ACYP2 contributes to malignant progression of glioma through promoting Ca²⁺ efflux and subsequently activating c-Myc and STAT3 signals

A total of 52 frozen surgical gliomas and 24 normal brain tissues from cerebral contusion and laceration patients were randomly obtained from the First Affiliated Hospital of Xi’an Jiaotong University. A part of the above tissues were taken, fixed in 10% formalin and embedded in paraffin for immunohistochemical analysis. None of these patients received any preoperative chemotherapy, radiotherapy or other biological therapy, and all patients signed an informed consent before the surgery. All of the tissues were histologically examined by two senior pathologists at the Department of Pathology of the Hospital based on World Health Organization (WHO) criteria, and study protocol was approved by the Institutional Review Board and Human Ethics Committee of the First Affiliated Hospital of Xi’an Jiaotong University.

RNA isolation, cDNA synthesis and RT-qPCR was carried out as described previously [16]. The mRNA expression of the indicated genes was normalized to 18S rRNA, and each sample was run in triplicate. The primer sequences were summarized in Additional file 1: Table S1.

Human glioma cell lines U251, SHG44, A172, U87, BT325 and SF295 were provided by Cell Bank of the Zhongshan University. A172 and BT325 was provided by Kunming Cell Bank of The Chinese Academy of Sciences. Cells were all routinely cultured at 37 °C in DMEM medium with 10% fetal bovine serum (FBS). All cell lines used in this study were authenticated by short tandem repeat (STR) analysis in Genesky Co. Ltd. (Additional file 1: Table S2), and the results was completely consistent with previous studies [16] and database (Cellosaurus: https://​web.​expasy.​org/​cellosaurus/​). In some experiments, cells were treated with 100 μM cell-permeable c-Myc-Max dimerization inhibitor 10,058-F4 (Selleck Chemicals) for 48 h to inhibit transcriptional activity of c-Myc. Cells were treated with 5 μM BAPTA-AM (Selleck Chemicals) for 6 h to chelate intracellular Ca²⁺. Cells were treated with 10 μM calpeptin (Selleck Chemicals) for 12 h to block calpain activity. Cells were treated with 10 μM sodium orthovanadate (Na₃VO₄) for 1 h to inhibit PTP1B activity. The same volume of the vehicle was used as the control.

Oligonucleotides of siRNAs targeting ACYP2, PMCA4 and PTP1B were obtained from Gene Pharma (Shanghai, China) and Ribobio (Guangzhou, China), respectively. The sequences were presented in Additional file 1: Table S3. Cells were transfected at 50% confluence using Lipofectamine 2000 (Invitrogen, Grand Island, NY) according to the instructions of the manufacturer, with a final siRNA concentration of 50 nM. All silencing experiments were carried out in triplicate. Two oligonucleotides with maximal knockdown efficiency were selected among three different sequences.

Open reading frame (ORF) of ACYP2 with stop codon was amplified and then cloned into pcDNA3.1(−) mammalian expression vector, termed pcDNA3.1(−)-ACYP2. The primer sequences were shown in Additional file 1: Table S4. Cells were transfected with the indicated constructs at 70% confluence using X-treme GENE HP DNA Transfection Reagent (Invitrogen, Grand Island, NY) according to the instructions of the manufacturer. Lentivirus encoding shRNA targeting ACYP2 and control lentivirus were obtained from HanBio Biotechnology Co., Ltd. (Shanghai, China). Cells were transfected at 50% confluence with a final lentivirus multiplicity of infection (MOI) of 50–100 according to the instructions of the manufacturer. Cells stably knocking down ACYP2 were selected by puromycin.

Cell proliferation was evaluated by MTT assay. Soft-agar assay was performed to assess colony formation ability. Cell apoptosis was evaluated by flow cytometer. Cell migration and invasion abilities were evaluated by transwell chambers. Each experiment was run in triplicate.

The detailed procedures were carried out as described previously [17]. Antibody information was summarized in Additional file 1: Table S5.

The intracellular cytosolic-free Ca²⁺ concentration was measured under the confocal microscope (Leica) or flow cytometer by using the Ca²⁺-sensitive dye Fluo-4 AM (Molecular Probes, Invitrogen). For the former, cells were seeded in Glass Bottom Culture Dishes (MatTek Corporation) before transfection with siRNAs targeting ACYP2 or control siRNA. After 48 h, the medium was removed, and cells were loaded with Fluo-4 AM (1 μmol/L) for 30 min at 37 °C with gentle shaking. Next, cells were washed and incubated for 20 min at 37 °C prior to experiments. Fluorescence intensity was determined at 494 nm excitation and 516 nm emission. For flow cytometer analysis, cells were trypsinized, washed and placed in Eppendorf tubes at 1 × 10⁶/mL, and incubated with Fluo-4 AM (1 μmol/L) at 37 °C for 30 min. Data were expressed as fluorescence intensity.

Activated calpain in the protein extract was measured using a calpain activity assay kit (Abcam, Cat. # ab65308). Protein lysates were extracted following the manufacturer’s instruction. Briefly, cytosolic protein extracts of glioma cells were prepared with the extraction buffer which prevents the auto-activation of calpain during the extraction procedure. The calpain activity was quantified by the measurement of fluorescence reading at λ max =505 nm using calpain substrate Ac-LLY-AFC. The activity was represented as relative fluorescence units (RFU)/mg protein.

Four-week-old male athymic nude mice were purchased from SLAC laboratory Animal Co., Ltd. (Shanghai, PR. China) and housed in a specific pathogen-free (SPF) environment. The mice were randomly divided into four groups (5 mice per group). SF295 cells stably knocking down ACYP2 or control cells (1 × 10⁷) were implanted in nude mice to establish tumor xenografts. From day 3 post-injection, BAPTA-AM (9 mg/kg), calpeptin (2 mg/kg) or vehicle were administered by intraperitoneal injection in 1.0% DMSO, and tumor size was measured every 2 days. Tumor volumes were calculated by the formula (length × width² × 0.5). Dosing was daily for 9 consecutive days. At the end of experiments, xenograft tumors were harvested and weighted. In addition, a part of tumor tissues were fixed in 15% formalin for 24 h, embedded in paraffin, and sectioned at 4 μm until use. Proliferation ability of xenograft tumors was assessed by quantification with Ki-67 immunohistochemistry. All animals’ experimental procedures were approved by the Laboratory Animal Center of Xi’an Jiaotong University.

The IHC assay was carried out to evaluate protein expression of ACYP2, c-Myc, NCL, phosphorylated STAT3 (Tyr705) and Ki67. The detailed protocol was carried out as described previously [18].

Gene expression in tumor tissues and control subjects were compared by the unpaired t test. Survival curves were constructed according to the Kaplan-Meier method, and statistical analysis was performed using the Log-rank test. All statistical analyses were performed using the SPSS statistical package (16.0, Chicago, IL). P < 0.05 were considered statistically significant.

We first examined mRNA and protein expression of ACYP2 in 52 gliomas and 24 normal brain tissues (control subjects) by qRT-PCR, immunohistochemistry and western blot assays. As shown in Fig. 1a-c, ACYP2 expression was significantly elevated in glioma tissues in comparison with control subjects at both mRNA and protein levels. Next, through analyzing the Cancer Genome Atlas (TCGA) dataset, we found that increased expression of ACYP2 was clearly associated with poor survival in low-grade glioma (LGG) patients (HR =0.23 with log-rank P < 0.0001; Fig. 1d), but not in glioblastoma (GBM) patients (data not shown).

A series of in vitro experiments were performed to determine biological role of ACYP2 in malignant phenotypes of glioma cells. First, we used two different siRNAs (si-ACYP2–482 and − 540) to knock down ACYP2 expression in U251, SF295 and U87 cells, and validated their knockdown efficiency by qRT-PCR (Additional file 2: Figure S1) and western blot analysis (Fig. 2a). The results showed that ACYP2 knockdown in U251 and SF295 cells significantly inhibited cell proliferation (Fig. 2b) and colony-forming ability on soft agar (Fig. 2c) compared to the control. In addition, we also tested the effect of ACYP2 depletion on the apoptosis of glioma cells. As shown in Fig. 2d, ACYP2 knockdown in these cells dramatically induced cell apoptosis in comparison with the control (9.7 ± 2.9% vs. 17.0 ± 1.1% in U251 cells, P = 0.004; 7.3 ± 0.1% vs.15.2 ± 0.2% in SF295 cells, P = 0.005). On the other hand, ectopic expression of ACYP2 in SHG44 and A172 cells clearly promoted cell proliferation and colony-forming ability in comparison with the control (Fig. 2e-g). To further validate the above conclusions, we knocked down or ectopically expressed ACYP2 in primary tumor cells from two glioma patients, and demonstrated that ACYP2 knockdown significantly inhibited cell proliferation, while ectopic expression of ACYP2 strongly promoted cell proliferation (Fig. 2h and i). Taken together, these data support the oncogenic role of ACYP2 in glioma cells.

The effect of ACYP2 on cell migration and invasion was next assessed in three glioma cell liens. As shown in Additional file 2: Figure S2a, ACYP2 knockdown in U251, SF295 and U87 cells significantly inhibited cell migration in comparison with the control. In addition, we also found that ACYP2-knockdown cells exhibited a relatively low ability of cell invasion compared to control cells (Additional file 2: Figure S2b). On the other hand, ectopic expression of ACYP2 in SHG44 and A172 cells clearly promoted cell migration and invasion in comparison with the control (Additional file 2: Figure S3). These results indicate that there is a strong link between aberrant expression of ACYP2 and metastatic phenotypes of glioma cells.

ACYP can catalyze the hydrolysis of not only low molecular weight substrates, such as carbamoyl phosphate and succinoyl phosphate, but also the acylphosphorylated intermediates of P-type ATPases, including plasma membrane Ca²⁺-ATPase (PMCA) [19]. To reveal molecular mechanisms underlying oncogenic role of ACYP2 in glioma cells, we first tested the effect of ACYP2 on intracellular Ca²⁺ concentration. As shown in Fig. 3a, under the fluorescence microscope, ACYP2 depletion in U251 and SF295 cells leaded to a significant increase of Ca²⁺ fluorescence intensity in the cytoplasm of tumor cells relative to the control. As supported, flow cytometry analysis showed rightward shifting of peak value in ACYP2-knockdown cells, indicating the increase of intracellular Ca²⁺ concentration (Fig. 3b). Besides, considering that endoplasmic reticulum (ER) Ca2+ pumps also belong to the P-type pump family, and there is evidence demonstrating that ACYP regulates Ca2 + −ATPase activity and Ca2+ transport on cardiac and skeletal ER [20, 21], thus we next tested the effect of ACYP2 on calcium pool in ER by loading the cells with the Ca2 + −sensitive fluorescent probe Mag-fura 2, which has relatively high Ca2+ affinity for ER. As expected, we found that ACYP2 knockdown in U251 and SF295 cells significantly decreased Ca2+ fluorescence intensity in ER relative to the control (Additional file 2: Figure S4). Given the above, we speculated that ACYP2 might act as its oncogenic role by altering intracellular Ca²⁺ homeostasis. To prove this, we treated U251, SF295 and U87 cells with 5 μM BAPTA-AM, a highly selective Ca²⁺ chelator, for 6 h. The results showed that inhibitory effect of ACYP2 knockdown on cell proliferation was effectively reversed upon BAPTA-AM treatment (Fig. 3c). This was also supported by cell migration assay (Additional file 2: Figure S5).

Calpains are a family of cytosolic cysteine proteinases whose enzymatic activities depend on Ca²⁺. Members of the calpain family are believed to function in various biological and pathological processes, including tumorigenesis [22‐24]. Thus, we suppose that ACYP2 may alter the activity of calcium-dependent calpains through modulating intracellular Ca²⁺ homeostasis. This will lead to changes of proteolytically cleavage of transcription factors in the cytoplasm and abnormal activation of some major signaling pathways, thereby promoting malignant progression of glioma. To prove this, we used Calpain Activity Assay Kit to determine the effect of ACYP2 depletion on calpain activity in U251, SF295 and U87 cells. Active calpain and calpain inhibitor calpeptin were used as positive and negative control, respectively. As shown in Fig. 3d, ACYP2 knockdown caused a significant increase of calpain activity in these cells in comparison with the control. Next, we treated the above glioma cell lines with 10 μM calpeptin for 12 h, and expectedly found that inhibitory effect of ACYP2 knockdown on cell proliferation was reversed upon calpeptin treatment (Fig. 3e). Similarly, calpeptin treatment also effectively reversed inhibitory effect of ACYP2 knockdown on cell migration in comparison with the control (Additional file 2: Figure S6). These data indicate that ACYP2 acts as an oncogenic function in glioma cells through altering intracellular Ca²⁺ homeostasis and calpain activity.

In addition to Ca2+/calpain signaling pathway, the activation of NFATc1, 2, 3 and 4 is also intracellular Ca²⁺-dependent, and there is evidence showing that NFATc1 as a key transcription factor can promote the migration and adhesion of GBM cells [25]. Thus, we also determined the effect of ACYP2 knockdown on the activity of NFATc1 by western blot and immunofluorescence assays. The results showed that ACYP2 knockdown almost did not affect NFATc1 expression in both cytoplasm and nucleus, cytoplasm-to-nucleus translocation of NFATc1, and the expression of its downstream target PTGS2 (Additional file 2: Figure S7). It has also been reported that intracellular Ca2+ can regulate NF-kB transcriptional activity in human aortic endothelial cells [26]. This was supported by our data that ACYP2 knockdown clearly decreased the expression of downstream targets of NF-kB signaling, Bcl-xL and Bcl-2, in U251 and SF295 cells (Additional file 2: Figure S8). The above data indicate that ACYP2 may be involved in various intracellular Ca2 + −related signaling pathways, and we mainly focused on Ca2+/calpain-related signaling pathway in this study.

Calpain is a proteinase with relatively poor specificity and functions in many cellular pathways through controlled proteolysis of various substrates [27]. Thus, we speculate that ACYP2/Ca2+/calpain signaling axis may act on some key transcription factors and affect their activity, thereby contributing to malignant phenotypes of glioma cells. There is evidence reporting that c-Myc proteins can be proteolytically cleaved by Ca2 + −dependent calpains at lysine 298, and this generates Myc-nick, a transcriptionally inactive cleavage product of c-Myc oncoproteins [28]. To prove this, we examined c-Myc levels in glioma cells ectopically expressing or knocking down ACYP2 by western blot using antibody against C terminus of c-Myc (9E10), which recognizes only full-length c-Myc [29]. As shown in Fig. 4a, ectopic expression of ACYP2 clearly elevated the levels of transcriptionally active c-Myc and its downstream target NCL (nucleolin) in SHG44 and A172 cells. Conversely, ACYP2 knockdown decreased their levels, while this effect could be effectively reversed by calpeptin treatment (Fig. 4b). This was also supported by the results of qRT-PCR assay (Fig. 4c). Besides, ACYP2 knockdown also significantly down-regulated the expression of its other two downstream targets, E2F2 and cyclin E, in U251 and SF295 cells compared to the control (Additional file 2: Figure S9). Similarly, BAPTA-AM treatment also reversed inhibitory effect of ACYP2 depletion on transcriptional activity of c-Myc (Fig. 4d). Next, we attempted to determine the potential role of c-Myc in oncogenic function of ACYP2 in glioma cells. The results showed that promoting effect of ACYP2 on glioma cell proliferation could be partially reversed by the treatment of c-Myc inhibitor 10,058-F4 (Fig. 4e).

Signal transducer and activator of transcription 3 (STAT3) has been reported to be aberrantly activated in gliomas, and be identified as a potential therapeutic target [30]. There is evidence showing that STAT3 activation is dependent on Ca2+ signal in breast cancer cells, and phosphorylation of STAT3 at Tyr705 by EGF can be substantially inhibited upon intracellular Ca2+ chelation [31]. Thus, we suppose that ACYP2 may promote phosphorylation and activity of STAT3 through modulating Ca2+/calpain signaling axis, contributing to malignant progression of glioma cells. The results showed that ACYP2 knockdown dramatically decreased the levels of STAT3 phosphorylation (Tyr705) in comparison with the control, while this effect could be effectively reversed by BAPTA-AM or calpeptin treatment (Fig. 5a and b). As expected, ACYP2 knockdown dramatically decreased the expression of its two downstream targets, c-Fos and c-Jun, in U251 and SF295 cells compared to the control (Additional file 2: Figure S9).

It should be noted that calpain-deficient breast cancer cells exhibit impaired invadopodia formation, which can be rescued by ectopic expression of protein tyrosine phosphatase 1B (PTP1B) [32]. In addition, PTP1B cleavage by calpain is considered as a general feature of platelet agonist-induced aggregation [33]. Thus, we speculate that ACYP2 may activate STAT3 by regulating Ca2+/calpain/PTP1B signaling axis. To prove this, we used the PTP1B inhibitor sodium aluminate (Na3VO4) to treat U251 and SF295 cells stably knocking down ACYP2 and control cells. As shown in Fig. 5c and d, Na3VO4 could reverse inhibitory effect of ACYP2 depletion on STAT3 phosphorylation and cell proliferation. Considering that Na3VO4 is a non-specific PTP1B inhibitor, we next used siRNA approach to knock down PTP1B, and observed the similar findings, as shown in Fig. 5e and f. Finally, we treated SHG44 and A172 cells with 1 or 5 μM STAT3 inhibitor Stattic, and found that Stattic partially reversed promoting effect of ACYP2 on cell proliferation in a dose-dependent manner (Additional file 2: Figure S10). The above findings indicate that ACYP2 can activate c-Myc and STAT3 signals through modulating intracellular Ca²⁺ homeostasis and calpain activity, thereby contributing to malignant phenotypes of glioma cells.

Given that PMCAs belong to the P-type pump family, which is characterized by the formation of a high-energy acylphosphorylated intermediate during the reaction cycle [34], thus we hypothesize that ACYP2 may regulate intracellular Ca2+ homeostasis in glioma cells by hydrolyzing phosphoenzyme intermediate of PMCAs. In mammals, PMCAs are the products of four distinct genes (PMCA1–4), and the expression of different PMCA isoforms is regulated in a developmental, tissue- and cell type-specific manner [35]. Thus, we first determined mRNA levels of PMCA1–4 in a panel of glioma cell lines by qRT-PCR assay, and found relatively high expression of PMCA1 and 4 in most of cell lines (Additional file 2: Figure S11). In addition, there is evidence indicating that the remodeling of PMCA4 in colon cancer can promote proliferative pathways, while avoiding increased sensitivity to apoptotic stimuli [36]. Thus, we first knocked down PMCA4 using two different siRNAs (si-PMCA4–1023 and − 2728) in U251 and SF295 cells, and found that PMCA4 knockdown significantly inhibited cell proliferation in comparison with the control (Fig. 6a and b). Next, we performed Co-IP assay to confirm the interaction between ACYP2 and PMCA4 in the above cells (Fig. 6c). In addition, our data also demonstrated that the promoting effect of ACYP2 on cell proliferation and the activities of c-Myc and STAT3 could be clearly reversed by PMCA4 knockdown (Fig. 6d and e).

Based on the above findings, we propose a simple model to illustrate the mechanism underlying oncogenic role of ACYP2 in glioma (Fig. 6f). Briefly, ACYP2 interacts with and activates PMCA4 to promote Ca²⁺ efflux through hydrolyzing its phosphoenzyme intermediate. This will lead to decreased activity of calpain and subsequent activation of c-Myc and STAT3, thereby promoting malignant phenotypes of glioma cells.

We also evaluated tumorigenic potential of ACYP2 in nude mice. Our data showed that the tumors induced by SF295 cells stably knocking down ACYP2 showed a much slower growth rate and smaller mean tumor volume than the tumors induced by control cells, while this effect could be clearly reversed by BAPTA-AM or calpeptin treatment (Fig. 7a and b). Next, we performed western blot analysis in xenograft tumors to further confirm the effect of ACYP2 on the activities of c-Myc and STAT3 in vivo. As shown in Fig. 7c, ACYP2 expression was significantly down-regulated in the ACYP2-knockdown tumors compared to control tumors; meanwhile, we also found that the levels of transcriptionally active c-Myc and its downstream target NCL, and STAT3 phosphorylation were dramatically decreased in the former relative to the latter, and this effect was similarly reversed by BAPTA-AM or calpeptin treatment. This was also supported by IHC assays (Fig. 7d). In addition, IHC results showed that ACYP2 knockdown leaded to low proliferative capacity of xenograft tumors relative to the control, as reflected by decreased percentage of Ki-67 positive cells, and this effect was similarly reversed by BAPTA-AM or calpeptin treatment (Fig. 7e). Collectively, these findings further support oncogenic role of ACYP2 in glioma cells by modulating intracellular Ca²⁺ homeostasis.