The meningioma cell line IOMM-Lee (ATCC Cat. No. CRL-3370, RRID: CVCL_5779) was kindly provided by Professor Jin-Hong Mei (Nanchang University, China) and was authenticated completely match with IOMM-Lee in the American Type Culture Collection (ATCC) short tandem repeat (STR) database without any cross-contamination of other human cell lines before and after this research. Cells were grown in Dulbecco's modified Eagle's medium (DMEM; HyClone, USA) supplemented with 10% fetal bovine serum at 37°C in a 5% CO₂ atmosphere.

The miR-221/222 mimics and inhibitors were chemically synthesized by RiboBio Co., Ltd. (Guangzhou, China) and were transfected into IOMM-Lee cells with riboFECT™ CP reagent according to the manufacturer's instructions. Scrambled oligonucleotides (GenePharma Co., Ltd., Shanghai, China) were also transfected as a negative control. The expression levels of miR-221 and miR-222 in transfected IOMM-Lee cells were identified by quantitative real-time PCR.

Irradiation was performed at room temperature in a linear accelerator (Varian600, Varian, USA) at a dose rate of 3.2 Gy/min (31, 33). Cells were plated into six-well plates and exposed to the specified dose (0, 2, 4, 6, and 8 Gy) of X-rays.

A clonogenic assay was applied to determine the radiosensitivity of IOMM-Lee cells. A predetermined number of viable cells (1,000 cells for 0, 2, and 4 Gy; 2,000 cells for 6 and 8 Gy) were seeded in six-well culture plates and incubated at 37°C for 24 h. Next, the cells were irradiated with different doses and then incubated for 7 days to allow colony growth. Then, colonies were stained with crystal violet, and those containing 50 or more cells were counted. The plating efficiency was calculated by dividing the average number of counted colonies by the number of seeded cells. Survival fractions (SFs) were calculated by normalization to the plating efficiency of the respective unirradiated controls (32). After estimation of the SF at different radiation doses, the survival curve (log of SF vs. the radiation dose) was plotted, and the D₀ value for each group was calculated using the following equation: SF = 1 – (1 – e^D/D0)ⁿ(32). The D₀ value, which represents the radiation dose required to reduce the SF from 100 to 37%, is considered a measure of the intrinsic radiosensitivity of cells (33). The sensitization enhancement ratio for each treated group was determined by the ratio of the D₀ of the control group to that of the treated group (33).

Cells were seeded into 96-well plates at a density of 2 × 10³ cells per well and cultured for 12 h. Cell proliferation was assessed using a Cell Counting Kit-8 assay (Fluorescence, Beijing, China) according to the manufacturer's instructions. Absorbance was measured at a wavelength of 450 nm on a Model 550 microplate reader (Bio-Rad Laboratories, Hercules, CA, USA).

The effects of miR-221/222 and irradiation on the cell cycle and apoptosis in IOMM-Lee cells were examined by flow cytometry. Pretreated IOMM-Lee cells in the log phase of growth were stained with Annexin V/fluorescein isothiocyanate (FITC) and propidium iodide (Beyotime, China). Cell cycle and apoptotic rate were examined with a fluorescence-activated cell-sorting (FACS) flow cytometer (BeamCyte, China), and the data were analyzed using CellQuest Software. The percentages of cells in G0/G1 phase and the apoptotic rate were measured by calculating the ratio of the number of corresponding cells and that of total cells. For each sample, 10,000 cells were measured.

The invasive potential of the pretreated cells was evaluated by measuring the number of cells that invaded Matrigel-coated Transwell chambers. Prior to the experiment, Transwell inserts with 8-μm pores were coated with Matrigel and reconstituted with fresh medium for 2 h. Cells (1 × 10⁵/ml) were seeded into the upper chambers in 200 μl serum-free DMEM, while DMEM supplemented with 10% fetal bovine serum (700 μl) was placed in the lower chamber. After incubation for 48 h, cells that degraded the Matrigel and invaded the lower surface of the Matrigel-coated membrane were fixed with 70% ethanol, stained with hematoxylin, and counted in five random fields at a magnification × 200 under an optical microscope.

The 3′-untranslated region (UTR) of phosphatase and tensin homolog (PTEN), which contains the predicted binding sites of miR-221/222, were cloned into the XhoI site of the psi-check2 reporter vector (Biomed, Beijing, China). For the luciferase reporter assays, IOMM-Lee cells were cultured in 24-well plates with three replicates, incubated for 24 h, and transfected with 500 ng of psi-check2-PTEN or psi-check2-control plasmids with/without 100 nM miR-221 mimics or miR-222 mimics using Lipofectamine 3000. Luciferase activity was measured 48 h after transfection using dual-luciferase reporter assay system (Promega, WI, USA) according to the manufacturer's procedures. Data were normalized by Firefly/Renilla luciferase activity.

Protein of IOMM-Lee cells from each subgroup was extracted using radioimmunoprecipitation assay (RIPA) buffer (Beyotime, Shanghai, China). Their concentration was determined using a BCA Protein Assay Kit (Beyotime, Shanghai, China). Equal amounts of protein (5 μg) were then subjected to 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) followed by transfer of protein to polyvinylidene fluoride (PVDF) membranes (Millipore, Darmstadt, Germany). Membranes were subsequently blocked in Tris-buffered saline containing 0.1% Tween-20 and 5% skimmed milk powder and were incubated with primary antibodies against PTEN (1:1,000 dilution) and β-actin (1:4,000 dilution) (Cell Signaling Technology, MA, USA) overnight at 4°C. Horseradish peroxidase (HRP)-conjugated secondary antibodies for PTEN (1:2,000 dilution) and β-actin (1:4,000 dilution) (Cell Signaling Technology, MA, USA) were used afterward. The blots were detected using Pierce™ ECL Western Blotting Substrate (Thermo Fisher Scientific, Rockford, USA), and the membranes were developed using a ChemiDoc MP imaging system (Bio-Rad, Hercules, CA).

All animal studies were conducted in accordance with an approved institutional animal care and use committee protocol of our hospital (202001014). IOMM-Lee cells (5 × 10⁶) were injected subcutaneously into the flank position of 5-week-old female BALB/c nude mice. When the tumors reached 5 mm in diameter, animals were randomly divided into 4 groups of 16 mice each and were, respectively treated with intratumoral injections of saline, scramble oligonucleotides, miR-221/222-3p agomirs, and miR-221/222-3p antagomirs (RiboBio, Guangzhou, China) every 4 days for a total of three doses (3 nmol/dose). Eight animals from each group were radiated with two doses of 4 Gy during the intervals between injections. The entire mouse body except the tumor area was covered with lead sheets to avoid exposure to radiation during treatments. Vernier caliper was used to measure the length and width of tumors on alternate days, and tumor volumes were calculated as π/6 × (length × width²). Regression in subcutaneous tumor growth was followed, and mice were euthanized when tumor rupture and hemorrhage were observed in unradiated-control group. Immediately after the removal of the tumors, half of each tumor was stored in liquid nitrogen for the subsequent quantification of miR-221 and miR-222 by using quantitative real-time PCR (qRT-PCR); the other half was fixed in buffered formaldehyde and was sectioned and subjected to the later H&E and immunohistochemical staining for PTEN.

The abovementioned experiments were performed at least in triplicate, and data are presented as the mean ± standard deviation. The effects of miR-221/222 expression level and radiation dose on IOMM-Lee cells in vitro and in vivo were tested with two-way analysis of variance. Simple effect and pairwise comparisons with Bonferroni posttest were performed if the interaction between the factors appeared significant; otherwise, main effect and multiple comparisons with Bonferroni posttest were performed. Further comparisons of invasive cell numbers between the inhibitor group and the control/scramble group exposed to different radiation doses were analyzed by independent-sample t-tests. Multiple comparisons of xenograft tumor volumes between different treatment groups were analyzed by ANOVA with Bonferroni posttest. All P values are two-sided, and significance was defined using a threshold of 0.05. Statistical analyses were performed with SPSS version 19.0 (IBM Corp. Armonk, New York, USA).

IOMM-Lee cells were transfected with miR-221/222 mimics or inhibitors. qRT-PCR revealed that no significant difference in miR-221 and miR-222 expression between the control and the scramble group (miR-221: P = 0.7640, miR-222: P = 0.0856). Compared with that in either the control or the scramble group, the expression of miR-221 and miR-222 increased significantly in the miR-221/222-mimic group (miR-221: vs. control, P < 0.0001; vs. scramble, P < 0.0001; miR-222: vs. control, P < 0.0001; vs. scramble, P < 0.0001), while it decreased significantly in the miR-221/222-inhibitor group (miR-221: vs. control, P < 0.0001; vs. scramble, P < 0.0001; miR-222: vs. control, P < 0.0001; vs. scramble, P < 0.0001) (Figures 1A,B).

With an increase in radiation dose or a decrease in miR-221/222 expression level, the absorbance and colony number of IOMM-Lee cells decreased gradually, while apoptotic percentage and G0/G1 phase percentage increased; however, their invasive cell number increased as the radiation dose increased and decreased as the miR-221/222 expression level decreased (Table 1, Figure 1). Two-way ANOVA revealed significant simple effects of radiation dose and miR-221/222 expression level and their significant interactions in the proliferation, colony formation, apoptosis, and invasiveness of IOMM-Lee cells, while it exhibited their significant main effects in the sub-G0/G1 population, yet without significant interactions (Table 2).

Increasing radiation dose and downregulating miR-221/222 expression have synergistic effects on inhibiting proliferation and promoting apoptosis of IOMM-Lee cells (Tables 1, 2). Significant decrease in cell absorbance and colony number appears at each step-up of irradiation dose or each fall of the miR-221/222 expression (Tables 3, 5 and Figures 1C,I), indicating that the proliferation-inhibiting effect of radiation can be significantly enhanced by downregulating miR-221/222 expression.

As to the apoptosis of IOMM-Lee cells, pairwise comparisons of different groups revealed that, by irradiating with the same dosage, the apoptosis rate was significantly increased with downregulation of miR-221/222 expression (Table 5 and Figures 1D,E). Meanwhile, (1) the radiation dose that initially significantly promoted cell apoptosis was much higher in the miR-221/222-mimic group (6 Gy) than in the other groups (2 Gy) compared to their respective unirradiated cells; (2) within irradiated IOMM-Lee cells, the significant increase in apoptotic rate caused by each step-up of irradiation dose, which can be seen in cells with regular or decreased miR-221/222 expression, was not observed as the expression of miR-221/222 promoted; (3) in the comparisons between two irradiated subgroups with an incremental gradient of 4 or 6 Gy, the apoptosis rate increased significantly in the control, scramble, and miR-221/222-inhibitor group, whereas no significant differences were detected in the miR-221/222-mimic group (Table 3 and Figures 1D,E). These findings, from different perspectives, suggest that the apoptosis-promoting effect of radiation can be significantly enhanced by downregulating miR-221/222 expression in IOMM-Lee cells.

Further analysis of cell cycle distribution exhibits that the sub-G0/G1 population was positively correlated with radiation dose but negatively correlated with miR-221/222 expression (Table 4, Figures 1F,G). No corresponding effects on the sub-G2/M population were found. Although a significant effect of radiation dose on the S phase population was presented with an interaction with the miR-221/222 expression level (Table 2), no obvious radiation dose-dependent trend was explored in pairwise comparisons (Table 3).

Increased radiation dose and downregulated miR-221/222 have antagonistic effects on cell invasiveness (Tables 1, 2). Pairwise comparison analysis revealed that (1) invasive cell number at 8 Gy was significantly higher than that at lower radiation doses in the control or the scramble group, while invasive cell number at 6 Gy was only significantly higher than that of the unirradiated/2 Gy-irradiated control groups and unirradiated scramble group, respectively; (2) in the miR-221/222-mimic group, the invasive cell number for cells irradiated with a dose no lower than 4 Gy was significantly higher than that at lower radiation doses. However, in the miR-221/222-inhibitor group, the invasive cell number was significantly increased only in the comparison of 0 vs. 8 Gy (Table 3 and Figure 1H); (3) the expression level of miR-221/222 had no significant effect on cell invasiveness at a low radiation dose (≤2 Gy) compared to that of the control or the scramble group. Only at high radiation doses did the high expression of miR-221/222 exhibit a significant invasion-promoting effect (≥4 Gy), while the low expression of miR-221/222 presented a significant invasion-inhibiting effect (≥6 Gy) (Table 5 and Figure 1H).

By further comparing the invasive cell numbers of the inhibitor group and the control/scramble group exposed to different irradiation doses, it was revealed that although failed to completely reverse the 8-Gy-promoted invasiveness to a low-dose radiation-induced level, downregulation of miR-221/222 expression can completely reverse the 6-Gy-induced cell invasiveness to a level, which is without significant increase compared with that of the low-dose-irradiated control groups or unirradiated/low-dose-irradiated scramble groups (Tables 1, 6).

The effect of genetic manipulation of miR-221/222 on radiosensitivity of IOMM-Lee cells was investigated using a clonogenic assay. The D₀ value of the control, scramble, miR-221/222-mimic, and miR-221/222-inhibitor groups are 5.4242, 5.0970, 5.6025, and 4.1296 Gy, respectively. The sensitization enhancement ratio (SER) was 1.0642, 0.9682, and 1.3135 for the scramble, miR-221/222-mimic, and miR-221/222-inhibitor groups, respectively (Table 7, Figure 1J). These results revealed a negative correlation between the SER and the miR-221/222 expression, which provides strong evidence that downregulation of miR-221/222 expression can promote the radiosensitivity of IOMM-Lee cells.

Dual luciferase reporter assay revealed that cotransfection of miR-221 or miR-222 mimics with psi-check2-PTEN significantly decreased luciferase activity compared to scramble or control-treated cells (miR-221: P < 0.0001; miR-222: P < 0.0001) (Figures 2A,B). Western blot analysis showed that PTEN was upregulated gradually as the miR-221/222 expression level decreased or the radiation dose increased (Figures 2C,D). All these data demonstrated that PTEN is a target gene of the miR-221/222 cluster.

Dramatic reductions in tumor volume were observed in irradiated control (38.66%), scramble (38.56%), miR-221/222-mimic (33.78%), and miR-221/222-inhibitor (20.08%) groups as compared with their respective unirradiated counterparts (Figure 3), indicating a significant inhibitory effect of irradiation on the volume of subcutaneous IOMM-Lee xenografts in nude mice (P < 0.0001, partial η² = 0.474). For unirradiated animals, tumor volume in the miR-221/222-inhibitor group decreased by 85.21, 80.38, and 90.15% as compared with the control, scramble, and miR-221/222-mimic groups. The corresponding reduction rates in irradiation groups were 80.74, 74.48, and 88.11%. These suggest that the tumor volume can be suppressed by inhibiting the expression of miR-221/222 in vivo (P < 0.0001, partial η² = 0.862). Furthermore, these two treatments have a synergistic effect on preventing tumor growth in vivo (P = 0.002, partial η² = 0.232).

In addition to further proving these abovementioned results, multiple comparisons of tumor volumes at the last measure also revealed that no significant difference between the radiated-mimics group and the unradiated-control or unradiated-scramble group, indicating the antagonistic effect between radiation and upregulated miR-221/222 expression in vivo (Table 8). A same situation was also found in the comparison between the unradiated-inhibitor group and the radiated-inhibitor group (Table 8). This may be explained by the observation that the effect of inhibiting the miR-221/222 expression on preventing tumor progression has already been too obvious to reflect the effect of radiation.

Immunohistochemical analysis of tissue sections of xenografts reflects a gradually increased immunoreactivity of PTEN as the miR-221/222 expression decreases (Figure 4); meanwhile, tissue sections from radiation-treated xenografts exhibited higher expression levels of PTEN compared to their corresponding unirradiated counterparts (Figure 4). These are consistent with the results of in vitro studies.

The radiation dose-dependent apoptosis-promoting and proliferation-inhibiting effects of radiation on IOMM-Lee cells provide theoretical bases for utilizing postoperative radiation therapy to control the growth of residual or recurrent meningiomas in the clinic (Tables 1–3, 8, Figures 1C–E, 3). However, the radiation-induced invasiveness of IOMM-Lee cells may explain the unsatisfactory recurrence-free survival or even some toxicities of clinical adjuvant radiotherapy (Tables 1–3 and Figure 1H).

It has been revealed in several cancer cells (including breast, lung, and liver cancer, and glioma cells) that ionizing radiation (IR) enhances their migratory and invasive properties by inducing the epithelial–mesenchymal transition (EMT) (35–40). This IR-induced EMT is mediated by EMT-inducing transcription factors (EMT-TFs) (e.g., Snail, ZEB, and Twist families) that are activated by a network of signaling pathways (41–44). These EMT-TFs possess two potentials in cancer cells: (1) prometastatic potential—the aforementioned IR-enhanced migration and invasiveness reflect their prometastatic role. They regulate the expression level of proteins that is implicated in cell polarity, cytoskeletal structural maintenance, cell–cell contact, and extracellular matrix degradation, and they suppress key epithelial genes (e.g., E-cadherin) (41–44); (2) oncogenic potential: they are implicated in inducing malignant transformation (41, 45), stemness properties (41, 45), and oncogenic metabolism (41, 44). Hence, it is logical to assume that the present radiation-enhanced invasiveness of IOMM-Lee cells may be caused by the IR-induced EMT. In addition, we revealed in our previous clinical study that malignant progressed atypical meningiomas are more likely to exhibit low connexin 43 expression in their preradiotherapeutic tissues (46). Malignant transformation is one of the toxicities of radiotherapy in meningiomas (2). Connexin 43, the most abundant connexin isoform in the central nervous system (47, 48), oligomerizes to form gap junctions between adjacent meningioma cells (49, 50). These two points, as well as the present radiation-induced invasiveness of IOMM-Lee cells, all correspond to the prometastatic and oncogenic capacities of EMT-TFs. Al-Mefty et.al discovered the same complex genetic alterations that they saw in histologically higher-grade meningiomas already apparent in the early, benign stages of those tumors (51). Arishima et al. reported that different subtypes of meningiomas express different levels of connexin 43 (52). These findings raises the possibility that meningioma cells' inherent expression levels of certain moleculars and the intrinsic regulation level of EMT-TFs may determine whether this meningioma will undergo invasiveness enhancement, tumor recurrence, or malignant progression after radiotherapy.

The radiosensitization of downregulating the miR-221/222 cluster has been certified in several human tumors: Zhang et al. successively discovered that tumor radiosensitivity could be promoted by the knockdown of miR-221 and miR-222 in gastric cancer cell line SGC7901 (31) and glioblastoma cell line U251, and demonstrated that PTEN is a target gene of the miR-221/222 cluster (31); Sun and Khoshinani confirmed that miR-221 (33) and miR-222 (32) mediated the radiosensitivity of colorectal cancer cells by regulating PTEN, respectively; consistent results were reported by Wu and his colleague in their study of nasopharyngeal carcinoma (34). The radiosensitivity enhancement of miR-221/222 downregulation and PTEN as the target gene of these two miRNAs were also confirmed in our present study of IOMM-Lee meningioma cells.

The PTEN gene, located at 10q23.3, was identified as one of the most commonly mutated tumor suppressor in human cancers, second only to p53 (53). Its encoded PTEN protein exhibits phosphoinositide 3-phosphatase activity toward phosphatidylinositol 3,4,5-trisphosphate and antagonizes phosphatidylinositol 3-kinase (PI3K) functions to negatively regulate cell proliferation and promote cell apoptosis (54). Loss-of-function mutations in the PTEN gene result in the inactivation of the PTEN protein, which further gives rise to oncogenic transformation of cells, resistance, and relapse in response to conventional therapeutic agents (55, 56).

IR exerts its therapeutic effect mainly by generating DNA damages (57). These IR-induced DNA damages, primarily double-strand breaks, trigger a number of DNA damage response and repair signaling cascades and subsequently phosphorylate p53 protein (58–61). Activated p53 upregulates the transcriptional and translational levels of several genes (including PTEN) to cause cell cycle arrest, apoptosis, autophagy, or senescence according to the severity of the DNA damage and the cell type (42, 62–65). Meanwhile, accumulation of PTEN, in turn, remarkably enhances p53 DNA binding and transcriptional activity by interacting with its C-terminal domain (66). Briefly, IR can induce PTEN accumulation to facilitate its therapeutic effects in some tumors. H460 cells obtained enhanced PTEN expression after irradiation in Il Lae Jung's previous research of nonsmall cell lung cancer (67). Similarly, the present radiation dose-dependent increase in PTEN in the IOMM-Lee cells suggests that the abovementioned mechanisms were activated during radiotherapy in meningiomas. Moreover, improved radiotherapeutic response in meningiomas can be achieved by further upregulation of PTEN through inhibiting the miR-221/222 expression.

The radiosensitization of PTEN has also been reported in previous literature: Rosser et al. identified forced expression of PTEN as a valuable approach to achieve radiosensitization in prostate cancer cells (68); multiple studies confirmed that the radioresistance of nasopharyngeal carcinoma could be enhanced by suppressing the expression of PTEN (69–71); consistent conclusions were obtained in the corresponding researches of non-small cell lung carcinoma (72), hepatocellular carcinoma (73), and esophageal cancer (74). Accordingly, the present observations indicate that the radiosensitization of miR-221/222 inhibition in IOMM-Lee cells was achieved by its further upregulation of PTEN expression on the basis of IR-induced PTEN accumulation.

The IR-induced invasiveness of IOMM-Lee cells enhanced significantly as radiation dosage increased. Downregulation of miR-221/222 could promote the expression of PTEN and reverse the IR-enhanced cell invasiveness. As previously described, the IR-enhanced cell invasiveness is associated with EMT. Aside from their abovementioned radiosensitivity-regulatory effect, miR-221 and miR-222 have also been revealed to promote EMT (75) and increase migration and invasion in several other tumors (76). As a target of the miR-221/222 cluster, PTEN has been verified to possess the ability of reversing EMT in Jin's radioresistant esophageal cancer cells study (74). Therefore, it is conceivable that miR-221/222 downregulation reverses the radiation-induced cell invasiveness and is achieved by the EMT-reversion effect of accumulated PTEN. However, the underlying mechanisms of PTEN-regulated EMT in meningiomas require further investigations.

In previous studies of IOMM-Lee cells, (1) Gogineni et al. indicated that radiation treatment (7 Gy) induced G2/M cell cycle arrest and a resultant decrease in the G0/G1 or S phase when evaluated against the unirradiated cells, as well as an insignificant cell-death-promoting effect (77). (2) However, by comparing the 5-Gy irradiated cells with unirradiated cells, Winson et al. exhibited a cell cycle distribution consist of an increased G0/G1, a decreased S, and an increased G2/M population, accompanied by an increment in apoptosis rate (78). The present results of radiation-induced G0/G1 cell cycle arrest and apoptosis-promoting effect are consistent with Winson's research while opposite to Gogineni's study (Tables 1–4 and Figures 1D–G). To determine the underlying causes of these opposites, we compared the data and discovered that the main difference is the dose rate, which was 0.71 and 3.2 Gy/min in the Gogineni's study (77) and the present study, respectively, while Winson et al. did not provide theirs (78).

According to Hall's revised and updated illustration of the dose-rate effect (79), the dose–response curve becomes progressively shallower as the dose rate reduces, indicating an increment in sublethal damage repair. Cells rest on their cell cycle phase without progression. However, a further reduction in dose rate in a limited range allows cells to progress through the cycle and accumulate in G2, resulting in the inverse dose-rate effect. The critical dose rate of IOMM-Lee initiating this effect has not been determined. With a higher dose rate in the present study, the capability of cells to repair sublethal damage was restrained, which further leads to an increase in apoptotic rate with a fixed cell cycle distribution. The dose rate used by Gogineni et al. is lower, which might have triggered the inverse dose-rate effect. This dose rate might not significantly increase cell death but have gradually accumulated cells to rest on G2 phase, and these may explain their cell cycle and apoptosis results.

In Kurpinski's research of differential effects of X-rays on human mesenchymal stem cells (80), it is proved that X-ray at a high dose rate (1 Gy/min) induces a significant increase in population of G0/G1 phase, a decrease in S phase, and no significant changes in G2/M phase in comparison with a low dose rate counterpart (0.1 Gy/min). We observed that the sub-G0/G1 population, which is referred to as an indicator of cell death, increased following a high dose rate radiation in IOMM-Lee cell (Tables 1, 2, 4 and Figures 1F,G). Combine with the dose-rate effect, these indicate that, within certain range, a higher dose-rate radiation treatment induces G0/G1 arrest and a relevant increased sub-G0/G1 population.