Long Noncoding RNAs and Hepatocellular Carcinoma

Gene expression pathways can be regulated at myriad levels to effectively modulate the activity of the encoded proteins. Investigations over the past few decades have focused principally on how various proteins regulate gene transcription, messenger RNA (mRNA) splicing and stability, translational efficiency, and posttranslational control mechanisms. More recently, control of gene expression by noncoding RNAs has become a novel area of great interest. Surprisingly, only ~1% of the human genome contains protein-coding genes; a greater percentage of the genome (~4%–9%) is transcribed into variable length RNAs with no (or few) identifiable protein-coding regions.¹ It has now become clear that these nonprotein-coding RNAs are of functional significance and do not simply represent “transcriptional noise.” Although only a few of the thousands of noncoding RNAs have been analyzed experimentally, many links to human disease are emerging, including various cancers² (eg, ovarian,³ pancreatic,⁴ and gastric⁵), cardiovascular disease,⁶ diabetes,⁷ and respiratory diseases.⁸ In addition, there are also strong links to hepatocellular carcinoma (HCC), which is among the most common and deadliest of all cancers.⁹ This background forms the basis of this editorial review, which examines an exciting paper related to promotion of liver cancer by a long noncoding RNA (lncRNA) published in this issue of Gastroenterology.¹⁰

Noncoding RNAs, which include small interfering RNAs (siRNAs) or micro RNAs (miRNAs) and other transcribed RNAs, can be classified by size; lncRNAs, the focus of this review, are generally considered as those >200 nucleotides in length.¹¹ These RNAs are transcribed by RNA polymerase II, have a 5’-methyl cap and are polyadenylated at the 3’ end, similar to protein-encoding mRNAs. lncRNAs can be localized to the nucleus or be found in the cytoplasm of cells, where they exert unique functions. Many lncRNAs are known to be expressed in particular developmental patterns, in a cell-type–specific fashion, and in distinct intracellular compartments, thus implicating them in a wide variety of biological processes. Moreover, lncRNAs have been classified according to their location in relation to protein-coding genes in genomic DNA, including intergenic, bidirectional, intronic, antisense, and sense.¹² lncRNAs act by several distinct mechanisms to regulate gene expression and thus modulate diverse biological functions. Some interact with chromatin and recruit histone-modifying enzymes, resulting in chromatin modification and altered DNA methylation patterns, effectively causing epigenetic silencing of target genes.¹³ Other lncRNAs regulate splicing of various pre-mRNAs¹⁴ or function as miRNA “sponges,” whereby the lncRNA binds the miRNAs and effectively inactivates them.¹⁵ LncRNAs may also interact directly with proteins and function as structural components, or modulate protein activity or alter protein localization. In the paper published in this issue of Gastroenterology, the authors make the novel discovery that a particular long intergenic, noncoding RNA (lincRNA-UFC1) functions as an oncogene in hepatocellular cancer cells by inducing progression through the cell cycle and promoting cell proliferation, while simultaneously inhibiting apoptosis. The authors go on to elucidate a novel mechanism by which this cytosolic lincRNA functions to promote tumor progression (as detailed herein).

HCC is the most common form of liver cancer, constituting >90% of clinical cases. It was reported that worldwide >700,000 cases of liver cancer are diagnosed yearly, resulting in approximately 600,000 deaths.¹⁶ The prognosis for prevention and treatment is extremely limited, because most patients are not diagnosed until a very late stage of disease progression and may experience recurrence and metastasis after partial hepatectomy. Several risk factors have been identified, including hepatitis B and C viral infections, afla-toxin B₁ exposure, tobacco smoking, and excessive alcohol consumption.¹⁷ Overall, these facts emphasize the need to develop novel diagnostic approaches and therapies to prevent and/or treat this devastating human disease. Over the past few decades, scientists have learned more about hepatic cancers, with a focus mainly on protein-coding genes, miRNAs, and classic epigenetic regulatory mechanisms involving DNA methylation and histone modification.¹⁸ Importantly, recent investigations have also begun to elucidate how lncRNAs are involved in the pathogenesis of HCC. A few examples of such are briefly outlined below, to place the current work of Cao et al¹⁰ in a proper context.

Many lncRNAs show increased expression in HCC, including H19, HOTAIR, HOTTIP, HULC, MALAT1, lncRNA-hPVT1, and others.^12,18,19 Several of these were shown to promote HCC cell growth and tumor progression (eg, H19, HOTAIR, HULC, and HOTTIP). Interestingly, HULC localizes to the cytoplasm where it co-purifies with ribosomes.²⁰ When HULC was silenced in HCC cells, it altered the expression of several genes known to play important roles in HCC development and progression. Importantly, HULC was detected in blood of HCC patients, suggesting that it could be a potential biomarker for disease diagnosis. Exemplifying a distinct mechanism of disease promotion, MALAT1 overexpression promoted tumor invasion.¹⁴ Other lncRNAs function in response to altered oxygen homeostasis, such as Linc-ROR, which promotes cell survival during hypoxic stress,²¹ and LncRNA-LET, which prevents hypoxia-induced HCC cell invasion (and is down-regulated in HCC).²² Another antisense lncRNA (PCNA-AS1) was recently shown to promote HCC by regulating cell-cycle progression via direct effects on the expression/activity of proliferating cell nuclear antigen (PCNA).²³ An additional recent publication showed that lncRNA-ATB (lncRNA activated by transforming growth factor-β), a mediator of transforming growth factor-β great interest. Surprisingly, only signaling, promotes the invasion–metastasis cascade.²⁴

The work by Cao et al¹⁰ describes another novel mechanism by which lncRNAs alter the pathogenesis of HCC, thus significantly advancing knowledge in this fledgling area of scientific investigation. In their paper, the authors provide convincing evidence that lincRNA-UFC1 directly enhances tumor progression in HCC, but it does so by a distinct mechanism. The initial observation was that lincRNA-UFC1 was more strongly expressed in tumor samples derived from HCC patients compared with adjacent, healthy tissue. This was initially discovered by performing microarray experiments using samples derived from 7 patients; it was subsequently confirmed by quantitative reverse-transcriptase polymerase chain reaction in samples from 49 patients (cohort 1). Moreover, using biological samples and clinical data from a large group of patients (n = 131; cohort 2), lincRNA-UFC1 expression was positively correlated with tumor size, clinical liver cancer stage, and patient outcomes. Laboratory studies were then utilized to determine whether there was a causal relationship between HCC progression and lincRNA-UFC1 overexpression. Transgenic expression of lincRNA-UFC1 in 4 HCC cell lines promoted cell-cycle progression and enhanced proliferation, but inhibited apoptosis. Conversely, inhibition of lincRNA-UFC1 expression by siRNA-mediated gene silencing had the opposite effect. Furthermore, xenograft tumors derived from cells overexpressing lincRNA-UFC1 in male BALB/c nude mice had larger mean volumes and formed more quickly than tumors grown from control cells. Tumors grown from lincRNA-UFC1 knockdown cells were smaller than controls. The authors thus provide convincing evidence, using correlative data in patients and complementary in vitro and in vivo experimental approaches, that lincRNA-UFC1 has direct effects on HCC progression. The next question was, what is the mechanism?

To address this question, additional experimentation was required. Using a combination of experimental techniques and different model systems, the authors provided a reasonable mechanistic explanation for how lincRNA-UFC1 overexpression promoted tumor progression in HCC patients. The first step was to establish correlations between lincRNA-UFC1 and other differentially expressed mRNAs. It was noted that β-catenin expression was increased in tumor tissue, which positively correlated with lincRNA-UFC1 expression in 23 HCC specimens, suggesting a possible link. Moreover, β-catenin mRNA expression increased in cells overexpressing lincRNA-UFC1, and was lower in lincRNA-UFC1 knockdown cells. Additionally, β-catenin protein accumulated in the nucleus of cells overexpressing lincRNA-UFC1 and transcription of some β-catenin target genes was increased (eg, c-myc, cyclin D1). Direct interactions between β-catenin and lincRNA-UFC1, however, could not be established. The authors thus reasoned that interactions could be mediated via other proteins, such as RNA-binding proteins. Indeed, RNA immunoprecipitation experiments revealed that β-catenin mRNA and lincRNA-UFC1 were both bound by HuR (an RNA-binding protein; also called ELAV-like RNA-binding protein 1). Further experiments suggested that lincRNA-UFC1 promoted HuR translocation from the nucleus to the cytoplasm in complex with β-catenin mRNA. The authors thus concluded that lincRNA-UFC1 promotes tumorigenesis by stabilizing the β-catenin transcript, in a HuR-dependent manner, ultimately leading to increases in functional β-catenin protein in the nucleus.

This investigation provides an intriguing explanation for how lincRNA-UFC1 promotes tumor progression in HCC. Presented data support the concept that lincRNA-UFC1 overexpression promotes a HuR/β-catenin mRNA complex to translocate from the nucleus into the cytosol, whereby interaction with lincRNA-UFC1 stabilizes the β-catenin transcript (Figure 1). This then increases β-catenin protein levels, leading to nuclear translocation (which may also be promoted by lincRNA-UFC1), which concomitantly then increases expression of some β-catenin target genes. The protein products of these genes then promote cell-cycle progression (cyclin D1) and cell proliferation, and decrease apoptosis (c-myc). These pathologic changes in gene expression then collectively promote tumor progression in HCC cells. Thus, lincRNA-UFC1 has been identified as a potential therapeutic target in HCC and could also serve as a prognostic marker (eg, if it were found in blood). Future investigations in more physiologically relevant, pre-clinical models of HCC would further strengthen these initial observations and provide rationale for eventually moving toward clinical trials in humans.

Figure 1.

Pathologic regulatory pathway. Long noncoding RNA (lncRNA)-UFC1 is increased in hepatocellular carcinoma (HCC) samples from patients (upper left). UFC1 promotes translocation of the β-catenin transcript bound to HuR from the nucleus into the cytosol. Here, the UFC1/β-catenin/HuR complex stabilizes the β-catenin mRNA, leading to increased translation and production of β-catenin protein. Increased protein levels promote nuclear translocation, assisted by UFC1, where transcription of β-catenin target genes is enhanced (eg, c-myc, cyclin D1). The c-myc and cyclin D1 proteins (and perhaps others) then presumably lead to progression through the cell cycle, increasing cell proliferation, and also inhibit apoptosis of HCC cells. The overall effect then is tumor (and disease) progression. There is evidence that, under normal circumstances, lncRNA-UFC1 expression can be attenuated by microRNA (miRNA)-34a. Expression of this miRNA, is however, suppressed during HCC, allowing high lncRNA-UFC1 expression.

This excellent investigation provides additional insight into the molecular pathogenesis of HCC. Because the β-catenin/WNT signaling cascade is well known to be involved in cancer cell metabolism,²⁵ the observations of Cao et al¹⁰ described in this issue of Gastroenterology are particularly intriguing. In fact, regulation of β-catenin/WNT signaling by lncRNAs has been implicated in the pathogenesis of other cancers as well.^26,27 The task now ahead of these and other investigators in this important area of research is to understand how the host of lncRNAs function in tandem with other traditional effectors of the cancer phenotype to promote disease progression and to develop novel therapeutics that can modify or eliminate these pathologic cellular aberrations that lead to uncontrolled proliferation and tumor development. Suggested therapeutic approaches include silencing of lncRNAs using RNAi technology or antisense oligonucleotides, and functional block of lncRNAs by small molecule inhibitors or antagonistic oligonucleotides which disrupt interaction with target proteins or mimic their secondary structure and thus compete for binding partners.²⁸