Clinical Potential of HOTAIR, MALAT1, and UCA1 lncRNAs As a Biomarker to Achieve More Accurate Prognostic Predictions for Liver Cancer

Esmaeil Mahmoudi Islamic Azad University Shahrekord Branch Mona Ebrahimi Islamic Azad University Shahrekord Branch Fatemeh Amini Chermahini Shahroud University of Medical Sciences Eskandar Hoseinnezhad Lazarjani Islamic Azad University Shahrekord Branch Hamidreza Kabiri Islamic Azad University Shahrekord Branch Hassan Jamshidian Isfahan University of Medical Sciences Asghar Arshi University of Arkansas Fayetteville Fatemeh Sadat Shari Islamic Azad University Shahrekord Branch Farzaneh Raeisi (  farzaneraisi@yahoo.com ) Shiraz University of Medical Sciences https://orcid.org/0000-0001-5721-5764

Methods and Results: The expression pro les of HOTAIR, MALAT1, and UCA1 lncRNAs were evaluated using qRT-PCR in the paired liver tumor and the adjacent non-tumor samples. After RNA extraction from tissue samples, cDNA synthesis and the RT-qPCR method were performed. Livak method (2 -ΔΔCt ) was used for calculating the expression level of lncRNAs. Principal-component analyses followed by receiver operating characteristic (ROC) curve analyses were performed to evaluate the diagnostic potential of the selected lncRNAs. Our results showed that HOTAIR, MALAT1, and UCA1 were overexpressed signi cantly in patients with liver cancer compared to the healthy groups (P < 0.001). Moreover, the expression of HOTAIR was enhanced signi cantly compared to the expression of MALAT1 and UCA1 in patients with liver cancer (P < 0.001). This study showed that there were no signi cant associations between lncRNAs expression and the clinical characteristics (P > 0.05). Signi cantly elevated circulating lncRNAs were found to be liver cancer-speci c and showed differentiation of liver cancer samples from the controls.
Kaplan-Meier analysis revealed no signi cant correlations between the lncRNAs expression and overall survival.
Conclusion: Based on our ndings, the studied lncRNAs were not correlated with clinicopathological characteristics of the liver cancer patients although the overexpression of these lncRNAs might provide novel molecular biomarkers in HCC cases.

Background
One of the most common types of cancer is liver cancer. The incidence of liver cancer differs vastly over the world and is more prevalent in sub-Saharan Africa and eastern Asia [1]. Most of the primary liver cancers (PLC) are derived from the intrahepatic bile ducts epithelial lining [2]. HCC encompasses nearly 90 percent of all cases of primary liver malignancy and is the biggest cause of cancer-related death [3,4].
Despite the advancement of various therapy strategies, such as surgical resection, radiation therapy, and chemotherapy, the outcomes are unsatisfactory as HCC molecular mechanism is still unknown [5].
Therefore, various novel and reliable biomarkers are required to identify, predict and treat liver cancer. As a result, many types of research on unraveling the molecular mechanisms of liver cancer have been conducted, and various mechanisms have been recognized [1].
Recent advances in cancer transcriptome pro ling and the documents supporting long non-coding RNAs (lncRNAs) function, several differentially expressed lncRNAs are correlated with various types of cancer including breast, lung, colorectal, prostate, and liver cancer [6]. Accumulating studies of cancer-associated lncRNAs have been reported on the critical role of lncRNAs in tumorigenesis and the development of various tumors, several cancer types, and metastasis [6-8]. Furthermore, numerous researches have shown the associations between some lncRNAs, such as prostate cancer-associated non-coding RNA 1 (PRNCR1), the metastasis-associated HOX antisense intergenic RNA (HOTAIR), metastasis-associated lung adenocarcinoma transcript 1 (MALAT1), urothelial cancer-associated 1 (UCA1), colon cancer-associated transcript 2 (CCAT2) and different cancers, such as prostate cancer [9], liver cancer [10], breast cancer [11], lung cancer [12], gastric cancer [13] and esophageal squamous cell carcinoma (ESCC) [14]. According to the literature, various cancer types have been affected by tumorous clinical-pathological features, such as age, gender, lymph node metastasis, clinical stage, and tumor size [15]. Recently, many researchers are searching for novel biomarkers that could help the diagnosis or prognosis of cancers. One of the main advantages of lncRNAs is their high stability as cancer diagnostic and prognostic biomarkers. Different studies have reported signi cant correlations between lncRNAs expression and clinical characteristics [13,[16][17][18]].
In the present study, we propose that the lncRNAs expression pro le may be used as a clinical marker for the diagnosis or prognosis of liver cancer therefore in this study we evaluated the potential usefulness of three lncRNAs (HOTAIR, MALAT1, and UCA1) in liver cancer.

Subjects
We studied a total of 15 patients. Fifteen liver tumor and non-tumor (healthy) tissue samples from patients without preoperative chemotherapy or radiotherapy were obtained from Isfahan General Hospital (Isfahan, Iran) and were histologically evaluated based on the type and the grade of cancer.

Rna Extraction And Cdna Synthesis
Samples were transferred to RNA later immediately after resection and stored at -20°C until used for RNA extraction. The RNX TM -Plus solution was used to extract total RNA (SinaClon, IRAN) except for an extended 1h treatment with DNaseI. Two methods were used to assess RNA purity, concentration, and integrity, Thermo Scienti c NanoDrop™ 1000 Spectrophotometer and electrophoresed on 2% agarose gel.
For reverse transcribed (1 mcg of RNA for complementary DNA), the random hexamer priming and PrimeScript TM -RT reagent kit (TaKaRa, Japan) was used according to the manufacturer's protocol. This cDNA was quanti ed using spectrophotometry.

Quantitative Real-time Pcr
All samples were analyzed using a rotor gene 6000 Corbett detection system and the qPCR was quanti ed using SYBR®Premix Ex Taq TM II kit (TaKaRa, Japan) according to the manufacturer's instructions. Thermal cycling conditions were applied for 5 min at 95°C followed by 40 cycles at 95°C for 15s and 60°C for 1 min. No template control (NTC) containing H 2 O was included in each run. To verify the speci city of PCR products, a melting curve analysis was performed. Besides, PCR products were analyzed in terms of size and speci city using agarose gel electrophoresis. For qPCR analysis, all samples were normalized to GAPDH. Forward and reverse primer sequences are indicated in Table 1. The qPCR tests were run in triplicate, and the results were provided as the mean ± the standard error of the mean (SEM). The relative lncRNA concentration was calculated using the mean value in each triple (ΔCt=Ct mean lncRNA-Ct mean GAPDH). To calculate expression fold changes, 2 −ΔΔCt methods were used [19,20].  [19,20].

Results
Expression of HOTAIR, MALAT1, and UCA1 in liver cancer The expression levels of three lncRNAs, (HOTAIR, MALAT1, and UCA1), in 15 pairs of liver cancer and matched adjacent non-cancerous tissues were measured by real-time PCR. The expression levels of HOTAIR, MALAT1, and UCA1 were upregulated compared to their average expression in healthy tissues (P<0.0001) (Fig. 1). Fig. 1 also indicates that the expression level of HOTAIR was increased signi cantly compared to the expression of MALAT1 and UCA1 in patients with liver cancer (P < 0.001).

Correlations Between Lncrnas Expressions
We used the Chi-square test and Fisher's exact test to evaluate correlations between the lncRNAs expressions. The patients were divided based on the median value of the lncRNAs expression into two groups of low and high lncRNAs expression cases. Table 2 shows the correlations between lncRNAs expressions for these two groups. The Chi-square test and Fisher's exact test indicated that there were no signi cant associations between lncRNAs expressions (P > 0.05).

Correlations Between Lncrnas Expressions And Clinical Characteristics
Tables 3, 4 and, 5 show the correlations between HOTAIR, MALAT1, and UCA1 expression and clinical characteristics, respectively. We found that there were no signi cant correlations between lncRNAs expressions and the clinical characteristics (P > 0.05).  Well differentiated 6.7 6.7 0 Well differentiated 6.7 6.7 0 Liver Cancer-speci c Tumor Marker ROC analysis was performed to distinguish the optimal cutoff value for lncRNAs to differentiate liver cancer cases from controls from which the sensitivities of circulating HOTAIR, MALAT1, and UCA1 were de ned to be 100%, 100%, and 100%, at the speci cities of 100%, 93.33%, and 100% with an area under the ROC curve of 1.000, 0.998 and 1.000, respectively (Fig. 2).

Correlation Between The Lncrnas Expression And Patient Survival
The log-rank test in liver cancer patients was applied. To evaluate the predictive value of the lncRNAs levels in liver cancer patients the Cox proportional hazards regression model was also used. Clinicopathological factors and OS were then analyzed in the high and the low lncRNAs expression groups, but no signi cant differences were found between groups (P > 0.05, Table 6 and Fig. 3).  [15,18]. These transcripts play critical roles in physiological processes [15]. LncRNAs were initially considered to be spurious transcriptional noise. However, in the past few years, thousands of lncRNAs were recognized, and the functional roles of them in epigenetics have been argued [13,16]. LncRNAs, functioning as regulatory agents, have been de ned for various complex cellular processes, including differentiation, cellular signaling, genomic imprinting, alternative splicing, angiogenesis, epigenetic regulation, cell death, cell proliferation, and growth [7,8]. Increasing evidence has indicated that dysregulation in expression levels of many lncRNAs are associated with developmental processes and disease states most notably in cancer [7,17].
One of the long intergenic non-coding RNA (lincRNA) is MALAT1 with >8,000 nts, located on chromosome 11q13 [32]. MALAT1 is correlated with high metastatic potential and poor patient prognosis in a variety of cancers, such as lung cancer, hepatocellular carcinoma, breast, prostate, uterus, and esophageal squamous [4,6,12,33]. This lncRNA is associated with cancer in terms of proliferation, metastasis, invasion, and apoptosis [32]. Association between high expression of MALAT1 and melanoma metastasis has been reported by Tian et al [34]. Moreover, Dong et al. revealed that MALAT1 can play an important role in tumor proliferation and metastasis via the phosphoinositide 3-kinase (PI3K)/Akt pathway [35]. It has been shown that overexpression of MALAT1 in tumor tissues or sera may cause advanced tumor stages and reduced overall survival of HCC patients and can signify a higher risk of tumor recurrence following liver transplantation [3,36]. In addition, the correlation between overexpression of MALAT1 in HCC and chemoresistance to multiple agents including 5-uorouracil, mitomycin C, and Adriamycin via a hypoxia-inducible factor (HIF)-1α-MALAT1-microRNA(mir)-216b pathway has been reported [37]. Also, the key role of MALAT1 in HCC progression by inducing serine/arginine-rich splicing factor 1 (SRSF1) upregulation and mammalian target of rapamycin (mTOR) activation was demonstrated [38].
reported the important role of UCA1 as one of signi cant mediators of radiation response in prostate cancer [41]. Moreover, according to Wen et al. UCA1 can perform as a potential biomarker for diagnosis and prognosis of osteosarcoma [40]. Ultimately, Wang et al. revealed the signi cance of UCA1 in predicting tumor lymph node metastasis [44]. According to Feng Wang et al. UCA1 overexpression in HCC tissues has been linked to a variety of factors, including TNM stage, metastasis, and postoperative survival. UCA1 has the ability to reverse the inhibitory effect of miR-216b on HCC cell growth and metastasis, which could be related to the derepression of broblast growth factor receptor 1 (FGFR1) expression acting as a miR-216b target gene and the activation of the ERK signaling pathway [15].
In this study, we rst assessed the expression of HOTAIR, MALAT1, and UCA1 in liver cancer tissues. From the results obtained in our study, HOTAIR, MALAT1, and UCA1 lncRNA in liver cancer tissues showed to have increased expression levels compared to the healthy tissues. Our ndings are in agreement with previous study studies [4,8,15,18]. In various reports, signi cant correlations between HOTAIR, MALAT1, and UCA1 expression levels and clinical-pathological characteristics, such as lymph node metastasis, clinical stage, and tumor size, in some cancers including gastric cancer [13], HCC [18], ESCC [33], and osteosarcoma [40] were found. However, our results did not show signi cant correlations between HOTAIR, MALAT1, and UCA1 expression levels and the clinical-pathological characteristics, our ndings also indicated HOTAIR, MALAT1, and UCA1 can serve as a biomarker for the diagnosis of liver cancer. According to various studies HOTAIR, MALAT1, and UCA1 may be as a biomarker in gastric cancer, papillary thyroid cancer, and osteosarcoma, respectively [13,40,45]. Our ndings are in line with these studies. Some studies have reported signi cant relationships between OS and the relative expression of these lncRNAs (HOTAIR, MALAT1, and UCA1) [4,12,15,18,33], but the OS of patients with high expression of these lncRNAs in liver cancer tissues was not signi cantly lower in our investigation.

Declarations
Ethics approval and consent to participate All applicable international, national, and institutional guidelines for the care of human were followed. All patients and legally authorized representative of patients who were dead during this study signed the informed consent. The study protocol was approved by the ethics committee of the Cellular and Molecular Research Center of Shahrekord University of Medical Sciences.

Consent for publication
Not applicable.

Availability of data and materials
The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.
EM, AA, and FR, designed the study and writing-original draft. ME, FACH, EHL, HK, HJ, FSSH, EM, AA, and FR performed the experiments and analyzed the data. All authors read, critically revised and approved the manuscript. Receiver-operating characteristic (ROC) curve analyses of three-lncRNAs signature to discriminate liver cancer patients from healthy controls