MicroRNA In Lung Cancer: Novel Biomarkers and Potential Tools for Treatment

Lung cancer is the leading cause of cancer death in men and women worldwide. The lack of specific and sensitive tools for early diagnosis as well as still-inadequate targeted therapies contribute to poor outcomes. MicroRNAs are small non-coding RNAs, which regulate gene expression post-transcriptionally by translational repression or degradation of target mRNAs. A growing body of evidence suggests various roles of microRNAs including development and progression of lung cancer. In lung cancer, several studies have showed that certain microRNA profiles classified lung cancer subtypes, and that specific microRNA expression signatures distinguished between better-prognosis and worse-prognosis lung cancers. Furthermore, microRNAs circulate in body fluids, and therefore may serve as promising biomarkers for early diagnosis of lung cancer as well as for predicting prognosis of patients. In the present review, we briefly summarize microRNAs in the development and progression of lung cancer, focusing on possible applications of microRNAs as novel biomarkers and tools for treatment.


Introduction
Lung cancer is the leading cause of cancer death in men and women worldwide, accounting for more than 1.5 million deaths per year [1]. Lung carcinoma is generally classified as either small-cell lung carcinoma (SCLC) (about 20% of all lung carcinomas) or non-small cell lung carcinoma (NSCLC) (about 80%). Within these groups, further distinctions are made, with NSCLC sub-divided into adenocarcinoma (about 50% of all lung carcinomas), squamous cell carcinoma (SqCC), and large cell carcinoma.
Although new molecular targeted therapies to some types of lung cancer have shown promising results, no potential targeted therapy can be applied to a large number of lung cancer patients. Despite improvements in early diagnosis of lung cancer, most lung cancers are diagnosed at an advanced stage. Therefore, the identification of novel diagnostic biomarkers or treatment strategies is critical and essential for the control of lung cancer.
MicroRNAs, which are small non-coding RNAs that range in size from 19 to 25 nucleotides, play important regulatory roles in animals and plants by targeting mRNAs for translational repression or degradation. MicroRNAs comprise one of the most abundant classes of gene regulatory molecules in multicellular organisms and likely influence the output of many protein-coding genes [2]. A growing body of evidence is emerging to suggest a wide range of fundamental cellular processes such as cell differentiation, proliferation, growth, mobility, and apoptosis, as well as carcinogenesis or cancer progression [3,4]. The expression patterns of microRNAs are likely to correlate with characteristic clinicopathological parameters in cancer subtypes [5], suggesting that microRNAs are potential biomarkers for different cancer subtypes, classified by origin, histology, aggressiveness, or chemosensivity [6,7].
Public database, miRBase (http://www.mirbase.org/) (last access: 9 February 2016) provides various aspects of microRNA information, and the annotated human microRNAs of 1881 have been registered in miRBase 21. In general, the list of microRNAs involved in lung cancer is not limited to several microRNAs, but much wider, as well as the potential diagnostic microRNAs in tissues.
Of importance, an abundance of undegradated microRNAs exists not only in tissues but also in body fluids, including blood, plasma, serum, and sputum [8]. This nature of easy availability makes microRNAs promising biomarkers for non-invasive liquid biopsy in cancer practice. Biomarkers analyzed by liquid biopsy [9,10] include circulating tumor cells and exosomes (containing DNA, mRNA, microRNA, etc.) [11,12] as well as circulating cell-free DNA [13], mRNA [14], and microRNA [15]. Although, non-invasive liquid biopsy is promising, it has not been employed for multiclass cancer diagnostics due to non-specificity of these blood-based biosources to pinpoint the nature of the primary cancer [9,13,14].
In addition, microRNAs are far less degradated in formalin-fixed paraffin-embedded (FFPE) samples than mRNAs. Therefore, accurate measurements of microRNAs can be performed from FFPE samples, which are usually collected and preserved in hospitals. An availability of archived FFPE samples for the accurate measurement of microRNAs is a great advantage to conduct microRNA research or apply microRNA profiles for clinical practice.
Due to the characteristic nature of microRNAs, microRNAs have a potential to be used for the development of diagnostics, prognostics, and targeted therapeutics. In terms of treatment, microRNA expression profiles can predict chemotherapeutic response and serve as important biomarkers for the stratification of patients for personalized therapeutic strategies. Furthermore, microRNAs have the potential to serve as molecular targeted agents. Although the initial results from human-based studies revealed the promise of microRNA targeted therapies, several obstacles need to be overcome prior to the therapeutic application of microRNAs to the clinic [16].
In this review, we summarize recent works of microRNAs in lung cancer, focusing on microRNAs as novel biomarkers and potential tools for treatment.

MicroRNA Biogenesis
Biogenesis of microRNA ( Figure 1) begins with the transcription of primary-microRNA by RNA polymerase II. Then, DROSHA/DGCR8 enzyme complex crops the primary-microRNA into precursor-microRNA, followed by exportin-5-mediated transport from the nucleus to the cytoplasm. Subsequently, DICER1 cleaves the precursor-microRNA to form the microRNA duplex. One strand of the microRNA duplex is selected to function as a mature microRNA and loaded into the RNA-induced silencing complex (RISC), whereas the partner microRNA* strand is preferentially degradated. The mature microRNA leads to translational repression or degradation of target mRNAs. Figure 1. MicroRNA biogenesis. MicroRNAs are initially transcribed by RNA polymerase II as primary-microRNAs with hairpin structures. DROSHA/DGCR8 enzyme complex then cleaves primary-microRNAs into precursor-microRNAs, which are transported to cytoplasm by Exportin 5, and cleaved by DICER to form microRNA duplexes. One strand is selected to function as a mature microRNA and loaded into the RNA-induced silencing complex (RISC), whereas the partner microRNA* is preferentially degraded. The mature microRNA with RISC binds to 3′UTR of target mRNA resulting in translational repression or degradation.

MicroRNAs as Tumor Suppressor Genes and Oncogenes
MicroRNAs play important regulatory roles in carcinogenesis not only as tumor suppressor genes but also as oncogenes. Several microRNAs are dysregulated in lung cancer. Table 1 shows principal microRNAs involved in the development or progression of lung cancer, their gene targets, and associated biological processes.

Let-7 Family
The let-7 family was the first identified microRNA in humans [17]. In lung cancer, let-7 has been shown to inhibit the expression of oncogenes involved in cellular proliferation, such as RAS, MYC, and HMGA2 [18,19]. Let-7 also inhibits the expression of CDK6, and the reduced expression of let-7, therefore, leads to the promotion of cell cycle progression [19]. Of interest, let-7 directly down-regulates DICER1 expression, suggesting that let-7 may regulate the global production of microRNAs [20]. MicroRNAs are initially transcribed by RNA polymerase II as primary-microRNAs with hairpin structures. DROSHA/DGCR8 enzyme complex then cleaves primary-microRNAs into precursor-microRNAs, which are transported to cytoplasm by Exportin 5, and cleaved by DICER to form microRNA duplexes. One strand is selected to function as a mature microRNA and loaded into the RNA-induced silencing complex (RISC), whereas the partner microRNA* is preferentially degraded. The mature microRNA with RISC binds to 3 1 UTR of target mRNA resulting in translational repression or degradation.

MicroRNAs as Tumor Suppressor Genes and Oncogenes
MicroRNAs play important regulatory roles in carcinogenesis not only as tumor suppressor genes but also as oncogenes. Several microRNAs are dysregulated in lung cancer. Table 1 shows principal microRNAs involved in the development or progression of lung cancer, their gene targets, and associated biological processes.

Let-7 Family
The let-7 family was the first identified microRNA in humans [17]. In lung cancer, let-7 has been shown to inhibit the expression of oncogenes involved in cellular proliferation, such as RAS, MYC, and HMGA2 [18,19]. Let-7 also inhibits the expression of CDK6, and the reduced expression of let-7, therefore, leads to the promotion of cell cycle progression [19]. Of interest, let-7 directly down-regulates DICER1 expression, suggesting that let-7 may regulate the global production of microRNAs [20].

miR-221/222
miR-221 and miR-222 are involved in the development and progression of lung cancer by targeting PTEN and TIMP3 tumor suppressor genes [34,35]. Overexpressed miR-221/222 inhibits apoptosis and promotes cell migration by down-regulating PTEN and TIMP3.

Diagnostic microRNAs in Tissues
Early detection of lung cancer is prerequisite to reduce lung cancer mortality, because lung cancers are often diagnosed at advanced stages, where clinical treatments are less (or least) effective. MicroRNA expression signatures of lung cancer have been reported by numerous studies, however the reported microRNA profiles were not so consistent. Vosa et al. performed a comprehensive meta-analysis of 20 published microRNA expression studies in lung cancer, including a total of 598 tumor and 528 normal lung tissues [36]. Using a recently published robust rank aggregation method, they identified a statistically significant microRNA meta-signature of seven up-regulated (miR-21, miR-210, miR-182, miR-31, miR-200b, miR-205, and miR-183) and eight down-regulated (miR-126-3p, miR-30a, miR-30d, miR-486-5p, miR-451a, miR-126-5p, miR-143, and miR-145) microRNAs.

Diagnostic microRNAs in Body Fluids
Importantly, microRNAs are present not only in tissues but also in body fluids, such as blood, plasma, serum, or sputum. By examining body fluids, we may be able to distinguish lung cancer patients from healthy individuals. Jiang's group conducted several studies to assess the usefulness of microRNAs in body fluids for the lung cancer screening [37][38][39][40]. Using sputum specimens, they examined the expression of miR-21, which is an overexpressed microRNA in lung cancer [37]. The expression of miR-21 in sputum was higher in NSCLC patients, and the detection of miR-21 expression produced 70% sensitivity and 100% specificity in the distinction between 23 NSCLC patients and 17 cancer-free individuals. In another study, using an independent set, they demonstrated that the expression profile of four sputum microRNAs (miR-21, miR-486, miR-375, and miR-200b) demonstrated 81% sensitivity and 92% specificity in the distinction between 64 NSCLC patients and 58 healthy individuals [40]. They also assessed the usefulness for plasma microRNAs as potential biomarkers for NSCLC [39]. They showed that the expression profile of four plasma microRNAs (miR-21, miR-126, miR-210, and miR-486-5p) yielded 86% sensitivity and 97% specificity in distinguishing 58 NSCLC patients from 29 healthy individuals. Furthermore, this panel of microRNAs produced 73% sensitivity and 97% specificity in identifying stage I NSCLC patients, suggesting the usefulness of plasma microRNAs as potential biomarkers to identify even early stage NSCLC patients [39]. In another study, they showed that the expression profile of three plasma microRNAs (miR-21, miR-210, and miR-486-5p) produced 75% sensitivity and 85% specificity in the distinction between 32 patients with malignant solitary pulmonary nodules (SPNs) and 33 individuals with benign SPNs. SPNs have been increasingly diagnosed with the advancement and widespread use of computed tomography (CT) scan. The combination of microRNA testing and CT scan may serve as a minimally invasive method of diagnosing individuals with SPNs. All of these four studies by Jiang's group [37][38][39][40] included miR-21 as a biomarker to distinguish lung cancer patients from healthy individuals. Because miR-21 is overexpressed in various types of cancer, further studies are warranted to examine miR-21 expression for the distinction between lung cancer patients and patients of the other cancers.

MicroRNAs as Biomarkers for Histological Classification
Recent advances in the treatment of NSCLC with new drugs require an appropriate histological subtyping at diagnosis to avoid hazardous side effects. For instance, bevacizumab (brand name: Avastin), a monoclonal antibody which blocks angiogenesis by inhibiting vascular endothelial growth factor A (VEGFA), cannot be used for SqCC patients due to serious hemorrhagic complications. Similarly, pemetrexed (brand name: Alimta), a chemotherapy drug belonging to a class of chemotherapeutic drugs known as folate antimetabolites, cannot be used for SqCC patients due to adverse responses. Several studies have been conducted to distinguish between SqCC and non-SqCC NSCLC, utilizing microRNAs profiles [41][42][43]. Lebanony et al. found that higher expression of miR-205 was specific to SqCC in a test set. In an independent validation set, miR-205 expression in FFPE samples yielded 96% sensitivity and 90% specificity in the distinction between SqCCs and non-SqCC NSCLCs [41]. As a replication study, Bishop et al. showed that the measurement of miR-205 expression in small biopsies/aspirates can distinguish between SqCCs and non-SqCC NSCLCs [42]. Recently, Hamamoto et al. have reported that the expression profile of three microRNAs (miR-205, miR-196b, and miR-375) distinguished between SqCCs and adenocarcinomas with 85% sensitivity and 83% specificity in a validation set [43].
SCLC, one of the neuroendocrine tumors, shows increased expression of ASCL1, which is a master gene of neuroendocrine differentiation. Nishikawa et al. demonstrated that miR-357 expression was induced by ASCL1 in lung neuroendocrine carcinoma [44]. They showed that the increased expression of miR-375 was prerequisite for the neuroendocrine differentiation by ASCL1, and suggested that miR-375 might reduce the YAP1-related proliferative arrest by inhibiting YAP1.

Investigation for Prognostic microRNAs
As is the case with mRNA, microRNA profiles have been investigated as potential prognostic biomarkers. In 2004, Takamizawa et al. focused on microRNA let-7 [45], and reported that let-7 expression is lower in lung cancer than in normal lung tissue, and that the lower expression of let-7 in lung cancer was associated with poor prognosis. In addition, overexpression of let-7 in A549 lung adenocarcinoma cell line inhibited cell growth. This study [45] is important with respect to the first report of reduced expression of let-7 and the potential clinical and biological effects of such a microRNA alteration in lung cancer. We examined let-7 expressions in adenocarcinoma in situ (AIS, the new name for BAC (bronchioloalveolar carcinoma) [46]) as well as in invasive adenocarcinoma to investigate the association of let-7 expression with the progression of lung adenocarcinoma [47]. Of note, even in AIS, the expression of let-7 was reduced in comparison to matched normal lung tissue, suggesting that let-7 expression was reduced in the early stage of lung carcinogenesis. AIS is categorized into two subtypes, non-mucinous and mucinous AIS. Interestingly, the expression of let-7 was lower in mucinous AIS than non-mucinous AIS. The differential expression of let-7 between two morphological subtypes of AIS suggests an association between microRNA expression and morphology (even in the same category of AIS). According to this observation, complexity of microRNA expressions depending on morphology as well as cancer progression or driver mutations can be presumed.
In 2006, the first microRNA profiling in lung cancer was reported by Yanaihara et al. [48]. Using 104 NSCLCs (65 adenocarcinomas and 39 SqCCs) and matched normal lung tissues, they conducted profiling analyses of microRNA expression by microarray. They identified the higher expression of miR-155 and the lower expression of let-7a-2 as biomarkers of poor prognosis for NSCLC patients, and confirmed the validity using an independent set by real-time RT-PCR. Subsequently, in 2008, Yu et al. [49] identified a five-microRNA signature (let-7a, miR-221, miR-137, miR-372, and miR-182*) that classified a training set of 56 NSCLC cases into good-prognosis and poor-prognosis group. Then, they confirmed that the five-microRNA signature divided 62 NSCLC testing cases into two groups with good and poor prognosis. Furthermore, they validated the five-microRNAs signature as a prognostic classifier using an independent cohort set of 62 NSCLCs. This five-microRNA signature worked as a prognostic classifier for NSCLCs even with the same stage or SqCCs as well as adenocarcinomas.
Importantly, abundant undegradaded microRNAs circulate in body fluids, such as blood, plasma, serum, and sputum [8]. There have been several studies to predict survival in NSCLC patients, by using microRNA profiles in body fluids. In 2010, Hu et al. reported that a four-serum-microRNA signature (miR-486, miR-30d, miR-1, and miR-499) was associated with overall survival in NSCLC, utilizing 120 cases for the training set and 123 cases for the testing set [52]. Subsequently, Wang et al. conducted pathway-based serum microRNA profiling and examined the prognostic association in advanced stage NSCLC patients [53]. They focused on microRNAs involved in TGF-β pathway, which plays crucial roles in control of cell proliferation, differentiation, apoptosis, and invasion. They identified 17 microRNAs which were significantly associated with two-year patient survival. Of these 17 microRNAs, miR-16 exhibited the most statistically significant association; high expression of miR-16 was associated with better survival (HR = 0.4, 95% confidence interval (CI), 0.3-0.5). They created 17-microRNA risk score to identify patients at the highest risk of death. NSCLC patients with a high-risk score had a 2.5-fold increased risk of death compared with those with a low-risk score (95% CI, 1.8-3.4; p = 1.1ˆ10´7).
6.3. Let-7, DICER1, and Survival of Lung Cancer DICER1 has an important role in microRNA biogenesis, converting precursor-microRNAs into mature microRNAs ( Figure 1). Inversely, microRNA let-7 directly down-regulates DICER1 expression [20]. According to the study by Karube et al. [54], the reduced expression of DICER1 was associated with poor prognosis in NSCLC, suggesting the involvement of reduced DICER1 expression in the progression of lung cancer.
There are several studies examining the association between microRNAs expression and driver mutations (especially EGFR and KRAS mutation). miR-21 has been suggested to be an EGFR-regulated anti-apoptotic factor in never-smokers' lung adenocarcinoma [68]. Recently, Li et al. have reported an association of miR-21 overexpression with acquired resistance of EGFR-TKI in NSCLCs [69]. Dacic et al. [70] examined the microRNA expressions in lung adenocarcinomas with different driver mutations, focusing on EGFR and KRAS mutation, using two EGFR-mutant, two KRAS-mutant, and two EGFR-wild-type/KRAS-wild-type adenocarcinomas and confirmed differentially expressed microRNAs in a validation set of 18 adenocarcinomas. They showed that miR-155, miR-25, and miR-495 were up-regulated only in the EGFR-wild-type/KRAS-wild-type, EGFR-mutant, and KRAS-mutant lung adenocarcinomas, respectively. In 2014, Bjaanaes et al. also examined microRNA profiles according to mutation status of EGFR and KRAS, using 154 lung adenocarcinomas by microarray [71]. The identified profiles were confirmed by real-time RT-PCR, using 103 lung cancer cases. They identified 17 microRNAs that were differentially expressed between EGFR-mutant and EGFR-wild-type adenocarcinomas, and three microRNAs differentially expressed between KRAS-mutant and KRAS-wild-type adenocarcinomas. There was no overlap between Dacic's and Bjaanaes' study, possibly due to the small sample size of Dacic's study, differences in methodologies, or by chance. Further researches with a large sample size are required to elucidate the specific microRNA profiles according to the mutation status of EGFR or KRAS.
Weiss et al. [72] reported that miR-128b could directly regulate EGFR. The loss of heterozygosity (LOH) of miR-128b occurred frequently in NSCLC and correlated significantly with clinical response and survival after EGFR-TKI. Another study showed that microRNA-146a targeted EGFR, suppressed its downstream signaling and regulation of cell growth, and enhanced the cytotoxic effect of EGFR-TKI in five NSCLC cell lines [73].
To understand the role of microRNAs in TKI-resistant NSCLCs, Garofalo et al. examined microRNA changes that were mediated by tyrosine kinase receptors [74]. They demonstrated that this resistance could be overcome by anti-miR-221/222 and anti-miR-30c, which recovered the expression of the pro-apoptotic protein BCL2L11 (BIM) and increased the gefitinib sensitivity of NSCLC both in vitro and in vivo.
Other aspects regarding the role of microRNAs in relation to EGFR-TKI deserve comment. Recently, plasma microRNA profiles (miR-21, miR-27a, and miR-218) have been identified for primary resistance to EGFR-TKIs in advanced NSCLCs with EGFR activating mutation [75]. Pak et al. have found unique microRNAs (miR-34c, miR-183, and miR-210) in lung adenocarcinoma groups according to major TKI sensitive EGFR mutation status [76]. A recent study by Wang et al. suggests that the modulation of specific microRNAs (miR-374a and miR-548b) may provide a therapeutic target to treat or reverse gefitinib resistance in NSCLC with high expression of the Axl kinase [77]. The review by Ricciuti et al. [78] is worth reading because it summarized the existing relationship between microRNAs and resistance to EGFR-TKIs, and also focusing on the possible clinical applications of microRNAs in reverting and overcoming such resistance.
For now, limited evidence can be available for the microRNA profiles of lung cancer with specific driver mutations. Further studies should be conducted to elucidate driver-mutation-specific microRNA profiles. MicroRNA signature may be used as a diagnostic biomarker of lung cancer with a specific driver mutation, and also the combination of TKI and microRNA-based treatment is promising.

Prospects from Basic to Clinical Application of microRNAs
Actually, despite enormous efforts, several obstacles remain to be solved for the transition of microRNAs from basic to clinical application as novel biomarkers or potential tools for treatment. These obstacles include the standardization of microRNA detection, improved understanding of how microRNAs interact with other components of the genome, and the development of non-toxic targeted delivery of microRNAs to the lung or metastatic lesions as well as the selection of the proper vehicle for delivery [79]. Therefore, we have a bunch of obstacles to clear away, however we are now trying to solve these issues strenuously [79].

Conclusions
In this review, we introduced recent works of microRNAs involved in the development and progression of lung cancer, with a particular interest in microRNAs as novel biomarkers and potential tools for treatment. Of importance, microRNAs are far less degradated in FFPE samples than mRNAs, and therefore, an easy availability of FFPE samples in hospitals enables us to measure accurate microRNA expressions. In addition, microRNAs are also present in body fluids, making microRNAs promising diagnostic biomarkers. Therefore, there exists a great potential in microRNA analyses for cancer research. Further studies are demanded in order to use microRNA profiles as diagnostic markers and conduct microRNA-based therapies in clinical practice.