An indirect role for oncomir-519b in the expression of truncated neurokinin-1 in breast cancer cells

Abstract

Neurokinin 1 (NK1) encodes full-length (NK1-FL) and truncated (NK1-Tr) receptors, with distinct 3' UTR. NK1-Tr exerts oncogenic functions and is increased in breast cancer (BC). Enhanced transcription of NK1 resulted in higher level of NK1-Tr. The 3' UTR of these two transcripts are distinct with NK1-Tr terminating at a premature stop codon. NK1-Tr mRNA gained an advantage over NK1-FL with regards to translation. This is due to the ability of miR519B to interact with sequences within the 3' UTR of NK1-FL, but not NK1-Tr since the corresponding region is omitted. MiR519b suppressed the translation of NK1-FL in T47D and MDA-MB-231 resulting in increased NK1-Tr protein. Cytokines can induce the transcription of NK1. However, our studies indicated that translation appeared to be independent of cytokine production by the BC cells (BCCs). This suggested that transcription and translation of NK1 might be independent. The findings were validated in vivo. MiR-519b suppressed the growth of MDA-MB-231 in 7/10 nude BALB/c. In total, increased NK1-Tr in BCCs is due to enhanced transcription and suppressed translation of NK1-FL by miR-519b to reduced tumor growth. In summary, we report on miRNA as a method to further regulate the expression of a spiced variant to promote oncogenesis. In addition, the findings have implications for therapy with NK1 antagonists. The oncogenic effect of NK1-Tr must be considered to improve the efficacy of current drugs to NK1.

Fig. 1. Nuclear run-on and Real-time RT-PCR for NK1 subtypes

(A) A non-radioactive method quantitated NK1-FL and NK-Tr. Fibroblast 1502 was stimulated with 10 ng/mL IL-1α or vehicle for 8 h. Nuclei were isolated and total RNA collected within a 24 h period for hybridization with cDNA probes and fluorochrome YoYo-1. The incorporated YoYo-1 was detected with a fluorometer. The results are presented as fluorescence intensity, mean±SD, n=7. (B) Real time was performed with primers that can discriminate NK1-FL and NK1-Tr with cDNA from IL-1α-stimulated fibroblasts. Control contained vehicle (PBS). The lowest value was normalized as 1 and the other experimental points were presented as relative fold change (mean±SD, n=4) as described [22]. * p<0.05 vs. IL-1α stimulated cells; ** p<0.05 vs. NK1-Tr; *** p<0.05 vs. NK1-Tr

Fig. 2. Nuclear run-on, real-time RT-PCR and western blots for NK1 subtypes in BCCs

(A) Unstimulated T47D and MDA-MB-231 were studied for NK1-FL and NK1-Tr using the nuclear run-on assay as for Fig. 1A. The results are presented as the mean fluorescence intensity±SD, n=4. (B) Total cell extracts from T47D and MDA-MB-231 were studied for NK1 subtypes by western blots. (C) Normalized band densities for NK1-FL and NK1-Tr, n=3±SD. (D) Real-time was performed for NK1-FL and NK1-Tr as for Figure 1B. The lowest value was normalized as 1 and the other experimental points were presented as relative fold change (mean±SD, n=4) as for Fig. 1D. * p<0.05 vs. NK1-FL

Fig. 3. MiR519b and miR609 levels in BCCs

(A) Shown are the 3′ UTR of NK1 (Tr and Fl) and the relative positions of miR519 and miR609. (B) Real-time PCR was performed for miR-519b and miR-609 with cDNA from T47D and MDA-MB-231. The data are presented as the relative fold change (mean±SD, n=5).

Fig. 4. Effects on reporter gene activity by pre-miR-519b and -miR609

MDA-MB-231 and T47D were co-transfected with pMIR/NK1-SR (Figure S4) and pre-miR-519b (A) or pre-miR-609 (B). After 48 h luciferase activities were quantitated with whole cell extracts and the results are presented as the mean±SD, n=6. * p<0.05 vs. pre-miR-transfectants. (C) BCCs were transfected with pre-miR519b and/or anti-miR519b. Controls were transfected with control pre-miR. After 48 h, real-time PCR was performed for pre-miR519b and the data presented as fold change (mean±SD, n=4) of the lowest value, which was assigned a value of 1. (D) The studies in A and B were repeated, except with co-transfection with pre- and/or anti-miR519b. Parallel cultures were transfected with control pre-miR. After 48 h, luciferase activities were quantitated and the results are presented as for A and B’ (mean±SD, n=4).

Fig. 5. MiRNAs in the expression of NK1 subtypes

(A) MDA-MB-231 was transfected with pre-miR609, -miR519b or control pre-miR. After 24 h, whole cell extracts were studied for NK1 subtypes by western blots. The membranes were stripped and reprobed with anti-β-actin. (B) The band densities in `A’ were normalized and the data from three different western blots are presented as the ratio of NK1-Tr/NK1-FL, mean±SD. MDA-MB-231 and T47D were transfected with anti-miR519b and then analyzed as for `A’. (C) MDA-MB-231 and T47D were transfected with anti-miR519b and then analyzed as for `A’. (D) The band densities in `B’ were normalized and the data from three different western blots are presented as for `B’. * p>0.05 vs. control; *** p<0.01 vs. anti-miR519

Fig. 6. Effects of anti-miR-519b on tumor growth

MDA-MB-231 cells were transfected with anti-miR519b or control anti-miR. The transfectants were validated by western blot for NK1-Tr (A) and real-time PCR for pre-miR519b (B). (C) Nude BALB/c (n=10) were injected with 10⁶ cells, subcutaneously in the dorsal flank. The tumor volumes, presented as mean±SD. The experimental group (anti-miR519b) were plotted as those with no palpable tumor (n=7) and those with tumor growth (n=3). *p<0.05 vs. control anti-miR

Fig. 7. Overall summary

Shown is an overall summary with other studies showing cytokine-mediated transcription of NK1 (*[14]). Cytokine production in BCCs, via autocrine stimulation, can induce NK1 transcription (*[5,15]) to produce NK1-FL and NK1-Tr. Mir-519b interact with NK1-FL to suppress its translation, thereby allowing for increased NK1-Tr. Since miR-519b can suppress the RNA-binding protein, HuR (**[19]), perhaps there is another mechanism in the suppression of NK1-FL translation.