Mouse‐specific up‐regulation of Ccnb1 expression by miR‐199a‐5p in keratinocyte

MicroRNA (miRNA) are a class of single‐stranded, small non‐coding RNA that regulate various biological processes, including skin and hair cycle regulation, by modulating the expression of specific genes at the post‐transcriptional level. Recently, several studies reported that miRNA directly or indirectly up‐regulate target genes. Previously, we performed microarray analysis to identify the target genes of miR‐199a‐5p in a mouse skin keratinocyte cell line and detected more than 200 genes whose expression was significantly increased by miR‐199a‐5p overexpression (> 1.5‐fold). In this study, we further investigated these genes and found that cyclin B1 (Ccnb1) expression was positively regulated by miR‐199a‐5p in keratinocyte. Moreover, Ccnb1 expression was inversely correlated with miR‐199a‐5p expression during the mouse hair cycle. Cell cycle analysis showed that the proportion of cells in S phase was slightly increased, while the proportion of cells in G2/M phase decreased by miR‐199‐5p. Using luciferase assay, we found that the 3′ untranslated region of Ccnb1 was a direct target of miR‐199a‐5p. We also found that the regulation of Ccnb1 expression by miR‐199a‐5p is mouse specific. CCNB1 expression was not affected in the human and monkey cell lines. These results provide a new relationship between Ccnb1 and miR‐199a‐5p in both mouse keratinocyte and miRNA biology.

In recent years, many investigators have studied the roles of miRNA in skin biology using keratinocytes and cancers, and found that miR-99b inhibits proliferation of human epidermal keratinocytes by downregulating IGF-1R expression, miR-378b promotes keratinocyte differentiation by targeting NKX3.1 [4,5]. miR-125b induces tumor initiation and promotes malignant progression by repressing differentiation and increasing survival of cancer cells, and miR-330-5p inhibits proliferation and migration of keratinocytes by targeting PDIA3 expression [6,7]. These studies suggest that miRNA are important regulators involved in the proliferation, differentiation, and migration of keratinocytes and in skin cancer.
The miRNA-199a hairpin precursor gene is located on human chromosome 19, and its orthologous gene resides on mouse chromosome 9 [8]. Previous studies reported that miR-199a-5p is involved in the regulation of biological processes in liver, stomach, testis, colon, and skin keratinocytes [9][10][11][12]. Recently, we performed microarray analysis to identify the target genes of miR-199a-5p in mouse skin keratinocytes and showed that Bcam and Fzd6 are new target genes in keratinocytes and human cutaneous squamous cell carcinoma [12]. Among the 393 putative target genes of miR-199a-5p, 232 genes were up-regulated by miR-199a-5p overexpression. In the current study, we focused on these up-regulated genes and found that the expression of the cyclin B1 gene (Ccnb1) is upregulated by miR-199a-5p. Moreover, we found that this regulation is mouse specific. These results reveal a new relationship between Ccnb1 and miR-199a-5p in mouse keratinocytes.

Mice
The BALB/C mice were bred in the barrier system under specific pathogen-free conditions with regulated light (07:00-19:00 h), temperature (23 AE 1°C), humidity (50 AE 5%), and ventilation (10-12 times per hour). All animal experiments were approved by the Institutional Animal Care and Use Committee of the Catholic University of Korea. All experiments were carried out in accordance with the guidelines for animal experimentation.
MiR-199a-5p-specific quantitative RT-PCR Total RNA from mouse dorsal skin was extracted from the cells using the Trizol reagent (Invitrogen) according to the manufacturer's instruction. A Mir-X TM miRNA First-strand synthesis kit (Clontech, MountainView, CA, USA) was used to synthesize complementary DNA (cDNA) following the manufacturer's protocol. Quantitative RT-PCR (qRT-PCR) was performed using miRCURY LNA TM MiR-330-5p-specific primer (Exiqon, Vedbaek, Denmark) following the manufacturer's instruction. The relative expression of miR-199a-5p was calculated against U6 small nuclear RNA expression using the comparative DDC t method [13].

RT-PCR and qRT-PCR
Total RNA were reverse-transcribed into cDNA using a PrimeScript 1st strand cDNA Synthesis kit (Takara, Tokyo, Japan) following the manufacturer's protocol. Thermal Cycler-100 (MJ Research, Waltham, MA, USA) and CFX96 (Bio-Rad Laboratories, Hercules, CA, USA) were used to perform RT-PCR and qRT-PCR, respectively. The primer sequences and cycling conditions used are listed in Table 1. The relative expression levels were normalized against glyceraldehyde-3-phosphate dehydrogenase gene expression using the comparative DDC t method [13]. Results represent the average of three independent experiments measured in duplicate.

Western blot analysis
Protein extracts from PAM212 cells were prepared from plates 72 h post transfection using radioimmunoprecipitation assay buffer (150 mM sodium chloride, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS, 50 mM Tris-HCl [pH 8.0]) according to the standard method. Then, lysates were subjected to 10% SDS/PAGE and transferred to a nitrocellulose membrane. The membrane was incubated with a rabbit polyclonal Ccnb1 antibody (1 : 2500; Santa Cruz, Santa Cruz, CA, USA) or a mouse polyclonal b-Actin antibody (1 : 5000; Santa Cruz) following the standard protocol. Protein bands were detected using an enhanced chemiluminescence system (Amersham Bioscience, Piscataway, NJ, USA).

Plasmid construction
The full-length 3 0 UTR cDNA of melanoregulin (Mreg), keratin 23 (Krt23), and Ccnb1 was amplified from cDNA generated from the total RNA of PAM212 cells by PCR using PrimeSTAR DNA Polymerase (Takara). The PCR product was cloned into pGEMT-easy vectors and subcloning into psiCHECK-2 vector DNA using the NotI cloning sites (Promega, Madison, WI, USA).

Cell cycle assay
MiR-199a-5p overexpressed PAM212 cells were harvested at 72 h post transfection and washed with PBS twice. Then, these cells were fixed in 70% ethanol at À20°C overnight. After washed with PBS, cells were resuspended in propidium iodide staining solution (40 lgÁmL À1 ). The percentage of cells in each phase of the cell cycle was measured by FACSCanto II (BD Biosciences, San Jose, CA, USA).

Statistical analysis
P values were determined using Student's t-tests and a value of P < 0.05 was considered statistically significant.

Ccnb1 is a target of miR-199a-5p in mouse keratinocyte
We have previously found that the expression of 232 genes was increased in PAM212 cells overexpressing miR-199a-5p (> 1.5-fold, P < 0.05) [12]. Among these genes, we selected Ccnb1 and the genes encoding Mreg and Krt23, because they are also associated with skin keratinocytes [14][15][16]. We validated the expression of these genes by qRT-PCR using total RNA originally used as templates for the mRNA microarray analysis and found that the expression of Mreg, Krt23, and Ccnb1 mRNA in PAM212 cells overexpressing miR-199a-5p was 3-5 times that in control cells ( Fig. 1A-C). To determine whether the up-regulation of these genes is due to direct targeting by miR-199a-5p, we used a luciferase assay system. While miR-199a-5p transfection did not affect the luciferase activity of Mreg and Krt23 reporters ( Fig. 1D-F), it increased the luciferase activity of the reporter containing the Ccnb1 3 0 UTR in comparison with cells transfected with the control miR (Fig. 1F). These results suggested that Ccnb1 is a target of miR-199a-5p in mouse keratinocytes.

Mir-199a-5p directly up-regulates Ccnb1 expression in mouse keratinocyte
To investigate whether Ccnb1 is a direct target of miR-199a-5p in mouse keratinocytes, we determined the expression of Ccnb1 in miR-199a-5p-overexpressing PAM212 cells at both mRNA and protein levels. qRT-PCR revealed that Ccnb1 mRNA expression was consistently higher in PAM212 cells transfected with miR-199a-5p than in cells transfected with the control RNA ( Fig. 2A). Western blot analysis showed that CCNB1 expression was also increased in miR-199a-5p-overexpressing PAM212 cells at both concentrations of the mimic (Fig. 2B). Overexpression of miR-199a-5p resulted in up-regulated CCNB1 expression by 2.02-and 2.70-folds at 50 and 100 nM mimic treatment, respectively (Fig. 2C). To further confirm these findings, we performed an inhibition experiment. Inhibition of endogenous miR-199a-5p expression using a miR-199a-5p inhibitor reduced the Ccnb1 mRNA expression (Fig. 2D).
Next, we used several online software programs MIRBASE TARGETS (www.mirbase.org), TAR-GETSCAN (www.targetscan.org), MicroRNA.org (www.MicroRNA.org), and RNAHYBRID (bibiserv. techfak.uni-bielefeld.de/rnahybrid) to predict the target site of miR-199a-5p in Ccnb1 mRNA. Only RNAHYBRID predicted a miR-199a-5p-binding site at 239 bp of the Ccnb1 3 0 UTR (Fig. 3A). To determine whether the predicted site is functional, we performed luciferase assay using a deletion mutant construct lacking this site. By comparing luciferase activity of the wild-type and deletion constructs (Fig. 1F), we did not find any inhibitory effect of the deletion (Fig. 3B). These results indicated that the predicted site is not responsible for the miR-199a-5p-dependent increase in CCNB1 expression, thus suggesting the presence of another site. Overall, the above results indicated that miR-199a-5p positively regulates Ccnb1 expression at the post-transcriptional level.
Correlation between expressions of Ccnb1 and miR-199a-5p in mouse hair cycle We investigated whether the up-regulation of Ccnb1 by miR-199a-5p occurs during the mouse hair cycle. First, the relative expression of Ccnb1 and miR-199a-5p was investigated at various stages of the hair cycle (P10-P28). qRT-PCR analysis revealed that the expression of both Ccnb1 and miR-199a-5p increased during the anagen phases and decreased at the following stages, with the lowest expression at telogen (Fig. 4A, B). These results showed that Ccnb1 expression is positively correlated with miR-199a-5p expression in the mouse hair cycle.
Since CCNB1 plays a role in the cell cycle as a mitotic cyclin that functions in the G2/M phase transition, we performed cell cycle assay to determine whether increased miR-199a-5p expression affects the cell cycle in PAM212 cells. We found that the number of cells in S phase slightly increased and the number of cells in G2/M phase decreased after transfection with the miR-199-5p mimic in comparison with transfection with the negative mimic (Fig. 4C, D).
Ccnb1 expression is up-regulated by miR-199a-5p in a mouse-specific manner Next, we investigated whether Ccnb1 expression was also up-regulated by miR-199a-5p in the immortalized human keratinocyte cell line HaCaT. In contrast to PAM212 cells, we found that the CCNB1 mRNA expression level was not affected by miR-199a-5p overexpression in HaCaT cells (Fig. 5A). To confirm this finding, we investigated the effect of miR-199a-5p overexpression on CCNB1 expression in other primate cell lines. We found that CCNB1 expression was not affected in the human colorectal cell lines SNU-C5 and Colo320 DM (Fig. 5B, C) and in the monkey fibroblast cell line Cos-1 (Fig. 5D). Interestingly,  miR-199a-5p overexpression increased the expression of Ccnb1 in another mouse cell line, 3T3-L1, similar to the effect in PAM212 cells (Fig. 5E). Furthermore, the Ccnb1 3 0 UTR responded to the miR-199a-5p mimic, as shown by the increased luciferase activity in 3T3-L1 cells (Fig. 5F). These data suggest that increased expression of Ccnb1 in response to miR-199a-5p is a mouse-specific phenomenon.
To analyze whether this differential regulation is caused by structural differences in the Ccnb1 3 0 UTR, we examined the alignment of mouse, monkey, and human 3 0 UTR sequences. Interestingly, we found that    6. Identification of the Ccnb1 3 0 UTR region responsible for miR-199a-5pinduced regulation of expression in the mouse. (A) Sequence alignment of Ccnb1 3 0 UTRs from monkey, human, and mouse; the alignment was generated using CLUSTALW (www.genome.jp/tools/ clustalw). (B, C) Dual luciferase reporter assays with constructs containing the R1 or R2 region of the Ccnb1 3 0 -UTR in (B) PAM212 and (C) 3T3-L1 cells. In both cell types, miR-199a-5p only activated luciferase activity of the R2 region. In contrast, the R1 region was not regulated by miR-199a-5p. Results are the average of three independent experiments conducted in duplicate. **P < 0.01; ***P < 0.001; NS, not significant. the mouse Ccnb1 3 0 UTR (939 bp) is much longer than those of human (622 bp) or monkey (569 bp) (Fig. 6A). This difference is also present in between the mouse and other species. Therefore, we hypothesized that mouse-specific regulation of Ccnb1 expression by miR-199a-5p depends on the mouse-specific 3 0 UTR region. To verify this hypothesis, we compared luciferase activities of the 3 0 UTR constructs containing either the evolutionarily conserved region of the 3 0 UTR (R1; 1-513 bp) or the mouse-specific region (R2; 491-939 bp). As expected, we found that luciferase activity of the R2-containing construct was significantly increased by miR-199a-5p in both PAM212 and 3T3-L1cells, while that of the R1-containing construct was not (Fig. 6B, C). Moreover, the increase in luciferase activity by miR-199a-5p conferred by the R2 region was similar to that of the full-length Ccnb1 3 0 UTR. From these results, we conclude that the upregulation of Ccnb1 by miR-199a-5p is mediated by the mouse-specific region of the Ccnb1 3 0 UTR.

Discussion
In general, the function of miRNA is to inhibit gene expression at the post-transcriptional level by binding to the 3 0 UTRs of specific target mRNA [2]. However, several studies have demonstrated that miRNA are also able to post-transcriptionally up-regulate their target genes. For instance, miR-466l increases IL-10 expression in Toll-like receptor-triggered macrophages by antagonizing the interaction between the RNAbinding protein tristetraprolin and IL-10 mRNA [17]. A microRNA, miR-145 promotes vascular smooth muscle cell differentiation in part by increasing myocardin protein expression [18]. Overexpression of miR-223 increases the total cellular level of glucose transporter type 4 protein in neonatal rat ventricular myocytes [19].
In this study, we showed that miR-199a-5p up-regulates CCNB1 expression in mouse keratinocytes and fibroblasts. Up-regulation of Ccnb1 by miRNA has been previously reported. Huang et al. [20] demonstrated that miR-744, miR-1186, and miR-466d-3p induce Ccnb1 expression by interacting with its promoter region in mouse cell lines. In contrast, we found that Ccnb1 up-regulation by miR-199a-5p is mediated through the 3 0 UTR of Ccnb1. Although we did not identify the precise binding site of miR-199a-5p in the Ccnb1 3 0 UTR because in silico sequence analysis showed no predicted miR-199a-5p target sites, dual luciferase assay suggested that the 3 0 UTR of Ccnb1 is a direct target of miR-199a-5p in mouse keratinocytes and fibroblasts.
We also found that the up-regulation of Ccnb1 expression by miR-199a-5p is mouse specific. Our data show that miR-199a-5p overexpression did not affect Ccnb1 expression in human and monkey cells. This was unexpected because most miRNA-binding sites on mRNA are conserved between species. However, there are some nonconserved miRNA-binding sites that cause species-specific miRNA-mRNA interactions. For instance, miR-351 and miR-298 regulate astrocyte activation by targeting genes involved in the tumor necrosis factor-alpha (TNF-a) signaling pathway in a mouse-and rat-specific manner [21]. FOXO1 regulates cell proliferation and invasion via miR-183 only in human cells [22]. These data suggest that changes in miRNA-mediated regulation of target genes occurred in a species-specific manner and contributed to phenotypic differences among various species.
Using a reporter assay, we demonstrated that miR-199a-5p up-regulates Ccnb1 expression by binding specific sequences in the mouse Ccnb1 3 0 UTR. There is a sequence variation in the mouse Ccnb1 3 0 UTR compared with those of other species. We found that Fig. 7. The effect of miR-199a-5p to CCNB1 expression in rat C6 cell. qRT-PCR revealed that CCNB1 expression was increased by miR-199a-5p in rat glial cell. Results are the average of three independent experiments. Results are the average of three independent experiments conducted in duplicate. ***P < 0.001. the expression of rat Ccnb1, which has a long 3 0 UTR similar to that of mouse Ccnb1, was increased by miR-199a-5p (Fig. 7). Interestingly, hamster, another rodent, has a Ccnb1 3 0 UTR of only 684 bp. Mouse and rat belong to the Muridae family, whereas hamster belongs to the Cricetidae. It would be interesting to see whether the regulation of CCNB1 by miR-199a-5p is present only in the Muridae or is common to all rodents. Further studies are required to determine the precise regulation mechanism.
Ccnb1 is well known to act in G2/M phase transition during the cell cycle [23,24]. It forms a complex with cyclin-dependent kinase 1 (Cdk1), and this complex (maturation-promoting factor) induces the early events of mitosis by controlling chromosome condensation, nuclear envelope breakdown, and spindle pole assembly. Interestingly, we also found that Cdk1 expression is concomitantly increased by miR-199a-5p in mouse keratinocytes and fibroblasts (Fig. 8). On the basis of this information, we speculated that miR-199a-5p regulates the mouse keratinocyte cell cycle by up-regulating the expression of Ccnb1 and Cdk1. Unexpectedly, the upregulation of Ccnb1 by miR-199a-5p did not markedly affect cell cycle phases in mouse keratinocytes (Fig. 3D). Thus, these data suggest that the increased expression of Ccnb1 and Cdk1 is not sufficient to change cell cycle. Alternatively, the up-regulation of Ccnb1 and Cdk1 expression by miR-199a-5p may play other, yet unidentified roles in mouse keratinocytes. This may be more plausible than a role in cell cycle, because our previous study showed that increased miR-199a-5p expression does not induce proliferation in PAM212 cells [12]. The precise role of Ccnb1/Cdk1 up-regulation by miR-199a-5p in mouse keratinocytes is unclear. Functional annotation analysis revealed that expression of the genes involved in not only cell cycle (50 genes) but also in cell division process (34 genes) was changed by miR-199a-5p in PAM212 cells (Table 2). This may have resulted in the combined effect of no proliferation of cells with the up-regulation of Ccnb1. Additional study is necessary to elucidate the mechanism and effect of this regulation in mouse cells.
In addition, miR-199a-5p and miR-199b-5p can potentially regulate the same transcripts because they have identical seed sequence. Since expression of miR-199b-5p in skin keratinocyte or hair follicle has not been documented, further studies are required to address this question.
In conclusion, our data indicate that Ccnb1 expression is increased by miR-199a-5p in a mouse-specific manner. Although further studies are required to understand the roles of miR-199a-5p and Ccnb1, these results reveal a new evolutionary relationship between Ccnb1 and miR-199a-5p in mouse keratinocytes and thus make a contribution to miRNA biology.

Conclusions
MiR-199a-5p up-regulates Ccnb1 expression in a mouse-specific manner. These results indicate an evolutionary relationship between Ccnb1 and miR-199a-5p in mouse keratinocytes.