miR-193b-3p Promotes Proliferation of Goat Myoblast Through Directly Targeting IGF2BP1 and Activating Its Transcription

As a well-known cancer-related miRNA, miR-193b-3p is enriched in skeletal muscle but dysregulated in muscle disease. However, mechanism underpinning has not been addressed so far. Here, we probed the impact of miR-193b-3p on myogenesis by mainly using goat tissues and skeletal muscle satellite cells (MuSCs), with combined methods including RNA-seq to prole the transcriptome affected by miR-193b-3p, cell-counting kit-8 (CCK-8) and 5-ethynyl-2'-deoxyuridine (EdU) for cell proliferation assay, and RNA-RNA dual-labeled uorescence in situ hybridization (FISH) for RNA colocalization. and brown adipocytes suggest the function of miR-193b-3p in myogenesis We found out that only levels of LC3B and BECN1 were elevated by ~1.5 fold when These results conrmed that positively regulates myogenesis of MuSC, which may be mainly attributed to the activation of proliferation ability. in mics C2C12 myoblasts reported previously [24]. These results suggest that miR-193b-3p /IGF2BP1 axis exert activation in goat myogenesis which is mainly attributed to its nucleus function.


Abstract Background
As a well-known cancer-related miRNA, miR-193b-3p is enriched in skeletal muscle but dysregulated in muscle disease. However, mechanism underpinning has not been addressed so far.
Results miR-193b-3p is highly enriched in goat skeletal muscles, and ectopic miR-193b-3p promotes MuSCs proliferation and differentiation. Moreover, insulin-like growth factor-2 mRNA-binding protein 1 (IGF2BP1) is the most activated insulin signaling genes when overexpression miR-193b-3p and the miRNA recognition element (MRE) within IGF1BP1 3 untranslated region (UTR) is indispensable for its activation caused by miR-193b-3p. Consistently, expression patterns and function of IGF2BP1 were similar to those of miR-193b-3p in tissues and MuSCs. While the overexpression of miR-193b-3p failed to induce pax7 expression and myoblast proliferation when IGF2BP1 knockdown. Furthermore, miR-193b-3p destabilized IGF2BP1 mRNA but unexpectedly promoted levels of IGF2BP1 heteronuclear RNA (hnRNA) dramatically. Moreover, miR-193b-3p could enhance y luciferase activity when inserted upstream of its promoter, and induce neighboring genes of itself. However, miR-193b-3p inversely regulated IGF2BP1 and myoblast proliferation in mouse C2C12 myoblast. These data unveil that goat miR-193b-3p promotes myoblast proliferation via activating IGF2BP1 by binding on its 3 UTR.

Conclusions
Our novel ndings highlight the positive regulation between miRNA and its target genes in muscle development, which further extends the repertoire of miRNA functions.
Here, mainly using tissues and MuSCs from goat, we found out that miR-193b-3p is located in enhancer loci and enriched in skeletal muscle. Furthermore, the insulin-like growth factor-2 mRNA-binding protein 1 (IGF2BP1), an anti-apoptotic gene, certainly promotes cell proliferation [33,34] and is directly targeted and activated by miR-193b-3p transcriptionally via seed-matching sites in the 3 UTR. Besides, the miR-193b-3p /IGF2BP1 axis mainly promotes myoblast proliferation. Our research provides a novel mechanism of miRNA activating its targeting gene transcriptionally.

Ethics declarations
All experimental protocols present here were strictly conducted following the Regulations for the Administration of Affairs Concerning Experimental Animals (Ministry of Science and Technology, China), and entirely approved by the Animal Care and Use Committee, Sichuan Agricultural University.

Animals and samples collection
A total of 12 female goat fetuses and kids aged 45, 60, and 105 days of gestation, and 3 days postnatal (n = 3 per stage) were randomly chosen and humanely sacri ced at the Jianzhou Big-Eared goats farm (Sichuan province, China). Tissue samples from six organs, including heart, liver, lung, spleen, kidney, and longissimus dorsi (LD) muscle were quickly collected and snap-frozen in liquid nitrogen for further study.
Finally, the puri ed Pax7 + MuSCs were stored in liquid nitrogen.
For the gain and loss function study, when cells at 80%~90% con uence, the GM was replaced with DMEM supplemented with 10% FBS. Then cells were transfected using Lipofectamine 3000 (Invitrogen, USA) with interfering RNA (si-NC, in-193b-3p and siIGF2BP1) or overexpression plasmid (pCtrl/pIGF2BP1) and chi-miR-193b-3p mimic (mi-193b-3p, RiboBio, China) at indicative concentrations, according to the manufacturer's direction. The transfected cells were kept in GM or replaced with DMEM supplemented with 2% FBS to induce differentiation and collected for RNA/protein assay at 48 h and immuno uorescence stained with EdU at indicative time points, or with MyHC (Myosin Heavy Chain) antibody at 7 th d post differentiation.
Given that culturing cells are easily contaminated by mycoplasma, which leads to severely spurious results, we monitored and even employed polymerase chain reaction (PCR) to directly detect mycoplasma contamination (TRansgen Biotech, China) in culturing media before transfection and harvesting [63], to ensure cells were mycoplasma-free (Supp Fig.4B).
For the luciferase reporter assays, goat MuSCs (~1×10 4 cells per well), mouse C2C12 myoblasts (1 × 10 5 cells per well, Thermo, USA), and HeLa cells (~2 × 10 5 cells per well, ATCC, USA) were cultured in GM on 24-well plates and cotransfected with vectors (640 ng/well for each vector) using Lipofectamine 2000 reagent when cells were grown to 80 %-90 % con uence. After transfection for 48 h, the transfected cells were lysed, and the activities of re y and Renilla luciferase were measured using a dual-luciferase reporter assay system (Promega, USA), according to the user's guidelines.
To provide solid interfering results for IGF2BP1, three small interfering RNA, including siRNA1, CAGCTCCTTTATGCAGGCT; siRNA2, CTCCTTTATGCAGGCTCCA; and siRNA3 CTTTATGCAGGCTCCAGAG, targeting goat IGF2BP1 mRNA were designed and synthesized in RiboBio (Guangzhou, China). We transfected them separately into MuSCs at two concentrations (50 nM and 100 nM) and then quanti ed IGF2BP1 mRNAs. The results suggest that IGF2BP1 levels are e ciently decreased at 100 nM for each siRNA (Supp Fig. 6), thus in the following experiment, we pooled them at 100 nM concentration. Besides, the miR-193b-3p miRNA mimic (mi-193b-3p) and control (mi-Ctrl), as well as inhibitor (in-193b-3p) and control (in-Ctrl) were ordered from RiboBio (Guangzhou, China).

RNA extraction and qPCR
Following the manufacturer's instruction, total RNAs were extracted from tissues or cultured cells using RNAiso Plus reagent (TaKaRa, Japan), and roughly quali ed by using 1.5% agarose gels electrophoresis and NanoDrop 2000c Spectrophotometer.
Then the quali ed RNAs (~1 mg) were reversely transcribed into cDNA by using the PrimeScript™ RT reagent Kit with gDNA Eraser or miRNA PrimeScript RT reagent Kit (Takara, Japan) separately for mRNA or miRNA assay. Then according to the manufacturer's guide, RNA levels of target genes in these cDNA were accurately measured by using real-time PCR (RT-qPCR) in the Bio-Rad CFX96 system (Bio-Rad, USA) with SYBR Premix Ex TaqTM II (Takara, Japan). Each treatment performed at least triply three independent times, and each sample triplicates in qPCR. Moreover, to enhance the accuracy, three housekeeping genes in goat ACTB (Actin Beta), SDHA (Succinate Dehydrogenase Complex Flavoprotein Subunit A), and PGK1(Phosphoglycerate Kinase 1)) and mouse (ACTB, GAPDH, and Hprt (Hypoxanthine Phosphoribosyltransferase 1)) were used as an internal control. The 2 −△△Ct or 2 −△Ct methods were employed to calculate the relative RNA levels of target genes. The detailed information for primers was listed in Supplementary Table 1.
Western Blotting and Immuno uorescence for protein analyses To analyze protein levels of the target genes via western blotting (WB), we extracted total proteins in lysed cells by using radioimmunoprecipitation assay (RIPA) (Beyotime, China). After quanti ed through the bicinchoninic acid (BCA) method (Beyotime, China), protein samples (~20 μg) were sequentially loaded for electrophoresis, transferred on polyvinylidene uoride (PVDF) membranes (Millipore, USA), incubated with primary antibodies for IGF2BP1, MyoG, MyoD, and PAX3 (Abcam, UK) at 4 °C overnight and secondary IgG (Beyotime, China) for 2 h. After the addition of horseradish peroxidase (HRP) substrate, protein bands were exposed via electrochemiluminescence (ECL) (Pierce, USA) and analyzed by using Image Software. GAPDH (BOSTER, China) worked as a loading control.

CCK-8 assay for cell proliferation
Cells were inoculated in 96-well plates with 2 × 10 3 cells per well initially (~ 30% con uency) and cultured in GM for 2 days, then transfected with different vectors according to protocols described above. Every 24 h, the absorbance value for each sample was measured at 450 nm by using a Microplate Reader (Model 680, Bio-Rad) after added 10 μL of Cell-Counting Kit-8 (CCK-8) reagents (Kumamoto, Japan) to the cells for 2 h.
Images were captured less than 5 h after hybridization by using Nikon DS-U3 in Nikon Eclipse Ti-SR and then analyzed by employing the CaseViewer software.

Nuclear-cytoplasmic fractionation
To accurately quantify expression levels of the target genes in the cytoplasm and nuclear separately, MuSCs proliferating for 2 d and differentiating for 7 d were fractionated into nuclear/cytoplasmic parts by using Nuclear/cytoplasmic fractionation & Nuclear RNA Puri cation Kit (Norgen Biotek, Canada), as described before [64]. In brief, the total RNA was separately extracted from both the supernatant (cytosol fraction) and pellet (nuclei), retro-transcribed into cDNA, and quanti ed targeted genes by using RT-qPCR. The nuclear-enriched U6 and cytoplasm-enriched 18S were used as control.
ActD analysis According to methods described previously [65], we analyzed the effect of miR-193b-3p on IGF2BP1 mRNA stability by using actinomycin D (ActD, Sigma) to block newly mRNA synthesis. In general, MuSCs were cultured in 12-well (~1 × 10 5 cells per well) plates and separately transfected with miR-193b-3p mimic, mimic NC, inhibitor, and inhibitor NC for 24 h. Then cells were treated with actinomycin D (5 μg/ml) and harvested at 0, 60, and 120 min post-treated. The total RNA was extracted from each sample and nally, the remaining levels of IGF2BP1 mRNA were quanti ed by using the canonical RT-qPCR method. At least each treatment contained three independent biological repetitions.

mRNA-seq and bioinformatic analyses
Library preparation and poly(A) selection mRNA-seq was performed at Novogene Company (Beijing, China). Brie y, using NEBNext® UltraTM RNA Library Prep Kit Illumina®, quali ed total RNA samples (200 ng) extracted from cells transfected with miR-193b-3p mimic or control (n=4 per group) were used to isolate polyA fraction (mRNA), following by fragmentation and generation of double-stranded cDNA. Libraries were sequentially evaluated by Qubit 2.0 Fluorometer, Agilent 2100 bioanalyzer, and RT-qPCR. Then quali ed libraries were sequenced in Illumina HiSeq 2500 platform ((Illumina, USA) with a 2x150 bp pair-end.
The raw data of each sample were ltered to obtain the clean data (>8 Gb per sample), then amount to 97.09 97.37 clean reads were quickly and accurately mapped onto the Capra_hircus _ARS1 reference genome (ftp://ftp.ncbi.nlm.nih.gov/genomes/all/GCF/001/704/415/GCF_001704415.1_ARS1/GCF_001704415.1_ARS1_genomic.fna.gz) using HISAT2, among which as high as 93.94 94.3 clean reads were uniquely mapped. The read count for each one was calculated using fragments per kilobase of transcript sequence per million base pairs sequenced (FPKM) value to evaluate gene expression. Differentially expressed genes (DEGs) between samples were canonically identi ed by DESeq2 R package vision Transcriptome clustering based on principal component analysis (PCA) indicates that miR_193b-3p mimics or control (mi-Ctrl)treated cells are distinctly separated. Function Enrichment Analyses of DEGs, including Gene Ontology (GO) enrichment analysis and KEGG pathway, were implemented by using the DESeq2 with P-adj < 0.05 (adjusted via Benjamini-Hochberg) was considered signi cantly enriched.

Statistical analysis
Unless stated otherwise, data presented here are shown as mean ± SD (standard deviation) from at least triplicate independent samples or animals. The unpaired two-tailed t-test in Graphpad Prism 7 was employed to evaluate means' difference, with signi cant threshold value set at ***P < 0.001 **P < 0.01, *P < 0.05, and § P < 0.1.
Although several studies unveil the adipogenesis-inducing function of miR-193b signaling [35,36] and anti-tumor regulation of miR-193b in many cancers [17,18,37], intriguingly, expression levels of MIR193BHG in human skeletal muscle ranked second among 53 tissues (Fig.1B, upper panel; Supplementary Fig.S1). In particular, hsa-miR-193b-3p is highly dominant than that of hsa-miR-193b-5p (Fig.1B, lower panel). Using RNAs extracted from goat tissues and cultured MuSCs isolated from newborn goat longissimus dorsi (LD) muscle, we also found out that compared with other internal organs, extinguished miR-193b-3p levels were presented in goat muscles from embryos at 75 d of gestation (E75) and maintained the high levels at birth (Fig.1C).
Besides, miR-193b-3p abundance continually increased in LD muscles from embryonic to newborn goats (Fig.1C). Moreover, miR-193b-3p levels were steady in MuSC culturing in GM but dramatically elevated when shifting to DM (Fig.1C). These suggest that miR-193b-3p most likely play critical roles in mammal skeletal muscle development.

miR-193b-3p targets and induces IGF2BP1 expression
To systematically anchor the overall effect of miR-193b-3p on global gene expression in myoblast, we performed transcriptome pro ling in MuSCs treated with miR-193b-3p mimic (mi-193b-3p, n=4) and blank control (mi-Ctrl, n=4). Total RNA was extracted, quali ed ( Supplementary Fig. S3A, B), and then sequenced to analyze protein-coding genes affected. Over 8.29 Gb clean bases were identi ed for each sample; the total_mapped reads and unique_mapped reads were as high as 97% and 94%, respectively ( Supplementary Fig. S3C); PCA analysis suggests that these samples were expectedly distinguished into two groups ( Supplementary Fig. S3D).
Among the detected 20,268 transcripts, 471 ones were differentially expressed (DE, padj<0.05, Fig 4A), and these DE genes were signi cantly enriched in molecular function GO terms, including several cell growth and cell cycle, as well as insulin-like growth factor binding (GO:0005520, Padj = 0.015, Fig 4B). Since insulin-like growth factors (IGFs) play critical roles in skeletal muscle hypertrophy and regeneration through interplaying with the myogenic regulator like MyoD and MyoG [41], we analyzed all 17 insulin family genes detected and found out that out of 14 miR-193b-3p-induced genes, IGF2BP1 was the most dramatically affected one (Fig 4C).

Function of miR-193b-3p on myoblast proliferation is IGF2BP1-dependent
Previously reported that IGF2BPs also named IMPs participate development of many tissues through cell proliferation, especially HMGA2-IGF2BP2 axis is critical in regulating activation of satellite cells toward myogenesis [42]. To address the function of IGF2BP1 in myogenesis, we ampli ed the CDS of IGF2BP1 and constructed an overexpressing vector (pIGF2BP1, 3 μg/mL), as well as designed small interfering RNAs against IGF2BP1 (siIGF2BP1, 100 nM, Supplementary Fig. S5) and transfected them into MuSC separately. As shown in Fig.6A, ectopic IGF2BP1 (pIGF2BP1) dramatically promoted mRNA (p<0.001) and protein levels of IGF2BP1, while siIGF2BP1 resulted in ~20% downregulation of IGF2BP1 mRNA (p<0.01) but no dominant protein density altered, which mainly resulted from the de ciency of endogenous IGF2BP1 in cells. Moreover, pIGF2BP1-treated cells contained higher transcripts of Pax7 (p<0.001), MyoD (p<0.01), and MyoG (p<0.001), while de ciency of IGF2BP1 (siIGF2BP1) decreased all myogenic genes expression insigni cantly (p>0.10) (Fig.6B). Additionally, the protein levels of PAX7 and MyoG elevated in pIGF2BP1-treated MuSCs con rmed the mRNA abundance altered (Fig. 6C). Correspondingly, EdU + cells were increased in pIGF2BP1 while decreased in siIGF2BP1 treated cells compared to the control group (p<0.01, Fig.6D). These results phenocopy overexpression or de ciency of miR-193b phenotype, respectively. Furthermore, to validate whether function of miR-193b-3p is mediated by IGF2BP1, we transfected miR-193b-3p mimic (mi-193b-3p) and simultaneously interfering IGF2BP1 (siIGF2BP1) in MuSC. As shown in Fig. 6E, the abundance of PAX7 and PCNA were insigni cantly altered (p>0.05) when IGF2BP1 de ciency, while transcripts of differentiation-related genes, including MyoD, MyoG, and Myomaker were still dramatically elevated (p<0.05). Furthermore, comparing with control, the fold change of EdU + cells at 24 h, 48 h, and 72 were insigni cantly altered in cotransfected cells (Fig.6E, p>0.10), suggesting that instead of myogenic differentiation, the proliferation of myoblasts induced by ectopic miR-193b-3p cloud be e ciently reversed by IGF2BP1 de ciency.

Goat miR-193b-3p activates IGF2BP1 transcriptionally
To address the spatial localization of miR-193b-3p and IGF2BP1 mRNA in cells, we designed uorescence-labeled oligonucleotide probes complementary to them and employed RNA-RNA dual-labeled uorescence in situ hybridization (FISH). We found out that miR-193b-3p and IGF2BP1 mRNA were dominantly distributed and almost entirely colocalized in the cytoplasm of MuSC culturing in growth (GM 2d) and differentiation media (DM 7d, Fig.7A). Typically, over 84% of miRNAs bind complementary target RNAs and trigger their destruction [26]. To further monitor the effect of miR-193b-3p on turnover of endogenous IGF2BP1 mRNA, we treated culturing MuSCs with actinomycin D (ActD, 5 μg/mL) to compromise transcription effectively [43], and then quanti ed the remaining levels of IGF2BP1 transcripts at three sequential time points. Expectedly miR-193b-3p mimic dramatically degraded IGF2BP1 mRNA in particular at 120 min (p<0.01), whereas interfering miR-193b-3p exhibited no effect on IGF2BP1 mRNAs stability ( Fig.7B and Supplementary Fig. S6), just like the typical relationship between cytoplasmic miRNA and its targets [44]. These suggest that otherwise mechanism implicates in the activation of miR-193b-3p on goat IGF2BP1.
Furthermore, using data from UCSC, we found that compared with that in GM12878, H1-hESC, the hsa-miR-193b region in human skeletal muscle myoblasts (HSMM) is enriched with H3K4Me3 and H3K27Ac histone mark ( Fig.7E upper panel, and Supplementary Fig. S7A), a signal used for a promoter and active enhancer discovery, respectively [47]. Moreover, TF factors such as YY1 (Yin and Yang 1) and CTCF (CCCTC-Binding Factor) that facilitates interactions including enhancer-promoter in a multi-gene cluster [48] are also enriched in hsa-miR-193b region (Data no shown). Whereas mouse miR-193b barely overlap with eRNA markers, including H3K4Me3 and H3K27Ac in limb from embryonic mouse aged 10 d (Supplementary Fig.S7B). On the other hand, both human and mouse IGF2BP1 3 UTR region lack histone modi cation ( Supplementary Fig. S8A, B), but the former enriches transcription factors including AGO2 (Argonaute RISC Catalytic Component 2), POLR2A (RNA Polymerase II Subunit A), CTCF (data not shown), suggesting that the activation of miR-193b-3p on IGF2BP1 is most likely attributed to the enhancer-related characteristics of goat miR-193b-3p.
To validate whether the goat miR-193b gene is enhancer-related, just like hsa-miR-193b-3p, we split it into two, ampli ed ~60 bp 5p and 3p region (Fig.7E lower panel), inserted them into a PGL3-Promoter vector separately (named pre-193b-5p and pre-193b-3p), and transfected these vectors into MuSC and HeLa. As shown in Fig.8F, comparing to the basic (pBasic) and control (pCtrl), pre-193b-3p e ciently promoted F-Luc/R-Luc levels in GM and DM myoblasts, as well as in HeLa cells (p<0.01 or 0.001), while miR-193b-5p enhanced F-Luc/R-Luc values greatly in DM cells (p< 0.001), implying that sequence of goat miR-193b could activate gene expression.

miR-193b-3p represses IGF2BP1 in mouse C2C12 myoblast
Although the mature sequence and the neighboring genes of miR-193b-3p are completely identical in goat and mouse (Fig.1A and Fig.8 A), previously reported that ectopic miR-193b blocks the shift of C2C12 myoblasts toward myogenesis [49]. To validate the function of miR-193b-3p in mouse myogenesis, we transfected the miR-193b-3p mimic (100 nmol) into mouse C2C12 myoblasts and found out that miR-193b-3p decreased IGF2BP1 mRNA levels (p<0.05); on the contrary, an inhibitor of miR-193b-3p increased transcripts of IGF2BP1 (p<0.10; Fig. 9A), which is opposite to those in goat. Meanwhile, levels of cell number-related genes, including PCNA, cyclin D1, and caspase1 mRNA, did not change signi cantly (Fig. 9A). Furthermore, EdU + cells were also correspondingly downregulated in miR-193b-3p mimic treated cells or upregulated when de ciency of miR-193b-3p, as shown in Fig. 9B. Moreover, overexpression or de ciency of miR-193b generally failed to altered expression of its neighboring genes correspondingly, including MKL2, PARN, and BFAR, in GM and DM mouse C2C12 signi cantly (Fig. 9C). These results suggest that at least miR-193b-3p exhibited no activation on IGF2BP1 in mouse myogenesis.

Discussion
Muscle formation is precisely mastered by a handful of hierarchical gene cascades covering protein-coding genes such as MRFs as well as non-coding genes like miRNAs. It is well-accepted that miRNAs play critical roles in muscle development through inversely ne-tune gene enrichment canonically [8,9]. Nevertheless, a few miRNAs have gained much attention for newly appreciated activation roles in directly targeted genes [50]. Here, we report that the anti-tumor gene miR-193b-3p induces proliferation and differentiation of goat MuSCs, phenocopying IGF2BP1 overexpression phenotype. Meanwhile, IGF2BP1 knockdown impairs myogenic proliferation but not differentiation of MuSCs induced by miR-193b-3p supplementation. Furthermore, miR-193b-3p in the nucleus directly targets 3 UTR of IGF2BP1 gene and blooms IGF2BP1 abundance transcriptionally and overexpression of goat miR-193b-3p induces its neighboring genes. Besides, miR-193b-3p destructs IGF2BP1 mRNA in cytoplasm, which is in line with the inverse regulation of miR-193b-3p on IGF2BP1 transcripts in mics C2C12 myoblasts reported previously [24]. These results suggest that miR-193b-3p /IGF2BP1 axis exert activation in goat myogenesis which is mainly attributed to its nucleus function.
anti-proliferation vs. pro-proliferation function of miR-193b in muscle In cancer tissues, miR-193b-3p functions oppositely as an anti-tumor [17,51] or tumor-inducer [52] under differed disease conditions. As for the normal organ development, miR-193b-regulated signaling induces adipogenesis for adipose-derived stem cells (Mazzu, Hu et al. 2017, Mazzu, Hu et al. 2019 or even shifted from C2C12 cells [24]. On the contrary, de ciency of miR-193b induces cells from mouse BAT to the muscle lineage with upregulated levels of muscle-speci c genes, including Cmk, Myf5, Myf6, MyoD, and MyoG [24]. Although muscle diseases like Duchenne muscular dystrophy 1 (DMD1) and DMD2 share similar symptoms, including gradually worsening muscle loss and weakness, miR-193b is accumulating in DMD1 patients [22] but downregulated in DMD2 [23], suggesting their miR-193b-related intrinsic mechanism differs. We found out that miR-193b-3p induces goat MuSCs proliferating and differentiating toward myotube, but using C2C12 myoblasts, we validate the suppressing effect of miR-193b-3p on myogenic proliferation reported previously [24]. Intriguingly, goat miR-193b-3p is predominantly located in the cytoplasm and triggers the destruction of IGF2BP1 mRNA, coinciding with the inverse regulation of mouse miR-193b-3p on IGF2BP abundance. Nevertheless, miR-193b-3p induces IGF2BP1 levels accumulating and goat MuSCs proliferating and differentiating toward myotube, which is differed from C2C12 myoblast. These suggest that cytoplasmic miR-193b-3p performs the canonical degradation of IGF2BP1 mRNA in myoblast but otherwise mechanism implicated in its activation on the goat IGF2BP1.
Activation of miR-193b on IGF2BP1 transcription miRNAs are tacitly assumed as cytoplasm located and post-transcriptional negative regulators of its target gene expression [26,27]. With the utilization of newly developed high-throughput assessment of miRNA, numerous mature miRNAs are currently identi ed in the nucleus [45]. Several models elaborating functions of nuclear miRNAs at both transcriptional (transcriptional gene silencing, TGS; transcriptional gene activation, TGA) have been constructed, in which miRNA targets and binds nascent RNA transcripts, gene promoter, or enhancer regions by a sequence complementary and exert further effects via recruiting additional epigenetic and(or) transcriptional factors or altering epigenetic modi cations [45]. For example, miR-483-5p promotes the IGF2 transcription by targeting the P3 5 UTR and promoter, accompanied by enrichment of activating-enhancer marks, including H3K4me3 and H3K27ac [32]. Similarly, the transcription activation of miR-373 on both E-cadherin [53], as well as miR-589 on COX-2 (Cyclooxygenase 2) [46], are via miRNA-promoter-related manner. Given the close relationship between gene enhancer and promoter [54,55], we speculate that most likely enhancer-related characteristics of miRNA-target pair, including enhancer-related miRNA or enhancer-related target, play a critical role in the activation of miRNAs on its target.
Currently, there are two distinct types of enhancer-related miRNAs: the super-enhancers (SEs)-miRNAs in which SEs neighbor celltype-speci c abundant miRNAs (like muscle-speci c miR-1 and miR-133 ), boost pri-miRNA processing, and regulates mRNA stability [56]; the enhancer-overlapped-miRNAs including miR-26a-1, miR-339, miR-3179 and miR-24-1/2 that transcribed from enhancer region, exert enhancer activity including activating its neighboring gene expression in cis as well as global genes in trans via triggering its targeted enhancers [31]. Given that goat miR-193b-3p inducing IGF2BP1 through the complementary binding on its 3 UTR that lack enhancer-related marks, we compared human and mouse data in the UCSC genome browser and found out that not the mouse but human miR-193b-3p gene locus is enhancer-overlapped (enriched with H3K27ac and H3K4Me3). Further, we found out that goat miR-193b-3p is partially nucleus-distributed and transcriptionally activates IGF2BP1 at both mRNA and protein levels, and this activation is completely abolished when mutated their base-pairing seed region.
Besides, both pre-miR193b-5p and 3p could activate gene expression when inserted upstream of F-Luc, and both miR-193b-3p mimics and pri-miR-193b gene sequence promotes the expression of its neighboring genes, including MKL2, BFAR, and BFAR in GM and DM MuSCs, coinciding with the synergistic function and same phenotype between a mature miRNA and its passenger strand reported previously [57]. Nevertheless, in mouse C2C12 myoblast, miR-193b-3p negatively regulates IGF2BP1 abundance and barely affects its neighboring gene expression. These suggest that goat miR-193b-3p is most likely function as an enhanceroverlapped miRNA in myogenesis, and the difference of histone modi cation in the miR-193b gene region may be the main contribution for the species-related inconsistency of miR-193b-3p function.
Other factors possibly involved in miR-193b-3p/IGF2BP1 axis IGFs play critical roles in myogenesis because they can uniquely promote muscle cell proliferation and differentiation, and induce MRFs mutually [41]. In goat MuSCs treated with miR-193b-3p mimics, IGF2BP1 was the most dramatically affected IGFs gene. IGF2BP1 promotes proliferation in mouse embryonic broblasts [58], and IGF2BP2 mediates translation of myogenic proliferation genes [59]. Similarly, we found out that IGF2BP1 could induce myogenic proliferation and differentiation of goat MuSCs, while cotransfection results imply that miR-193b-3p/IGF2BP1 axis mainly affects proliferation. Given the many-to-many [60] and stage-speci c [14] targeting manner between miRNAs and their targets, as well as other factors like the RNA A-to-I editing modi cation of miR-193b-3p [50] and m6A-dependent manner of IGF2BPs on the stability and storage of their target mRNAs [61], we speculate that part of these factors is likely involving in miR-193b-3p/IGF2BP1 pathway in promoting myoblast proliferation.

Conclusions
Taken together, here we proposed that similar to previously reported miR-24, the enhancer-overlapped miRNA-193b-3p carries a dual function ability: activating its target gene within the nucleus as well as repressing mRNA of the target gene in the cytoplasm. It is worth noting that miR-193b-3p in the nucleus exerts its function on promoting goat MuSCs' myogenesis by directly targeting 3 UTR of the IGF2BP1 locus and inducing its expression, which is beyond previously reported models of nucleus-miRNA activation on targets [45]. Our research further extends the repertoire of miRNA functions.

Consent for publication
Not applicable.

Competing interests
The authors declare that no any type of potential con ict of interest implicated.

Availability of data and materials
The accession number for the RNA sequencing data reported here is NCBI BioProject: PRJNA665306.

Funding
This research was nancially supported by the National Natural Science Foundation of China (31802048 and 31672402).

Authors' Contributions
The study was mainly conceived by Li   The sequence conservation and abundance of miR-193b-3p in human and goat tissues. A, miR-193b-3p is highly conserved among 100 verts in UCSC genome browser; the 10th nucleotide of human miR-193b-3p is different from that of other animals, including goat, mouse, rat, bovine, and chicken. B, expression of human MIR193BHG, miR-193b-5p, and miR-193b-3p in multiple tissues from GTEx. An arrow marks the skeletal muscle. C, levels of miR-193b-3p in tissues from embryonic 75 days (E75) and 3 d postnatal (B3) goat kids, or longissimus dorsi (LD) muscle from E45 to B3, as well as in cultured skeletal muscle satellite cells (MuSCs). RNA levels are quanti ed by employing RT-qPCR and calculated by the 2−△△Ct methods, with normalization to β-actin (ACTB) and values of the rst tissue or timepoint set to 1. Data are shown as mean ± SD (standard deviation).

Figure 2
Overexpression of miR-193b-3p promotes the proliferation and differentiation of goat skeletal muscle satellite cells (MuSCs). A and B, miR-193b-3p elevates the abundance of myogenic genes. The cultured goat MuSCs were transfected with miR-193b-3p mimics or control (mi-193b-3p or mi-Ctrl, 50 nM), and cells were kept in growth media (GM) for 48 h or continually induced to differentiation for 48 h. The RNA levels of the target gene are quanti ed via RT-qPCR and calculated using the 2−△△Ct, with βactin (ACTB) as an internal control and values of the mi-Ctrl set to 1. Using antibodies against PAX7, MyoG, and GAPDH (glyceraldehyde-3-phosphate dehydrogenase), western blotting (WB) assay was typically performed to detect protein levels of MyoD and MyHC in MuSCs treated with miR-193b-3p or mi-Ctrl. GAPDH works as a loading control. C, Cell proliferation is measured using Cell-Counting Kit-8 (CCK-8). Cells in 96-well plates (~2 × 103 cells per well) were transfected (mi-193b-3p or mi-Ctrl, 50 nM). And then every 24 h post-transfection (PT), the absorbance (OD) value at 450 nm is measured individually after adding 10 μL of CCK-8 reagents for 2 h. D, miR-193b-3p promotes EdU+ cells and myotube formation. Left panel, Representative immuno uorescence images of EdU and MyHC staining cells after transfected mi-193b-3p or mi-Ctrl (50 nM). Cells are cultured in GM consisting of 10 μM EdU, stained with anti-DAPI (blue) and anti-MyHC (red). Scale bar = 100 μm. Right panel, The EdUstained cells (ratio of EdU+ myoblasts to all) are evaluated using randomly selected elds and normalized to control. Each treatment is at least tripled. Data are shown as mean ± SD. An unpaired two-tailed t-test is used to evaluate the means difference. Data are shown as mean ± SD. *** P < 0.001, ** P < 0.01, * P < 0.05, and § P < 0.1.

Figure 3
De ciency of miR-193b-3p retards myogenesis of skeletal muscle satellite cells (MuSCs). A and B, de ciency of miR-193b-3p suppresses abundance of myogenic genes. The cultured goat skeletal muscle satellite cells (MuSCs) were transfected with interfering RNA against miR-193b-3p or control (in-193b-3p or in-Ctrl, 50 nM) when reached to ~50 % con uency. Then before be sampled to extract total RNA and protein, cells are kept in growth media (GM) for 48 h or induced to differentiation (DM) for 48 h. The RNA levels of the target gene are quanti ed via RT-qPCR and calculated using the 2−△△Ct, with β-actin as an internal control and values of the mi-Ctrl set to 1. Using antibodies against PAX7, MyoG, and GAPDH (glyceraldehyde-3-phosphate dehydrogenase), western blotting (WB) assay is typically performed to detect protein levels of MyoD and MyHC in MuSCs treated with in-193b-3p or in-Ctrl. GAPDH works as a loading control. C, de ciency of miR-193b-3p delays absorbance (OD) value of proliferation cells. Cells in 96-well plates (2 × 103 cells per well) are transfected (in-miR-193b-3p or in-Ctrl, 50 nM). Every 24 h post-transfection (PT), the absorbance (OD) value at 450 nm is measured individually after the addition of 10 μL of Cell-Counting Kit-8 (CCK-8) reagents for 2 h. D, de ciency of miR-193b-3p retards EdU+ cells and myotube formation. Representative images of EdU and MyHC staining cells transfected with in-193b-3p or in-Ctrl (50 nM). Cells are cultured in GM consisting of 10 μM EdU, stained with anti-DAPI (blue) and anti-MyHC (red). Scale bar = 100 μm. The EdU-stained cells are evaluated using randomly selected elds and normalized to control. Each treatment is at least tripled. Data are shown as mean ± SD. An unpaired two-tailed t-test is used to evaluate the means difference. Data are shown as mean ± SD. ** P < 0.01, * P < 0.05, and § P < 0.1.

Figure 4
Insulin-like growth factor 2 binding protein 1 (IGF2BP1) is the most miR-193b-3p-affected insulin genes. A, a volcano plot displays log2foldchange (log2FC) against -log10padj from the t-test for all the transcript detected in mRNA-sequencing. B, Functional enrichment analyses of differential expression genes (DEGs) were implemented using the clusterPro ler R package (v. 3.4.4) with Padj < 0.05. A red box marks insulin-like growth factor binding (GO:0005520, Padj = 0.015). C, a volcano plot for the total 17 insulin family genes detected in RNA-seq. The green dot indicates that gene expression is decreased by miR-193b-3p mimics, while others are elevated compared to mimics control (mi-Ctrl).  IGF2BP1 promotes myoblast proliferation. A, levels of IGF2BP1 are successfully disturbed. mRNA and protein levels of IGF2BP1 were measured in MuSCs 48 h post-transfection of pIGF2BP1(3 μg/mL) or siIGF2BP1(100 nM). The RNA Levels of IGF2BP1 were quanti ed via RT-qPCR and calculated using the 2−△△Ct. B and C, the effect of IGF2BP1 on myogenic genes. RNA levels were quanti ed via RT-qPCR and calculated using the 2−△△Ct, normalized to β-actin (ACTB), with values of control set to 1. Protein levels of PAX7, MyoD, and MyoG are quanti ed using typical western blotting (WB) assay, with GAPDH as a loading control. D, IGF2BP1 elevates EdU+ cells. Representative immuno uorescence images of EdU staining cells treated with pCtrl and pIGF2BP1(3 μg/mL, red) or siCtrl and siIGF2BP1(100 nM, green), cultured in GM consisting of 10 μM EdU (red or green) for 48 h, stained with anti-DAPI (blue). Scale bar = 100 μm. The fold change of EdU+ cells are evaluated using randomly selected elds and normalized to control. E, interfering IGF2BP1 retards miR-193b-3p-induced myogenic proliferation. Total RNA in cells cotransfected with miR-193b-3p mimics (mi-193b-3p, 50 nM) and interfering RNA against IGF2BP1 (siIGF2BP1,100 nM) or control were extracted 48 h post-transfection of in growth media (GM) or continually differentiated for 48 h (DM). RNA levels of target genes were quanti ed via RT-qPCR and calculated using the 2−△△Ct. Representative immuno uorescence images of EdU and DAPI (blue) staining cells post-transfection for 24, 48, and 72 h. Scale bar = 100 μm. The fold change of EdU+ cells are evaluated using randomly selected elds and normalized to control. Data are shown as mean ± SD. An unpaired two-tailed t-test is used to evaluate the means difference. Data are shown as mean ± SD. *** P < 0.001, ** P < 0.01, and * P < 0.05. Figure 9 miR-193b-3p negatively regulates IGF2BP1 and suppresses the proliferation of mouse C2C12 myoblast. A and B, miR-193b-3p decreases transcripts of IGF2BP1 but barely affects the abundance of its neighboring genes in C2C12. RNA Levels of genes are measured in C2C12 transfected with miR-193b-3p mimic (mi-Ctrl and mi-193b-3p, 50 nM) or interfering RNA (in-Ctrl and in-193b-3p, 50 nM) for 48 h in GM or 48 h in DM with qPCR. All values are normalized to geomean of ACTB, GAPDH, and Hprt (Hypoxanthine Phosphoribosyltransferase 1). C, miR-193b-3p deduces EdU+ cells in C2C12. Representative immuno uorescence images of EdU staining cells in miR-193b-3p-disturbed C2C12 (mi-Ctrl and mi-193b-3p, in-Ctrl and in-193b-3p, 50 nM) (left panel).
C2C12 are cultured in GM consisting of 10 μM EdU (red or green) for 48 h, then stained with anti-DAPI (blue). Scale bar = 100 μm. The fold change of EdU+ cells (ratio of EdU+ myoblasts to all) are evaluated using randomly selected elds and normalized to control. Each treatment is repeated ten times. Data are shown as mean ± SD. An unpaired two-tailed t-test is used to evaluate the means difference. Data are shown as mean ± SD. ** P < 0.01, * P < 0.05, and § P < 0.1.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.