Interplay between K-RAS and miRNAs

Abstract

K-RAS is frequently mutated in cancers, and its overactivation can lead to oncogene-induced senescence (OIS), a barrier to cellular transformation. Feedback onto K-RAS limits its signaling to avoid senescence while achieving the appropriate level of activation that promotes proliferation and survival. Such regulation could be mediated by miRNAs, as aberrant RAS signaling and miRNA activity coexist in several cancers, with miRNAs acting both up- and downstream of K-RAS. Several miRNAs both regulate and are regulated by K-RAS, suggesting a noncoding RNA-based feedback mechanism. Functional interactions between K-RAS and the miRNA machinery have also begun to unfold. This review comprehensively surveys the state of knowledge connecting K-RAS to miRNA function and proposes a model for the regulation of K-RAS signaling by noncoding RNAs.

Keywords: K-RAS; cancer; feedback regulation; miRNA.

Figure 1.. miRNAs up and downstream of K-RAS signaling

K-RAS is a GTPase that is activated through guanine nucleotide exchange factors (GEFs)-mediated binding to GTP. Its inactivation is through GTP hydrolysis facilitated by GTPase activating proteins (GAPs). Proper functioning of GEFs, such as SOS, requires upstream signaling activation of EGFR and subsequent recruitment of SHC and GRB2 to the plasma membrane. An increase in K-RAS-GTP level potentiates downstream signaling such as RAF/ERK, PI3K/AKT, and RalGDS pathways [1]. Various miRNA act up and downstream of the K-RAS signaling pathway. Several miRNAs have been identified to directly inhibit K-RAS translation, including let-7, miR-143/145, miR-193a, etc. [23-26]. Dysregulation of these miRNAs are frequently detected in cancers and potentially exhibit tumor-suppressive effects due to their targeted suppression of K-RAS. Indirect inhibition of K-RAS activity can also be achieved by miRNAs targeting key regulators such as GAPs. miR-21 and miR-31-mediated suppression of RASA1, a prominent GAP, enhances K-RAS activation by decreasing its hydrolysis potential [28-30]. Hyperactivation of K-RAS signaling also induces drastic changes in downstream miRNA levels, including up-regulation of miR-155, miR-450b, miR-31, and miR-21 [30,36,46,47]. These changes in miRNA level and activity induced by K-RAS could contribute to shaping a pro-tumorigenic environment or serve as a negative regulator that dampens the oncogenic signals.

Figure 2.. miRNA-mediated feedback regulation of K-RAS

(a) Extensive survey of all past literature yields 166 human miRNAs as downstream to K-RAS, as their expressions are reported to be dysregulated upon hyperactivation of K-RAS signaling at least once. A total of 39 miRNAs are classified as K-RAS targeting miRNAs with experimental evidence of direct miRNA targeting induced K-RAS suppression in various cancer cell lines. Overlapping the two cohorts presents 10 miRNAs that act both up and downstream of K-RAS: let-7a, miR-127, miR-143, miR-145, miR-16, miR-181, miR-193a, miR-200C, miR-27B, and miR-4689. (b) Three distinct modes of feedback regulation emerge upon integration of K-RAS-induced effects on miRNAs with miRNA-mediated direct/indirect regulation of K-RAS. miR-16 and miR-200c are not included here due to conflicting reports of their expression changes upon hyperactivation of K-RAS. Expression of K-RAS targeting miR-181a is up-regulated upon expression of oncogenic K-RAS, forming a negative feedback loop [26,34,35]. On the other hand, miR-21, miR-31, and miR-30c target RAS GAPs such as RASA1 and NF1 [28-30]. Their expressions are up-regulated upon K-RAS hyperactivation [30,36], suggesting that these miRNAs can positively feedback onto K-RAS by suppressing its negative regulators to further enhance the signaling activation. Furthermore, several K-RAS targeting miRNAs (let-7a, miR-127, miR-143, etc.) are down-regulated by K-RAS, establishing a mutual suppressive relationship [26,37,38]. Activity of these miRNA and K-RAS signaling level likely form a delicate equilibrium during homeostasis and its disruption in various pathologies could exacerbate the hyperactivation of oncogenic K-RAS.

Figure 3.. AGO2 phosphorylations by K-RAS pathway

Proper phosphorylations of AGO2 are critical for its function. Several of these sites are regulated by members of the RAS pathway. AGO2 Ser387 phosphorylation is critical for its association with TNRC6 to form the RNA-induced silencing complex (RISC) at the processing body (p-body). Additionally, Ser387 phosphorylation inhibits both miRNA-mediated AGO2 cleavage of target mRNA transcripts and exosome sorting of AGO2/miRNA complexes. This critical phospho-site is regulated by AKT3, a kinase in the PI3K/AKT pathway downstream of K-RAS [88-90]. Therefore, hyperactivation of K-RAS could elevate AGO2 Ser387 phosphorylation to regulate its function. AGO2 Tyr393 represents another functional phospho-site that regulates AGO2:Dicer binding and miRNA loading to AGO2, both of which are required for proper AGO2 function. Its phosphorylation is regulated by EGFR, an upstream activator of K-RAS signaling that is susceptible to MAPK-induced negative feedback [3,57]. Thus, it is reasonable that the negative feedback suppression of EGFR triggered by oncogenic K-RAS signaling would undermine phosphorylation of AGO2 Tyr393, facilitating its function by enhancing AGO2 interaction with Dicer and miRNA. Additionally, PTP1B, the phosphatase governing AGO2 Tyr393, is also regulated by K-RAS. An increase of K-RAS signaling upregulates reactive oxygen species (ROS). This in turn inhibits PTP1B function and up-regulates AGO2 Tyr393 phosphorylation [92], suggesting a multifaceted role of K-RAS in regulating AGO2 Tyr393.