Resveratrol decreases the levels of miR-155 by upregulating miR-663, a microRNA targeting JunB and JunD

Read Also

Our study provides the very first data concerning the effects of resveratrol on miRNAs and its ability to both upregulate miR-663 and downregulate oncogenic pro-inflammatory miR-155. It provides strong support that: [i] miR-663 targets JunB and JunD transcripts, consequently decreasing JunB and JunD levels as well as AP-1 activity; [ii] miR-663 also potentially targets transcripts encoding several other factors needed for the innate immune response; [iii] the downregulation of AP-1 activity by resveratrol is miR-663-dependent; [iv] resveratrol impairs the upregulation of miR-155 by LPS at least in part by upregulating the anti-inflammatory miR-663.

Transcripts from the LOC284801 locus on chromosome 20, which represent miR-663 primary RNAs, showed the highest statistically significant level change following resveratrol treatment of THP-1 cells, as deduced from Affymetrix microarray analysis. Although no probe is presently available to follow miR-663 accumulation by qRT-PCR, due to its more than 92 % CG content, miRNA microarrays, RNase-protection and luciferase assays provided independent confirmations of the upregulation of mature miR-663 by resveratrol. As it is the case for many other miRNAs, miR-663 upregulation was not proportional to that of its primary transcripts, which probably reflects post-transcriptional regulations.

Our results confirm the downregulation of AP-1 activity by resveratrol. However, while previous reports mainly focused on resveratrol effects on c-Jun and c-Fos, our results show that resveratrol can specifically target JunB and JunD by increasing miR-663 levels, and suggest that the same miRNA could also target FosB transcripts. This is an important result, for AP-1 factors include c-Jun, JunB, JunD, FosB, Fra-1 and Fra-2, as well as Jun dimerization partners JDP1 and JDP2 or the closely related ATF2, LRF1/ATF3 and B-ATF, so that potentially about 18 different dimeric combinations may be formed. Thus, the capability of resveratrol to upregulate miR-663 to specifically target a subset of AP-1 dimers might have profound effects on the levels of expression of promoters to whom different AP-1 factors can compete to bind to. Due to the many roles of AP-1 factors both in inflammation (37) and cancer (38,39), the specific targeting of genes encoding a subset of AP-1 factors, by changing the composition of AP-1 dimers on key promoters, may possibly explain some of resveratrol anti-inflammatory and anti-cancer properties. However, while 663-I increased luciferase activity produced from Luc-JunB and Luc-JunD constructs in the presence of resveratrol, it failed to impair the downregulation of JunB by resveratrol, indicating that miR-663 independent effects were also taking place. This was probably due to the fact that resveratrol downregulates the pro-inflammatory NF-kB activity (32), which is needed for JunB activation (40). Also, 663-I failed to increase JunB levels following LPS challenge. This is probably due in part to the fact that LPS challenge decreased miR-663 levels. It is also possible that JunB might have already reached its maximum possible levels under LPS challenge in control experiment, that LPS may limit miR-663 access to JunB transcripts, or that miR-663 might target transcripts encoding an unidentified repressor of JunB.

Importantly, miR-663 also potentially targets transcripts encoding factors playing a role not only in TLR signaling and the innate immune response to LPS, but also in granulopoiesis, monopoiesis, INFg signaling, Th17 lymphocytes differentiation or TCR signaling. The functions of miR-663 are thus probably not restricted to the monocyte/macrophage lineage.

Our results also show that miR-663 impairs the upregulation of miR-155 by LPS. This effect may be due to the targeting of JunB or FosB, which have been shown to transcriptionally activate BIC, namely miR-155 host gene (36). Also, miR-663 potentially targets KSRP transcripts, which encode a RNA binding protein implicated in the LPS-induced miR-155 maturation from its primary transcripts BIC (41). Given the facts [i] that AP-1 factors participate to the activation of many pro-inflammatory cytokine genes in partnership with NFAT factors (37) and are implicated in the upregulation of miR-155 by LPS, [ii] that miR-155 upregulation is a hallmark of innate immune response (21,22), and [iii] that miR-663 levels decreased at the very beginning of LPS challenge (this manuscript), miR-663 is likely to work along with miR-146a and miR-146b to control the intensity of the innate immune response. Furthermore, beside its roles in myeloid differentiation and innate response, miR-155 has many other functions in the immune system, such as the regulation of myelopoiesis and erythropoiesis, Th1 differentiation, B cell maturation, IgG1 production, somatic hypermutations, gene conversion, class switch recombination, or B and T cell homeostasis (19). It will thus be very interesting to determine if miR-663 can also modulate miR-155 levels and functions in these different cell types.

Finally, the fact that resveratrol decreases both AP-1 activity and miR-155 levels at least in part through the upregulation of miR-663 could be critical in optimizing the use of resveratrol as an anti-inflammatory but also as an anti-tumor agent. Namely, several reports have recently established a direct link between elevated levels of miR-155 expression and the formation and development of tumors such as leukemias and breast, lung or gastric cancers (19). Elevated miR-155 levels are linked with enhanced cell proliferation (16,17), transgenic mice with B cells overexpressing miR-155 develop B-cell leukemia (25), and a sustained expression of miR-155 in hematopoietic stem cells causes a myeloproliferative disorder (26). These transgenic mice also showed that one of the many effects of miR-155 is the enhancement of cell proliferation. Of note, resveratrol impaired the upregulation of cyclins B1 and D3 by LPS (results not shown). This suggests that, beside opposing the proliferative effects of AP-1 factors, resveratrol may also oppose the proliferative activity of miR-155 through the upregulation of miR-663. Importantly, miR-663 was found to be downregulated in hormone refractory prostate cancer cells, along with miR-146a and miR-146b (42). Unfortunately, miR-663 is not present in mouse genome, thus impairing direct experimentation using knock-out animals.

Nevertheless, while resveratrol anticarcinogenic potential has been linked with data primarily from human cell culture systems, evidence that resveratrol can inhibit carcinogenesis in several organ sites emerged from results of cancer prevention and therapy studies in laboratory animal models (29). However, previous results have shown that, at least when it comes to the immune system, the expression and thus potentially the function of miRNAs cannot be directly transposed from one species to another one. For example, the upregulation of anti-inflammatory miR-146a and miR-146b during the innate immune response has been observed only in human cell lines, such as THP-1 (20) or A549 cells (24), but not in mouse cell lines or in mouse spleen macrophage (22,43). As miR-663 is found only in primates, our results come as a warning that studies in animal may not always allow to predict accurately the molecular effects of resveratrol in human.