Augmented expression levels of lncRNAs ecCEBPA and UCA1 in gastric cancer tissues and their clinical significance.

OBJECTIVES: As the second cause of cancer death, gastric cancer (GC) is one of the eminent dilemmas all over the world, therefore investigating the molecular mechanisms involved in this cancer is pivotal. Unrestricted proliferation is one of the characteristics of cancerous cells, which is due to deficiency in cell regulatory systems. Long non-coding RNAs (lncRNAs) have emerged as critical regulators of the epigenome. lncRNA extra coding CEBPA (ecCEBPA) is involved in DNA methylation. This lncRNA reduces CEBPA promoter methylation by interacting with DNA methyltransferase 1. lncRNA UCA1 (urothelial carcinoma-associated 1) elevates cell proliferation through the PI3K/Akt signaling pathway which has a critical role in cell growth and apoptosis. The aim of this study was to examine the expression of ecCEBPA and UCA1 genes in GC tissues as well as their clinical significance.
MATERIALS AND METHODS: Total RNA extraction, cDNA synthesis, and quantitative real-time PCR were performed for cells and 80 paired GC tissues. Furthermore, clinical relevance of UCA1 expression was investigated in TCGA cohort data.
RESULTS: Our results showed ecCEBPA and UCA1 over-expression in GC tissues. Furthermore, lncRNAs associations with clinicopathological features were demonstrated both in the current and TCGA cohort. Kaplan-Meier analysis indicated that patients with higher UCA1 expression had a worse overall survival in the case of pancreatic and lung adenocarcinomas but not other solid cancer types including GC.
CONCLUSION: These data demonstrate UCA1 and ecCEBPA involvement in GC and suggest that these lncRNAs might be useful as diagnostic/ prognostic biomarkers in cancer.

Introduction

Gastric cancer (GC) terminates the lives of a plenty of people every year and is still the second most prevalent cause of cancer deaths worldwide. Stomach cancer progression is a multistep process including alteration in various genes. In spite of developments in diagnostic methods, GC is usually recognized late, therefore examining molecular biomarkers and mechanisms is pivotal for early GC detection (1). Progression in transcriptome analysis has revealed that about 70% of human genome is transcribed into RNAs that do not act as templates for proteins. These RNAs that are referred to as non-coding RNAs (ncRNAs) are classified into different subsets based on their length. lncRNAs are classified as ncRNAs with at least 200 nucleotides length that lack an open reading frame of significant length. lncRNAs have emerged as regulatory players in abundant biological functions such as gene regulation, epigenetic regulation, transcription, mRNA splicing, and translation (2).

Located on human 19p13.12, Urothelial carcinoma associated 1 (UCA1) is a lncRNA with three exons (3). UCA1 over-expression has been reported in different cancer types (4-9). It has been shown that UCA1 affects p27 expression by interacting with heterogeneous nuclear ribonucleoprotein I and inhibits the p27 protein resulting in elevation of tumor growth by increasing proliferation in breast cancer (10). CREB (cAMP responsive element binding protein) transcription factor, which is involved in augmenting cancer prog-ression, is activated by UCA1 through AKT kinase in the PI3K/AKT pathway (11). Furthermore, it has been demonstrated that UCA1 over-expression stimulates cell cycle progression and tumor growth in colorectal cancer cells (12). One of the studied molecules that has a binding site on UCA1 promoter is CCAAT/enhancer-binding protein α (C/EBPα) that increases UCA1 expression, which in turn induces cell viability and reduces cell apoptosis in bladder cancer (13). UCA1 up regulation leads to cyclin D1 over expression which promotes cell cycle progression in GC (14). The interaction between UCA1 and miR-182 has been reported in glioma tissues and cell lines (15). Moreover, a recent study on GC has shown that UCA1 is negatively associated miR-27b expression (16).

Extra coding CEBPA (ecCEBPA) is a non-polyadenyl-ated lncRNA that is located on the upstream region of the CEBPA locus on chromosome 19. ecCEBPA is involved in DNA methylation. This lncRNA interacts with DNA methyltransferase 1 and diminishes CEBPA promoter methylation leading to CEBPA up-regulation. ecCEBPA expression has been shown in Hl-60 and U937 cell lines (17).

According to these findings, we designed a study to evaluate UCA1 and ecCEBPA expression patterns in GC specimens as well as their correlation with clinicopathological parameters. Furthermore, we analyzed UCA1 gene expression and clinicopatho-logical characteristics data of this lncRNA in GC from the Cancer Genome Atlas (TCGA) database.

Materials and Methods

Tumoral and non-tumoral tissues

The gastric tissues were acquired from Iran Tumoral Bank (Tehran, Iran). Biological materials were provided by Iran National Tumor Bank which is funded by Cancer Institute of Tehran University for Cancer Research (18-20). All tissue specimens were examined for gene expressions which consisted of 40 tumoral and 40 non-tumoral paired tissue samples. The scheme of this experiment was approved by Ethics Committee of Isfahan University of Medical Sciences. Additionally, written informed consents were obtained from all patients, preceding their participation, by Iran Tumoral Bank.

RNA sequence data sets and differential expression

The data set from an independent cohort in the TCGA database (http://cancergenome.nih.gov) was utilized for the evaluation of UCA1 lncRNA gene expression and its clinicopathological relevance. The lncRNA reads per kilobases per million reads (RPKM) expression value in TCGA database was downloaded through The Atlas of Noncoding RNAs in Cancer (TANRIC), which contains 285 GC and 33 non-tumoral tissues (21, 22). Clinical information about these 318 patients was also downloaded from TCGA database. Furthermore, TCGA overall survival (OS) data were retrieved from OncoLnc database (23).

Cell culture

The human cell lines HEK-293 (human embryonic kidney 293 cells), HUVECs (human umbilical vein endothelial cells), SKBR3 (human breast cancer cells), A542 (human pulmonary carcinoma cells), MCF7 (human breast cancer cells), and NT2 (human embryonic carcinoma cell line, NTERA2) were cultured in high glucose DMEM (Gibco Life Technologies, Karlsruhe, Germany), supplemented with 15% fetal calf serum, 100 U/ml penicillin, and 10 μg/ml streptomycin. The human cancer cell line HepG2 (hepatocellular carcinoma cell line) was cultivated in RPMI-1640 (Gibco Life Technologies, Karlsruhe, Germany), enriched with 10% fetal calf serum, 100 U/ml penicillin, and 10 μg/ml streptomycin.

RNA extraction, DNase I treatment and cDNA synthesis

RNA extraction for cells and tissues was performed by using TRIzol® reagent (Invitrogen, Carlsbad, California, United states) according to the provided instructions by the manufacturer. RNA integrity was assessed using 1% agarose gel electrophoresis and RNA concentration was checked by the Nanodrop instrument (NanolytiK, Duesseldorf, Germany). DNase I treatment was performed using DNase set (Fermentas, Vilnius, Lithuania) in order to prepare DNA-free RNA prior to RT-PCR. cDNA was synthesized by using PrimeScript™RT reagent Kit (TaKaRa, Kusatsu, Shiga, Japan).

Quantitative real-time PCR

The relative expression of lncRNA UCA1 and ecCEBPA were measured by quantitative real-time RT-PCR with specific primers designed using the GeneRunner software package, version 4.0 (Table 1). Primers for amplification of the GUSB (β-Glucuronidase) gene (as an internal control) were taken from another study (24). PCR was performed using RealQ Plus 2x Master Mix Green (high Rox) (Ampliqon, Odense M, Denmark) on an Applied Biosystems StepOnePlus™ instrument. The PCR cycling conditions consisted of a first denaturation step at 95 °C for 10 min, 40 cycles of denaturation at 95 °C for 15 sec, annealing at 61°C for lncRNA UCA1 and ecCEBPA, and at 60 °C for GUSB genes and then extension for 15 sec at 72 °C. Additionally, the specificity of PCR amplicons was verified by Sanger sequencing using Applied Biosystems 3730XL sequencer (Macrogen, Seoul, South Korea).

Table 1

Primer sequences for amplification of ecCEBPA, UCA1, and GUSB*

Primers	Sequence	Amplicon size
hecCEBPA-F1	TTGGCGAGGCTTCTTATCTG	133 bp
hecCEBPA-R1	GCTGCAGCTGTAGGTGATTTG
hUCA1-F1	ATCGGATCTCCTCGGCTTAG	145 bp
hUCA1-R1	TGATCGTCCAGCTAGGGTGTC
hGUSB-F1	CACGACACCCACCACCTACATC	121 bp
hGUSB-R1	GACGCACTTCCAACTTGAACAG

Primer sequences were derived from this reference (24)

Statistical analysis

Relative gene expression was calculated using the ΔCt method (Ct of lncRNA minus Ct of housekeeping gene). All experiments were replicated at least 2-3 times and acquired data are represented as mean± standard error of mean (SEM). Kolmogorov-Smirnov test was implemented in order to find out the normal distribu-tion of samples. The results were analyzed using Student’s t-test, ANOVA (analysis of variance), and chi-square. Kaplan-Meier and Cox regression analyses were utilized to assess the association between lncRNA UCA1 and overall survival. The SPSS software, version 16.0 (SPSS Inc., Chicago IL), was used for statistical analysis. Figures were made by GraphPad Prism 6 (GraphPad Software Inc., San Diego, CA, USA). A P-value of less than 0.05 was considered a significant difference.

Results

Expression profile of UCA1 and ecCEBPA in various cell lines

Optimization of UCA1 was performed on the HepG2 cell line, as previously reported in a study (9), UCA1 is expressed in these cells. Agarose gel electrophoresis showed a specific band with the expected size. Addi-tionally, real-time RT-PCR reaction for the examined genes showed a unique melting curve without primer dimers. A few PCR products were further sequenced to confirm lncRNAs specific amplification (data not shown). UCA1 expression was detected in HUVECs, SKBR3, A542, NT2, and HepG2 cell lines whereas it was not observed in HEK-293 and MCF7 cells. Moreover, ecCEBPA expression evaluation on the mentioned cell lines revealed their expression in HUVECS, A542, HepG2, HEK-293, and MCF7 cells but not in SKBR3 and NT2 cultured cells (Figure 1).

Figure 1

Expression of ecCEBPA, UCA1, and GUSB transcripts in various cell lines. Electrophoresis results of ecCEBPA, UCA1, and GUSB PCR products on the agarose gel are presented in the figure

Augmented expression of lncRNA UCA1 and ecCEBPA in gastric cancer tissues

The expression levels of UCA1 and ecCEBPA were measured by quantitative real-time PCR in 80 pairs of GC and their adjacent non-tumoral tissues. Specific primers were used for both lncRNAs and GUSB (as a reference gene). The ∆Ct method was applied to examine the relative expression levels of UCA1 and ecCEBPA. As presented in Figure 2, UCA1 relative expression showed an increase in tumoral tissues (P-value=0.036) compared with the adjacent non-tumoral ones (6.867±1.03 versus 10.00±1.05, respectively). As shown in Figure 3, the relative expression status of ecCEBPA was significantly elevated in tumoral tissues (P-value=0.001) compared with the adjacent non-tumoral ones (11.187±0.82 versus 14.254±0.44, respectively). Note that a lower ΔCt indicates a higher relative expression.

Figure 2

Relative expression of UCA1 in tumoral and non-tumoral gastric tissue samples (n=40). A lower ΔCt value shows higher expression levels. Values shown represent the mean±SEM. Asterisk represents a statistically significant difference (P≤ 0.05) Error bars stand for standard error of mean (SEM)

Figure 3

Relative expression of ecCEBPA in tumoral and non-tumoral gastric tissue samples (n=40). A lower ΔCt value shows higher expression levels. Values shown represent the mean ± SEM Asterisk represents a statistically significant difference (P≤ 0.05) Error bars stand for standard error of mean (SEM)

Association of UCA1 and ecCEBPA with various clinicopathological parameters

As shown in Tables 2 and 3, data analysis showed no meaningful correlation between UCA1 and ecCEBPA expression levels with clinicopathological character-ristics, except for significant associations between the expression level of lncRNA UCA1 and tumor grades (P-value=0.01), tumor type (P-value=0.04), and M classification (P-value=0.05). Furthermore, GC samples were classified into high (#20) and low (#20) expression groups based on median value of lncRNAs expression levels in tumoral specimens. As presented in Tables 4 and 5, results demonstrate no correlation between UCA1 and ecCEBPA expression levels and any of the clinicopatho-logical parameters. Furthermore, a Pearson correlation analysis between UCA1 and ecCEBPA relative expression levels in GC tissues revealed a moderate positive correlation (P-value =0.000, r=0.46) (Figure 4).

Table 2

Relationship between UCA1 mean expression levels (ΔCt*) and clinicopathological characteristics of tumoral gastric specimens

Characteristics	Number (#40)	Mean ± SEM	P-value
Sex			0.29
Male	24	7.50±1.4
Female	16	5.90±1.52
Age (years)			0.24
≥70	18	7.92±1.61
<70	22	6.0±1.33
Depth of invasion			0.31
T2	3	9.19±3.81
T3-4	37	6.67±1.08
N classification			0.30
NX-N0	7	5.11±2.37
N1	20	7.94±1.56
N2-3	13	6.39±1.74
M classification			0.050**
MX	8	4.46±2.23
M0	24	8.47±1.27
M1	8	4.44±2.41
TNM stages			0.21
I-II	22	7.94±1.43
III	11	9.2±1.92
IV	7	4.44±2.41
Perineural invasion			0.34
Negative	10	6.75±2.53
Positive	30	6.9±1.11
Lymphatic invasion			0.12
Negative	14	5.36±1.83
Positive	26	7.67±1.24
Tumor size			0.30
≥5	31	6.58±1.18
<5	9	7.85±2.18
Tumor grades			0.01**
I	12	5.11±2.0
II	9	4.32±2.05
III	19	9.44±1.32
Tumor types			0.04**
Diffuse	19	8.35±1.44
Intestinal	21	5.519±1.43

A lower ΔCt value indicates a higher expression level

Statistically significant

Table 3

Relationship between ecCEBPA mean expression levels (ΔCt*) and clinicopathological characteristics of tumoral gastric specimens

Characteristics	Number (#40)	Mean± SEM	P-value
Sex			0.45
Male	24	11.22±1.33
Female	16	11.13±1.06
Age (years)			0.15
≥70	18	12.25±1.09
<70	22	10.31±1.19
Depth of invasion			0.25
T2	3	10.18±2.82
T3-4	37	11.26±0.87
N classification			0.20
NX-N0	7	9.16±2.02
N1	20	11.83±1.20
N2-3	13	11.35±1.39
M classification			0.24
MX	8	10.50±1.90
M0	24	10.89±1.04
M1	8	12.76±2.02
TNM stages			0.21
I-II	22	11.01±1.17
III	11	10.36±1.42
IV	7	12.76±2.02
Perineural invasion			0.20
Negative	10	14.57±0.63
Positive	30	10.89±0.93
Lymphatic invasion			0.23
Negative	14	10.32±1.57
Positive	26	11.65±0.95
Tumor size			0.30
≥5	31	10.84±1.95
<5	9	12.36±4.12
Tumor grades			0.35
I	12	11.11±1.67
II	9	9.66±1.97
III	19	12.07±0.99
Tumor types			0.24
Diffuse	19	10.97±1.10
Intestinal	21	11.37±1.23

A lower ΔCt value indicates a higher expression level

Table 4

Relationship between UCA1 expression levels (as divided into two groups based on the median of ΔCt) and clinicopathological characteristics of tumoral gastric specimens

Characteristics	Number (#40)	lncRNA UCA1 expression		P-value
		Low (higher ΔCt than median) (#20)	High (smaller ΔCt than median) (#20)
Sex				0.37
Male	24	13	11
Female	16	7	9
Age (years)				0.50
≥70	18	9	9
<70	22	11	11
Depth of invasion				0.50
T2	3	2	1
T3-4	37	18	19
N classification				0.21
NX-N0	7	3	4
N1	20	12	8
N2-3	13	5	8
M classification				0.06
MX	8	2	6
M0	24	15	9
M1	8	3	5
TNM stages				0.17
I-II	22	13	9
III	11	5	6
IV	7	2	5
Perineural invasion				0.07
Negative	10	4	6
Positive	30	16	14
Lymphatic invasion				0.16
Negative	14	5	9
Positive	26	15	11
Tumor size				0.50
≥5	31	15	16
<5	9	5	4
Tumor grades				0.13
I	12	5	7
II	9	3	6
III	19	12	7
Tumor types				0.26
Diffuse	19	11	8
Intestinal	21	9	12

Table 5

Relationship between ecCEBPA expression levels (as divided into two groups based on the median of ΔCt) and clinicopathological characteristics of tumoral gastric specimens

Characteristics	Number (#40)	lncRNA ecCEBPA expression		P-value
		Low (higher ΔCt than median) (#20)	High (smaller ΔCt than median) (#20)
Sex				0.36
Male	24	13	11
Female	16	7	9
Age (years)				0.37
≥70	18	10	8
<70	22	10	12
Depth of invasion				0.50
T2	3	2	1
T3-4	37	18	19
N classification				0.22
NX-N0	7	2	5
N1	18	10	8
N2-3	15	8	7
M classification				0.12
Mx	8	3	5
M0	24	11	13
M1	8	6	2
TNM stages				0.13
I-II	21	9	12
III	11	5	6
IV	8	6	2
Perineural invasion				0.13
Negative	10	7	3
Positive	30	13	17
Lymphatic invasion				0.50
Negative	14	7	3
Positive	26	13	17
Tumor size				0.135
≥5	30	13	17
<5	10	7	3
Tumor grades				0.46
I	12	6	6
II	9	4	5
III	19	10	9
Tumor types				0.50
Diffuse	19	9	10
Intestinal	21	11	10

Figure 4

Correlation analysis of lncRNAs UCA1 and ecCEBPA gene expression levels (P-value=0.000, r=0.46)

Association of UCA1 expression level with clinical data in the TCGA stomach cancer cohort

The authors explored the expression levels of UCA1 lncRNA in a TCGA stomach cancer (STAD) cohort. We found that UCA1 was significantly overexpressed in gastric tumoral tissues compared with normal tissues (P<0.0001, Figure 5). Further analysis based on UCA1 mean expression data showed that there is a significant correlation between UCA1 gene expression levels and M classification (P=0.01) and various Lauren`s classes (P=0.0009) (Table 6). Patients were then divided into low and high UCA1 expression groups according to the median value. The results demonstrated that there were significant associations between UCA1 gene expression level and M classification (P=0.04), TNM stages (P=0.01), tumor grades (P=0.0002), and Lauren`s classes (P=0.0002) (Table 7).

Figure 5

Relative expression of UCA1 in tumoral and non-tumoral gastric tissue samples in the TCGA database. UCA1 was found to be highly overexpressed in tumoral tissues compared with normal tissues in the TCGA RNA-seq data (P<0.0001)

Table 6

Relationship between UCA1 mean expression levels and clinicopathological characteristics of tumoral gastric specimens in the TCGA cohort

Characteristics	Numbers (%)	Mean±SEM	P-value
Sex			0.20
Male	174 (61.05)	5.25±0.79
Female	111 (38.95)	3.64±.079
Age (years)			0.07
≥70	113 (39.65)	5.46±1.01
<70	167 (58.60)	3.98±.070
NA*	5 (1.75)
Depth of invasion			0.20
T1	13 (4.56)	2.018±0.66
T2	72 (25.26)	4.90±1.18
T3	113 (39.65)	5.84±1.05
T4	78 (27.37)	2.82±0.82
TX	9 (3.16)	6.52±4.23
N classification			0.07
N0	94 (32.98)	4.27±0.92
N1	78 (27.37)	4.53±0.89
N2	47 (16.49)	6.63±1.88
N3	53 (18.60)	3.86±1.45
NX	13 (4.56)	3.83±2.86
M classification			0.01**
M0	253 (88.77)	4.06±0.50
M1	18 (6.32)	6.30±4.05
MX	14 (4.91)	12.68±4.99
TNM stage			0.40
I	40 (14.03)	4.45±1.44
II	99 (34.74)	4.20±0.85
III	101 (35.44)	4.86±0.99
IV	25 (8.77)	5.69±2.93
NA	20 (7.02)
Tumor grades			0.11
G1-GX	12 (4.21)	2.75±1.28
G2	91 (31.93)	4.20±1.69
G3	182 (63.86)	3.49±0.97
Tumor types			0.0009**
Diffuse	51 (17.89)	3.81±1.54
Intestinal	95 (33.33)	4.53±0.90
Mixed	136 (47.72)	5.06±0.85
NA	3 (1.05)

Not available

Statistically significant

Table 7

Relationship between UCA1 expression levels and clinicopathological characteristics of tumoral gastric specimens of the TCGA cohort based on median gene expression level

Characteristics	Numbers (#285)	UCA1 expression		P-value
		A higher expression than median (#142)	A lower expression than median (#143)
Sex				0.35
Male	174	91	83
Female	111	51	60
Age (years)				0.15
≥70	113	62	51
<70	167	76	91
NA*	5	4	1
Depth of invasion				0.29
T1	13	8	5
T2	72	41	31
T3	113	57	56
T4	78	33	45
TX	9	3	6
N classification				0.22
N0	94	51	43
N1	78	39	38
N2	47	27	20
N3	53	20	33
NX	13	5	8
M classification				0.04**
M0	253	127	126
M1	18	5	13
MX	14	10	4
TNM stage				0.01**
I	40	24	16
II	99	52	27
III	101	49	52
IV	25	8	17
NA	20	9	11
Tumor grades				0.0002**
G1-GX	12	7	5
G2	91	61	30
G3	182	74	108
Tumor types				0.0002**
Diffuse	51	13	38
Intestinal	95	58	37
Mixed	136	70	66
NA	3	1	2

Not available

Statistically significant

High expression levels of UCA1 were correlated with unfavorable survival in patients with pancreatic and lung adenocarcinomas

We furthermore examined whether UCA1 expression level correlates with patient overall survival time across multiple solid cancer types (based on P-values from the univariate Cox proportional hazards model and the log-rank test and visualization through a Kaplan-Meier plot). As shown in Figure 6, elevated UCA1 expression was contributed to a significant poorer survival in patients with pancreatic (P=4.5×10-6) and lung adenocarcinomas (P=7.3×10-3) but not with survival time of other solid cancer types including GC patients (data not shown).

Figure 6

Kaplan-Meier curves for overall survival of TCGA cohort patients with pancreatic, lung, and gastric adenocarcinomas categorized according to UCA1 expression: significantly poorer overall survival was observed in patients with high UCA1 expression than in those with low UCA1 expression (a and b). There was no significant association between UCA1 expression and overall survival of TCGA cohort gastric cancer patients (c)

Discussion

In the present study, we evaluated and quantified the expression level of the ecCEBPA gene in tumoral and non-tumoral gastric tissues as well as various cultured cell lines by using quantitative real-time PCR. To our knowledge, the expression profiling of ecCEBPA has not been previously studied in GC. Our results indicate that ecCEBPA expression level was elevated in tumoral tissues compared to the non-tumoral samples. We observed no associations between this lncRNA expression pattern and any of the clinicopathological features. According to the only published study on ecCEBPA conducted by Di Ruscio et al. (17), ecCEBPA is a non-polyadenylated lncRNA which is transcribed from the CEBPA locus and is enriched in nuclear fraction. It has been stated that ecCEBPA associates physically with DNMT1 which leads in blocking DNA methylation in CEBPA promoter and maintains the mRNA expression on from this locus. According to that study, CEBPA promoter DNA demethylation, which is mediated by ecCEBPA, is almost selective for CEBPA locus. Furthermore, they have reported ecCEBPA expression in Hl-60 and U937 cell lines by using strand specific reverse transcriptase PCR and northern blot analysis.

Furthermore, our study demonstrated UCA1 over-expression in tumoral gastric tissues in comparison to non-tumoral ones. Additionally, in the present study, we analyzed stomach cancer datasets based on the TCGA platform and showed that the expression of UCA1 was significantly higher in GC tissues. Clinically, increased expression of UCA1 was a predictor of OS in pancreatic and lung adenocarcinoma patients in the TCGA corhort. We could further detect a significant association between UCA1 expression levels and tumor grades, tumor type, and M classification. UCA1 involvement has been shown in many cancer types. It has been reported that this lncRNA accelerates cell cycle progression and proliferation and also suppresses apoptosis in colorectal cancer cells (12). UCA1 increases tumor growth and metastasis by raising cell proliferation through repressing p27 in breast cancer tissues (10). According to a recent study in GC cells, UCA1 promotes epithelial-mesenchymal transition (EMT), an essential early step in tumor metastasis (25). Our data on association between UCA1 expression levels and M classification, as an indicator of distant metastasis is in agreement with the role of UCA1 in EMT and metastasis regulation (25).

Researchers (26) have investigated UCA1 expre-ssion pattern in tumoral and non-tumoral gastric tissues (5 pairs) and also in plasma samples of patients with GC and in their pair-matched plasma (20 pairs) by using RT-PCR. According to their study, UCA1 is up-regulated in tumoral gastric tissues and plasma with a positive correlation between UCA1 expression in cancerous gastric tissues and plasma. Associations between UCA1 expression level and lymph node metastasis and staging have also been shown in that study. In another study by Zheng et al. (27), expression profile of UCA1 was measured in tumoral and non-tumoral gastric tissues and juice. Their results demonstrate over-expression of UCA1 in tumoral samples; they also showed associations between UCA1 expression and differentiation, tumor size, invasion depth, and TNM stages. Moreover, it was indicated that patients with a high expression level of UCA1 are likely to have shorter overall and disease-free survival than patients with lower expression. Overall, our study is consistent with the mentioned studies whereas we could show a correlation between UCA1 expression level and tumor grades, types, stages, and M classification. Up-regulation of UCA1 has been previously documented in several other cancer types including bladder carcinoma (3), non-small cell lung cancer (4), tongue squamous cell carcinoma (5), esophageal squamous cell carcinoma (6), ovarian cancer (7), melanoma (8), and hepatocellular carcinoma (9). According to the present information, UCA1 can serve as an oncogenic lncRNA in a variety of cancers, which makes it a competent therapeutic target.

Another interesting finding of the current study was a moderate positive correlation between UCA1 and ecCEBPA expression levels, which is consistent with the data and hypotheses of Di Ruscio et al. (17) and Xue et al (13). Generally, our investigations showed up-regulation of both ecCEBPA and UCA1 in GC. Moreover, the associations between UCA1 and tumor grades, types, stages, and M classification were significant. The elucidation of the molecular mechanisms modulated by these lncRNAs needs further investigation.

Conclusion

This study demonstrates UCA1 and ecCEBPA up-regulation in GC tissues. Higher expression of UCA1 was associated with tumor grades, types, stages, and M classification. Furthermore, analyzing data taken from TCGA database for UCA1 expression, showed over-expression of UCA1 in GC. Moderate co-expression of lncRNAs, ecCEBPA, and UCA1 were also shown in this research. These data indicate UCA1 and ecCEBPA involvement in GC and suggest that these lncRNAs might be useful as diagnostic/prognostic biomarkers in GC.