ADAR1: a promising new biomarker for esophageal squamous cell carcinoma?

Abstract

Esophageal Squamous Cell Carcinoma (ESCC) is a heterogeneous tumor with enormous genetic and epigenetic changes. RNA editing is an epigenetic mechanism that serves as an additional layer of ‘RNA mutations’ in parallel to DNA mutations. The most frequent type of RNA editing, A-to-I (adenosine-to-inosine) editing catalyzed by Adenosine DeAminase that act on RNA (ADARs), modulates RNA transcripts with profound impact on cellular functions. RNA editing dysregulation has been found to be associated with cancers. Our recent study demonstrated that among all the three RNA editing enzymes, only ADAR1 was overexpressed in primary ESCCs compared with matched non-tumor specimens. In this review, we will discuss current views on the involvement of abnormal A-to-I editing in cancer development, more specifically on the ADAR1-mediated editing in ESCC. Although much is not yet learned about the role of ADAR1 in ESCC, ADAR1 may present an attractive option as a new biomarker for ESCC and as a new molecular therapeutic target.

Keywords::

• ADAR1
• biomarker
• esophageal squamous cell carcinoma
• RNA editing

Esophageal squamous cell carcinoma (ESCC), accounting for more than 90% of all esophageal cancer, is the leading cause of cancer death worldwide and is associated with a poor prognosis [1,2]. Recent advances have proved largely ineffective in treatment, thus studying the molecular pathogenesis of ESCC is the current trend that might potentially lead to the identification of biological markers for early diagnosis and targeted therapies. Up to now, most cancer scientists have focused on studying DNA mutations that convey irreversible changes in the genome. However, discoveries in last decade completely changed our views on RNA and the enormous diversity that can be generated at the RNA level. RNA editing may lead to tumor-specific ‘RNA mutation’, exceeding the number of genomic mutations. In humans, the most frequent type of editing is the conversion of adenosine to inosine (A-to-I), which is catalyzed by the dsRNA-specific Adenosine DeAminase that act on RNA (ADAR) family of protein [3]. The imbalance of ADARs expression/activity is found to be associated with a variety of human diseases, such as amyotrophic lateral sclerosis [4], systemic lupus erythematosus [5,6], neurological disorders [7,8] and cancer [9]. Recently, we have provided the first analyses of expression profiles of ADARs in human ESCC and demonstrated that among all three RNA editing enzymes ADAR1, ADAR2 and ADAR3 [10], only ADAR1 was significantly overexpressed in ESCC tumors, which has great prognostic value and diagnostic potential for ESCC; ADAR1 functions as an oncogene in the development of ESCC; the tight link between the differential expression of ADAR1 and an altered gene-specific editing pattern was investigated to illustrate how the A-to-I RNA editing balance is deregulated in ESCC; and a new functional recurrent RNA editing event, resulting in an amino acid substitution of the AZIN1 gene (antizyme inhibitor 1), was specifically regulated by ADAR1 and found to confer more aggressive tumorigenic behaviors [11]. In this article, we discuss recent studies connecting the differentially expressed ADAR genes with RNA editing alteration in human cancers, with a specific focus on ADAR1-mediated RNA editing in carcinogenesis.

ADAR1 expression & localization

The developmental and cell type-specific modulation of A-to-I RNA editing is tightly linked to ADAR expression and localization. ADAR1 and ADAR2, ubiquitously expressed in many tissues, catalyze all currently known A-to-I editing sites. In contrast, the brain-specifically expressed ADAR3 protein has no documented deaminase activity. ADAR1 protein is the largest of the three family members and its transcripts generate two major ADAR1 isoforms, a full-length interferon-inducible ADAR1 p150 and a shorter and constitutive N-terminally truncated ADAR1 p110. It has been recently reported that both ADAR1 p150 and ADAR1 p110 shuttle between the nucleus and cytoplasm [12]. The nuclear ADAR1 p110 is considered to catalyze A-to-I editing of double-stranded primary RNA transcripts mainly in the nucleus. ADAR1 p150 is detected mainly in the cytoplasma, possibly targeting a different class of dsRNA substrates, such as endoribonuclease-prepared siRNAs [13] and viral RNAs [3,14]. It has been reported that interferons induce the upregulation of ADAR1 [15], therefore raising the possibility that ADAR1 serves as an antiviral defense mechanism against viral infection and inflammation. The expression of ADAR1 is also downregulated by miRNA-1 (miR-1) [16,17], and the tight regulation of ADAR1 by miR-1 is likely to be critical for heart development [18,19]. In ESCC, we found that ADAR1 gene, which is mapped to chromosome 1q21, was amplified in approximately 70% of ESCC tumors using fluorescent in situ hybridization, and there was a positive correlation between overexpression of ADAR1 and the genomic amplification of ADAR1 gene in ESCC specimens [11], suggesting another mechanism of ADAR1 overexpression in tumors.

ADAR1-regulated RNA editing in cancer

The imbalance in expression of ADARs enzymes is highly correlated with cancer development and progression [9,20]. Despite the fact that more than 85% of RNAs are found to be edited in noncoding and/or coding sequences [21], the edited mammalian transcripts have been only well studied in central nervous system and include transcripts encoding the subunits GluR-B, GluR-C, and GluR-D of the α-amino-3-hydroxy-5-methyl-4-isoxazolepropionate receptor, the subunits GluR-6 and GluR-5 of the kainite receptor, and the serotonin receptor (5-HT_2C) [21,22]. However, investigating link between the altered RNA editing activity and cancer progression is only the initial step. In brain tumors, all of three editing enzymes were found to be downregulated in brain tumors [23], and ADAR1 was frequently reduced in metastatic melanomas [16]. As a common editing target by ADAR1 and ADAR2, the decreased editing frequency of glioma-associated oncogene 1 was observed in basal cell carcinoma tumor samples when compared to normal skin, and RNA editing of glioma-associated oncogene 1 transcription factor are involved in Hedgehog signaling [24]. In contrast, ADAR1 was found to be upregulated in tumor tissues, such as hepatocellular carcinoma (HCC) [25], ESCC [11] and breast cancer [26]. It has been reported that homodimer formation may be necessary for ADAR1 and ADAR2 to act as active deaminases [27]. In pediatric glioblastoma multiforme, the overexpression of ADAR1 p110 may cause the formation of an inactive ADAR1/ADAR2 heterodimer or the sequestration of ADAR2 from editing substrates, leading to the altered ADAR2 editing activity and the subsequent decreased editing frequencies [28]. In prostate cancer (CaP), high expression of both enzymes ADAR1 and ADAR2 in androgen-independent CaP cells (DU145 and PC3) resulted in the higher number of RNA editing mutation when compared with androgen-dependent CaP cells (LNCaP and 22Rv1) [29]. Due to the interferon-responsive ADAR1 activity, ADAR1 p150 was upregulated in chronic myeloid leukemia patient samples, which was correlated with BCR/ABL (oncogenic gene fusion protein) amplification and the increased A-to-I editing level [30]. Similarly, ADAR1 was also found to be required for the survival of leukemia cells [31]. Among different types of pediatric acute leukemias, only the B-cell acute lymphoblastic leukemia subgroup exhibited a significant overexpression of ADAR1 p110, with a dramatic decrease in its level in patients achieving complete remission [32].

ADAR1, a prognosis marker in ESCC

Our recent studies have reported that the differential expression of ADAR1 and/or ADAR2 was observed in HCC and ESCC tumors, leading to a gene-specific hyper or hypo-editing phenotype [11,25,33]. In ESCC, ADAR1 and ADAR2 were abundantly expressed, whereas ADAR3 was undetectable. Among all three RNA editing enzymes, ADAR1 was the only RNA editing enzyme found to be significantly upregulated in primary ESCC tumors [11]. The overexpression of ADAR1 in two ESCC cell lines, KYSE180 and EC109 cells, demonstrated more aggressive tumor behavior than control cells, as manifested by the accelerated growth rate, higher frequencies of focus formation and colony formation in soft agar, and the increased migrative and invasive capabilities. As reported previously, ADAR1 could catalyze the deaminase reaction of target genes AZIN1 and FLNB (filament B, β) in HCC cells [23]. The wild-type AZIN1 protein is a short-lived protein and shares high homology with ornithine decarboxylase (ODC). Antizyme binds to ODC and AZIN1, but AZIN1 possesses higher binding affinity to antizyme than ODC [34]. AZIN1 can reduce antizyme-mediated ODC degradation by sequestering antizyme from ODC. Compared with wild-type AZIN1 protein, the edited form binds to antizyme with a higher affinity, and the resultant higher protein stability could promote cell proliferation via neutralizing the antizyme-mediated degradation of ODC and other oncoproteins [25]. Due to the overexpression of ADAR1 in ESCC tumors, the editing frequencies of AZIN1 and FLNB were significantly higher than those in non-tumor specimens. Moreover, the edited form of AZIN1 conferred a ‘gain-of-function’ phenotype associated with aggressive tumor behavior, suggesting that AZIN1 is likely to be one of the downstream editing targets that are responsible for the ADAR1-induced tumorigenicity during ESCC progression.

RNA editing enzyme levels could be used as prognostic markers for identifying ESCC patients at an early stage, selecting treatment modalities for individual patients, and determining post-therapeutic outcomes. Elevated level of ADAR1 was found to be associated with shorter overall survival time of ESCC patients [19]. Moreover, the tumoral overexpression of ADAR1 was shown to be the independent prognostic factor for the overall survival [19]. Overall, ADAR1 is of great clinical value as a new diagnostic, therapeutic indicator and prognostic prediction based on the molecular mechanism underlying esophageal carcinogenesis.

Conclusions

We speculate that monitoring expression level of ADAR1 represents a useful new biomarker for the detection of disorders in cancer before clinical symptoms become apparent. Better understanding the ADAR1 expression of ESCC may lead to more effective management of ESCC by precise prognostic indicators and effective personalized therapy. RNA editing dysregulation can be rectified by restoring ADAR balance by overexpressing or silencing ADARs by shRNAs (small hairpin RNAs) or clustered regularly interspaced short palindromic repeats [35]. However, due to the widespread activity of RNA editing enzymes, reinstating a specific hyper-edited or hypo-edited transcript by introducing a specific RNA-binding peptide [36] or locked nucleotide acids [37] would be more effective.

Financial & competing interests disclosure:

This research is supported by the National Research Foundation Singapore and the Singapore Ministry of Education under its Research Centres of Excellence initiative. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.