Long non-coding RNA TUG1 acts as a miR-26a sponge in human glioma cells

Long non-coding RNA TUG1 acts as a miR-26a sponge in human glioma cells

Accepted Manuscript Long non-coding RNA TUG1 acts as a miR-26a sponge in human glioma cells Jun Li, Meng Zhang, Qingfang Ma, Gang An PII: S0006-291X(...

1MB Sizes 0 Downloads 0 Views

Accepted Manuscript Long non-coding RNA TUG1 acts as a miR-26a sponge in human glioma cells Jun Li, Meng Zhang, Qingfang Ma, Gang An PII:

S0006-291X(16)31056-7

DOI:

10.1016/j.bbrc.2016.06.129

Reference:

YBBRC 36045

To appear in:

Biochemical and Biophysical Research Communications

Received Date: 23 June 2016 Accepted Date: 26 June 2016

Please cite this article as: J. Li, M. Zhang, Q. Ma, G. An, Long non-coding RNA TUG1 acts as a miR-26a sponge in human glioma cells, Biochemical and Biophysical Research Communications (2016), doi: 10.1016/j.bbrc.2016.06.129. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT 1

Long non-coding RNA TUG1 acts as a miR-26a sponge in human glioma cells Jun Li 1,*, Meng Zhang 1, Qingfang Ma 1, Gang An 1

2 3

1

4

Dongnan University, Xuzhou, Jiangsu, China

Department of Neurosurgery, The Affiliated Xuzhou Hospital, College of Medicine,

5 6

Running title: TUG1, a sponge inhibitor of miR-26a in human glioma

8

*Corresponding author:

9

Jun Li, M.D

RI PT

7

Address: Department of Neurosurgery, The Affiliated Xuzhou Hospital, College of Medicine,

11

Dongnan University, 199 South Jiefang Rd., Xuzhou, Jiangsu Province 221009, People’s

12

Republic of China.

13

TEL: +86-0516-83956861

14

E-mail: [email protected]

M AN U

SC

10

15

AC C

EP

TE D

16

1

ACCEPTED MANUSCRIPT Abstract

2

Background: Long non-coding RNA taurine upregulated gene 1 (TUG1) acts as an important

3

regulator in cancer pathogenesis; however, its functional mechanism in glioma development

4

remains unclear. This study aims to explore the potential function of TUG1 in glioma by

5

sponging miR-26a.

6

Methods: The expression of TUG1, miR-26a, and phosphatase and tensin homolog (PTEN)

7

in 20 paired glioma tissues was detected by quantitative real-time PCR and subjected to

8

correlation analysis. Bioinformatics analysis was performed by using DIANA Tools.

9

Abnormal TUG1 expression was conducted in two glioma cells to analyze its regulation on

RI PT

1

miR-26a and PTEN using real-time PCR, western blot, and luciferase reporter assay.

11

Results: TUG1 expression was confirmed to be upregulated in glioma tissues, and showed an

12

inverse correlation with downregulated miR-26a. TUG1 could negatively regulate the

13

expression of miR-26a in glioma cells. The bioinformatics prediction revealed putative

14

miR-26a binding sites within TUG1 transcripts. Further experiments demonstrated the

15

positive regulation of TUG1 on the miR-26a target, PTEN, wherein TUG1 could inhibit the

16

negative regulation of miR-26a on PTEN by binding its 3’UTR. Additionally, the expression

17

of PTEN was also upregulated in glioma tissues, showing a positive or negative correlation

18

with TUG1 or miR-26a, respectively.

19

Conclusion: TUG1 could serve as a miR-26a sponge in human glioma cells, contributing to

20

the upregulation of PTEN. This study revealed a new TUG1/miR-26a/PTEN regulatory

21

mechanism and provided a further understanding of the tumor-suppressive role of TUG1 in

22

glioma development.

23

Keywords: Glioma; TUG1; miR-26a sponge; PTEN

TE D

25

EP

24

M AN U

SC

10

Introduction

27

Glioma is the most common primary brain tumor that associated with fatal clinical outcomes [1].

28

Although great progress has been made in the diagnosis and treatment of glioma [2], the prognosis

29

of patients with glioma is still poor [3], partially due to the lack of reliable biomarkers and

30

effective therapeutic targets [4]. Up to date, increasing efforts to improve this prognosis have

31

focused on exploring potential biomarkers and targets by using advanced technologies [5-7].

32

However, the emerging role of various molecules showing aberrations in glioma have not been

33

characterized yet, and the clinical value of these molecules remains to be validated.

34

Recently, long non-coding RNAs (lncRNAs), a group of non-coding genes with more than 200

35

bases [8], have been discovered to be closely correlated with various types of malignant tumors

36

[9], including glioma [10]. Further studies reveal that lncRNAs can act as oncogenes or tumor

37

suppressors by regulating the expression of tumor-related genes and the function of tumor-related

38

pathways [11, 12]. Taurine upregulated gene 1 (TUG1), which was initially characterized by a

39

genomic screening study in mouse retinal cells [13] and then proved to act as a tumor suppressor

AC C

26

2

ACCEPTED MANUSCRIPT or oncogene depending on cancer type [14-17], has been shown to potentially regulate tumor cell

2

proliferation, apoptosis, and invasion. The expression of TUG1 has been reported to be

3

downregulated in human glioma cells with doxorubicin treatment [18]. Our previous studies also

4

revealed the downregulation of TUG1 in human glioma tissues and cells [19], wherein it might

5

function as a tumor suppressor by promoting cell apoptosis. However, the underlying mechanism

6

of the tumor-suppressive role of TUG1 in glioma remains to be clarified.

7

MicroRNAs (miRNAs) are another novel class of non-coding, single-stranded RNAs, with a

8

length of 18 to 22 nucleotides, and mainly function as gene regulators at the post-transcriptional

9

level by binding their targets [20]. Moreover, the role of miRNAs in glioma has been

10

characterized in a similar fashion to lncRNA [21]. miR-26a is an miRNA that is particularly

11

amplified in high-grade glioma tissues [22]; it has been reported to promote the tumorigenesis of

12

glioma by targeting the expression of the phosphatase and tensin homolog (PTEN) [23], a

13

negative regulator in the Akt pathway [24, 25], or of prohibitin, another potential tumor suppressor

14

[26]. An increasing number of studies have also reported another novel function of lncRNAs in

15

tumorigenesis by sponging specific miRNAs due to their direct interaction [27].

16

In the present study, bioinformatics analysis revealed a potential interaction between TUG1 and

17

miR-26a, indicating that TUG1 might also function as a miR-26a sponge in glioma. Therefore, the

18

expression of TUG1 and miR-26a were detected in glioma tissues, and the relationship between

19

them (speculated to be negative) was further analyzed. More experiments were performed to

20

validate the functional role of TUG1 in the regulation of PTEN expression by sponging miR-26a

21

in two human glioma cell lines.

TE D

22

M AN U

SC

RI PT

1

Materials and Methods

24

Tissue collection

25

Twenty paired tissues were collected from glioma patients receiving no chemotherapy or

26

radiotherapy before surgery in the Affiliated Xuzhou Hospital of the College of Medicine of

27

Dongnan University. The study was approved by the local ethics committee and performed

28

according to the Declaration of Helsinki Principles. Written, informed consent was obtained

29

from patients or their family members for sample collection.

30

Quantitative real-time PCR assay

31

Total RNA was isolated from tissues or cultured cells by using TRIzol reagent (Invitrogen,

32

Carlsbad, CA, USA), and then reverse-transcribed into cDNA by using PrimeScript RT

33

Reagent Kit (Takara, Dalian, China) following the manufacturer’s protocol. Real-time PCR

34

assay was performed on an ABI 7500 real-time PCR system (Applied Biosystems, Foster City,

35

CA, USA) by using the SYBR PrimeScript RT-PCR kit (Takara) with specific primers.

36

GAPDH was used as an internal control for TUG1 expression analysis. For miR-26a

37

expression detection, reverse transcription was performed following a TaqMan MicroRNA

38

Assay protocol (Applied Biosystems), and U6 snoRNA was used as the normalizer. All results

39

were obtained from at least three experiments with triplicate reactions.

AC C

EP

23

3

ACCEPTED MANUSCRIPT Cell culture

2

The U251 and SHG-44 cells, both human glioma cell lines, were obtained from the Chinese

3

Academy of Science Cell Bank (Shanghai, China). Cells were grown in the Dulbecco Modified

4

Eagle Medium (DMEM; Invitrogen-Gibco, Shanghai, China), containing 10% fetal bovine

5

serum, 2 mM glutamine, and 100 U/ml penicillin, and incubated at 37°C under 5% CO2.

6

Plasmids, small-interference RNAs, and miRNAs

7

Plasmids for TUG1 overexpression, containing TUG1-pCDNA and empty pcDNA vector, and

8

three independent small-interference RNAs (siRNAs) for reducing TUG1 expression were

9

obtained as described in a previous study [19]. TUG1 mutant plasmid was obtained by

10

site-directed mutagenesis, according to the bioinformatics analysis described below. MiR-26a

11

mimic and inhibitor were purchased from GenePharma (Shanghai, China), as were the

12

respective negative controls.

13

Transfection

14

Cells were seeded onto six-well plates and pre-incubated to 40% to 60% confluence before

15

transfection. Lipofectamine 2000 (Invitrogen) was used to transfect plasmids, siRNAs, or

16

miRNAs into cells according to the manufacturer’s instructions. Two days after transfection,

17

cells were harvested for further analysis.

18

Bioinformatics

19

DIANA Tools (http://diana.imis.athena-innovation.gr/DianaTools) was used for in silico

20

prediction of miR-26a binding sites within the TUG1 transcript. On this basis, it was assumed

21

that binding domains existed within the TUG1 transcript and miR-26a. Additionally, TUG1

22

mutant transcript was generated according to the binding sites of miR-26a within the TUG1

23

transcript.

24

Luciferase reporter assay

25

PTEN 3’-UTR cDNA fragments containing miR-26a binding sites were amplified and

26

sub-cloned into the pGL3 luciferase reporter vector (Promega, Madison, WI). The resulting

27

constructs, or control luciferase construct, were co-transfected into cells, together with

28

miR-26a mimic or TUG1 plasmids. After 24 h of transfection, luciferase activity was

29

measured by using the Dual Luciferase Reporter 1000 Assay System (Promega). Luciferase

30

activity results were normalized to those of the control group. All reactions were repeated in

31

triplicate through at least three independent experiments.

32

Western blotting

33

Protein concentrations in the whole-cell lysates were determined by using a BCA Protein

34

Assay Kit (Beyotime Biotechnology, Nanjing, China). Then, ~30 µg protein samples were

35

separated by 10% SDS–PAGE and then electro-transferred onto poly (vinylidene difluoride)

36

membranes (Amersham Pharmacia, Little Chalfont, UK). The immunoblots were conducted

37

by incubation with specific antibodies for PTEN and GAPDH. All antibodies were purchased

38

from Santa Cruz Technology (Santa Cruz, CA, USA) and diluted for use according to the

39

manufacturer’s

AC C

EP

TE D

M AN U

SC

RI PT

1

suggestions.

Immune

bands 4

were

visualized

by

an

enhanced

ACCEPTED MANUSCRIPT chemiluminescence (ECL) kit (Beyotime Biotechnology) and analyzed on ChemiDoc XRS

2

(Bio-Rad Laboratories, Hercules, CA).

3

Statistical analysis

4

All results are presented as mean ± SD. Correlations were evaluated by Pearson’s method.

5

Differences between the two groups were analyzed by using a one-way analysis of variance

6

(ANOVA) or Student’s t-test (two-tailed) with SPSS 17.0 software (IBM/SPSS; Armonk, NY,

7

USA). P<0.05 was considered statistically significant.

RI PT

1

8 9

Results

TUG1 expression showed negative correlation with miR-26a in human glioma tissues

11

The expression levels of TUG1 and miR-26a were determined through qRT-PCR analysis in

12

glioma tissues and adjacent normal tissues from 20 glioma patients, whose clinical features

13

are shown in Table 1. The average expression level of TUG1 was 0.027 after the value was

14

normalized to GAPDH in cancerous tissues, significantly lower than the level (0.044) in

15

adjacent normal tissues (Fig. 1 A; P<0.0001). For miR-26a, the average expression level was

16

0.098 in glioma tissues, strongly higher than the level (0.066) observed in normal specimens

17

(Fig. 1 B; P<0.0001). Further correlation analysis revealed the negative correlation of TUG1

18

and miR-26a expression in glioma tissues (Fig. 1 C; R2=0.5149, P<0.001), indicating the

19

positive regulatory mechanism between them.

20

TUG1 could regulate miR-26a expression in human glioma cells

21

To further explore the potential relationship between TUG1 and miR-26a, gain- and loss-of

22

function studies were performed in two human glioma cell lines, U251 and SHG-44. As

23

shown in Fig. 2, an approximately 0.4-fold decrease or 1.7-fold increase of miR-26a

24

expression was detected in TUG1 overexpressing or inhibiting cells obtained by

25

TUG1-pCDNA (TUG1-plasmid; Fig. 2 A) or siTUG1s (si-TUG1-1, si-TUG1-2, or si-TUG1-3;

26

Fig. 2 B) transfection, respectively. These results confirmed that TUG1 showed negative

27

regulation of miR-26a expression in human glioma cells, and indicated that TUG1 might

28

function as an inhibitor of miR-26a by sponging miR-26a through their potential interaction.

29

TUG1 regulated PTEN expression by sponging miR-26a

30

As expected, bioinformatics analysis for the predicted interaction revealed potential binding

31

domains within TUG1 transcripts and miR-26a (Fig. 3 A). Then, the potential regulation of

32

TUG1 on the expression of PTEN, one validated miR-26a target, was investigated in cells

33

with abnormal TUG1 expression. Results showed that a more than 1.9-fold increase or less

34

than 0.5-fold decrease of PTEN mRNA expression was observed in TUG1 overexpressing or

35

inhibiting cells, respectively (Fig. 3 B), in line with the tendency shown in the corresponding

36

changes in protein expression (Fig. 3 D). Moreover, luciferase reporter assays indicated that

37

TUG1 overexpression could inhibit miR-26a–induced decrease of PTEN 3’UTR activity,

38

while no obvious change was obtained by further overexpression of TUG1 mutant in miR-26a

39

overexpressing cells (Fig. 3 C). Thus, it was concluded that TUG1 could regulate PTEN

AC C

EP

TE D

M AN U

SC

10

5

ACCEPTED MANUSCRIPT expression by sponging miR-26a.

2

PTEN expression showed significant correlation with TUG1 and miR-26a

3

In order to further validate the correlation between PTEN and TUG1 or miR-26a, PTEN

4

expression was also detected in glioma tissues as well as in adjacent normal tissues.

5

Quantitative real-time PCR assays showed that the average expression level of PTEN was

6

0.074 in cancerous tissues, markedly lower than the 0.10 level observed in normal tissues (Fig.

7

4 A; P<0.007). Further correlation analysis revealed the negative or positive relationship

8

between PTEN and miR-26a (Fig. 4 B; R2=0.4798, P<0.001) or TUG1 (Fig. 4 C; R2=0.5816,

9

P<0.001), respectively. These results supported the role of TUG1 as a miR-26a sponge in

10

RI PT

1

glioma pathogenesis.

11

Discussion

13

In the present study, 20 paired tissues with obviously downregulated expression of TUG1 in

14

glioma were analyzed on the basis of our previous study [19]. Meanwhile, markedly

15

upregulated expression of miR-26a was observed in glioma tissues, and further analysis

16

demonstrated a negative correlation between them. In vitro functional studies revealed that

17

miR-26a expression can be negatively regulated by TUG1, while miR-26a binding sites were

18

also found within TUG1 transcripts. Furthermore, TUG1 also showed positive regulation on

19

PTEN, which has been defined as a direct and functional downstream target of miR-26a. In

20

addition, the expression of PTEN was confirmed to be positively or negatively correlated with

21

that of TUG1 or miR-26a in glioma tissues, respectively.

22

LncRNAs are emerging as important regulators in cell biology [28], and increasing evidence

23

has shown the link between their abnormal expression and diverse human diseases [29],

24

especially in cancer [30, 31]. Mounting studies have suggested that lncRNAs are closely

25

associated with various types of malignant tumors, including glioma [12]. For example, H19

26

is one highly upregulated lncRNA in gliomas, and can drive tumor transformation by binding

27

transcription factor c-Myc, which contributes to tumorigenic phenotypes [32]. Maternally

28

expressed gene 3 (MEG3) is another lncRNA with markedly decreased expression in glioma

29

tissues, and its overexpression in glioma cells can inhibit cell proliferation and promote cell

30

apoptosis [33]. In the present study, the role of TUG1, a taurine-induced lncRNA initially

31

found in developing mouse retinal cells [13], was investigated to explore its underlying

32

mechanism in regulating glioma development, since its role as a tumor suppressor was

33

revealed in our previous study [19].

34

Previous studies have reported that TUG1 can be induced by p53, one common tumor

35

suppressor, and bind to polycomb repressive complex 2 (PRC2), evidence of its role in

36

repressing specific genes involved in cell-cycle regulation [34]. Recent studies also reveal that

37

p53-regulated TUG1 can modulate tumor growth partly through epigenetic regulation on

38

homeobox B7 (HOXB7) [16]. In the previous study, it was found that TUG1 could also

39

promote cell apoptosis in glioma cells, but with an unclear mechanism [19]. Thus, to further

AC C

EP

TE D

M AN U

SC

12

6

ACCEPTED MANUSCRIPT understand the role of TUG1 in glioma pathogenesis, the potential relationship between

2

TUG1 and miR-26a was investigated based on the fact that miR-26a expression is highly

3

upregulated in glioma [22], which is contrary to the alteration of TUG1 expression [19].

4

Twenty paired glioma tissues and adjacent normal ones were collected to analyze the

5

expression of TUG1 and miR-26a, with results showing a negative correlation between them.

6

Increasing evidence has shown that lncRNA may function as a competitor to endogenous

7

RNA or molecular sponge in regulating the expression and biological functions of miRNA

8

when the expression of lncRNA and miRNA shows an inverse correlation [35]. This study

9

verified the potential negative regulation of TUG1 on miR-26a expression in glioma cells

10

with abnormal TUG1 expression. Moreover, bioinformatics prediction revealed miR-26a

11

binding domains within TUG1 transcripts, indicating that TUG1 may serve as a miR-26a

12

inhibitor through direct interaction between miR-26a and its binding sites within TUG1

13

transcripts. As far as we know, miRNAs can inhibit gene expression at the post-transcriptional

14

level by binding the 3’-untranslated regions (3’UTR) of their target mRNA [20], so these

15

lncRNAs serving as miRNA sponges can also regulate the expression and biological functions

16

of miRNA targets [36].

17

Currently, several miR-26a target genes, including PTEN [23], prohibitin [26], and

18

ataxia-telangiectasia mutated (ATM) [37], have been identified in glioma development or

19

treatment, suggesting the role of miR-26a as a tumor-promoter or radio-sensitizer. PTEN is a

20

tumor suppressor that often exhibits reduced expression in many cancers [38], and plays a

21

critical role in inhibiting tumor cell proliferation through suppression of the Akt pathway [24,

22

25]. Thus, in order to illustrate the role of TUG1 as a miR-26a sponge, the potential

23

regulation of TUG1 on PTEN expression was also analyzed in glioma cells with abnormal

24

TUG1 expression. Expression analysis and the luciferase reporter assay showed that TUG1

25

negatively regulated the inhibition of miR-26a on PTEN expression and also miR-26a binding

26

with PTEN 3’UTR, with the TUG1 mutant transcript not being capable of binding to miR-26a,

27

and thereby exerting little influence on miR-26a–induced effects. These results demonstrated

28

a positive regulation of TUG1 on the miR-26a target, PTEN, which was not consistent with a

29

recent study revealing the role of TUG1 as one PTEN sponge-lncRNA [39]. In addition, a

30

strongly upregulated expression of PTEN was also confirmed in glioma tissues and

31

respectively shows positive or negative correlation with that of TUG1 or miR-26a. Therefore,

32

it was concluded that TUG1 may regulate the cell apoptosis of glioma cells through the

33

upregulation of PTEN, partially by sponging miR-26a.

34

Conclusion

35

In summary, a new TUG1/miR-26a/PTEN regulatory mechanism in glioma pathogenesis was

36

revealed in the present study, providing a further understanding of the role of TUG1 in glioma

37

development. Moreover, these results indicate that TUG1 might regulate cell apoptosis in

38

glioma cells through the upregulation of PTEN, leading to the reduction of cell proliferation

39

and promotion of cell apoptosis. Although more studies are needed to clarify the TUG1

AC C

EP

TE D

M AN U

SC

RI PT

1

7

ACCEPTED MANUSCRIPT 1

function in animal experiments and following clinical trials, this study revealed the

2

underlying mechanism of TUG1 as a tumor suppressor in glioma, supporting its potential

3

value in the treatment and prognosis of glioma.

4

6

Conflicts:

7

None declared.

RI PT

5

8 9 10

References

11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

1.

gliomas. Clinical and Developmental Immunology, 2013. 2013. 2.

SC

Wei, J., K. Gabrusiewicz, and A. Heimberger, The controversial role of microglia in malignant Stupp, R., et al., High-grade malignant glioma: ESMO Clinical Practice Guidelines for diagnosis,

treatment and follow-up. Ann Oncol, 2010. 21 Suppl 5: p. v190-3.

Ostrom, Q.T., et al., CBTRUS statistical report: Primary brain and central nervous system tumors

M AN U

3.

diagnosed in the United States in 2006-2010. Neuro-oncology, 2013. 15(suppl 2): p. ii1-ii56. 4.

Hochberg, F.H., et al., Glioma diagnostics and biomarkers: an ongoing challenge in the field of

medicine and science. Expert review of molecular diagnostics, 2014. 14(4): p. 439-452. 5.

Gupta, K. and P. Salunke, Molecular markers of glioma: an update on recent progress and

perspectives. Journal of cancer research and clinical oncology, 2012. 138(12): p. 1971-1981. 6.

Huse, J.T. and K.D. Aldape, The evolving role of molecular markers in the diagnosis and

management of diffuse glioma. Clin Cancer Res, 2014. 20(22): p. 5601-11. Kros, J.M., et al., Circulating glioma biomarkers. Neuro Oncol, 2015. 17(3): p. 343-60.

8.

Ponting, C.P., P.L. Oliver, and W. Reik, Evolution and functions of long noncoding RNAs. Cell,

2009. 136(4): p. 629-641. 9.

TE D

7.

Zhang, H., et al., Long non-coding RNA: a new player in cancer. J Hematol Oncol, 2013. 6(1): p.

1-7.

EP

10. Zhang, X., et al., Long non-coding RNA expression profiles predict clinical phenotypes in glioma. Neurobiol Dis, 2012. 48(1): p. 1-8.

11. Park, J.Y., et al., Roles of Long Non-Coding RNAs on Tumorigenesis and Glioma Development.

AC C

Brain Tumor Res Treat, 2014. 2(1): p. 1-6. 12. Zhang, X.Q. and G.K. Leung, Long non-coding RNAs in glioma: functional roles and clinical perspectives. Neurochem Int, 2014. 77: p. 78-85. 13. Young, T., T. Matsuda, and C. Cepko, The noncoding RNA taurine upregulated gene 1 is required for differentiation of the murine retina. Current biology, 2005. 15(6): p. 501-512. 14. Zhang, Q., et al., Down-regulation of Long Non-coding RNA TUG1 Inhibits Osteosarcoma Cell Proliferation and Promotes Apoptosis. Asian Pacific Journal of Cancer Prevention, 2013. 14(4): p. 2311-2315. 15. Xu, Y., et al., Upregulation of the long noncoding RNA TUG1 promotes proliferation and migration of esophageal squamous cell carcinoma. Tumor Biology, 2014: p. 1-9. 16. Zhang, E.B., et al., P53-regulated long non-coding RNA TUG1 affects cell proliferation in human non-small cell lung cancer, partly through epigenetically regulating HOXB7 expression. Cell Death Dis, 2014. 5: p. e1243. 8

ACCEPTED MANUSCRIPT 17. Han, Y., et al., Long intergenic non-coding RNA TUG1 is overexpressed in urothelial carcinoma of the bladder. J Surg Oncol, 2013. 107(5): p. 555-9. 18. Liu, Q., et al., Altered expression of long non-coding RNAs during genotoxic stress-induced cell death in human glioma cells. J Neurooncol, 2015. 122(2): p. 283-92. 19. Li, J., et al., LncRNA TUG1 acts as a tumor suppressor in human glioma by promoting cell apoptosis. Exp Biol Med (Maywood), 2016. 241(6): p. 644-9. 20. Lewis, B.P., C.B. Burge, and D.P. Bartel, Conserved seed pairing, often flanked by adenosines,

RI PT

indicates that thousands of human genes are microRNA targets. Cell, 2005. 120(1): p. 15-20.

21. Brower, J.V., et al., MicroRNAs in cancer: glioblastoma and glioblastoma cancer stem cells. Neurochem Int, 2014. 77: p. 68-77.

22. Visani, M., et al., Expression of 19 microRNAs in glioblastoma and comparison with other brain neoplasia of grades I-III. Mol Oncol, 2014. 8(2): p. 417-30.

SC

23. Guo, P., et al., C-Myc negatively controls the tumor suppressor PTEN by upregulating miR-26a in glioblastoma multiforme cells. Biochem Biophys Res Commun, 2013. 441(1): p. 186-90. 24. McDowell, K.A., G.J. Riggins, and G.L. Gallia, Targeting the AKT pathway in glioblastoma. Curr Pharm Des, 2011. 17(23): p. 2411-20.

M AN U

25. Molina, J.R., et al., Loss of PTEN binding adapter protein NHERF1 from plasma membrane in glioblastoma contributes to PTEN inactivation. Cancer Res, 2010. 70(17): p. 6697-703. 26. Qian, X., et al., MicroRNA-26a promotes tumor growth and angiogenesis in glioma by directly targeting prohibitin. CNS Neurosci Ther, 2013. 19(10): p. 804-12.

27. Xia, T., et al., Long noncoding RNA associated-competing endogenous RNAs in gastric cancer. Sci Rep, 2014. 4: p. 6088.

28. Hung, T. and H.Y. Chang, Long noncoding RNA in genome regulation: prospects and mechanisms.

TE D

RNA Biol, 2010. 7(5): p. 582-5.

29. Shi, X., et al., Long non-coding RNAs: a new frontier in the study of human diseases. Cancer Lett, 2013. 339(2): p. 159-66.

30. Gibb, E.A., et al., Human cancer long non-coding RNA transcriptomes. PLoS One, 2011. 6(10): p. e25915.

EP

31. Qiu, M.T., et al., Long noncoding RNA: an emerging paradigm of cancer research. Tumour Biol, 2013. 34(2): p. 613-20.

32. Barsyte-Lovejoy, D., et al., The c-Myc oncogene directly induces the H19 noncoding RNA by allele-specific binding to potentiate tumorigenesis. Cancer Res, 2006. 66(10): p. 5330-7.

AC C

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

33. Wang, P., Z. Ren, and P. Sun, Overexpression of the long non-coding RNA MEG3 impairs in vitro glioma cell proliferation. J Cell Biochem, 2012. 113(6): p. 1868-74. 34. Khalil, A.M., et al., Many human large intergenic noncoding RNAs associate with chromatin-modifying complexes and affect gene expression. Proc Natl Acad Sci U S A, 2009. 106(28): p. 11667-72.

35. Zhao, X., et al., Gas5 Exerts Tumor-suppressive Functions in Human Glioma Cells by Targeting miR-222. Mol Ther, 2015. 23(12): p. 1899-911. 36. Salmena, L., et al., A ceRNA hypothesis: the Rosetta Stone of a hidden RNA language? Cell, 2011. 146(3): p. 353-8. 37. Guo, P., et al., MiR-26a enhances the radiosensitivity of glioblastoma multiforme cells through targeting of ataxia-telangiectasia mutated. Exp Cell Res, 2014. 320(2): p. 200-8. 38. Sansal, I. and W.R. Sellers, The biology and clinical relevance of the PTEN tumor suppressor 9

ACCEPTED MANUSCRIPT 1 2 3 4

pathway. J Clin Oncol, 2004. 22(14): p. 2954-63. 39. Du, Z., et al., Integrative analyses reveal a long noncoding RNA-mediated sponge regulatory network in prostate cancer. Nat Commun, 2016. 7: p. 10982.

5

Figure legends

7

Figure 1. Expression of TUG1 and miR-26a from 20 pairs of human glioma tissues.

8

A. The levels of TUG1 expression in glioma tissues (Cancer) were significantly lower than

9

those in non-tumorous tissues (Control). P<0.0001. B. miR-26a expression showed

10

significantly higher in glioma tissues (Cancer) compared to non-tumorous tissues (Control).

11

P<0.0001. C. Analysis for the correlation between TUG1 and miR-26a expression. R2=0.5149;

12

P<0.001.

SC

RI PT

6

13

Figure 2. Regulation of TUG1 on miR-26a expression through gain- and loss-of function

15

studies.

16

Abnormal TUG1 expression was induced in U251 and SHG-44 cells through transfection

17

with TUG1-overexpression plasmid (TUG1-pCDNA3.1) or siRNAs (siTUG1s), respectively.

18

A. miR-26a expression was downregulated in TUG1-overexpressing cells, with miR-26a

19

inhibitor transfection serving as the positive control. B. miR-26a expression was upregulated

20

under TUG1 inhibition. Control: cells with pCDNA3.1 (A) or si-NC (B) transfection;

21

TUG1-plasmid: cells with TUG1-pCDNA3.1 transfection. **P<0.01.

22

TE D

M AN U

14

Figure 3. Potential regulation of TUG1 on PTEN expression through miR-26a–mediated

24

targeting of PTEN.

25

A. Bioinformatics analysis for the prediction of the interaction between TUG1 transcript and

26

miR-26a. The sequences “UGA” and “UUG” within the two binding domains were replaced

27

by “ACC” and “AAC”, respectively, to generate TUG1 mutant transcript. B. PTEN mRNA

28

expression in cells with abnormal TUG1 expression, with miR-26a inhibitor transfection as

29

positive control. C. Luciferase report analysis for the activity of PTEN 3’-UTR under

30

miR-26a mimic transfection or co-transfection with TUG1 plasmids containing normal

31

(TUG-plasmid) or mutant (mutTUG-plasmid) TUG1 transcript. D. PTEN protein expression

32

under abnormal TUG1 expression. **P<0.01.

AC C

33

EP

23

34

Figure 4. PTEN expression in human glioma tissues and its correlation with miR-26a or

35

TUG1 expression.

36

A. Results for PTEN expression levels show significantly lower expression of PTEN in

37

glioma tissues (Cancer) than in non-tumorous tissues (Control). P=0.007. B. The negative

38

correlation between PTEN and miR-26a expression in glioma tissues. R2=0.4798; p<0.001. C.

39

The positive correlation between PTEN and TUG1 expression in glioma tissues. R2=0.5816; 10

ACCEPTED MANUSCRIPT 1

P<0.001.

2

AC C

EP

TE D

M AN U

SC

RI PT

3

11

ACCEPTED MANUSCRIPT Table 1 Characteristic of individuals Variable

Patients

Number of cases

20

Sex Male/Female

12/8 50.2

RI PT

Age at diagnosis (mean) Clinical pathology features Extent of resection Subtotal

8

12

Radiographic pattern Solitary lesion

SC

Total

14

M AN U

Invasive and multifocal lesions WHO grading I–II

6

9

11

III–IV KPS score ≥80 <80

TE D

15 5

AC C

EP

WHO: World Health Organization, KPS: Karnofsky performance score

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT 1. TUG1 expression was inversely correlated with miR-26a expression in glioma tissues. 2. miR-26a expression was negatively regulated by TUG1 in two glioma cells. 3. Putative miR-26a binding sites were revealed within TUG1 manuscripts. 4. TUG1 could positively regulate PTEN expression by sponging miR-26a.

AC C

EP

TE D

M AN U

SC

RI PT

5. PTEN expression also correlated with TUG1 and miR-26a expression in glioma tissues.