Downregulation of GPR155 as a prognostic factor after curative resection of hepatocellular carcinoma

Hepatocellular carcinoma (HCC) ranks at the third most common cause of cancer-related death in the world [1, 2]. Although liver resection has been the mainstay of treatment for HCC, the recurrence rate after curative resection remains high at approximately 70% [2‐4]. Complete cure of this disease is quite challenging even though various therapeutic modalities have been developed. A realistic initial goal is the establishment of methods for accurate risk stratification and prediction of recurrence sites after liver resection to provide appropriate perioperative management according to each individual patient’s circumstances [5]. The TNM classification system has been broadly employed as a tumor staging method to predict postoperative outcomes concisely but can be inaccurate [6, 7]. For example, patients with an earlier tumor stage sometimes have unfavorable prognosis. Extrahepatic recurrences, such as lung, bone, and brain metastases, can be a cause of an unexpected and rapidly deteriorating patient course; however, no methods for predicting the likelihood of extrahepatic recurrences of HCC are currently available [8, 9]. Conversely, some patients are long-term survivors after resection of advanced HCC without adjuvant therapy. To address these clinical issues, development of a novel molecular marker able to reflect potential characteristics of the tumor is required [10].

G protein-coupled receptors (GPCRs) are reportedly cell surface signaling proteins that have important roles in various physiological functions, and in initiation and progression of cancer [11]. The G protein-coupled receptor 155 gene (GPR155), present on 2q31.1, encodes a 97 kDa transmembrane receptor protein that is a member of the GPCR family [12]. Although there has been a report that GPR155 expression is suppressed in neoplasms of the thyroid, the oncologic roles of GPR155 in HCC remain unclear [13, 14]. We focus on GPR155 because it is recognized as a transmembrane marker possibly associated with the transport of growth factors and anticancer drugs, and no published data of GPR155 expression in HCC.

The aims of this study were to evaluate the clinical significance of GPR155 expression, explore the factors that regulate GPR155 transcription, and assess the performance of GPR155 as a potential prognosticator of HCC.

In the present study we evaluated the expression of GPR155 and its predictive value in HCC. The GPCR superfamily of membranous receptors, of which GPR155 is a member, has a variety of roles in intracellular signal transduction [21, 22]. When various ligands are recognized by GPCRs, GDP is converted to GTP and the α subunit and βγ subunit, acting as individual effector molecules, dissociate from the GPCR and are reported to be involved in multiple processes of cancer progression [11, 22, 23]. GPR155 harbors an auxin efflux carrier domain, a pleckstrin/G protein-interacting region, and a winged helix repressor DNA-binding domain; however, the function of the receptor is poorly understood [12, 14]. There have been some reports of GPR155 expression in mouse models, such as aberrant expression of GPR155 in UV-induced melanoma and Huntington’s disease models [12, 24]. With respect to human neoplasms, only one microarray analysis indicated suppression of GPR155 in thyroid tumor, and to our best knowledge this study is the first to evaluate GPR155 expression in digestive cancers, including HCC [14].

We found that GPR155 mRNA expression was decreased in 89% of HCC cell lines compared with the control non-tumorigenic cell line. As promoter hypermethylation is recognized as one of the prominent regulatory mechanisms of gene transcription [25], we conducted bisulfate sequence analysis to determine mechanisms of GPR155 suppression; however, no methylation was detected at the CpG island within the promoter region of GPR155 gene in any of the cell lines tested. We then performed copy number analysis to explore an alternative mechanism of GPR155 transcription because analysis of copy number variations on a genomic scale has been reported to be useful for assessing cancer progression and identifying congenital genetic abnormalities. Moreover, accumulating evidence indicates that loss of heterozygosity, mutations, and homozygous deletions are frequently present at human chromosome 2q31, the location of the GPR155 gene [26, 27]. We found copy number alterations in five (56%) HCC cell lines that showed reduced expression levels of GPR155 mRNA. These results indicated that copy number alteration might be one of the major regulatory mechanisms of GPR155 transcription. However, some HCC cell lines with decreased GPR155 mRNA expression did not show copy number alterations. When referring to The Cancer Genome Atlas database for HCC via the cBioPortal (http://​www.​cbioportal.​org/), mutations and copy number alterations were found 0.8% and 2% of HCC tissues, respectively, though our data showed more frequent copy number alterations in HCC cell lines. Further investigation of other molecular modifications, such as acetylation of histone and microRNA expression, is expected to increase our understanding of GPR155 regulation in HCC.

In clinical samples, GPR155 mRNA levels were decreased in HCC tissues compared with the corresponding non-cancerous tissues, consistent with the results in cell lines. GPR155 mRNA expression levels were equivalent among normal liver, hepatitis, and cirrhosis as background liver status. These findings suggested that alteration of GPR155 expression may represent a specific event that occurs in the final stage of the initiation of HCC or during disease progression. Downregulation of GPR155 was associated with more aggressive phenotypes of HCC, and subsequently linked to poorer postoperative survival. GPR155 protein was successfully detected by immunohistochemical staining and we found a close correlation between GPR155 protein and mRNA expression, which allowed us to evaluate the clinical significance of GPR155 mRNA levels in a quantitative manner. Furthermore, this result may emphasize the clinical utility of GPR155 because immunohistochemical staining is a convenient and popular method commonly available in most hospitals. Both liver biopsy samples and surgically-resected specimens can be applicable in this context.

HBV and HCV infection have been recognized as major causes of HCC [2, 28]. In the latest decade, the incidence of HCV-related HCC has been dramatically declining due to increased adoption of precautions and the introduction of a direct-acting anti-HCV agent [3, 29]. Accordingly, nonBnonC-HCC arising from chronic hepatic disease, including nonalcoholic steatohepatitis and nonalcoholic fatty liver disease, is becoming increasingly important in clinical practice [7, 30‐32]. In this study, we conducted a subset analysis according to hepatitis virus infection. No significant differences in GPR155 expression levels were observed among the nonBnonC, HBV, and HCV groups for both HCC and non-cancerous tissues. In previous literature it has been reported that nonBnonC-HCC is more prevalent in male patients, has relatively low transaminase levels, larger tumor size, advanced disease stage at the time of diagnosis, and a worse prognosis compared with HBV/HCV-related HCCs [7, 31, 33]. We found that the prognostic impact of GPR155 expression was equivalent in nonBnonC and HBV/HCV-related HCCs. These findings highlight the clinical utility of GPR155 expression as a prognosticator regardless of hepatitis virus infection.

Another notable finding of our study was that GPR155 expression was associated not only with overall survival but also with initial recurrence patterns. The fact that downregulation of GPR155 had a more remarkable effect on overall survival than disease-free survival motivated us to investigate the association between GPR155 expression and initial recurrence patterns. Recurrence sites represent a serious issue in the management of HCC. In cases with liver-confined recurrences, repetition of liver resection is applicable and long-term survival can be expected [2]. In contrast, the prognosis of patients with extrahepatic recurrences is dismal due to the lack of effective systemic chemotherapy [3, 9, 34]. To date, there are no biomarkers for prediction of the recurrence patterns of HCC. Our findings indicate that physicians can make a risk stratification of distant recurrences and poor prognosis by determining the expression levels of GPR155 using liver biopsies or surgical samples. Moreover, the expression levels of GPR155 may serves as a biomarker to establish a criterion for determining an appropriate therapeutic strategy such as topical therapy or systemic chemotherapy. For future consideration, external validation is necessary.

This study was limited by its lack of sufficient functional analysis of GPR155, which tempers the conclusion that it acts as a tumor suppressor in HCC. Further studies including pathway analysis and functional analysis by forced expression experiments are expected to clarify the molecular mechanisms underlying the biological activities of GPR155 in HCC.

Taken together, our results indicate that downregulation of GPR155 might be a prognostic factor and a predictor of initial recurrence sites, independent of hepatitis virus infection. Evaluation of GPR155 expression might improve patient follow-up and treatment after liver resection, possibly leading to better prognosis.