Let-7 as biomarker, prognostic indicator, and therapy for precision medicine in cancer

During carcinogenesis, cells acquire capabilities termed the hallmarks of cancer [1]. Abnormal microRNA (miRNA) regulation has been attributed to all phases of cancer and affects several of the cancer hallmarks [2, 3]. Discovered in C. elegans, let-7 (lethal-7) miRNA family functions as an important regulator of differentiation [4, 5]. In mammals, let-7 is known as the keeper of differentiation, and its abnormal regulation and expression has been associated with cancer initiation and progression [6]. The functions of all members are generally thought to be overlapping because of sequence similarity [7]. Figure 1 includes a diagram of let-7 structure with seed sequence highlighted. Because let-7 targets several oncogenes, its repression in cancer is most often associated with poor patient prognosis [8]. The human genome contains 13 let-7 family members encoding 9 mature miRNAs. Let-7a1, a2, a3 are encoded from different transcripts, producing identical mature sequence; the same is true for let-7f1, f2. With the exception of let-7i and let-7g, which are encoded individually, transcripts of different let-7 members are located in clusters along with other miRNAs [9‐12]. Due to different genomic loci, transcriptional regulation varies between individual let-7 family members. In this review, we discuss let-7′s involvement in patient survival, focusing on its function as diagnostic, prognostic and therapeutic, in isolation as well as in combination with current therapy regimens. We review let-7 effects on cellular phenotype, and explain it by molecular mechanisms. We also discuss instances of let-7 oncogenic functions, differences between regulation and function of different let-7 family members, and its importance for understanding its effect in cancer biology and its therapeutic potential.

The pleiotropic effects of let-7 include repression of oncogenes, suppression of epithelial-to-mesenchymal transition, induction of chemosensitivity, controlling cell signaling pathways, and decreasing cellular proliferation.

Let-7 effect on cancer observed in clinical, in vivo, and in vitro studies can be explained by several functional aspects. One way let-7 acts as a tumor suppressor is via repression of oncogenes resulting in a decrease in stemness [60, 74]. Let-7 levels inversely correlate with percentage of cancer stem cells (CSC), and its overexpression reduces CSC markers nestin and CD133 in glioblastoma and ALDH1 in breast cancer [40, 73]. To determine the presence of cancer stem cells functionally, spheroid (mammosphere in breast cancer) formation and colony formation assays are used. Spheroids are enriched for tumor initiating cells and have lower let-7 levels, and in mammospheres that are allowed to differentiate, let-7 levels increase [52]. Let-7 over-expression inhibits stemness, resulting in reduced sphere formation [40, 52, 60, 64, 73]. Cancer cells with a stem-like phenotype are also able to form colonies, measured as clonogenicity. Up-regulation of let-7 results in decreased clonogenicity [47, 58, 59, 75].

Let-7 targets oncogenes and genes important for tumor initiation and progression including Myc, RAS, E2F1, E2F5, LIN28, ARID3B, PBX3, HMGA2 and long non-coding RNA H19 [42, 59, 70, 76]. Silencing these genes causes the functional tumor suppressive effects mediated by let-7. LIN28A is a well-known pluripotency marker that is present in embryonic stem cells (ESCs) and decreases upon differentiation [77]. In prostate cancer, LIN28 increases aggressiveness and results in increased tumor burden in vivo [78]. ARID3B and HMGA2 transcriptionally activate OCT-4 and SOX2, respectively, both of which are pluripotency factors highly expressed in ESCs [39, 67, 69, 70, 79‐81]. Repression of H19 results in methylation of promoters of several other genes due to up-regulation of DNMT3b [37, 42, 48, 72, 73, 81, 82]. PBX3 is an oncogene that induces epithelial-to-mesenchymal transition and promotes invasiveness and metastasis of gastric cancer [42, 60, 70, 71, 83]. Thus, let-7 can repress the function of a number of factors that can be recruited in oncogenesis. These examples illustrate the specific effects of let-7 demonstrated to result in functional changes relevant to cancer.

Besides repression of oncogenes, let-7 also plays a role in controlling cell signaling pathways. Overexpression of the let-7 family member miR-98 results in reduced phosphorylation and down-regulation of Akt and Erk, which have been implicated in carcinogenesis [42, 59, 64, 72, 84]. In Ewing’s sarcoma, let-7 directly represses signal transducer and activator of transcription 3 (STAT3) and results in a less aggressive cancer phenotype [85]. The STAT3 pathway regulates genes related to cell cycle and cell survival and is often linked to cancer progression. STAT3 activity correlates with chemo- and radioresistance and poor survival [86]. In breast cancer, let-7 targets estrogen receptors, which activate WNT signaling and promote stemness and cancer aggressiveness [40]. Let-7 down-regulates WNT signalling activity by targeting estrogen receptors in breast cancer and TCF-4 (a transcription factor downstream of WNT) in hepatocellular carcinoma. WNT pathway is a major regulator of cell proliferation, differentiation, and migration, and has been shown to promote tumor growth and contribute to cancer stem cell phenotype [60, 64, 87]. Cumulatively, let-7′s effects on cell signaling pathways impede the aggressive phenotype.

Another way that let-7 exerts tumor suppressive effects is via inhibition of epithelial-to-mesenchymal transition (EMT). EMT is a normal process during embryonic development as well as wound healing. Cancer development and metastasis are associated with abnormal occurrence of EMT in somatic cells. During EMT, epithelial cells gain the ability to invade and metastasize [88]. Reduced let-7 levels correlate with an increase in EMT markers Twist, Snail, vimentin, and N-cadherin, resulting in increased cancer aggressiveness, as assessed by spheroid formation, migration, invasion, mesenchymal appearance, and resistance to chemotherapy [44, 62, 72]. Over-expression of let-7 reduces expression of Snail and N-cadherin, while increasing E-cadherin; these effects are proposed to be via HMGA2 [42, 59].

Let-7 induction of chemosensitivity seen in vitro and in vivo is due to inhibition of LIN28A/B, STAT3, E2F1, IMP1 and chemoresistance genes MDR1, ABCG2, and MMP9 [33, 37, 62, 63, 65, 89]. In EOC, let-7 down-regulates BRCA1, RAD51, PARP, and IGF1, resulting in increased sensitivity to cisplatin, and longer progression free survival and overall survival [34, 59]. BRCA1, RAD51, PARP, and IGF1 proteins contribute to DNA double strand break repair, which is induced by cisplatin. Inhibiting those enzymes decreases the ability of cancer cells to survive [34, 59, 90]. Nanoparticle delivery of let-7 together with paclitaxel results in an increase in sensitivity, resulting in apoptosis [59]. In blood cancers, up-regulation of the let-7 family member miR-98 results in increase of BAX and p21 in acute myeloid leukemia and increased sensitivity to adriamycin [37]. Figure 2 represents overall let-7 tumor suppressive function.

Let-7 decreases the cellular proliferation rate due to a decreased proportion of cells in S phase of the cell cycle [56, 64, 75, 76, 91, 92]. Let-7 also represses negative regulators of histone H2b monoubiquitylation (H2Bub1). H2Bub1 loss correlates with cancer progression and poor prognosis while up-regulation causes a decrease in number of breast cancer cells in S phase and cell migration [93]. Inhibition of cancer cell growth by let-7 is also due to increased apoptosis via up-regulation of bak and bax and reduced bcl-xL [37, 42, 58, 60, 67]. Table 4 summarizes targets of let-7 family members stating which have been reporter assay-validated.

Unexpectedly, let-7 can also have detrimental effects. Even though let-7 has been demonstrated to have tumor suppressive effects in various cancer types, emerging data suggest that, counterintuitively, in some cases let-7 may act as an oncogene. Several groups have demonstrated that the let-7a3 locus is highly methylated in normal tissues, but hypomethylated in lung and ovarian tumors, with higher expression of mature let-7a in cancers [94, 95]. Over expression of let-7a3 in lung cancer cells results in increased aggressiveness of cells, assessed via anchorage independent assay and increase in gene expression associated with cell proliferation, as well as down-regulation of genes associated with adhesion, relevant to tumor progression and metastasis [95]. Higher let-7a3, let-7b, and let-7c levels in ovarian and hepatic cancers are correlated with poor prognosis and decreased overall survival [43, 94, 96]. Ma et al. demonstrated that let-7e is increased in and positively affects migration and invasion of esophageal squamous cell carcinoma cells, possibly via targeting ARID3a [97]. Since ARID3a negatively correlates with pluripotency, decreasing it could contribute to stemness [98]. Let-7f and let-7e have been shown to be upregulated in tongue squamous carcinoma, and let-7c, let-7d, and let-7f are upregulated in aggressive relative to non-aggressive tumors [99]. Mir-98 has been shown to increase chemoresistance via indirect repression of mir-152 by targeting Dicer1. Mir-152 controls RAD51 expression, contributing to the poor prognosis of EOC patients with increased levels of mir-98 [100]. In certain in vitro conditions such as starvation, let-7 paradoxically induces expression of HMGA2 [101]. All of these indicate the complexity of the relationship between let-7 and cancer cell aggressiveness, and illustrate the fact that the actions of any miRNA are context dependent. The set of genes expressed in a particular cell determines the available let-7 targets. Thus, it is important that let-7 overexpression treatment strategies be tailored towards individualized clinical scenarios based on specific miRNA expression profiles, as opposed to overarching treatment schemas spanning across multiple malignancy types.

Tumor microenvironment and stroma are also important to consider when developing new therapies. Baer et al. demonstrated that increased let-7 expression in tumor associated macrophages (TAMs) results in conversion into the M2 phenotype. While tumor infiltration by TAMs with M1 phenotype have pro-inflammatory activity and better prognosis, the M2 phenotype is associated with increased angiogenesis and increased tumor burden [102]. Let-7 delivery as a therapeutic regimen therefore has to be specific to cancer cells due to its oncogenic functions in tumor immune cells. Even though a few studies demonstrated let-7 as having oncogenic functions and correlating with poor prognosis, the vast majority of evidence suggests otherwise. Therefore, let-7 remains a potential therapeutic target.

In this review, we emphasize the importance of miRNA let-7 in cancer. We focus on the potential to use let-7 in precision medicine for screening and diagnosis of cancer, for its prognostic value, and as a therapeutic agent. We review the complex regulation and function of the let-7 family members, and focus on their abnormal regulation in cancer, which leads to abnormal and/or loss of function. let-7 miRNAs have been referred to as tumor suppressors, but it is important to consider that there is evidence to support their oncogenic functions in vitro and in clinical subjects. Our goal is to demonstrate the importance of let-7 during treatment decisions for chemo- and radiotherapy, to enable its use as precision medicine, and to deliver optimal results for patients.

Let-7 remains a promising cancer therapy and warrants more research; but even before all details of its therapeutic use are worked out, tumor let-7 levels can be used to choose the best therapy options for each individual. Low or high tumor let-7 levels can point to the most effective therapy regimens, and its levels in bodily fluids show potential for use as an aid to diagnosis, therapy monitoring, and prognosis.

Many questions remain unanswered. Knowledge of levels of all let-7 family members in each type of cancer can provide a more precise overview of its regulation, and provide more specific diagnostic/prognostic tools. Functional studies may reveal that upregulation of a specific let-7 member offers the most beneficial effect as a therapeutic regimen. Combination of standard therapy with let-7 over-expression has to be well studied in order to avoid toxicity and unwanted interactions. More in vivo models are needed to develop let-7 into a safe and effective therapy regimen that will provide the rationale for clinical trials.