ The poor prognosis with a high rate of local recurrence and metastasis is because of this rare undifferentiated subpopulation (<1% of the total tumor cell population) of cells that use their self-renewal properties to promote tumor growth and proliferation. Moreover, oral tumors shrink and sometimes become unresponsive to the conventional anticancer therapies only to recur after 5 years of treatment as the CSCs remain occult during treatment. The CSCs are ostensible to the resistance to standard HNSCC treatment making it an alluring target for novel treatment modality. CD44 is one of the most recognized CSC surface markers in many cancers including oral cancer. ⁹ It is broadly distributed as transmembrane surface glycoprotein (80–250 kDa) that is involved in the hyaluronan–CD44 interaction where it anchors matrix metalloproteinase-9 (MMP-9) to facilitate cancer cell migration and adhesion. ¹⁰ Prince et al. for the first time accomplished the isolation of a pool of HNSCC cells with high CD44 expression that exhibited stem cell–like characteristics like self-renewal, generation of differentiated progeny, lack of differentiation markers, and expression of immature cell markers. These CD44^high cells were shown to have exclusive tumorigenic capacity when introduced in immunosuppressed mice. ⁹

Like normal stem cells, CSCs are also anticipated to possess intrinsic and exclusive self-renewal propensity that allows sustained repopulation. ¹⁰ The canonical Wnt/β-catenin signaling is a crucial developmental pathway involved in self-renewal, differentiation, migration, proliferation, and tumorigenesis. ¹¹ The significance of Wnt/β-catenin signaling in controlling the tumorigenesis and self-renewal of CSCs in HNSCC is well established. It has been documented that overexpression of β-catenin in HNSCC led to a dedifferentiated stem-like state where its downregulation inhibited self-renewal capacity, stemness-associated gene expression, and in vivo tumorigenicity. ¹² Apoptosis resistance, one of the key hallmarks of cancer, has been demonstrated prominently in CSCs. ¹³ The side population (SP) isolated from Ho-1-N-1 oral squamous cell carcinoma cell line is highly resistant to 5-fluorouracil and carboplatin as compared to the non-SP owing to the enhanced expression of antiapoptotic protein CFLAR, BCL2, and BCL2A1. ¹³ The Wnt/β-catenin signaling induces resistance to apoptosis by inhibiting tumor necrosis factor (TNF)/c-Myc and anoikis which are dependent on the death receptor signaling pathway in HNSCC. ¹⁴ Moreover, HNSCC CD44⁺ cells showed upregulation of Bcl-2 and inhibitor of apoptosis (IAP) family genes compared with CD44⁻ cells.^15,16

Abrus agglutinin (AGG), a heterodimeric plant lectin of 134-kDa, was isolated from the seeds of Abrus precatorius, an Indian medicinal plant. AGG is a heterodimer comprising two 30-kDa toxic A-chain and two 31-kDa B-chains linked through disulfide bridges; A chain has ribosomal RNA (rRNA) N-glycosidase activity and irreversibly inhibits protein synthesis through inactivation of ribosomes, and B chain binds to the cell surface. ¹⁷ It has specificity toward (gal (β 1→3) gal NAc) and belongs to type II ribosome inactivating protein (RIPII) family with a protein synthesis inhibitory concentration (IC₅₀) of 0.469 µg/mL and a lethal dose (LD₅₀) of 5 mg/kg body weight in mice. ¹⁷ The anticancer effects of AGG have been elucidated in several tumor models at sub-lethal doses by direct killing of tumor cells through extrinsic and intrinsic apoptosis. ¹⁷ In addition, AGG showed immunostimulatory properties and has a strong mitogenic activity to lymphocytes. It exhibits both humoral and cellular immunity with a propensity to stimulate the innate effector arms like macrophage and natural killer cells by activating splenocytes leading to Th1 response.^17–20 Moreover, heat-denatured and tryptic-digested AGG shows potent antitumor and immunomodulatory activities in normal as well as in tumor bearing mice.^17–20 Recently, we have documented that AGG inhibits Akt/PH domain to induce endoplasmic reticulum stress–mediated autophagy-dependent cell death in cervical carcinoma. ¹⁹ Our recent data showed that AGG induced potent anticancer activity through p73-dependent pathway in oral cancer. ²⁰ The remarkable antineoplastic activity of AGG in different bulk cancer cells is well established.^17–20 However, its efficacy in eliminating CSCs in oral cancer has not been yet deciphered. Here, we investigated the growth inhibitory potential of AGG in orospheres present in oral cancer cells. Our findings showed that AGG could inhibit the growth of orospheres via the reactive oxygen species (ROS)-mediated induction of caspase-dependent apoptosis in FaDu cells. Moreover, AGG found to regulate the plasticity of orospheres by inhibiting the canonical Wnt signaling pathway in oral cancer cells.

Material and methods

Chemical and reagents

6-Diamidino-2-phenylindole dihydrochloride (DAPI), dimethylsulfoxide (DMSO), and N-acetyl cysteine (NAC) were purchased from Sigma-Aldrich (India). Fetal bovine serum (sterile-filtered, South American origin), minimal essential medium (MEM), Dulbecco’s Modified Eagle Medium:Nutrient Mixture F-12 (DMEM/F-12), trypsin, and antibiotic–antimycotic solution were purchased from Invitrogen, India. Annexin V from Immunotools, Germany was purchased. Antibodies against β-catenin, phosphorylated glycogen synthase kinase 3β (pGSK3β), Bcl-2, Bax, poly(adenosine diphosphate–ribose) polymerase (PARP), cleaved caspase 3, and actin were purchased from Cell Signaling Technology (USA); CD44 was purchased from BD Biosciences (USA).

Cell culture and sphere culture

Hypopharyngeal cells (FaDu) was cultured in MEM, supplemented with antibiotic–antimycotic (1 ×) and 10% fetal bovine serum. The tongue squamous cell carcinoma (SSC4 and SSC25) were cultured in DMEM/F-12 supplemented with 400 ng/mL hydrocortisone. For sphere-forming culture, FaDu cells were cultured in serum-free medium supplemented with 1% N₂ supplement (Gibco, USA), 2% B27 supplement (Gibco), 20 ng/mL basic fibroblast growth factor-2 (bFGF-2; BD Biosciences), and 20 ng/mL epidermal growth factor (EGF; BD Biosciences) and plated in ultra-low attachment plates (Corning, USA). Cells were grown in these conditions as non-adherent spherical clusters of cells (usually named spheres or orospheres). All the cultures were maintained at 37°C in a humidified 5% CO₂ incubator. After 14 days of culture, the number of orospheres after AGG treatment was counted under a microscope. ²¹

Quantification of CD44^- population

Bulk populations of SCC25, SCC4, and FaDu were treated with the different doses of AGG (0.1, 0.5, and 1 µg/mL). After 24 h treatment, cells were washed with 0.1% 1  × phosphate-buffered saline-F (PBS-F). Cells were stained with 200 µL of CD44 (1:200) and were suspended in 500 µL PBS-F for further analysis by FACSCalibur.^22,23

Quantification of cell viability by trypan blue exclusion method

About 2000 cells/well were cultured in an ultra-low adherent six-well plate at 37°C and exposed to different concentrations of AGG (0.1, 0.5, and 1 µg/mL) for 24 h. All experiments were performed in triplicate, and the relative cell viability was expressed as the percentage relative to the untreated control cells.

Expression of CD44 and β-catenin by confocal immunofluorescent staining

The orospheres were plated in chamber slide and were treated with different doses of AGG (0.1, 0.5, and 1 µg/mL). After 24 h, cells were fixed in 10% formaldehyde, washed with PBS, and permeabilized with 0.2% Triton X-100 for 20 min at room temperature (RT) along with primary antibody incubation of CD44 (1:500) and β-catenin (1:500). Cells were then washed with PBS and incubated with secondary antibody for 6 h followed by DAPI counterstaining. Cell images were captured at 1000 × magnification on a Leica TCS SP8 (Wetzlar, Germany) using 561/488/405 nm laser wavelengths to detect CD44 (red), β-catenin (green), and DAPI (blue) emissions, respectively.

Apoptosis analysis by Annexin V/propidium iodide staining

The oral cancer-initiating/stem cells were seeded in six-well plates and were treated with AGG (0.1, 0.5, and 1 µg/mL) for 24 h and processed and evaluated for Annexin V binding as described.^19,20

ROS analysis

AGG-treated (0.1, 0.5, and 1 µg/mL) orospheres were incubated with Dihydrorhodamine 123 (DHR123; 2.5 µg/mL) for 30 min. After incubation, the cell pellet was suspended in 500 µL PBS and analyzed through flow cytometry.^17,20

Western blotting

Orospheres were treated with AGG (0.1, 0.5, and 1 µg/mL) for 24 h. Treated and untreated orospheres were subjected to the preparation of whole-cell lysates followed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) for β-catenin, pGSK3β, Bcl-2, Bax, PARP, cleaved caspase 3, and actin (1:1000, mouse monoclonal; BD Biosciences) at 4°C overnight, which were further incubated with horseradish peroxidase (HRP)-conjugated secondary antibody at RT for 1 h (1:5000, mouse monoclonal; BD Biosciences) to examine the protein levels as described and visualized using ECL prime reagent.^18,21

HNSCC cancer xenograft mouse model

Xenograft studies were performed as previously described.^18,24 All mice were weighed before the start of the experiment. Animal experiments were conducted in accordance with Singapore NACLAR Guidelines (Law as of November 2004) for laboratory animal use and care. Briefly, 4-week-old athymic BALB/c nude male mice (Biolasco, Taiwan) weighing 16–18 g were randomized into the following treatment and control group (n = 5). FaDu cells were subcutaneously injected (5 × 10⁶ cells/mouse) to each mice. The control group was treated with normal saline (0.9%) and treatment group mice received intraperitoneal plant lectin agglutinin at doses of 50 µg/kg five times a week. The treatment was continued for 2 weeks from the date of randomization. The mice body weight and tumor sizes were recorded every day and the tumor size was determined by Vernier caliper and calculated using the formula (length × (width) ² )/2. At the end of 2 weeks, mice were sacrificed under isoflurane anesthesia followed by cervical dislocation, and tumor volume was measured and weighed. A part of the tumor tissue was fixed in phosphate-buffered formalin for histology, and the remaining tissue was stored in liquid nitrogen for future analysis. ²⁴

Immunohistochemical staining and scoring

For the evaluation of immunohistochemistry (IHC) results, formalin-fixed and paraffin-embedded specimens of 3–4 µm thickness were sectioned and staining was performed with anti-CD44 (BD Biosciences, India) and β-catenin (BD Biosciences, India) as described previously. IHC analysis was performed by finding out the percentage of positive cells by counting the number of positive stained cells (weak, moderate, and strong) and the total number of cells from control under 40 × magnification. IHC score was analyzed for control and treated as 0+ (no staining), 1+ (weak staining), 2+ (moderate staining), 3+ (strong staining), and 4+ (very strong staining).^18,19

Statistical analysis

Data were presented as mean ± SD and evaluated with Student’s t-test; *p value < 0.05, **p value < 0.01, and ***p value < 0.001 were considered significant.

Results

AGG significantly downregulates CD44 expression and preferentially targets self-renewal potential in orospheres

CD44 is an important biological marker to screen oral CSCs in bulk tumor cell population.^25–27 CSC population in oral cancer comprises CD44⁺ cells with aggravated proliferation and higher tumorigenic potential than the CD44⁻ cells. To explore the significance of AGG in inhibition of CD44⁺ population, we treated bulk cell populations of FaDu, SCC4, and SCC25 with different doses of AGG for 24 h. The flow cytometry analysis corroborated our hypothesis with increased CD44⁻ cell populations in FaDu cells from 9 ± 1.1 in control to 10 ± 1.5, 13 ± 1.3, and 18 ± 2.5 in respective doses of AGG. The CD44⁻ cell populations in SCC25 increased from 27 ± 4.1 in control to 58 ± 6.2, 60 ± 5.3, and 61 ± 4.1, while SCC25 cells displayed from 4 ± 1 in control to 13 ± 2.5, 17 ± 1.5, and 18 ± 1.0 in a dose-dependent manner (Figure 1(a) and (b); *p < 0.05; **p < 0.01). Since the oral stem/progenitor cells are enriched in non-adherent spherical clusters, in order to evaluate the effect of AGG on the formation of orospheres, FaDu cells (2000 cells/well) were enriched in a six-well plate for 14 days after 24 h treatment with various doses of AGG. Our results demonstrated that AGG treatment could decrease the number of spheres as well as reduced sphere size in FaDu cells (Figure 2(a) and (b)). The plasticity of CSCs is maintained by self-renewal pathways including Wnt signaling, Hedgehog signaling, and Notch signaling that play a provoking role in a cancer cell transformation contributing to tumorigenicity and its progressive metastasis. Our confocal microscopy data showed that AGG treatment significantly decreased the expression of β-catenin along with CD44 expression in FaDu-derived orospheres (Figure 3(a)). To evaluate the activation of Wnt canonical pathway, FaDu orospheres were treated with AGG for 24 h, and the expression of β-catenin, the transcriptional activator of Wnt signaling, was analyzed. The β-catenin expression was decreased to 0.61 ± 0.25, 0.58 ± 0.17, and 0.34 ± 0.15 from control in a dose-dependent manner in the presence of AGG (Figure 3(b) and (c); *p < 0.05; **p < 0.01). According to the reports, the cytoplasmic β-catenin level is regulated by a multifaceted kinase including glycogen synthase kinase 3β (GSK3β). The GSK3β at its dephosphorylation state promotes the ubiquitin-proteasome degradation of β-catenin through phosphorylation at Ser33/Ser37/Thr41 residues.^12,14,15 Our data showed that AGG inhibited the expression of pGSK3β to 0.85 ± 0.13, 0.58 ± 0.20, and 0.45 ± 0.14 compared to control in a dose-dependent manner in the presence of AGG without altering the expression of total GSK-3β and could manifest its role in destruction complex preventing aberrant accumulation of β-catenin (Figure 3(b) and (d); *p < 0.05; **p < 0.01). LiCl, an activator of Wnt pathway, inactivates GSK-3β through Ser9 phosphorylation which in turn diminishes β-catenin degradation. The orospheres from FaDu cells when treated with AGG in the presence of LiCl showed increased expression of β-catenin to 1.25 ± 0.11 as compared to 0.74 ± 0.21 in only AGG-treated group, while pGSK3β showed increased expression to 1.27 ± 0.18 as compared to 0.79 ± 0.24 in only AGG-treated set (Figure 3(e)–(g); ^#,*p < 0.05). ²¹ Henceforth, AGG effectively restrained the Wnt/β-catenin canonical pathway which is important for self-renewability of CSCs and tumor development.

Figure 1.

AGG significantly downregulates CD44 expression. (a and b) The oral squamous cell carcinomas (FaDu, SCC4, and SCC25) were treated with different concentration of AGG (0.1, 0.5, and 1 µg/mL). After 24 h treatment, cells were washed with 0.1% 1 × PBS-F. Cells were stained with 200 µL of CD44 (1:200) and were suspended in 500 µL PBS-F for further analysis by FACSCalibur to check the CD44⁺ and CD44⁻ populations. Data reported as the mean ± SD of three independent experiments and compared against PBS control. (*p value < 0.05 and *p value < 0.01 were considered significant as compared with control.)

Figure 2.

AGG inhibits formation of FaDu orospheres. (a and b) About 2000 cells/well were cultured in a serum-free media of an ultra-low adherent six-well plate at 37°C and exposed to different concentrations of AGG (0.1, 0.5, and 1 µg/mL) for 24 h. All experiments were performed in triplicate, and the relative cell viability was expressed as the percentage relative to the untreated control cells. After 14 days, the orospheres were observed in microscope, photographed, and quantified. Data are reported as the mean ± SD of three independent experiments and compared against PBS control. (*p value < 0.05 and **p value < 0.01 were considered significant as compared with control.)

Figure 3.

AGG preferentially targets self-renewal potential in orospheres. Orospheres derived from FaDu cells were treated with AGG (0.1, 0.5, and 1 µg/ml) for 24 h and analyzed for the expressions of β-catenin (1:500) along with CD44 (1:500). Cells were then washed with PBS and incubated with secondary antibody for 6 h followed by DAPI counterstaining. (a) Cell images were captured at 1000 × magnification on a Leica TCS SP8, Wetzlar, Germany using 561/488/405 nm laser wavelengths to detect CD44 (red), β-catenin (green), and DAPI (blue) emissions, respectively. (b and e) Cell lysates obtained after AGG treatment in the presence or absence of LiCl were analyzed for expression of β-catenin and GSK-3β by western blot. (c and d; f and g) Densitometer of proteins is represented as histogram plotted on the basis of actin to the target protein ratio by ImageJ. Data reported as the mean ± SD of three independent experiments and compared against PBS control. (*p value < 0.05 and **p value < 0.01 were considered significant; ^#p value < 0.05 indicates data compared against AGG treatment.)

AGG inhibits cell proliferation and induces apoptosis in orospheres

To target the proliferative propensity of CSC in FaDu, we studied the cell viability assay of orospheres upon treatment with different doses of AGG for 24 h. The data showed that AGG decreased the percentage of viable cells from 94 ± 4.5 to 71 ± 6.5, 50 ± 6.2, and 13 ± 1.5 in FaDu-derived orospheres (Figure 4(a); *p < 0.05; **p < 0.01). To decipher the mechanism of action for apoptosis caused by AGG, we performed Annexin V/propidium iodide (PI) staining analysis through flow cytometry after 24 h of AGG treatment. The data showed that there was an increase in apoptotic population from 20 ± 1.7 in control to 24 ± 1.7, 34 ± 2.6, and 41 ± 2.8 in the instance of 0.1, 0.5, and 1 µg/mL of AGG, respectively (Figure 4(b) and (c); *p < 0.05). We further examined the changes in expression of the Bcl-2 and Bax proteins in FaDu cancer sphere cells that regulate the intrinsic apoptosis pathway. AGG decreased the antiapoptotic Bcl-2 to 0.99 ± 0.18, 0.69 ± 0.22, and 0.55 ± 0.16 compared to control, while increased the pro-apoptotic Bax protein to 1.02 ± 0.24 1.03 ± 0.31, and 1.12 ± 0.38 compared to control in a dose-dependent manner (Figure 5(a)–(c); *p < 0.05). Increased Bax to Bcl-2 protein ratio to 1.03 ± 0.23, 1.49 ± 0.25, and 2.03 ± 0.42 compared to control led to the increased activation of pro-caspase 3 to activated caspase 3 expression to 1.28 ± 0.14, 1.87 ± 0.21, and 2.57 ± 0.18 compared to control in different doses of AGG (Figure 5(a) and (e); *p < 0.05). The FaDu-derived orosphere showed increased expression of cleaved PARP to 1.23 ± 0.12, 1.88 ± 0.20, and 2.78 ± 0.13 compared to control with increased dose of AGG (Figure 5(a) and (d); *p < 0.05). These results suggest the pivotal role of AGG in mitochondria-dependent (intrinsic) apoptosis in AGG-mediated elimination of CSC in FaDu cells.

Figure 4.

AGG inhibits cell proliferation and induces apoptosis in orospheres. (a) Orospheres were treated with different concentrations of AGG (0.1, 0.5, and 1 µg/mL) for 24 h and the cell viability assay through trypan blue was evaluated. (b and c) The Annexin V/PI staining analysis was performed through flow cytometry after 24 h of AGG (0.1, 0.5, and 1 µg/mL) treatment. Data reported as the mean ± SD of three independent experiments and compared against control. (*p value < 0.05 and **p value < 0.01 were considered significant.)

Figure 5.

AGG upregulates apoptotic molecules in orospheres. Orospheres were treated with AGG (0.1, 0.5, and 1 µg/mL) for 24 h. Treated and untreated orospheres were subjected to the preparation of cell lysates. (a) Target protein expressions of Bax, Bcl-2, cleaved PARP, and cleaved caspase-3 were analyzed by western blot. (b–e) Densitometer of proteins is represented as histogram plotted on the basis of actin to target protein ratio by ImageJ. Data reported as the mean ± SD of three independent experiments and compared against PBS control. (*p value < 0.05 was considered significant compared to control.)

ROS induced by AGG regulates apoptosis and Wnt signaling

CSCs have low levels of ROS and possess an antioxidant expression profile to maintain quiescence and self-renewal. ROS generation by anticancer therapy may disrupt the redox stability and selectively induces apoptosis in cancer cells without significant toxicity to normal cells. To define the role of ROS in apoptosis, orospheres were treated with AGG for 24 h and analyzed for ROS generation. The cells were incubated with DHR123 which was rapidly taken up by the cells and converted to Rhodamine 123 (Rh123) in the presence of ROS. Flow cytometry analysis showed increased level of ROS after AGG treatment as compared to control. Pre-treatment with an ROS scavenger NAC (10 mM) for 2 h recorded reduction in ROS production compared to only AGG (1 µg/mL)-treated orospheres (Figure 6(a)). Importantly, AGG-induced caspase activity was declined in the presence of NAC to 0.76 ± 0.11 from 1.28 ± 0.23 in only AGG-treated FaDu orospheres. In addition, AGG-modulated β-catenin expression was rescued in NAC-pretreated group from 1.26 ± 0.16 as compared to 0.65 ± 0.14 in AGG-treated orospheres, suggesting the implication of AGG-induced ROS in caspase-dependent apoptosis and self-renewal activity in oral CSCs (Figure 6(b)–(d); ^#,*p < 0.05).

Figure 6.

ROS induced by AGG regulates apoptosis and Wnt signaling. (a) Orospheres treated with AGG (1 µg/mL) in the presence of N-acetyl cysteine (NAC, 10 mM) were incubated with Dihydrorhodamine 123 (2.5 µg/mL) analyzed for ROS generation through flow cytometry. Orospheres were pre-treated N-acetyl cysteine (NAC, 10 mM) followed to AGG treatment (1 µg/mL) and analyzed for expression of cleaved caspase-3 and β-catenin by western blot. (c and d) Densitometer of proteins is represented as histogram plotted on the basis of actin to target protein ratio by ImageJ. Data reported as the mean ± SD of three independent experiments and compared against PBS control. (*p value < 0.05 was considered significant; ^#p value < 0.05 indicates data compared against AGG treatment.)

Effect of AGG on the expression of CD44 and β-catenin in FaDu xenograft tissue

To further corroborate our in vitro finding under in vivo, next we examined the effect of AGG on expression of CD44 and β-catenin in FaDu xenograft tissues. Earlier, we studied that AGG strongly (50 µg/kg body weight) significantly inhibited the FaDu xenograft growth in athymic nude mice. The tumor tissues were subjected to IHC analysis to check the status of molecules involved in Wnt canonical pathway. FaDu xenograft tissue analyses revealed that there was a significant decrease in the expression of β-catenin and CD44 in AGG-treated group as compared to control group (Figure 7(a)). The IHC score significantly distinguished the expression of β-catenin in control and treated as high and low, while the expression of CD44 in control and treated as moderate and low according to their staining intensities (Figure 7(c) and (d)). Together, these findings suggested that AGG suppressed the expression of the CSC phenotypes in vivo as a part of its anticancer efficacy against HNSCC.

Figure 7.

AGG inhibits the expression of CD44 and β-catenin in FaDu xenograft tissue. AGG strongly (50 µg/kg body weight) significantly inhibited the FaDu xenograft growth in athymic nude mice. (a) Tumor tissues were harvested followed by fixation with formalin, and paraffin-embedded sections were immunostained for β-catenin and CD44 in control and treated groups. (b–d) The IHC score significantly distinguished the expression of β-catenin in control and treated groups as high and low, while the expression of CD44 in control and treated groups as moderate and low. (e) Diagrammatic representation of mechanism of AGG inhibition of tumor spheres in oral cancer. (*p value < 0.05 was considered significant compared to control.)

Discussion

In recent years, advanced HNSCC tumors contain a rare small subset of CSCs, which is characterized by therapeutic resistance and frequent tumor relapse. ²⁵ Treatment failures resulting in the development of a second primary tumor have urged for the novel non-toxic natural approaches to target CSCs for treatment of cancer.^{27–30,31,32} Lectins have undergone advanced research in Asia comprising China and India and Europe as an alternative tumor therapy, ¹⁹ but more work has to be done to target CSCs. This work underpins the antineoplastic potential of AGG, an RIPII family member that inhibits proliferation and plasticity of orospheres and triggers ROS-mediated caspase-dependent apoptotic type-I programmed cell death. Here, we report for the first time the novel in vitro and in vivo effects of AGG on oral CSCs.

CD44 expressions appear to exhibit CSC properties in many cancers, including HNSCC. CD44 is a family of transmembrane receptors found on a number of different benign and malignant cells. Although CD44 alone is not sufficient for precisely isolating CSC in head and neck cancer cells, the CD44 high cells show elevated self-renewal, proliferation, and differentiation, thus expressing enhanced tumorigenicity. Palagani et al. ²⁶ showed significant decrease in the expression of CD44 after chemotherapy independent of any drug combination or tumor type. Zerumbone inhibited the EGF-induced CD44 expression through inhibition of the signal transducer and activator of transcription 3 (STAT3) pathway. ²⁷ Similar to these results, our work showed regressed CD44 expression by AGG treatment in a dose-dependent manner. Downregulation of CD44⁺ cells restricted the aberrant CD44 expression in FaDu, SCC25, and SCC4 cells. Furthermore, our confocal data showed that CD44 expressed in orosphere decreased in dose-dependent manner with AGG treatment, indicating a positive correlation between AGG and decline in CD44⁺ populations in oral squamous cell carcinoma.

Wnt/β-catenin signaling pathway has a wide range prospects in developmental biology including maintenance of stem cell compartments in adult tissue. The signaling pathway is derailed in CSCs contributing to its plasticity and therapy resistance. The underlying mechanism behind these two salient properties of CSCs is very little known. Honokiol, an active compound of Magnolia officinalis, has come up strongly with its potential to eliminate human oral CSCs by inhibiting Wnt/β-catenin signaling axis and apoptosis induction. ²⁹ It has also reciprocated its potential in suppressing the proliferation as well as angiogenesis in the xenograft growth of oral CSC-like cells. Resveratrol, a natural polyphenolic compound abundantly found in plant foods, inhibited the proliferation of breast cancer stem cells (BCSCs) followed by induction of autophagy via suppressing Wnt/β-catenin signaling pathway. ³⁰ The Wnt/β-catenin pathway is also responsible for the clonogenic propensity involved during survival and development of mixed lineage leukemia (MLL) leukemic stem cells (LSCs). Furthermore, cordycepin selectively reduced β-catenin stability by regulating GSK-3β to eradicate leukemia via elimination of LSCs. ³³ In this study, we proposed that the activated Wnt/β-catenin pathway might be critically responsible for the development and aggressiveness of oral cancers and could potentially be excellent candidates for targeting cancer-initiating/stem cells. Our molecule successfully restricted the self-renewal property of CSCs as evidenced by the reduced orosphere formation by inhibiting the Wnt canonical pathway through downregulation of β-catenin and GSK-3β expression. The confocal data further supported our hypothesis with decline in the β-catenin–CD44 interaction in orospheres with AGG treatment.

There are several natural compounds that have restored cell sensitivity to death stimuli, suppressing CSC self-renewal and tumor metastasis. Recently, sulforaphane, cucurbitacin I, resveratrol, quercetin, and curcumin have been reported to suppress the CSCs by inducing apoptosis.^{22,28,30,31,32} The proliferation of colon CSCs was eliminated by 20 (s)-ginsenoside Rg3 by induction of apoptosis through caspase-9 and caspase-3 pathways. ³⁴ In CD133⁺ rectal CSCs, curcumin was found to induce apoptosis and significantly increased its radiosensitivity. ³⁵ Earlier, our group has shown that AGG treatment in different cancer cells triggered ROS generation followed by programmed cell death.^18–20 Here, we observed that AGG treatment in orospheres succinctly generated ROS in a dose-dependent manner. Orospheres have high Bax/Bcl-2 ratio and increased Annexin V positive cells that correlated with the low caspase 3 cleavage in the presence of NAC. ROS-mediated apoptosis targeted the orospheres since AGG disturbed the redox balance that directs the selective killing of cancer cells. Flow cytometric distribution explained about the increase in the CD44⁻ population in different oral cancer cells upon AGG treatment in dose-dependent manner. Thus, eradicating the orospheres by phytoproducts like AGG has emerged as a valued complementary strategy in successful treatment of oral cancer.

Conclusion

The antineoplastic effect of AGG in oral CSCs has been documented for the first time in vitro. In FaDu cells with mutated p53, our data further connoted downregulation of Wnt/β-catenin signaling axis responsible for the maintenance of oral CSCs, followed by apoptosis induction. Our study further demonstrated that AGG induced stress and initiated the ROS generation, subsequently leading to apoptosis in hypopharyngeal CSCs. Thus, AGG can be a promising alternative therapeutic agent in the treatment of oral cancer by targeting Wnt/β-catenin as well as antiapoptotic signaling in orospheres.