Background

Glioma is one of the most common types of adult brain tumors, with high morbidity and mortality [1,2,3,4]. Although multi-dimensional treatments, including surgical resection, radiotherapy and chemotherapy, have been developed, the infiltrating growth pattern of glioma leads to poor prognostic outcomes. The median survival time in patients with glioblastoma multiform (GBM), which is the most malignant tumor type of glioma, is less than 15 months [5]. Thus, a better understanding of molecular pathogenesis and mechanisms of glioma generation is urgently needed to develop more effective treatments.

The spontaneous regression of malignancies is a phenomenon in which a proven tumor partially or completely disappears without any substantial treatment [6]. Since Everson and Cole first reported the spontaneous tumor regression in 1967 [7], more than 1000 case reports of spontaneous tumor regression have been published, these articles reported on common associated tumors, including glioma [8,9,10], choriocarcinoma [11], neuroblastoma [12], renal cell carcinoma [13], melanoma [14], sarcoma [15], bladder carcinoma [16] and osteogenic sarcoma [17]. Recent studies have reported that almost all tumors can undergo spontaneous regression [18,19,20,21,22,23,24,25]. Spontaneous tumor regression may be triggered by immune regulation, induced differentiation, hormone mediation, elimination of carcinogens, psychological factors, apoptosis and epigenetic mechanisms [6].However, the underlying mechanism causeing a potentially lethal tumor strikingly shift course without any intervention remains unknown.

Due to low incidence of spontaneous tumor regression and the difficulty of establishing a model, most reports are case studies and phenomenon analyses, and studies on the molecular mechanism of establishing tumor regression models of spontaneous tumor regression are rare. Current available methods for CTVT and surgical transplantation are limited by specific tumor models, high technical threshold and difficulties in the surgical transplant [26, 27]. Therefore, the establishment of a simple, reproducible spontaneous tumor regression model with a high success rate is necessary to study the mechanism of spontaneous tumor regression.

Long non-coding RNA (lncRNA) is a non-encoded RNA with a length greater than 200 nucleotides. Based on different lncRNA profiles during mammalian development and in various diseases [28, 29], researchers have demonstrated that lncRNAs perform important tasks throughout the course of life. However, lncRNAs responsive to spontaneous tumor regression have rarely been reported. Therefore, it is necessary to investigate the biological roles of lncRNAs in spontaneous tumor regression.

Materials and methods

Cell culture

U87MG cells (human malignant glioblastoma multiforme cell line) were purchased from ATCC (Manassas, Virginia, USA), G422 cells (mouse malignant glioblastoma multiforme cell line) were obtained from Land biotechnology (Guangzhou, China). The cells were grown and cultured with DMEM supplemented with 10% FBS (Invitrogen, USA) in 5% CO2 incubator at 37 °C.

Tumor regression models

BALB/c mice (male, 4–6 weeks old, approximately 20 g) were obtained from the Experimental Animal Center at Southern Medical University (Guangzhou, China) and maintained in a sterile environment in accordance with standard animal care guidelines. The experiments were performed according to national regulations. The mice were administered cyclosporine soft capsule [30,31,32] (HuaDong Medicine Co., Ltd, China) (40 mg/kg/day), and the U87 (6 × 106) cells were subcutaneously inoculated on the right back of mice three days later. After 8 days of inoculation, the mice were randomly divided into two groups (A and B) when the tumor was successfully established. Group A continued to receive treatment with cyclosporine soft capsules (40 mg/kg/day), while group B ceased to receive treatment. Digital calipers (Mitutoyo, Japan) was used to measure tumor length (L) and short diameter (W) every three days (tumor volume V = LW2/2), and the IVIS Spectrum Imaging System (Bruker, USA) was used to monitor the growth process of fluorescent labeled U87 tumor. The mice were sacrificed, and the tumor tissue was collected when group B showed tumor regression of approximately 50%.

RNA extraction

Total RNA was extracted from U87 tumor sample according to the manufacturer’s instructions of Trizol®LS (Invitrogen, Stockholm, Sweden) and the quantity and quality was confirmed by a NanoDrop 1000 Spectrophotometer. The total amount of RNA obtained was not significantly different between the samples.

LncRNA microarray assay

Isolation of total RNA from tumor sample was performed as described above. Sample labeling and array hybridization were performed according to the Arraystar microarray-based gene expression analysis protocol (Agilent Technology). The Arraystar Mouse LncRNA Microarray V3.0 which is designed for the global profiling of mouse lncRNA and protein-coding transcripts can detected approximately 35,923 lncRNAs and 24,881 coding transcripts. Agilent Scanner G2505B was uded to scan the arrays, and Agilent Feature Extraction software (version 11.0.1.1) was used to analyze the acquired array images, and GeneSpring GX v12.1 software (Agilent Technologies) was used to quantile normalization and subsequent data processing. The microarray assay was carried out by Kangchen Bio-tech (Shanghai, China) and The procedures were conducted as previously described [33].

Quantitative reverse transcription-polymerase chain reaction (RT-qPCR)

To initiate cDNA synthesis, the PrimeScript™ RT Reagent Kit with gDNA Eraser (Perfect Real Time) (TaKaRa, Japan) was used to reverse transcribe (RT) 2 µl of total RNA according to the manufacturer’s instructions. The real-time monitoring of the DNA amplification reaction was performed using a Real-Time PCR Detection System (Light Cycler 480, Roche) and SYBR®Premix Ex Taq™ (Tli RnaseH Plus) (TaKaRa, Japan). β-actin was performed as the internal control and used to normalize the results. The primers used for RT-qPCR are listed in Table 1.

Table 1 The primers used in RT-qPCR analysis
Full size table

Gene ontology (GO) and kyoto encyclopedia of genes and genomes (KEGG) pathway analysis

GO categories and GeneSpring GX software (http://www.geneontology.org) were used to determine processes or functional categories which were differentially expressed [34]. The GO annotation system uses a controlled and hierarchical vocabulary to assign function to genes or gene products in any organism [35]. The lower the P-values, the more significant the GO terms (P ≤ 0.05 was considered statistically significant).

We performed pathway analysis to map genes to KEGG for the differentially expressed mRNAs (http://www.genome.jp/kegg/). The P-value (EASE-score, Fisher P-value or Hypergeometric P-value) denotes the significance of the pathway correlated to the conditions (the recommended P-value cut-off is 0.05).

RACE and vector construction

LncRNA-135528 is located on chromosome fifth, starting at txstart:92331840-txend:92346770, with a total length of 1815 bases. To clone the full-length cDNA of lncRNA-135528, we performed the 5′-RACE and 3′-RACE analyses using a SMARTer™ RACE cDNA Amplification Kit (Takara, Japan) according to the manufacturer’s instructions. Then, the lncRNA-135528 cDNA was used as the template for PCR amplification and the recovery of PCR products by DNA gel recovery kit (Dongsheng Biotech, China). Then the recovery products and pLVX-mCMV-ZsGreen-puro vector were digested with Bamhi/noti. T4 DNA Ligase (Takara, Japan) was used to link the target fragment with pLVX-mCMV-ZsGreen-puro vector of the enzyme-cutting recovery and to form the vector pLVX-mCMV-ZsGreen-puro-lncRNA-135528 and Sequencing verification.

SiRNA screening

G422 cells were seeded into 6-well plates (CORNING, China) at a density of 5 × 104 cells/well, and maintained at 37 °C in a humidified incubator containing 5% CO2. When the rate of cell confluence reaches 40%, the cells were transfected with siRNA for 6 h. After that, the old medium was replaced with the new medium. 24 h later, the cells was collected and siRNAs efficiency was detected by quantitative PCR. Three small interfering RNAs were designed and synthesized to target various region of lncRNA-135528 gene expression by Sigama, USA and NC_AS was designed as control group (Table 2). β-actin was performed as the internal control and used to normalize the results. PCR primers are shown in Table 1.

Table 2 siRNAs
Full size table

Transfection

To obtain lncRNA135528 silence and overexpression of G422 cells. siRNA, siRNA-NC, pLVX-mCMV-ZsGreen-puro, or pLVX-mCMV-ZsGreen-puro-lnc-RNA135528 was transfected into G422 cells. The siRNAs and vectors were tra-nsfected into G422 cells by using Lipofectamine 2000 (Invitrogen, USA) in accordance with the manufacturer’s instructions.

TUNEL assay

Apoptosis analysis in tumor tissue was detected by TUNEL method. Mice were killed after 14 days of inoculation and tumor tissue samples were removed and fixed in 4% paraformaldehyde. After paraffin embedding, slicing and dewaxing, the TUNEL reaction was performed according to the manufacturer’s instructions of In situ Apoptosis Detection Kit (Abcam, UK). The color of TUNEL-positive cells were brown.

Immunostaining

Proliferation analysis in tumor tissue was tested by Ki67 immunostaining. Mice were sacrificed after 14 days of inoculation and tumor tissue samples were removed and fixed in 4% paraformaldehyde. Paraffin sections were incubated with anti-Ki-67 (1:50) overnight at 4 °C, which was followed by secondary antibodies [IFKine Red AffiniPure Donkey Anti-Rabbit IgG (H + L)] at 20–37 °C for 1 h. Photos were taken by fluorescence microscope (DMI6000B, leica, Germany).

CCK-8 assay

Cell proliferation in vitro was performed using CCK-8 assays. The five experimental groups (cell, NC, lncRNA, siRNA-NC and siRNA) of G422 cells were seeded into 96-well plates (CORNING, China) at a density of 1 × 104 cells/well, with 3 replicate wells per group. The cells were cultured in a humidified 5% CO2 incubator at 37 °C for 3 days in succession. Then 10 µl CCK-8 solution was added to each well. After 4 h, the spectrophotometric absorbance was measured at 450 nm for each sample and the mean value was calculated. Cell Counting Kit-8 was purchased from Invitrogen, USA.

Cell apoptosis assay

Cell apoptosis in vitro was assessed using a cell apoptosis kit, Annexin V-FITC Apoptosis Detection Kit (keygen, China). The experimental groups were harvested and washed twice in PBS, and re-suspended in Annexin V-FITC for 15 min in the dark. Then binding after 1× binding buffer and Propidium Iodide were added. Cell apoptosis was analyzed by BD caliber (BD, USA).

Cell cycle analysis

Cell cycle progression were analyses by Flow-cytometric. G422 cells were collected 48 h after transfection and washed twice in PBS and fixed in 70% ice-cold ethanol at 4 °C overnight. After washing with PBS, add 500 µl PBS containing 50 µg/ml bromide (PI), 100 µg/ml RNase A, 0.2% Triton X-100, 4 °C to incubate for 30 min, data were collected using a BD Calibur flow cytometer.

Western blot analysis

Western blotting was performed as per standard protocols. Tissue or cells were lysed with RIPA buffer (Beyotime, China),. and protein concentration were measured by BCA Protein Assay Kit (KeyGEN BioTECH, China). Soluble proteins were loaded into 8% SDS-PAGE and electrophoresis. Then the proteins were transferred onto PVDF membrane (Millipore,USA) and blocked for 2 h with Blocking Buffer (Beyotime, China) in room temperature. Primary antibodies were incubated Overnight, 4 °C and secondary antibodies for 1 h, room temperature. For visualization of bands, we used the BeyoECL Plus Western Blotting Detection Reagents (Beyotime, China). GAPDH(aksomics, CHINA) as loading control. The following antibodies were used in this study: Anti-CXCL10 Antibody (Sigma, USA), Goat Anti-Rabbit IgG (H + L) Antibody (Sigma, USA), Anti-Ki67 antibody (abcam, UK), IFKine Red AffiniPure Donkey Anti-Rabbit IgG (H + L) (Abbkine, CHANA), Anti-JAK1 antibody (abcam, UK), Anti-JAK2 antibody (abcam, UK), Anti-STAT2 antibody (abcam, UK), Anti- STAT3 antibody (abcam, UK), Goat Anti-Rabbit IgG (H + L) (Mouse/Human ads-HRP) (Abbkine, CHANA). Expression ratios were calculated with Image J.

Statistical analysis

Data were presented as mean ± SEM, and Statistical significance was determined by one-way analysis of variance (ANOVA) followed by LSD or Dunnett’s T3 test in SPSS software (version 19.0, SPSS, Chicago, IL, USA). All samples were run in triplicate and experiments were repeated at least three times unless specified. A P values of < 0.05 was considered statistically significant.

Results

Tumor regression model

The Fig. 1 shows that the tumor was obviously uplifted on the skin, with a volume of approximately 250 mm3 after 8 days. As shown in Fig. 2, after stopping the administration of cyclosporine, the tumor in group A continued to increase, and the tumor in group B began to subside, on day 8, there was no statistically significant difference between the two groups (P > 0.7), but on days 11 and 14, there was a statistically significant difference between the two groups (P < 0.05, P < 0.05). Additionally, the fluorescence intensity of group A increased with increasing tumor size, and the fluorescence intensity of group B decreased with tumor decline (Fig. 3). Repeated experiments generated similar results. These results show that the tumor regression models were successful.

Fig. 1
figure 1

Morphology and size of tumor tissues in vivo and in vitro

Full size image
Fig. 2
figure 2

Changes of tumor volume in the spontaneous glioma regression model. a represents the progression group, b represents the spontaneous regression group. The number of animals for each experiment were 5

Full size image
Fig. 3
figure 3

Changes of fluorescence value of mice subcutaneous tumor in the spontaneous glioma regression model. a represents the progression group, b represents the spontaneous regression group. a1, a2, a3, a4, a5 represent different mice in Group A, b1, b2, b3, b4 and b5 represent different mice in Group B

Full size image

Proliferation and apoptotic analysis in tumor tissue

The expression of ki67 in the regression group was significantly decreased compared with the progression group (Fig. 4a). TUNEL assay shows that cell apoptosis significantly increased in the regression group compared with the progression group (Fig. 4b). These results further validated the tumor regression models were set up successfully.

Fig. 4
figure 4

Proliferation and apoptotic analysis in tumor tissue. a immunofluorescence staining analysis of Ki67 in tumor tissues. b Representative images for TUNEL assay to determine apoptosis cell in tumor tissues. Yellow arrow, proliferating cells; red arrow, apoptotic cells. (Color figure online)

Full size image

Differentially expressed lncRNAs and mRNAs in spontaneous regressive and progressive tumors

Figure 5 showed the differently expressed IncRNA and mRNA associated with spontaneous glioma regression. The lncRNA expression profiles revealed that approximately 15,791 lncRNAs (7605 up- and 8186 down-regulated) were differentially expressed (≥ twofold-change, P ≤ 0.05), and 67 lncRNAs (66 up- and 1 down-regulated) displayed greater than 50-fold changes (Table 3). The mRNA expression profiles revealed approximately 11,173 mRNAs (4929 up- and 6244 down-regulated) that were significantly differentially expressed (≥ twofold-change, P ≤ 0.05), and 97 mRNAs (95 up- and 2 down-regulated) and displayed greater than 50-fold changes (Table 4).

Fig. 5
figure 5

RNA expression profiles in spontaneous regressive U87 tumors and progressive tumors. a, b Hierarchical clustering of lncRNAs (a) and mRNAs (b). The red and the green shades indicate the expression above and below the relative expression, respectively, across all the samples. c Box plots of lncRNAs to visualize the distribution of a dataset. d Scatter plot of the lncRNAs to assess the variation between chips. e Volcano plot of the lncRNAs. “Red” indicates up-regulated expression, “green” indicates down-regulated expression, and “black” represents transcripts with undifferentiated expression. (Color figure online)

Full size image
Table 3 The differentially expressed lncRNAs (long non-coding RNAs) in the spontaneous regressive U87 tumors implanted in BALB/c mice compared to progressive controls
Full size table
Table 4 The differentially expressed mRNAs (messenger RNAs) in the spontaneous regressive U87 tumors implanted in BALB/c mice compared to progressive controls
Full size table

Real-time quantitative PCR validation

To independently validate the chip results, quantitative real-time polymerase chain reaction (RT-qPCR) was performed to analyze the expression of 7 lncRNAs and 5 mRNAs (the detailed microarray analysis results was shown in Additional file 1: Table S1) randomly selected from the differentially expressed lncRNAs and mRNAs. As shown in Fig. 6, lnRNA uc009bpw.2, AK154404, ENSMUST00000124148, ENSMUST00000125649 and mRNA Cldn3, Mybpc2, Myoz1 were up-regulated, lnRNA ENSMUST00000131925, ENSMUST00000132304, NR_038007 and mRNA Scg2, Hs6st2 were down-regulated, these results were consistent with the microarray analysis.

Fig. 6
figure 6

Validation of lncRNA microarray data by RT-qPCR. A total of seven differentially expressed lncRNAs and five differentially expressed mRNAs were randomly selected from the microarray to validate the accuracy of the microarray results by using RT-qPCR

Full size image

GO analysis and pathway analysis

The genes corresponding to the down-regulated mRNAs included 5325 genes involved in biological processes, 5387 genes involved in cellular components and 5,280 genes involved in molecular functions. In contrast, the genes corresponding to the up-regulated mRNAs included 4398 genes involved in biological processes, 4460 genes involved in cellular components and 4360 genes involved in molecular functions. As shown in Fig. 7a, “T cell chemotaxis” (GO: biological processes), “nuclear telomere cap complex” (GO: cellular components), and “telomerase activity” (GO: molecular functions) were the highly enriched GOs targeted by up-regulated transcripts, while “inactivation of MAPK activity” (GO: biological processed), “PRC1 complex” (GO: cellular components) and “JUN kinase binding” (GO: molecular function) were the highly enriched GOs targeted by down-regulated transcripts (Fig. 7b). Among these terms, T cell chemotaxis was highly relative to immunity, while inactivation of MAPK activity was highly associated with glioma apoptosis [36,37,38], suggesting tumor suppression in spontaneous glioma regression. Simultaneously, we identified 5923 genes (3036 up- and 2887 down-regulated) associated with cell proliferation and cell apoptosis (Additional file 2: Table S2). Among these genes, 284 genes were involved in the regulation of neuron apoptotic process, 21 genes regulated cell proliferation in the brain, 8 genes were associated with regulation of glial cell proliferation, and 582 genes were closely associated with the apoptotic or proliferative processes of immune cells.

Fig. 7
figure 7

The Gene ontology (GO) project provides a controlled vocabulary to describe gene and gene product attributes in any organism. a GO up-regulated. b GO down-regulated

Full size image

We performed a pathway analysis of the differentially expressed mRNAs by using the KEGG database. The pathway analysis revealed 61 pathways corresponding to the up-regulated transcripts (P ≤ 0.05), and 7 pathways were related to immune responses and inflammatory reactions. In contrast, 78 pathways corresponded to the down-regulated transcripts (P ≤ 0.05), and 12 pathways were associated with cancer pathways (Additional file 3: Table S3).

The expression of lncRNA-135528 and CXCL10 in tumors of spontaneous regression model

To verify the differential expression of lncRNA-135528 and CXCL-10 in tumor regression models, the expressions of lncRNA-135528 and CXCL-10 in the different groups was analyzed by qRT-PCR and Western blotting. As shown in Fig. 8a, the expression levels of lncRNA-135528 in the tumor regressive group was significantly decreased comparing with the tumor progression group (P < 0.05).As shown in Fig. 8b–d the expression of CXCL-10 mRNA and protein in the tumor regression group was significantly decreased comparing with the tumor progression group (P < 0.05, P < 0.05). The results showed that lncRNA-135528 and CXCL-10 expression were up-regulated in the tumor regression group, while stable in the tumor progression group. This was consistent with the results of microarray analysis.

Fig. 8
figure 8

The expression of lncRNA-135528 and CXCL10 in tumors of spontaneous regression model. a The expression of lncRNA-135528 in tumor tissues of different groups. b The expression of CXCL10 mRNA in tumor tissues of different groups. c, d The expression of CXCL10 protein in tumor tissues of different groups. The data represent the means ± SEM. *p < 0.05 represent the significant differences between the progression group and the regression group

Full size image

LncRNA-135528 full-length clone and Vector construction

The cDNA fragment of lncRNA-135528 was obtained by RACE technology. The full-length lncRNA-135528 cDNA was cloned into the plvx-mcmv-zsgreen-puro vector, and BLAST analysis showed that lncRNA-135528 was successfully cloned into the plvx-mcmv-zsgreen-puro vector, and blast 100% was consistent with known sequences on NCBI (Additional file 4: Fig. S1). Thus, this construct can be used in subsequent experiments.

SiRNA3 (25 nM) had the highest inhibition rate

We designed 3 small interfering RNAs targeting different regions of lncRNA-135528 sequencing. The glioma cell line G422 was infected with different concentrations of siRNAs. At 24 h after infection, all three lncRNA-135528 siRNAs reduced lncRNA-135528 expression comparing with the negative control group. The inhibition rate of siRNA1-25 nM, siRNA1-50 nM, siRNA3-100 nM, siRNA2-25 nM, siRNA2-50 nM, siRNA2-100 nM, siRNA3-25 nM, siRNA3-50 nM, siRNA3-100 nM was 25%, 5%, 32%, 3%, 43%, 42%, 75%, 26% and 26% respectively, but siRNA3-25 nM had the highest efficiency, with an inhibition rate of 75% (Fig. 9). Thus, we used siRNA3-25 nM for further biological experiments.

Fig. 9
figure 9

SiRNA screening Notably, 1, 2, and 3 represent different siRNAs, and 25, 50 and 100 nM represent different concentrations. The data represent the means ± SEM. *p < 0.05 represent the significant differences between the treated group and the cell group

Full size image

LncRNA-135528 inhibited cell proliferation, promoted cell apoptosis and arrested cell cycle progression

To investigate the biological roles of lncRNA-135528 in glioma, we stably silenced or overexpressed lncRNA-135528 in G422 cells by transfecting siRNA against lncRNA-135528 or the lncRNA-135528 overexpression plasmid. As shown in Fig. 11a, the expression of lncRNA-135528 in the lncRNA group (lncRNA-135528 overexpression group) was significantly increased (P < 0.05) compared with the cell group, while the expression of lncRNA-135528 in the siRNA group (lncRNA-135528 knockdown group) was significantly decreased (P < 0.05), indicating successful transfection.

In order to study the effect of lncRNA-135528 on cell proliferation, we conducted the cck-8 assay. As shown in Fig. 10a, b, the proliferation of G422 cells in the lncRNA group was drastically inhibited (P < 0.05) compared with the cell group, while the proliferation of G422 cells in the siRNA group was significantly enhanced (P < 0.05), suggesting that the ectopic expression of lncRNA-135528 inhibited G422 cell proliferation.

Fig. 10
figure 10

Effects of lncRNA-135528 on cell proliferation (a, b), cell apoptosis (c, e) and cell cycle (d, f). a MTS detection of lncRNA-135528 gene sensitivity in G422 cells. b Inhibitory rate of lncRNA-135528 gene sensitivity in G422 cells. c, e Apoptotic results for the cell group, lncRNA group, NC group, siRNA group and si-NC group. d Cell cycle results for the cell, lncRNA, NC, siRNA and si-NC groups. f Percentage of G1, G2 and S-phase cells in the cell cycle. The data represent the means ± SEM. *p < 0.05 represent the significant differences between treated groups and the cell group

Full size image

Apoptosis was detected by flow cytometry. As shown in Fig. 10c, e, comparing with the cell group, the apoptosis of G422 cells in lncRNA group was significantly enhanced (P < 0.05), while siRNA group decreased significantly (P < 0.05), suggesting that lncRNA-135528 overexpression promoted apoptosis of G422 cells.

In order to investigate the effect of lncRNA-135528 on cell cycle, we performed cell cycle analyses by flow cytometry. As shown in Fig. 10d, f, comparing with the cell group, the percentage of G1 cells decreased and the percentage of S-phase cells increased in the siRNA group, while the lncRNA group showed the opposite results. No significant changes in the percentage of cells were observed in the G1 or S phase in the NC, siRNA-NC and Blank groups (P > 0.05). These results suggest that lncRNA-135528 can arrested cell cycle progression.

Taken together, these findings suggest that lncRNA-135528 plays a role in the proliferation, apoptosis and cell cycle progression of G422 cells.

LncRNA-135528 regulates the JAK/STAT pathway by up-regulating CXCL10

As shown in Fig. 11e, f, the expression of CXCL10 mRNA and protein in the lncRNA group was significantly increased comparing with the cell group (P < 0.05, P < 0.05), while the expression of CXCL10 mRNA and protein in the siRNA group was decreased significantly (P < 0.05, P < 0.05). As shown in Fig. 11b–d, compared with the cell group, the expression of JAK1, JAK2, STAT2, STAT3 mRNA and protein was significantly increased in the lncRNA group (P < 0.05, P < 0.05, P < 0.05, P < 0.05), while in the siRNA group was reversed. The schematic diagram was performed as described before [39] and shown in Fig. 12.

Fig. 11
figure 11

Effects of lncRNA on CXCL10 and JAK/STAT pathways. a The expression differences of lncRNA-135528 in the different groups. The differential expression of mRNA (b) and protein (c, d) associated with JAK/STAT pathway in the different groups. The expression of CXCL10 mRNA (e) and CXCL10 protein (f) in the different groups. Cell group represents the cells that have not been treated, lncRNA, NC, siRNA, si-NC represent cells transfected by pLVX-mCMV-ZsGreen-puro-lncRNA135528, pLVX-mCMV-ZsGreen-puro, siRNA or siRNA-NC respectively. The data represent the means ± SEM. *p < 0.05 represent the significant differences between treated groups and the cell group

Full size image
Fig. 12
figure 12

Mechanism diagram of lncRNA-135528 inhibiting glioma progression. Overexpression of lncRNA-135528 can up-regulate CXCL10, followed by the JAK/STAT signal pathway activation, and then inhibited cell proliferation, promoted cell apoptosis and arrested cell cycle progression of tumor cells, and finally inhibited tumor progression

Full size image

Discussion

Spontaneous tumor regression occurs in many types of human cancers. The mechanisms underlying spontaneous tumor regression may include immune mediation, tumor inhibition by growth factors and/or cytokines, hormonal mediation, induction of differentiation, tumor necrosis and/or angiogenesis inhibition, apoptosis and epigenetic mechanisms [6]. The present study reported the expression profile of lncRNAs and mRNAs in spontaneous U87 tumor regression models and the inhibitory effect of lncRNA-135528 on glioma activity and its molecular mechanism. The data obtained in the present study suggested that immune activation and tumor inhibition were possible underlying mechanisms. This finding is consistent with the results of other studies. Hicks et al. showed that spontaneous regression mice resisted high doses of cancer cells, and that this resistance was mediated by the rapid infiltration of leukocytes, including natural killer cells, macrophages, and neutrophils [40]. Moreover, Halliday et al. indicated that the overexpression of CD4+ T lymphocytes was related to the spontaneous regression of human skin tumors [41]. Rogozinskj et al. reported that the spontaneous regression of human papillomavirus-induced plane warts reportedly relies on immunological responses [42].

Due to low incidence of spontaneous tumor regression and the difficulty of establishing a model, most reports are case studies and phenomenon analyses [43, 44], it’s necessary to establish a simple, reproducible spontaneous tumor regression model. In this study, U87 cells were subcutaneously inoculated on the right back of BALB/c mice to establish a spontaneous tumor regression model, and cyclosporin soft capsules were used to suppress the immune system in mice. Then, cyclosporin treatment was stopped after successful inoculation, and over time, the tumor began to undergo spontaneous regression. This model was reproducible and easy to operate, showing a high success rate.

In recent years, researchers have gradually unveiled the hidden functions of lncRNAs in complex gene networks. LncRNAs are no longer transcriptional noise in the transcriptome, but rather take center stage in both normal cellular processes and disease pathogenesis [45]. Owning to the abnormal lncRNA profiles in various tumors, lncRNAs play important roles in cancer biology. For example, increasing genome-wide profiling studies have demonstrated that lncRNAs are dysregulated in gliomagenesis and could be used to predict clinical phenotypes and prognosis in gliomas [46,47,48]. Moreover, several lncRNAs have been demonstrated to possess functional significance in glioma, lncRNA HOTAIR acts as a prognostic factor for glioma patient survival [49]. Another well-known lncRNA H19 promoted invasion in glioma cells [50]. In this study, we found that silencing lncRNA-135528 inhibits the proliferation and promotes the apoptosis of glioma cells. However, this study used mouse microarray and mouse glioma cell line (G422) to study the effects of the immune system on the regression of tumors during the course of tumor regression. Future studies could use human microarray and human glioma cell line (U87) to verify this finding and to further elucidate the molecular mechanism of tumor regression.

In the present study, we performed comprehensive lncRNA and mRNA expression profiling on the U87 tumors grown in BALB/c mice with spontaneous regression using microarray technology. The results showed that more than half of transcripts were involved in the regulation of cell proliferation or apoptosis, consistent with the direct cause of spontaneous tumor regression, suggesting that the rate of tumor cell proliferation was slower than that of tumor cell apoptosis. Moreover, among these proliferation-/or apoptosis-associated genes, approximately one-tenth of the transcripts were involved in the proliferation or apoptosis of immune cells. This suggests that there may be some factors triggering the immune system and then induce spontaneous tumor regression. Identifying the key molecular elements will bring us one step closer to effectively curing or preventing cancer.

CXCL10 has important physiological and pathological effects and is involved in inflammatory, tumor, and immunological diseases [51]. Studies have shown that CXCL10 can play a role in tumor inhibition by promoting the aggregation and infiltration of lymphocytes in tumor tissue and activating lymphocyte function [52,53,54,55,56]. However, there are few studies on the role of CXCL10 in inhibiting glioma and its molecular mechanism. Chip information shows that CXCL10 was up-regulated 49 times in spontaneous tumor regression, and software predictions show that lncRNA-135528 has an interaction relationship with CXCL10. Thus, we selected lncRNA-135528 as the focus of the present study. This study showed that the overexpression of lncRNA-135528 can significantly up-regulate CXCL10 and that the JAK/STAT signal pathway was activated, while silencing lncRNA-135528 causes the opposite effect. Additionally, the overexpression of lncRNA-135528 inhibited cell proliferation, promoted cell apoptosis and arrested cell cycle progression, while silencing lncRNA-135528 inhibited the proliferation and promoted the apoptosis of glioma cells. This suggests that LncRNA-135528 inhibits tumor progression by up-regulating CXCL10 through JAK/STAT pathway.

Conclusion

lncRNA-135528 can inhibit tumor progression by up-regulating CXCL10 through the JAK/STAT pathway.