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

Abstract

Abnormal regulation and expression of microRNAs (miRNAs) has been documented in various diseases including cancer. The miRNA let-7 (MIRLET7) family controls developmental timing and differentiation. Let-7 loss contributes to carcinogenesis via an increase in its target oncogenes and stemness factors. Let-7 targets include genes regulating the cell cycle, cell signaling, and maintenance of differentiation. It is categorized as a tumor suppressor because it reduces cancer aggressiveness, chemoresistance, and radioresistance. However, in rare situations let-7 acts as an oncogene, increasing cancer migration, invasion, chemoresistance, and expression of genes associated with progression and metastasis. Here, we review let-7 function as tumor suppressor and oncogene, considering let-7 as a potential diagnostic and prognostic marker, and a therapeutic target for cancer treatment. We explain the complex regulation and function of different let-7 family members, pointing to abnormal processes involved in carcinogenesis. Let-7 is a promising option to complement conventional cancer therapy, but requires a tumor specific delivery method to avoid toxicity. While let-7 therapy is not yet established, we make the case that assessing its tumor presence is crucial when choosing therapy. Clinical data demonstrate that let-7 can be used as a biomarker for rational precision medicine decisions, resulting in improved patient survival.

Introduction

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.

Fig. 1.

LIN28/TUT4(7) let-7 binding during post-transcriptional processing. Modified from Nam et al. and Faehnle et al. [111, 112]. CSD and ZFD are abbreviations for cold shock domain and zinc finger domain of LIN28 respectively. LIM and CM are abbreviations of LIN28 interactive module and catalytic module of TUT4(7) respectively. Sequence of mature let-7a demonstrates position of let-7 family members in the stem along with highlighted seed region

Let-7 as tumor suppressor

Use as a screening tool

Levels of let-7 family members can serve as biomarkers to assist with cancer diagnosis, and monitoring. Detecting differential let-7 levels in bodily fluids has the potential to allow early detection of cancer using minimally invasive procedures, minimizing risks associated with biopsy. Increased plasma let-7 levels are seen in patients with breast, prostate, colon, renal, liver, gastric, thyroid, and ovarian cancer [13–19]. Elevated urine let-7 levels can also be detected in renal cancer [20]. Some studies have also reported decreased serum let-7 levels in colon, lung, prostate, gastric, ovarian, and breast cancers [21–29]. Patients with colorectal carcinoma have decreased levels of let-7 in stool samples compared to healthy controls, providing a less invasive tool to aid with diagnosis [22]. These studies, with some apparently contradictory results, point out the need for further study, but the use of serum let-7 appears to be a promising biomarker. For each cancer type, results are consistent. Table 1 provides a summary of abnormal let-7 levels in plasma (liquid biopsy) based on the type of cancer. Plasma let-7 levels have the potential to serve as a monitoring system for therapy, and may predict accelerated reproliferation of lung cancer, which would assist in providing personalized treatment options for patients [30]. Let-7 levels are directly influenced by therapy, illustrated by let-7c in acute promyelocytic leukemia: its levels increase in blasts after chemotherapy, then decrease again upon relapse [31]. Chemoresistant epithelial ovarian cancers, lung cancers, and acute myeloid leukemia have reduced let-7 levels relative to chemo-sensitive cells, resulting in non-response to chemotherapy [32–37]. These are examples of ways that monitoring let-7 levels could be used to predict drug response or recurrence. Thus, much more work remains in order to understand the diagnostic value of let-7 levels in blood and urine.

Table 1.

Levels of serum let-7 relative to normal control in patients with different types of cancer

Let-7 expression in liquid biopsy
Cancer	Increased	Decreased
Breast	let-7a [13] let-7b [16, 17] let-7c [17] let-7i [17]	let-7c [28]
Prostate	let-7a [13]	let-7a [24]
Colon	let-7a [13, 14]	let-7a [21, 22] let-7b [21] let-7c [21] let-7f [22] let-7i [21]
Renal	let-7a [13]
Lung		let-7a [23] let-7b [23] let-7c [29] let-7f [26] let-7ia [15]
Gastric	let-7f [19] let-7i [19]	let-7a [25]
Liver	let-7b [15] let-7f [18]
Ovarian	let-7b [134]	let-7f [27] let-7i [135]
Thyroid	let-7e [136]

In these experiments, liquid biopsies (from blood) were sampled

^aAssociated with smoking

Use as a diagnostic tool for therapy selection

Let-7 is repressed in many different types of human cancer. Table 2 summarizes abnormal let-7 expression obtained from patient tumors, obtained by solid biopsy, and cultured cancer cell lines (including cases where let-7 is up-regulated). Mechanisms for loss of let-7 are incompletely understood, however studies in ovarian cancer suggest that let-7 repression is due to genomic deletions and abnormal transcription, rather than loss of processing mechanisms involving Dicer and Drosha [38]. While 31.2% of epithelial ovarian cancers (EOC) demonstrate let-7a3 deletions, only 3.1% had alterations in copy number of let-7i [36, 38]. Abnormal expression of let-7 family members correlates with patient prognosis. Decreased expression of let-7 correlates with aggressive, high-grade tumors, and poor prognosis; accordingly, high let-7 levels are associated with better prognosis and prolonged patient survival [36, 39–46]. Low post-surgical tumor let-7 levels indicate poor prognosis for lung cancer patients, with reduced overall survival [44, 47]. The picture is similar for breast, pancreatic, colorectal, liver, and ovarian cancer (the only exception is let-7b and c in ovarian cancer) [17, 36, 39–43, 45, 46]. Of note, high let-7b levels in high grade serous EOC positively correlate with markers of invasiveness and worse prognosis. Let-7 family members are expressed at lower levels in metastatic sites compared to primary tumor in gastric, breast, liver, and lung cancers [48–51]. In vivo, let-7 over-expression in breast cancer resulted in reduced lung and liver metastasis, while let-7 repression resulted in increased in metastasis [51, 52]. Therefore, tumor let-7 levels correlate with and can be used as a prediction for distal metastasis. Thus, while important exceptions must be noted, loss of let-7 in most cancers closely correlates with poor prognosis.

Table 2.

Levels of let-7 family members relative to normal tissue in different types of human cancer

Let-7 expression in tumors
Let-7	Decreased	Increased
Let-7a	Hepatoblastoma [70] Glioma [67] Ewing sarcoma [85] Gastric [48] Nasopharyngeal [70] Lung [47] Liver [42, 96] Melanoma [75] Endometrial [137] Cervical [107, 128] Prostate [138, 139] Ovarian [43]
Let-7b	Hepatoblastoma [66] Liver [42, 96] Melanoma [75] Prostate [138, 139]	Ovarian [43]
Let-7c	Prostate [58, 139] Acute promyelocitic leukemia [31] Liver [42, 96] Lung [44] Endometrial [137] Prostate [138]	Ovarian [43]
Let-7d	Oral [62] Liver [42] Melanoma [75] Prostate [138] Ovarian [140]	Acute promyelocytic leukemia [31]
Let-7e	Melanoma [75] Endometrial [137] Prostate [138] Ovarian [140]	Tongue [99] Esophageal [97]
Let-7f	Liver [42] Endometrial [137] Prostate [138] Ovarian [43, 140]	Tongue [99]
Let-7 g	HCC [42] Prostate [138] Ovarian [43]
Let-7i	Tongue [99] Ovarian [141] Melanoma [75] Cervical [107]
Mir-98	Glioma [71] Salivary adenoid cystic carcinoma [72] Prostate [138]	Ovarian [100]

In these experiments, tumors were sampled

While let-7 levels in body fluids can possibly assist in diagnosis, let-7 levels in tumors can be used to create a personalized optimal treatment plan including both chemotherapy and radiation. Determining levels of tumor let-7 as well as its targets is expected to be useful to deliver personalized treatment when considering therapy options. Colorectal carcinoma patients with KRAS mutation and with high levels of let-7 can benefit from anti-EGFR therapy, while patients with low levels of let-7 have impaired responses [45]. A study by Lu et al. examined let-7a expression levels and response to chemotherapy in patients treated with platinum-based chemotherapy with or without paclitaxel. The patients in this study were treated between 1991 and 2000 [53]. The platinum/paclitaxel doublet became the first line standard of care in advanced EOC after the publication of results of GOG 111 in 1996 [54]. In the study by Lu and colleagues, patients with ovarian cancer and high let-7a levels have better prognosis than those with low let-7a levels when treated with platinum-based therapy alone, but counterintuitively, high let-7a levels correlate with poor response when platinum is combined with paclitaxel. The reverse was true for patients with low let-7a levels in tumors [53]. These observations lead to the hypothesis that, for patients with high tumor let-7a levels, forgoing paclitaxel results in improved outcomes. Paclitaxel inhibits microtubule polymerization, thus affecting rapidly dividing cells. Let-7 has anti-proliferative functions (described below), providing a possible explanation why patients with high tumor let-7 levels did not respond well to paclitaxel [55]. Therefore, knowledge of tumor let-7a levels is expected to be an important contributor to decisions about chemotherapy. However, this will require careful consideration and further retrospective trials followed by robust clinical trials, as currently doublet platinum based or combination intravenous/intraperitoneal chemotherapy are recommended for advanced ovarian cancer in the front line setting (NCCN Guidelines Ovarian Cancer Version 2.2018 3/14/2018, http://www.nccn.org, accessed 1/6/2019).

However, what is true for ovarian cancer may not apply to other malignancies: breast cancer cell lines that have low let-7 levels respond better to taxol treatment [56]. The discrepancy between results obtained by the studies in ovarian and breast cancer can be attributed to differences in biology of breast and ovarian cancer as well as the study design. While Lu et al. [53] compared clinical data and tumor let-7a levels from ovarian cancer patients that had undergone different courses of treatment, Sun et al. [56] used breast cancer cell lines for in vitro studies. Tumor let-7 levels can also predict response to other therapies. Breast cancer patients with low tumor let-7 levels do not respond to epirubicin; therefore, choosing an alternative therapy may prolong survival [57]. In prostate cancer patients, tumors with decreased let-7c levels are resistant to androgen therapy, and let-7 delivery to tumors provides promising therapy [58]. Table 3 summarizes best therapy options based on let-7 levels in several types of cancer.

Table 3.

Therapy of choice based on tumor let-7 levels in different types of cancer

Rational therapy choice based on let-7 expression
Cancer	Let-7 levels	Additional	Therapy of choice
Colorectal	High	KRAS mutation	Anti-EGFR therapy [45]
Ovarian	High		Platinum [53]
	Low		Platinum with Paclitaxel [53]
Breast	Low		Taxol [56] No response to Epirubicin [57]
Prostate	Low		Resistant to androgen therapy [58]

Let-7 replacement as a therapeutic

Tumor delivery of let-7 is a potential therapy, as a strategy for reversing stemness and chemoresistance, in combination with chemotherapy [59]. Let-7 over-expression results in increased sensitivity to chemotherapy and radiation therapy in ovarian cancer, hepatocellular carcinoma, oral squamous carcinoma, breast cancer, lung cancer, and myeloid leukemia, while inhibiting let-7 results in acquisition of resistance [35–37, 57, 59–65].

Higher tumor let-7 levels contribute to an increase in sensitivity to therapy [36, 57, 58]. This decrease in resistance can allow treatment with lower dose of chemotherapy to obtain the same therapeutic benefit. This represents an opportunity to avoid severe side effects of cancer treatments by using lower chemotherapy dosages. The ability to use lower drug dosages to obtain equivalent therapeutic benefit may lead to lower levels of toxicities and chemotherapy related adverse events, allowing for better quality of life for the patients undergoing treatment. Also, there would likely be fewer instances of chemotherapy discontinuation due to lower instances of dose-limiting toxicities.

The feasibility of using let-7 as therapy has been demonstrated by successful in vivo studies. Let-7 overexpression in animal model studies results in reduction of tumor size, metastasis, and prolonged survival [35, 42, 52, 58, 59, 62, 66–68]. These results are explained via functional assays in vitro, where let-7 decreases cellular proliferation, migration, and invasion [37, 42, 58, 59, 67–73]. Let-7 overexpression has been accomplished in pre-clinical murine models via let-7 mimics, demonstrating its efficacy. As miRNA will rapidly degrade in plasma, advanced let-7 delivery methods are required. In order to avoid tissue toxicity and delivery to other cells within tumor stroma, strategies for delivery of mimics specifically to cancer cells must be developed (see below). Dai et al. utilized polyethyleneglycol (PEG) nanoparticles to deliver let-7 together with paclitaxel in vivo, and they observed successful repression of tumor burden without animal toxicity [59].

Molecular aspects governing functional phenotype

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.

Fig. 2.

Overview of tumor-suppresive let-7 effect on cancer functional phenotype

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.

Table 4.

Validated and non-validated direct Let-7 targets

Human let-7 targets	Validated	Human let-7 targets	Validated
HMGA2 [39, 57, 67, 70, 89, 142]	Yes	TARBP2 [40, 81]	Yes
HMGA1 [81]	Yes	ZC3H3 [40, 81]	Yes
LIN28A [33, 69, 115]	Yes	Etv2 [143]	Yes
c-MYC [129, 144]	Yes	Acvr1b [145]	Yes
LIN28B [33, 81, 89]	Yes	Zbtb16 (PLZF) [146]	Yes
STAT3 [63, 85]	Yes	Cyclin D1 [64, 75]	Yes
N-RAS [72, 81, 147]	Yes	Cyclin A [75]	Yes
K-RAS [147]	Yes	IMP1 [65, 89]	Yes
H-RAS [57, 147]	Yes	MAP4K3 [44]	Yes
Dicer1 [100]	Yes	ITGB3 [44]	Yes
IL-6 [119]	Yes	HIF-1A [148]	Yes
Cyclin D [35, 91]	Yes	IGF2BP1 [144, 149]	Yes
IGF1 [150]	Yes	IGF2BP2 [151]	Yes
ARID3A [97]	Yes	RSU1P2 [128]	Yes
ARID3B [80]	Yes	NEDD9 [106]	Yes
TCF-4a [60]	Yes	DOCK3 [106]	Yes
MMP1 [152]	Yes	NGF [153]	Yes
NTN1 [154]	Yes	GHR [155]	Yes
INSR [115]	Yes	Twist [62]	No
IGF1R [115, 150]	Yes	Snai1 [62]	No
IRS2 [115]	Yes	Vimentin [62]	No
Pik3ip1 [115]	Yes	N-Cadherin [62]	No
AKT2 [115]	Yes	IMP2 [89]	No
TSC1 [115]	Yes	ATXN7L3 [93]	No
RICTOR [115]	Yes	USP44 [93]	No
LOX1 [156]	Yes	USP42 [93]	No
PBX3 [71]	Yes	BCL11A [157]	No
ERα [40]	Yes	TGF-βR1 [158]	No
EZH2 [35, 139]	Yes	TGF-βR3 [158]	No
E2F2 [40, 73, 81]	Yes	SMAD2 [158]	No
E2F5 [81]	Yes	FIGN [89]	No
CPSF1 [40, 81]	Yes	CDC34 [89]	No
DDX18 [40, 81]	Yes	NME6 [89]	No
EiF4A1 [40, 81]	Yes	MED6 [89]	No
EiF2C2b [40, 81]	Yes	COL4A5 [89]	No
LSM6 [40, 81]	Yes	NAP1L1 [89]	No
PABPC4 [40, 81]	Yes	PIGA [89]	No
RBM38 [40, 81]	Yes	SLC25A24 [89]	No
PLAGL2 [159]	Yes	E2F1 [37]	No
AURKB [160]	Yes	E2F1 [37]	No
PLAGL2 [159]	Yes

^aLet-7 inhibits at the promoter region

^bLet-7 increases expression

Let-7 as oncogene

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.

Let-7 regulation

Transcriptional regulation

Let-7 promoters are activated by the stem cell renewal and pluripotency factor OCT-4, and are repressed by the proto-oncogene MYC, some mutant forms of p53, and in cases of cellular stress (e.g. radiation), by wild type p53 [81, 103–105]. Let-7 repression by wild type p53 during stress is important when considering choice of therapy. p53 is activated by radiation, and in turn, p53 represses let-7 transcription. Thus, radiotherapy could induce acquired radio-resistance stemming from the of loss of let-7. Lung tumors in which let-7 levels are low correlate with low proliferation levels prior to radiotherapy. These tumors tend to exhibit accelerated reproliferation posttreatment. Thus, tumor let-7 levels in lung cancer patients may inform the clinician whether radiotherapy would be counterproductive in some cases. Because p53 is involved in many cellular processes and acts differently upon different stimuli, more research is needed to study this phenomenon. The epithelial-mesenchymal transition (EMT) factor Twist also represses the let-7 promoter in cooperation with BMI1 [106].

Epigenetic regulation

Abnormal let-7 expression is also due to epigenetic mechanisms. Let-7 is repressed by a single nucleotide polymorphism (SNP) in the let-7i promoter region, correlating with increased susceptibility to cervical squamous cell carcinoma [107]. Let-7 repression is also achieved by inhibiting let-7e promoter demethylation by JARID1B in urothelial cancer, promoter methylation by COX2/PGE2 signaling, and histone modifications of miR-125b in breast cancer [91, 108, 109]. MiR-125b and let-7a2 share the same promoter, suggesting that let-7a2 is repressed by this mechanism as well.

Post-transcriptional regulation

RNA binding proteins LIN28A and LIN28B represent a major post-transcriptional let-7 regulation pathway. LIN28 blocks let-7 maturation with high specificity at pre- and pri- stages [110]. The cold shock domain (CSD) of Lin28 interacts with the pre-E loop, and the CCHCx2 domain with the GGAG motif at the 3′ end of let-7, inhibiting let-7 processing [111]. Let-7 monouridylation by terminal uridyltransferases TUT4(7) stabilizes let-7 precursors for further processing, and LIN28 binding results in polyuridylation, which is a signal for degradation [112]. Figure 1 illustrates simplified let-7 binding by LIN28 and TUT4(7). LIN28B represses let-7 less effectively than LIN28A due to its nuclear localization, where terminal uridyltransferase, a mediator of let-7 repression, is not present [113]. LIN28A is present at high levels during early embryonic development, is progressively lost as cells differentiate, and is absent in somatic cells. It aberrantly increases in cancer, repressing let-7. Elevation of LIN28 has been attributed to loss of transcriptional regulation [78].

Although these two factors, LIN28 and let-7, appear mutually exclusive, there is evidence that they can coexist. Both mature let-7 and LIN28 are present in ESCs, fine-tuning each other [114]. As let-7 and LIN28 co-exist in ESCs, they also coexist in normal fully differentiated cells, the balance of which is important for proper control and function, as illustrated by glucose metabolism: repression of LIN28 and let-7 upregulation results in insulin resistance and impaired glucose metabolism in vivo [115]. It is also important to note that LIN28 function is not exclusively controlled by let-7. Balzer et al. demonstrated let-7 independent LIN28 function during neurogliogenesis [116]. LIN28 plays an important role during terminal differentiation of mouse skeletal muscle and is detected in mouse muscle tissues, demonstrating co-expression with let-7 [104, 117].

Let-7 overexpression also illustrates the precise balance necessary to maintain homeostasis. While loss of let-7 leads to oncogenesis, aberrantly high expression of let-7 also leads to toxicity indicating that homeostasis requires a precise level of expression. Wu et al. demonstrated that let-7 overexpression by 20-fold resulted in liver damage and dysfunction [66]. Based on this observation and co-expression of let-7 with LIN28 in ESCs, LIN28 is considered an important regulator of let-7 even in somatic cells. Furthermore, changes in LIN28 levels may alter normal cellular processes via let-7 repression or up-regulation. Parisi et al. demonstrated let-7 independent LIN28 increase upon exit from pluripotency [118].

Cellular signaling, including NFkB, STAT3, and MAPK-Erk pathways are also involved in let-7 regulation. While MAPK-Erk signaling positively regulates let-7 by inhibition of LIN28, NFkB and STAT3 cause both LIN28A up-regulation and let-7 repression [85, 119, 120]. Tsanov et al. demonstrated LIN28 stabilization via phosphorylation by MAPK-Erk, which had no effect on let-7 levels, in contrast to results obtained by Liu et al. showing MAPK-Erk-mediated let-7 activation. The discrepancy obtained by the two groups is possibly due to differences in biology of the cell types used and experimental procedures. Liu et al. and Tsanov et al. used mouse and human embryonal carcinoma cells respectively [120, 121]; a species difference could explain the conflicting findings. While Liu et al. used a knockin LIN28 mutant to demonstrate the effect of phosphorylation, Tsanov et al. used overexpression of the mutant.

In normal cells, wild type p53 helps maintain let-7 levels by disrupting the inhibitory effect of LIN28 and facilitating loading of mature let-7 onto Ago2. Mutation and loss of p53 in cancer are associated with let-7 repression [122, 123]. ADAR1 (adenosine deaminase acting on RNA), an RNA-binding protein, negatively regulates let-7 biogenesis by altering let-7 secondary structure at DROSHA and Dicer cleavage sites. ADAR1 expression is positively regulated by JAK2 signaling, and is overexpressed in CML and presumably in other cancers where JAK2 signaling is increased [124]. Let-7 is also inhibited post-transcriptionally by (DCAMKL-1) in colorectal cancer [125].

Aside from repression at the level of transcription and post-transcriptional processing, other RNAs can also inhibit let-7. MiR-107 forms complexes with let-7 and increases its degradation [126]. Long non-coding (lnc) RNA H19, linc-ROR, CCR492, and lnc RSU1P2 inhibit let-7 function by acting as sponges. MiR-107 is over-expressed in some breast cancers, linc-ROR in pancreatic ductal adenocarcinoma, and lnc RSU1P2 in cervical cancer, where they contribute to cancer progression and poor prognosis [49, 127–129]. In glioblastoma, insulin-like growth factor 2 binding protein 2 (IMP2) blocks let-7 function by binding to miRNA recognition elements of let-7 targets. There is a lack of negative correlation between let-7 and its targets in spheroids due to the protective effect of IMP2. In cancers expressing IMP2, its repression may be necessary together with let-7 up-regulation to obtain the desired tumor suppressive effect [130]. Table 5 lists factors that regulate let-7 at transcriptional, post-transcriptional, and functional levels.

Table 5.

Let-7 regulators on transcriptional, post-transcriptional, and functional levels

Let-7 regulation
Inhibitor	Family member	Context	Mechanism
JARID1B [91]	let-7e	Breast cancer
p53 mutant [81]	let-7i	Lung cancer
DCMAKL-1 [125]	let-7a	Colorectal cancer
MYC [105]	let-7a-1, f-1, d	Hepatocellular carcinoma
OCT-1 [156]	let-7g	Aorta smooth muscle cells
COX2* [161]	let-7b	Urothelial cancer
TWIST [106]	let-7i	Head and neck cancer
BMI1 [106, 162]	let-7i	Head and neck cancer
KDM2B [163]	let-7b	Embryonic fibroblasts	Promoter methylation
LIN28 [110, 113, 133]	let-7a1, a2, b, c, d, e, g, f1, f2, I, Mir-98	Mouse ESCs, Hela cells
STAT3 [85]	let-7a	Ewing sarcoma	NFkB activation
NFkB [119]	let-7a, b, c, d, f	Breast cancer	LIN28 up-regulation
MiR-107 [126]	let-7a	Breast cancer
LncRNA H19 [49]	let-7a, b	Breast cancer
Link-ROR [127]	let-7i, b, e, c	Pancreatic cancer
IMP2 [130]	Seed	Glioblastoma stem cells	Target stabilization
LncRNA RSU1P2 [128]	let-7a	Cervical caner
ADAR1 [124]	let-7d	Leukemia stem cells
LncRNA CCR492 [129]	Seed	Mouse embryonic fibroblast
eEBPa [164]	let-7a2	Lung cancer
SNP rs10877887 [107]	let-7i	Cervical cancer
P53 [104]	let-7a, b	Colon cancer	Cellular stress

Activator	Family member	Context
ZEB1 [97]	let-7e	Esophageal cancer
OCT4 [103]	let-7a-2	Cervical cancer
NF-kB [165]	let-7a-3/b	HeLa, 293T
ESE3/EHF [78]	let-7b	Prostate cancer
P53 [122]	let-7 a, b, c, e, f, g, i, Mir-98	Colon cancer
Tritetraspolin [123]	let-7b, f	Ovarian cancer
MAPK-Erk [120]	let-7a, g	Mouse embryonic carcinoma

*Indirect by inducing promoter methylation

Differences between individual family members

Functional differences

Since mature miRNA let-7 family members have nearly identical sequences, in general, it is assumed that they function similarly and have common targets, due to off target binding for which miRNAs are notorious. However, there is some evidence that different members of the let-7 family do have different functions, most likely due to unique target preferences, and therefore cannot be considered as one. In hepatocellular carcinoma, it has been demonstrated that overexpression of different let-7 family members affects cell viability to different extents: let-7a has the greatest effect [60]. It has been demonstrated that when over-expressed together, let-7i and let-7g had a greater effect on hepatoma cell division and apoptosis than overexpression of individual miRNAs, suggesting that members of this family may act in synergy to deliver tumor suppressive actions and other physiological functions [131]. Takamizawa et al. demonstrated that while both let-7a and f reduce the ability of lung cancer to form colonies, let-7f is able to do so to a greater extent [47].

Regulatory differences

Since let-7 family members are located in different clusters, transcriptional regulation is different in each case. During neural differentiation, let-7a1, a2, d, f2, and i are active in several cell types and constitutively transcribed, while let-7a3, b, c, e, and g show dynamic transcription. This difference may be due to the number of transcription start sites (TSS) present in their promoter regions. Multiple TSS produces dynamic expression because more transcription factors are involved in regulation [132]. Another way let-7 family members differ from each other is via post-transcriptional regulation. One study has demonstrated that miR-107, which contributes to metastasis of breast cancer by inhibiting let-7, binds to different let-7 members with different efficiency [126]. Different let-7 family members are repressed by LIN28 to different degrees, and in fact let-7a3 bypasses repression by LIN28 altogether due to a different sequence in the preE region of the bulge [133].

Let-7 over-expression has been widely investigated as a therapeutic agent to inhibit progression of many cancers in vitro and in animal models. It is important to consider which mature let-7 family member would be the most beneficial to patient survival before developing it into therapy.

Conclusion and future direction

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.

Abbreviations

ABCG2

ATP Binding Cassette Subfamily G Member 2

ADAR1

adenosine deaminase 1

AGO2

argonaute 2

Akt

protein kinase B

ARID3A

AT-rich interaction domain 3A

BAX

BCL2 associated X protein, apoptosis regulator

BCL-xL

apoptosis inhibiting protein

BRCA1/2

DNA repair associated protein 1/2

CCR492

cell-cycle regulated long non-coding RNA

CDK4

cyclin dependent-kinase 4

COX2

stem cell factor

CSC

cancer stem cell

CSD

cold shock domain

DCAMKL-1

doublecortin and CaM kinase-like 1

DICER

ribonuclease

EMT

epithelial-to-mesenchymal transition

EOC

epithelial ovarian cancer

ESC

embryonic stem cell

GHR

growth hormone receptor

H2Bub1

histone H2B monoubiquitylation

HMGA2

high mobility group AT-Hook 2

IGF1

insulin like growth factor 1

IGF1R

insulin like growth factor 1 receptor

IGF2BP1

insulin like growth factor 2 binding protein 1

IGF2BP2

insulin like growth factor 2 binding protein 2

IMP2

insulin-like growth factor 2 binding protein 2

JAK2

janus kinase 2

JARID1B

jumonji AT-rich interactive domain 1B

LIN28A/B

LIN28 homolog A/B

Lnc H19

long non-coding RNA H19

Lnc ROR

long non-coding RNA regulator of reprogramming

Lnc RSU1P2

long non-coding RNA Ras suppressor protein 1 pseudogene 2

MDR1

multi-drug resistance-1

MIRLET7

miRNA lethal-7

miRNA

microRNAs

MMP1

matrix metallopeptidase 1

MMP9

matrix metallopeptidase 9

MYC

proto-oncogene protein

N-Cadherin

neuronal cadherin

NGF

nerve growth factor

NTN1

netrin-1

OCT-4

octamer-4 embryonic gene

P21

cell cycle regulator

PARP1/2

poly (ADP-ribose) polymerase 1/2

PEG

polyethyleneglycol

PGE2

prostaglandin E2

PLAGL2

pleiomorphic adenoma gene-like 2

RAD51

DNA repair protein

SNP

single nucleotide polymorphism

STAT3

signal transducer and activator of transcription-3

TAM

tumor associated macrophage

TCF-4

transcription factor 4

TSS

transcription start site

TUT4(7)

terminal uridylyl transferase 4(7)

WNT

signaling pathway

Authors’ contributions

EC and JJU performed the literature review and analyzed the data that informed this review, and drafted the review. All authors contributed to the writing of the review. All authors read and approved the final manuscript.

Funding

This work was supported by Loma Linda University School of Medicine, Grants to Promote Collaboration and Translation, and by the Department of Pathology and Human Anatomy.

Availability of data and materials

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.