Effect of Inhibition of GLUT1 Expression and Autophagy Modulation on the Growth and Migration of Laryngeal Carcinoma Stem Cells Under Hypoxic and Low-Glucose Conditions

Background: Enhanced glucose uptake and autophagy are means by which cells adapt to stressful microenvironments. We investigated the roles of glucose transporter-1 (GLUT-1) and autophagy in laryngeal carcinoma stem cells under hypoxic and low-glucose conditions. Methods: CD133 + Tu212 laryngeal carcinoma stem cells were puried by magnetic-activated cell sorting and subjected to hypoxic and/or low-glucose conditions. Proliferation was evaluated using a cell-counting kit and a clone-formation assay, and migration was evaluated through a Transwell assay. Autophagy was assessed via transmission electron microscopy. GLUT-1 and beclin-1 expression were silenced using an shRNA and autophagy was manipulated using rapamycin, 3-MA, or chloroquine. Gene expression levels were evaluated by quantitative reverse transcription-polymerase chain reaction and protein concentrations were assessing via Western blotting. Results: Compared to CD133 – stem cells, CD133 + cells showed increased proliferation and migration, and reduced apoptosis, under hypoxic or low-glucose conditions. They also showed increased expression of GLUT-1 and autophagy markers. Finally, GLUT-1 knockdown or autophagy inhibition reduced their proliferation and migration. Conclusions: Enhanced glucose uptake and autophagy maintain the functions of CD133 + laryngeal carcinoma stem cells under hypoxic and low-glucose conditions.


Page 3/17
CD133 + CSCs show greater proliferation and a lower frequency of apoptosis compared to CD133 − cancer cells in multiple types of tumors, particularly under hypoxic or nutrient-deprived conditions [14,15].
Interestingly, autophagy induction reportedly promotes the conversion of non-stem pancreatic cancer cells into CD133 + stem-like cells under intermittent hypoxia [16]. The expression of glucose transporter-1 (GLUT-1) is reportedly associated with autophagy activation in CSCs under hypoxic or nutrient-deprived conditions [17][18][19][20]. Autophagy may also affect cellular glucose uptake [21][22][23]. Beclin-1, an autophagy marker, plays an important role in the initiation of autophagy [24]. It promotes the localization of other autophagy-related proteins to autophagosomes, thus promoting the formation and maturation of autophagosomes. High expression of beclin-1 is typically accompanied by enhanced autophagy, increased GLUT-1 expression, and increased glucose uptake in certain types of tumors, such as nonsmall-cell lung carcinoma and breast cancer [25][26][27]. However, in one study, beclin-1 and GLUT-1 expression were negatively correlated in 29 cases of head-and-neck squamous cell carcinoma [28], indicating that the association between autophagy and glucose metabolism may be cancer-type related.
In this work, we investigated the regulation by GLUT-1 and autophagy of the functions of laryngeal carcinoma stem cells under hypoxic or low-glucose conditions, as well as the underlying mechanisms.

Methods
Cell culture and treatment Sorting and identi cation of Tu212 laryngeal carcinoma cells Magnetic sorting. Brie y, 1 × 10 7 Tu212 cells were resuspended in phosphate-buffered saline (PBS). The resuspended cells were added to 100 µL FcR Blocking Reagent and 100 µL CD133 MicroBeads, mixed, and incubated at 4 °C for 30 min. Next, 2 mL PBS was added, and the cells were centrifuged at 300 × g for 10 min; the supernatant was discarded. Subsequently, the cell pellet was resuspended in 500 µL PBS. The magnetic separation (MS) column was clipped to the magnetic separator and 500 µL PBS was added to moisten the column. Next, the cell suspension was added to the MS sorting column. The MS separation column was washed three times with 500 µL PBS to remove unbound cells. The column was removed from the magnetic separator, 1 mL PBS was added, and the cells were expelled from the column using a push rod. After centrifugation at 300 × g for 10 min, the supernatant was discarded, and the cells were resuspended in 1 mL PBS and enumerated.
Determination of the purity of Tu212 CD133 + cells. Cells (2 × 10 5 ) were removed before and after separation, centrifuged, and the supernatant was discarded. Next, 80 µL PBS and 10 µL anti-human CD133-PE were added. The sample was gently mixed using a micropipette and incubated at 4 °C for 10 min. The cells were centrifuged at 1000 rpm for 5 min and the supernatant was discarded. Precooled PBS (1 mL) was added, and unbound excess antibody components were removed by two centrifugation and washing steps. After adding 4% paraformaldehyde and incubation at 4 °C for 20 min, the supernatant was centrifuged. The cells were transferred to a ow tube and stored at 4 °C protected from light. Flow cytometry was performed using the standard procedure (Beckman, Fullerton, CA).

Clonogenic assay
The cell suspension was dispersed; the percentage of individual cells was greater than 95%. Next, cells were counted, and the cell density was adjusted to 250/mL by adding culture medium. The cell suspension was added to the wells of a six-well plate (2 mL per well), and the plate was gently shaken. The plate was placed in an incubator for 2 to 3 weeks, and the medium was replaced every 3 days. The culture was terminated when clones became visible. The medium was discarded, and the cells were gently washed twice with PBS, stained with 1% crystal violet at room temperature for 1 h, and photographed.
Cell-Counting Kit-8 assay Cells were incubated at 37 °C in an atmosphere containing 5% CO 2 for 48 h. Next, 20 µL Cell Counting Kit-8 (CCK-8) solution was added, and the cells were incubated in the dark for 1 h. The absorption at 450 nm of the suspension was measured using a Spectra Plus Microplate Reader (Molecular Devices, Sunnyvale, CA).

Flow cytometry
Brie y, 10 × Binding Buffer was diluted 1:10 with deionized water. Cells were collected by centrifugation for 5 min, exposed to reagents, digested, and resuspended in 500 µL Annexin V binding buffer. Next, 5 µL uorescein isothiocyanate and 10 µL propidium iodide (Sigma Aldrich Co., St. Louis, MO) were added for 10 min in darkness at RT. Finally, the proportions of non-apoptotic and apoptotic cells were determined in triplicate by ow cytometry with ModFit LT software (Becton Dickinson, Mountain View, CA) Transwell assay Cells were digested with trypsin and the culture medium was discarded. Next, the cells were washed once or twice with PBS and resuspended in serum-free medium (containing 0.2% bovine serum albumin) to a density of 1 × 10 6 /mL. Cell suspension (200 µL) was added to the upper Transwell chamber and 600 µL FBS was added to the lower chamber. The cells were incubated in a 5% CO 2 atmosphere at 37 °C for 24 h.
The Transwell chamber was removed, and the culture medium was discarded. Then the chamber was washed twice with calcium-free PBS, xed in formaldehyde for 30 min, air-dried, and the cells were stained with 0.1% crystal violet for 20 min. Finally, the upper layer of unmigrated cells was gently removed using cotton swabs and washed three times with PBS.

Quantitative real-time polymerase chain reaction
The cells were collected, washed three times with precooled PBS, centrifuged at 1500 rpm for 3 min, and lysed on ice in the presence of TRIzol. Total RNA was extracted from the cells according to the manufacturer's instructions. Brie y, 1 µg RNA was reverse-transcribed using a First-Strand cDNA Synthesis Kit (K1622; Fermentas, Burlington, ON, Canada) and ampli ed by PCR using a SYBR Green qPCR Kit (Merck, Darmstadt, Germany). The PCR program was 37 °C for 60 min, 85 °C for 5 min, and 4 °C for 5 min. The ampli cation products were stored at − 20 °C. The primers for GLUT-1, Beclin-1, Atg7, Atg5, and LC3 were designed and synthesized by Sangon Biotech ( Table 1). The 2 ΔΔCt method was used to calculate relative gene expression levels. Table 1 The primers for GLUT-1, Beclin-1, Atg7, Atg5, and LC3 forward reverse Western blotting Total proteins were extracted from cells and tumor tissues in radioimmunoprecipitation assay buffer. The cells were collected, washed three times with precooled PBS, and centrifuged at 1500 rpm for 3 min. An appropriate volume of cell lysate was added, and the cells were left on ice for 30 min. The supernatant was centrifuged at 1200 rpm at 4 °C for 30 min and stored at − 80 °C. After assaying the protein concentration, samples were added to 4⋅ sodium dodecyl sulfate loading buffer, boiled for 5 to 10 min, and centrifuged at 12,000 ⋅ g for 1 min. Proteins (30 µg) were subjected to SDS-polyacrylamide gel electrophoresis (SDSPAGE) and transferred to a polyvinylidene di uoride membrane (Millipore). Primary antibodies against GLUT-1 (Abcam), beclin-1, LC3 (Proteintech), Atg7, Atg5, and β-actin (Abcam) were added and incubated at 4 °C overnight; β-actin served as the internal control. After washing three times with Tris-buffered saline/Tween 20, the secondary antibodies were added for 1 h at room temperature. The signal was developed using an enhanced chemiluminescence assay kit (Beyotime Biotech) and analyzed semi-quantitatively using the ChemiDoc XRS + System (Bio-Rad).

Transmission electron microscopy
Cells were collected, washed in PBS, xed in 2.5% glutaraldehyde, post-xed in 1% osmium tetroxide, and dehydrated in ethanol and acetone. After embedding in epoxy resin, sections were cut and stained with uranyl acetate and lead citrate. Autophagy was visualized by transmission electron microscopy (TEM; Thermo Fisher Scienti c, Waltham, MA).

Results
The proliferation, migration and apoptosis of CD133 + laryngeal carcinoma stem cells under oxygen-or glucose-deprived conditions To investigate the role of GLUT-1 in LCSCs, we rst sorted CD133 + cells from laryngeal carcinoma Tu212 cells by MACS. Cell purity was examined by ow cytometry. The results showed that the proportion of CD133 + Tu212 cells was only ~ 8% before sorting, and reached more than 90% after sorting. Moreover, the isolated Tu212 CD133 + cells showed good viability as demonstrated by SSC/FSC plots (Fig. 1a).
Next, we investigated whether the growth, proliferation, apoptosis, and migration of laryngeal carcinoma Tu212 CD133 + cells under hypoxia and low glucose were signi cantly higher than those in CD133 − cells or CD133 + under normal condition. Clonal formation test showed that the clonal numbers of Tu212 CD133 + cells were signi cantly higher than that of Tu212 CD133 − cells under hypoxia (1%O 2 ), low glucose (2.5 mM glucose) and hypoxia + low glucose conditions. However, the clonal-forming ability was similar between Tu212 CD133 + cells and Tu212 CD133 − cells under normal culture condition (Fig. 1b). CCK-8 assay also revealed that compared with CD133 − cells, the proliferation of Tu212 CD133 + cells was signi cantly increased under hypoxia, low glucose and hypoxia + low glucose conditions (Fig. 1c). In addition, ow cytometry showed that the apoptotic rate of Tu212 CD133 + cells was signi cantly decreased compared to Tu212 CD133 − cells under hypoxia, low glucose and hypoxia + low glucose conditions (Fig. 1d). Transwell assay demonstrated that the migratory ability of Tu212 CD133 + cells was also higher than that of Tu212 CD133 − cells under the above three stress conditions (Fig. 1e). In contrast, the proliferation and apoptosis were comparable between Tu212 CD133 + cells and Tu212 CD133 − cells under normal culture condition, with CD133 + cells showing only a marginal increase in the migratory ability. Hence, CD133 + laryngeal carcinoma stem cells showed increased proliferative and migratory capacity than CD133 − cells under stressful conditions.
Thus, we explored whether the increased proliferation and migration of laryngeal carcinoma CD133 + Tu212 cells were associated with high level of GLUT-1 expression and cell autophagy. To this end, we rst investigated the expression of GLUT-1 and autophagy in Tu212 cells under hypoxia and low glucose conditions. qRT-PCR results showed that the expression of GLUT-1 mRNA in Tu212 CD133 + cells was signi cantly higher than that in Tu212 CD133 − cells under normal, hypoxia, low glucose, and hypoxia + low glucose conditions. The expression of GLUT-1 mRNA of Tu212 CD133 + cells under hypoxia, low glucose, and hypoxia and low glucose was higher than that of Tu212 CD133 + cells under normal culture condition (Fig. 2a). Western blotting also showed that the protein level of GLUT-1 in Tu212 CD133 + was signi cantly higher than that in Tu212 CD133 − cells under the above four conditions (Fig. 2b, S1).
In terms of cell autophagy, we found that hypoxia or low glucose increased the expression of Beclin-1, Atg7, Atg5, and LC3 in Tu212 CD133 + cells, but to a lesser extent in Tu212 CD133 − cells. Compared with Tu212 CD133 − cells, the levels of these autophagy markers were signi cantly higher in Tu212 CD133 + cells under hypoxia, low glucose, and hypoxia + low glucose conditions (Fig. 2a, b, S1). Transmission Electron Microscopy (TEM) imaging also showed that the number of autophagosomes in Tu212 CD133 + cells was signi cantly higher compared with that in Tu212 CD133 − cells under the above three conditions, indicating the enhanced autophagy in stressed-laryngeal carcinoma stem cells (Fig. 2c). Thus, the increased survival and migration of laryngeal carcinoma stem cells under hypoxia and low glucose maybe associated with high GLUT-1 expression and cell autophagy.
The association between the expression of GLUT-1 and autophagy markers in CD133 + laryngeal carcinoma stem cells To further study the association between the levels of GLUT-1 and autophagy marker genes. We adopted Beclin-1, Atg7, Atg5 and, LC3II/LC3I ratio in Tu212 CD133 + . On the other hand, RAPA treatment signi cantly increased the expression of Beclin-1 and LC3II/LC3I ratio in Tu212 CD133 + cells (Fig. 3a, b,   S2). The above results suggested that the expression of GLUT-1 and autophagy markers is closely associated in stressed laryngeal carcinoma stem cells. In addition, TEM imaging showed that silencing of GLUT-1 and Beclin-1, or inhibiton of autophagy signi cantly decreased the number of autophagosomes in Tu212 CD133 + cells under hypoxia + low glucose condition. In contrast, activation of autophagy by RAPA treatment signi cantly enhanced the number of autophagosomes in Tu212 CD133 + cells (Fig. 3c).
GLUT-1 knockdown or autophagy inhibition reduced the proliferation and migration of CD133 + laryngeal carcinoma stem cells In the functional analysis, we found that silencing of GLUT-1 markedly decreased the clonal-forming capacity, cell proliferation, and migration of Tu212 CD133 + cells under oxygen-and glucose-deprived condition (Fig. 4a-c). In contrast, the apoptotic rate of Tu212 CD133 + cells was signi cantly increased by the silencing of GLUT-1 (Fig. 4d). Similarly, Beclin-1 silencing or autophagy inhibitor (3-MA, CQ) treatment also signi cantly decreased the above malignant behaviors of CD133 + cells. Importantly, upon GLUT-1 silence or autophagy inhibition, the survival and migratory advantages of CD133 + cells over CD133 − counterparts were greatly compromised. To our surprise, autophagy activator Rapamycin also reduced the malignant behaviors of CD133 + cells, suggesting that the excessive autophagy may be harmful for stem cells as well (Fig. 4a-d). Taken together, the enhanced glucose uptake and autophagy are responsible for maintaining the growth and migration of the stressed laryngeal carcinoma stem cells.
A stressful extracellular microenvironment, such as hypoxia or low glucose, may upregulate GLUT-1 expression [37]. Indeed, hypoxia promotes the proliferation, migration, and chemoresistance of CSCs [4-7, 14, 15]. In this study, the growth and migration of CD133 + Tu212 cells were greater than those of CD133 − Tu212 cells under hypoxic and low-glucose conditions; this was associated with increased mRNA and protein levels of GLUT-1 in CD133 + Tu212 cells. These ndings are consistent with previous reports that high GLUT-1 expression facilitates glucose uptake to meet the energy demand of cancer cells. This adaptive response enables cancer cells to overcome external stresses, such as hypoxia or nutrient deprivation [6,38], thereby suppressing apoptosis.
The levels of autophagy markers in Tu212 CD133 + cells were increased by hypoxia and low glucose, whereas GLUT-1 silencing reduced the levels of these proteins. These results suggest mutual regulation of glucose uptake and autophagy in stressed laryngeal carcinoma stem cells. Autophagy modulates various metabolic pathways, including that centered on glucose [23]. Hypoxia and glucose deprivation may induce autophagy, promoting glucose uptake by upregulating GLUT-1 expression to increase the glycolytic ux and maintain nutrient uptake under stress conditions [21-23, 40, 41]. In airway progenitor cells, a lack of GLUT-1 impacts its recycling but not its expression, facilitating glucose uptake [40]. In mouse embryonic broblasts, however, autophagy enhances glucose uptake by increasing GLUT-1 expression and promoting GLUT-1 tra cking [21]. In this study, silencing of GLUT-1 decreased the levels of the autophagy markers beclin-1, Atg7, and Atg5, as well as the LC3II/LC3I ratio. However, inhibition or activation of autophagy by the beclin-1 shRNA/3-MA/CQ or rapamycin did not affect GLUT-1 expression. By contrast, rapamycin-induced autophagy activation increased the frequency of apoptosis of laryngeal CSCs, consistent with reports that excessive autophagy induces cell death.
This study had some limitations. First, we did not assay glucose metabolism, instead using GLUT-1 as a surrogate for glucose uptake. Second, we did not explore the in vivo implications of our ndings using animal models. Third, the signaling mechanisms responsible for the enhanced GLUT-1 expression and autophagy in laryngeal carcinoma stem cells under hypoxic and low-glucose conditions warrant further investigation.

Conclusions
In summary, hypoxia and low glucose increased the growth and migration of CD133 + laryngeal carcinoma stem cells by enhancing the expression of GLUT-1 and activating autophagy.

Availability of data and materials
The data that support the ndings of this study are available from the corresponding author upon reasonable request. The expression of GLUT-1 and autophagy markers in stressed-CD133+ laryngeal carcinoma stem cells.
Indicated Tu212 cells were challenged with hypoxia or low glucose, either alone or in combination. (a, b) The mRNA and protein levels of GLUT-1/Beclin-1/Atg7/Atg5/LC3 were evaluated by qRT-PCR (a) and Western blot (b), respectively. (c) Cell autophagy was examined by transmission electron microscopy. Data were mean ± SD and were representative of at least 3 independent experiments. *P<0.05; **P<0.01; Figure 3 The relationships between GLUT-1 expression and cell autophagy in stressed-CD133+ laryngeal carcinoma stem cells. Indicated Tu212 cells were trasnfected with GLUT-1 and Beclin-1 siRNA, or treated with autophagy inhibitor (3-MA, CQ) or activator (Rapamycin). Then challenged with hypoxia or low glucose, either alone (a-c) or in combination (d). The mRNA and protein levels of GLUT-1/Beclin-1/Atg7/Atg5/LC3 were evaluated by qRT-PCR (a) and Western blot (b), respectively. (c) Cell autophagy was examined by transmission electron microscopy. Data were mean ± SD and were representative of at least 3 independent experiments. *P<0.05; **P<0.01;