Introduction

MicroRNAs (miRNAs) are a class of non-coding RNAs that function as endogenous triggers of the RNA interference pathway. In general, aberrant expression of microRNAs has been linked to all human cancers. miRNAs are dysregulated in a plethora of diseases, including cancers; poorly differentiated tumors displayed lower miRNA expression compared to tumors with a higher differentiation. The Slack laboratory showed one of the first pieces of evidence that lethal-7 (let-7) plays a key role in cancer. By showing that RAS is regulated by the let-7 miRNA family in Caenorhabditis elegans and by highlighting let-7 complementary sites in human RAS 3′UTRs, they first validated that the let-7 miRNA family negatively regulates RAS in two different C. elegans tissues and human cell lines and lung tissue. This was the first report of the mechanism of action of a miRNA acting as a tumor suppressor [1, 2].

The heterochronic miRNA let-7 display tumor suppressive properties at least in lung, colon, breast, and leukemia cancers. It is categorized as a tumor suppressor because it reduces cancer aggressiveness, chemoresistance, and radio-resistance. Particularly, studies inferred from human cancer databases report let-7 family member expression associated with poor overall survival in some cancers; but regretfully with unsignificant results in glioblastoma multiforme (GBM) [3].

By contrast, many in vitro studies report that let-7 inhibits the malignant behavior of glioma cells and stem-like cells [4,5,6,7,8,9,10,11,12,13] (Table 1). Given the promiscuous and context-specific nature of miRNA targeting, many mechanisms of interactions remain to be elucidated. Moreover, regulation of RAS protein level and RAS/MAPK cascade are regulated by a myriad of miRNAs family without a clear mechanistic link [11, 12, 14,15,16]. Three recent reports focused on miRNAs targeting RAS in GBM and showed that miR-143-3p directly targets NRAS [17], let-7a-5p directly targets KRAS [7], and both R-Ras and N-Ras (related RAS viral oncogene homolog, HRAS homolog) are direct targets of miR-124-3p [18].

Table 1 let-7 target genes in glioma cancer
Full size table

let-7 is a bona fide tumor suppressor gene, but this categorization is rarely straightforward, since some miRNA can have both oncogenic and tumor-suppressive mechanisms, depending on the context. Let-7 targets multiple oncogenes (RAS, c-MYC, HMGA2 to date) and the prominent mechanisms by which let-7 exerts a tumor suppressive role is by repressing the translation of the three RAS proteins (HRAS, NRAS, and KRAS) and c-MYC, a downstream effector of RAS-ERK signaling [1, 23,24,25,26,27]. Moreover, oncogenic functions for let-7 have been also reported in some cases [28, 29]. Indeed, the functions of let-7 have been reported not only in cancer but also in other diseases, including viral infection, immune diseases, neurological diseases, diabetes, and cardiovascular diseases. In vivo studies with Cre-inducible let-7-transgenic mice have reported a strong phenotypic read-out both in diabetic retinopathy and in diabetes [30, 31].

As a proto-oncogene that regulates several oncogenic pathways, KRAS has always been considered a central signaling modulator. Thus, there is a flourishing literature on studying miRNAs that regulate K-RAS expression, with let-7 being of prime importance. Our knowledge of how miRNAs can modulate activation of the RAS-ERK signalling pathway continues to grow as potential miRNA-mRNA regulatory networks are identified using different strategies [32]. From these studies, three main paradigms of miRNA-mediated RAS-ERK regulation have emerged. miRNAs can affect the translation of (i) core components of the RAS-ERK pathway (e.g., let-7 targets HRAS, NRAS and KRAS) [1, 33], (ii) critical proteins that regulate the pathway and are required for proper spatial and temporal control of RAS-ERK signalling [29, 34, 35], and (iii) upstream drivers and downstream effector/regulator molecules [36, 37]. A myriad of miRNAs regulates RAS-ERK pathway activity in a variety of cancer contexts [38, 39].

In general, there is a very clear correlation between the loss of let-7 expression and the development of poorly differentiated and aggressive cancers, this is the case for let-7a in lung carcinomas. In prototypic studies in non-small cell lung cancer (NSCLC) models, let-7 expression was analysed in vitro and in vivo and its role in KRAS-mediated NSCLC tumorigenesis was demonstrated [40]. let-7 inhibits tumour development and the RAS-ERK signalling pathway in an autochthonous model of NSCLC driven by activated KRAS (KRASG12D) [41, 42]. In xenograft models of NSCLC, let-7 exerts a tumour suppressive role [24] and increased expression of let-7a significantly reduces tumour burden in a K-Ras lung cancer model in mice [43]. Furthermore, let-7 counteracts the maintenance, survival and self-renewal of cancer stem-like cells (CSCs) in ovarian and breast cancer, and this suppressive activity correlated with reduced expression of RAS and HMGA2 [43,44,45,, 45, 46, 47]. Thus, by suppressing RAS expression, let-7 can attenuate RAF-MEK-ERK signalling and its dependent oncogenic phenotypes independent of RAS mutation status. Functional studies have confirmed that miRNA let-7 dysregulation is causative in many cancer cases, highlighting its potential impact on RAS-ERK signalling from a mechanism perspective.

Main text

Let-7 expression in brain tumors

Brain tumours comprise a broad spectrum of over 120 histologically, demographically, clinically and molecularly different diseases (overview in [48]). Glioblastoma is the most common malignant primary brain tumour in adults and remains incurable [49]. Standard treatment for most brain tumours remains focused on maximal surgical resection, radiotherapy and chemotherapy with temozolomide (TMZ) as first-line therapy. Indirect targeting of the tumour by anti-angiogenics (e.g. bevacizumab) and immunotherapies (vaccines, adoptive therapies, immune checkpoint inhibitors and oncolytic viruses) have shown mixed efficacy (or inactivity) in preclinical studies [5048,49,50,51,, 51, 52, 53, 54]. Alternative studies focusing on molecular profiling of GBM identified neurofibromin-1 (NF1) as targeting mutations that contribute to activated KRAS signalling [55, 56].

The members of the let-7 family are direct and strong regulators of K-RAS, N-RAS and H-RAS mRNAs through their 3′UTR sequences [15, 57, 58]. A single miRNA can control entire cellular signalling pathways by interacting with a broad spectrum of target genes. For example, let-7 inhibits GBM tumour growth by interacting with a broad spectrum of target genes such as Ras, c-Myc, Stat3, Cyclin D1 [4, 6, 7, 10, 22] (see Table 1). Mature let-7 can be blocked by the LIN28 protein, increased levels of which are associated with poorer survival in gliomas, and let-7b can serve as a marker for chemoresistance [20, 21]. In irradiated human glioblastoma cells, let-7 mediated resistance to radiotherapy by regulating its relative expression [19]. Remarkably, the resulting loss of let-7 enhances the expression of oncogenic targets such as RAS in loss-of-function and gain-of-function experiments [36].

The detection of miRNA has rapidly emerged as a potential biomarker in patients with glioblastoma [59, 60]. In particular, decreased let-7b has recently been associated with poor prognosis in gliomas [61]. Based on several high-throughput genomic technologies, The cancer genome atlas (TCGA) has defined RAS/MAPK as one of the central pathways involved in GBM [62] and let-7 miRNA family expression levels are not reduced in GBM (summarised by [11, 12]). Regardless of its expression level, let-7 miRNA can impair glioblastoma growth and cell migration through RAS inhibition [7, 10]. Specifically, forced expression of let-7 miRNA reduced the expression of pan-RAS, N-RAS and K-RAS, thereby reducing proliferation and migration as well as tumour size in xenograft-transplanted GBM in nude mice [10]. In another study, overexpressed let-7a inhibited glioma cell malignancy by directly targeting KRAS independently of PTEN [7], and let-7b in turn inhibits malignant behaviour (proliferation, migration and invasion) of glioma cells and stem-like glioma cells [6]. Indeed, focal deletion of members of the let-7 family (let-7a-2 and let-7e) has been found in medulloblastoma (MB) [63] and the let-7 family has been validated in spontaneous and radiation-induced MB [64]. Conversely, miR-let7g was found to be upregulated in anaplastic and differentially expressed in desmoplastic MB [65, 66], although the functional consequences of its dysregulation have not yet been investigated. Recently, let-7 miRNA activity has been identified as a prognostic biomarker of SHH medulloblastoma [67]. Furthermore, in a paediatric brain tumour (Diffuse intrinsic pontine gliomas, DIPG), RAS signalling has recently been identified as a novel therapeutic vulnerability and the RNA-binding protein LIN28B is overexpressed in DMG and suppresses the let-7 family of microRNAs [68, 69]. Numerous oncogenes and signalling pathways besides RAS (e.g. MYC) have been shown to be targets of let‐7 miRNAs, and KRAS is the target of many other miRNAs in gliomogenesis [5, 16]. Which of these targets genes determine the overall phenotype in glioblastoma remains to be investigated.

RAS oncogenes in brain tumors

Comprehensive molecular profiling has dramatically changed the diagnostic neuropathology of brain tumours. Several of the key molecular alterations that are critical for glioma classification involve epigenetic dysregulation at a fundamental level and involve areas of biology not previously thought to play an important role in glioma pathogenesis [48]. The biological functions of the RAS family [the viral oncogene homologue of Harvey rat sarcoma (HRAS), the viral oncogene homologue of Kirsten rat sarcoma (KRAS) and the viral oncogene homologue of neuroblastoma (NRAS)] have been studied in detail for decades. Although only 1% of GBM tumours exhibit RAS mutation or amplification, 10% of GBM tumours contain inactivating genetic alterations of neurofibromin 1 (NF1) that lead to hyperactive RAS activity by enhancing intrinsic GTPase activity [62, 70]. Remarkably, dysfunctional signalling in tumours arises not only from gene mutations but also from epigenetic changes or pathway rewiring, which probably explains why there appear to be no dominant driver mutations in certain tumour types, most notably glioblastoma. Furthermore, no evidence of oncogenic mutations affecting NRAS, KRAS, HRAS, BRAF or PDGFR was found in medulloblastoma [71].

Deregulated RAS signalling is an important step in carcinogenesis, with activating RAS mutations playing a role in 30% of all cancers [72]. In contrast to many human tumours, RAS mutation is not common in human gliomas, with some exceptions such as cerebellar GBM [73]. Hyperactive RAS signalling alone is sufficient to generate gliomas that closely resemble human tumours in glioma mouse models. The RAS signalling pathway is therefore of central importance for human gliomagenesis. Primary GBMs are associated with impaired RAS signalling and expression of the oncogenic HRAS leads to a malignant phenotype in glioma cell lines [74, 75, 76, 77]. In particular, the KRAS oncogene is strongly involved in tumourigenesis in glioblastomas [75, 78, 79, 80], although KRAS mutations are almost absent in malignant gliomas [81]. Therefore, the observed deregulation of the Ras-RAF-ERK signalling pathway in gliomas is generally attributed to its upstream positive regulators, including EGFR and PDGFR, which are known to be highly active in many malignant gliomas [70]. It is likely that mechanisms other than mutations contribute to the activation of the RAS-MAPK signalling pathway in wild-type cancer. Indeed, epigenetic modifications have been described to enhance this activation in human tumours [82], and dysregulation of physiological miRNA activity has been shown to play an important role in gliomagenesis, but the functional significance of this regulatory level is currently unknown [11, 12, 75,76,77,78,, 84, 85, 86, 87].

Therapeutic and diagnostic potential of mirnas

Clinical studies on the benefits of miRNAs in diagnosis and therapy are already underway [88, 89]. MiRNAs are abundant non-coding RNAs, and their short length increases their stability compared to longer RNA molecules. In addition, miRNAs are secreted alone into the extracellular fluid or encapsulated by vesicles such as microvesicles and exosomes. Subsequently, the secreted miRNAs are found in the circulation. In summary, cancer-specific and circulating miRNAs are attractive diagnostic markers. Besides diagnostic miRNAs, miRNAs that can be used to predict drug efficacy and patient prognosis will greatly assist the advancement of precision cancer medicine. Despite the increasing interest in miRNA therapies as potential approaches for cancer treatment, the actual deployment of these molecules into solid tumors remains a formidable obstacle [89]. Stabilty of let-7 mimics for cancer therapy have been improved [90], but delivery issues remain [91].

Therapeutic anti-miRs are currently being developed for cancer therapy, such as miR-155 for treating leukemia and lymphoma [92]. Several clinical trials of improved miRNA drug strategies, such as synthetic RNA molecules and advanced delivery technologies, are ongoing despite the failure of the first-in-human clinical trial of miRNA cancer therapy [93]. Thus far, one clinical trial in gliobastoma to assess miR-10b expression in patients with several subtypes of brain cancer is still ongoing (recruiting, NCT01849952). A miR-10b inhibitor, RGLS5579, was developed and preclinically tested for glioblastoma multiforme (GBM) [94, 86,87,, 96, 97].

Since there is currently only one miRNA drug in clinical trials, there is little direct evidence of that miRNAs can be applied with minimal side-effects. In fact, their mechanism of action is tuning expression rather than blunting their targets which reasonably should be less detrimental to healthy tissues. In contrast to the selectivity of enzymatic protein inhibitors, miRNA drugs are developed with the idea of controlling multiple gene-components in the same or overlapping signaling-pathways. Such gene-products are not limited to proteins with enzymatic activity but could include any deregulated genes or proteins in a given disease.

Conclusion

The role of KRAS, when activated through canonical mutations, has been well established in cancer. Research on RAS-driven cancers has focused almost exclusively on RAS coding mutations. However, KRAS without canonical mutations is still largely unexplored. KRAS regulation by miRNA is now emerging as a new layer of regulation [98]. Since the discovery of let-7 miRNA, work is currently in its pre-clinical era. The control of KRAS expression by the let-7 family of miRNAs is well documented. Low expression of let-7 in human cancer correlates with high KRAS expression (at least in lung, breast and colorectal). The control over KRAS expression and activity in glioblastomas suggests that let-7 can regulate RAS activity even without oncogenic mutation and that this may be a general phenomenon related to the interactions between tumour suppressor genes (let-7) and proto-oncogenes (KRAS). Open questions remain to be answered: (i) is the functional output of KRAS signaling essential for promoting tumorigenesis?; (ii) is the suppression of KRAS the dominant signalling effect in glioblastoma?; (iii) what are the general physiological effects of let-7 miRNA dysregulation in brain tumours (if any)?; (iv) are oncogenic inhibitors currently in clinical trials for KRAS-driven cancers suitable for therapeutics in KRAS wild-type tumours? Moreover, a large fraction of miRNAs bind their targets independent of seed match complementarity at nucleotides 2–7 [99], this observation greatly affects our ability to accurately predict miRNA targets using existing algorithms. Besides the 3′UTR, miRNAs have been demonstrated to target the 5′UTR and coding sequence of mRNAs as well as other RNA species such as lncRNAs, pseudogenes, rRNAs and tRNAs [100]. We still have little work on the roles of other miRNAs and ncRNAs in brain tumors. While let-7 shows promise as a therapeutic in brain tumors, we are left with the question of how this therapy will be implemented to help patients.