Mice were maintained and bred in specific pathogen-free facilities of the Department of Laboratory Animal Resources at the Yonsei University College of Medicine. C57BL/6 mice were purchased from Jackson Laboratory (Bar Harbor, ME, USA) and Orient Bio (Seongnam, Republic of Korea). Animal care and experiments were conducted according to the guidelines and protocols approved by the Institutional Animal Care and Use Committee (IACUC) of Yonsei University College of Medicine. Only healthy male mice at 6 to 12 weeks of age were used throughout this study.

Chinese hamster ovary (CHO) cells (CHO-S cells; Gibco, Life Technologies, Carlsbad, CA, USA) were cultured in DMC7 medium composed of DMEM containing L-glutamine, high glucose, and pyruvate (HyClone, Logan, UT, USA) and 7% fetal calf serum (FCS; HyClone) supplemented with 1× solutions of non-essential amino acids and antibiotic-antimycotic (HyClone). The following conjugated antibodies were purchased from BioLegend (San Diego, CA, USA): APC-Cy™7-conjugated anti-I-A/I-E, anti-CD11c, anti-B220/CD45R, anti-CD45, anti-Rat IgG2b κ Isotype Control; Alexa Fluor® 647-conjugated anti-CD117; APC-conjugated anti-B220/CD45R, anti-PDCA-1 (BST2, CD317), anti-CD135; PE-Cy™7-conjugated anti-Ly6G, anti-CD3, anti-CD11c, anti-CD19, anti-Ter119, anti-DX5, anti-I-A/I-E, anti-NK1.1, anti-Gr-1, anti-Sca-1; PerCP-Cy™5.5-conjugated anti-Ly6C, anti-CD11c, anti-CD24, anti-CD117, anti-CD172α (Sirpα); PE-conjugated anti-SiglecH, anti-CD11c, anti-CD115, anti-CD117, anti-CD135, anti-Armenian Hamster (AH) IgG Isotype Control; FITC-conjugated anti-CD172α; Alexa Fluor® 488-conjugated anti-Ly6C, anti-CD172α, anti-PDCA-1, anti-CD24, anti-CD45.2; Brilliant Violet™ 421 (BV421)-conjugated anti-CD11c, anti-CD45.1, anti-CD45.2, anti-CD115, anti-AH IgG Isotype Control; biotin-conjugated anti-NK1.1, anti-DX5, anti Ter119, anti-Ly6G, anti-CD3, anti-CD19, anti-CD135. LIVE/DEAD® fixable dead cell stain kit (Life Technologies) and propidium iodide (Roche, Indianapolis, IN, USA) were purchased and used according to the manufacturers' instructions.

The cDNA of mouse Flt3L was cloned by RT-PCR of total splenic RNA from C57BL/6 mice, and was used to generate a construct encoding soluble FLAG and OLLAS tagged Flt3L, internal ribosomal entry site (IRES), and enhanced green fluorescence protein (EGFP), i.e., SFO.Flt3L-IRES-EGFP. The GenBank accession number for the sequence including the extracellular domain of mouse Flt3L is GU168042, and the IRES-EGFP sequence is from pIRES-EGFP plasmid (Clontech, Mountain View, CA, USA). CHO cells were then transfected using Lipofectamine 2000 (Life Technologies), with a mammalian expression vector plasmid encoding SFO.Flt3L-IRES-EGFP under CMV promoter and a neomycin resistance gene, constructed with the backbone of pEGFP-N1 (Clontech). CHO/Flt3L cells stably expressing the SFO.Flt3L-IRES-EGFP were produced by the following steps: (i) treatment of transfected CHO cells with G418 (1.5 mg/ml) for 1 week; (ii) enrichment of EGFP-positive CHO cells with FACSAria II cell sorter (BD Biosciences, San Diego, CA, USA); (iii) generation of clonal cells by limiting dilutions of FACS-sorted EGFP-high CHO/Flt3L cells; and (iv) selection of CHO/Flt3L clones after evaluating both levels of EGFP expression by FACS analysis and Flt3L secretion by anti-OLLAS Western blot analysis (22). Chosen CHO/Flt3L cells were cultured in cell culture flasks with DMC7 to produce CHO/Flt3L-conditioned medium.

Different amounts of supernatant from CHO/Flt3L cell culture were mixed with an equal volume of 2×SDS PAGE sample buffer and boiled at 95℃ for 5 min. The samples were then separated in 12% SDS-PAGE and transferred onto PVDF membranes (Thermo Fisher Scientific, Rockford, IL, USA) before being incubated with anti-OLLAS monoclonal antibody (22). Anti-OLLAS antibody-reactive bands on the blots were examined after incubation with peroxidase-conjugated anti-rat IgG antibody (Southern-Biotech, Birmingham, AL, USA) and subsequent visualization with SuperSignal™ West Pico Chemiluminescent Substrate (Thermo Fisher Scientific) and ImageQuant™ LAS 4000 mini (GE Healthcare Life Sciences, Pittsburgh, PA, USA). Graded quantities (0~160 ng) of purified OLLAS-tagged Gag p41 protein (2) were evaluated in parallel for quantification of OLLAS-tagged Flt3L bands.

Mice were sacrificed by asphyxiation in a CO₂ chamber. Whole splenocyte suspension was prepared by cutting and mincing extracted spleens followed by grinding with frosted glasses and cell strainers (BD Biosciences). Whole BM cell suspension was prepared by flushing out femurs and tibias excised from the hind legs of mice in sterile conditions as described previously (23). Then, the single cell suspension was cultured in either 48- or 24-well tissue culture plates at 1×10⁶ or 2×10⁶ cells per well, respectively, with DMC7 containing different doses of Flt3L, i.e., CHO/Flt3L conditioned medium described above. During the culture, half of the medium in each well was carefully removed and replenished with fresh DMC7 containing Flt3L every 2 days until harvest for use in subsequent experiments.

Single cell suspensions from the harvest of mouse organ tissues or cultures thereof were incubated with 2.4G2 (Fc blocker) hybridoma supernatant and washed with FACS buffer (2% FCS, 2 mM EDTA, 0.1% sodium azide). Then, each sample was incubated with the appropriate mixture of fluorochrome-conjugated monoclonal antibodies and live/dead staining dye for 30 min at 4℃ and washed twice with FACS buffer. The samples were then analyzed with FACSVerse flow cytometer (BD Biosciences) or sorting with FACSAria II cell sorter (BD Biosciences). As for sorting, the isolated cells with 90% or higher purity were utilized for subsequent experiments. Gating criteria for respective precursor and DC populations were as follows. With the lineage (Lin) markers of CD3, CD19, Ly6G, NK1.1, DX5, Ter119: for pDCs, Lin⁻ CD11c⁺ B220⁺ PDCA-1⁺ SiglecH⁺; for cDC1, Lin⁻ CD11c⁺ MHCII⁺ B220⁻ CD24^highCD172α⁻; for cDC2, Lin⁻ CD11c⁺ MHCII⁺ B220⁻ CD24^intCD172α⁺. With the lineage markers of CD3, CD19, Ly6G, NK1.1, DX5, Ter119, MHCII, B220: for MDP, Lin⁻ CD11c⁻ CD115⁺ CD135⁺ CD117^high; for CDP, Lin⁻ CD11c⁻ CD115⁺ CD135⁺ CD117^int. Flow cytometric data were analyzed using FlowJo software (FlowJo, Ashland, OR, USA).

Total RNA was isolated from sorted cells from cultured or uncultured mouse primary cells using TRIzol® Reagent (Life Technologies) following the manufacturer's instructions.

For miRNA expression profiling, 100 ng of purified total RNA was used for reverse transcription using Taqman® MicroRNA Reverse Transcriptase Kit (Applied Biosystems, Foster City, CA, USA). The resultant cDNA was used in combination with Taqman® MicroRNA Assays (Applied Biosystems) for respective miRNA and U6 control transcripts and Taqman® Universal Master Mix II (Applied Biosystems) for PCR according to the manufacturer's instructions. The amplification and detection of products were performed in a Light Cycler 480 II (Roche) with an initial denaturation at 95℃ for 10 min, followed by 40~60 cycles of amplification at 95℃ for 15 sec and 60℃ for 60 sec, before cooling. The threshold cycle (Ct) of miR-124 expression was automatically defined, located in the linear amplification phase of the PCR, and normalized to the control U6 (ΔCt value). The relative difference in expression levels of miR-124 in the sorted cells (ΔΔCt) was calculated and presented as the fold induction (2 ^−ΔΔCt).

For pri-miR-124 profiling, 100 ng of purified total RNA was used for reverse transcription using PrimeScript™ RT Reagent Kit (TaKaRa Bio Inc., Ohtsu, Japan). The resultant cDNA was used in combination with primer oligonucleotides designed for DC-related transcription factors (5) and pri-mmu-miR-124-1/-2/-3 (24) and SYBR® Premix Ex Taq II (TaKaRa) for PCR according to the manufacturer's instructions. Sequences for the primer oligonucleotides, synthesized by Cosmo Genetech (Seoul, Korea), are as follows: TCF4, 5'-TGAGATCAAATCCGACGA-3' forward, 5'-CGTTATTGCTAGATCTTGACCT-3' reverse; Batf3, 5'-AGACCCAGAAGGCTGACAA-3' forward, 5'-CTGCACAAAGTTCATAGGACAC-3' reverse; Irf8, 5'-AAGGGCGTGTTCGTGAAG-3' forward, 5'-GGTGGCG TAGAATTGCTG-3' reverse; pri-miR-124-1, 5'-GCCTCT CTCTCCGTGT-3' forward, 5'-CCATTCTTGGCATTCA-3' reverse; pri-miR-124-2, 5'-AGAGACTCTGCTCTCCG TGT-3' forward, 5'-CTCCGCTCTTGGCATTC-3' reverse; pri-miR-124-3, 5'-GGCTGCGTGTTCACAG-3' forward, 5'-ATCCCGCGTGCCTTA-3' reverse; GAPDH, 5'-ACA GTCCATGCCATCACTGCC-3' forward, 5'-GCCTGCTT CACCACCTTCTTG-3' reverse. The amplification and detection of products were performed in a Light Cycler 480 II (Roche) with an initial denaturation at 95℃ for 5 min, followed by 40~60 cycles of PCR at 95℃ for 5 sec and 50℃ for 30 sec, before melting and cooling. The relative difference in expression levels for pri-miR-124-1/-2/-3 in the sorted cells was calculated and presented as described above but with normalization to the control GAPDH.

The predicted target genes of miR-124 were identified using a public database (miRWalk2.0, http://www.umm.uniheidelberg.de/apps/zmf/mirwalk). Sections containing the miR-124 binding site of the 3'UTR of TCF4 and Zbtb46 were cloned into pmirGLO Dual-Luciferase miRNA Target Expression Vector (Promega, Fitchburg, WI, USA) independently. HeLa cells were seeded at 2.5×10⁴ cells per well in a 24-well plate. After 48 hrs, the pmirGLO vector containing the target mRNA binding site for the miR-124 was co-transfected with mmu-miR-124 mimic (Applied Biosystems) or the negative control using Lipofectamine 2000 (Life Technologies). Luciferase activity was measured after 48 hrs using Dual Luciferase Assay System (Promega) according to the manufacturer's instructions on a luminometer (Promega). Renilla luciferase activity was used to normalize the transfection efficiency. Each assay was repeated at least 3 times.

Experiments with multiplicated samples were presented as mean±SEM from at least three independent experiments. Statistical comparisons between different groups were analyzed using unpaired Student's t-test using SigmaPlot (Systat Software, San Jose, CA). Statistical significance is denoted by the p values equal or below 0.05 (^*), 0.01 (^**), and 0.001 (^***). Data were plotted for graphs with SigmaPlot.

We generated the extracellular domain sequence of mouse Flt3L fused with the sequences of signal peptide, FLAG tag, and OLLAS tag. This soluble FLAG and OLLAS tagged Flt3L (SFO.Flt3L) sequence was devised to express with IRES and EGFP under CMV promoter (Fig. 1A) following transfection into CHO cells. CHO cells expressing high levels of Flt3L were enriched by sorting for EGFP, cloned, and named CHO/Flt3L cells (Fig. 1B). The conditioned medium from CHO/Flt3L cells was produced in large volume, sterilized by filtration, and blotted with anti-OLLAS antibody for the determination of Flt3L concentration before being added into the culture of BM cells. According to the anti-OLLAS Western blot analysis, the CHO/Flt3L conditioned medium was evaluated to contain approximately 10 µg/ml of Flt3L (Fig. 1C). To verify the efficacy of Flt3L produced, graded doses of CHO/Flt3L conditioned medium were used to culture BM for the generation of DCs in vitro. Higher concentrations of Flt3L in BM culture induced DC differentiation more efficiently (Fig. 1D). Especially, when the culture media were composed of 10% or higher amount of the Flt3L conditioned medium (up to 50%), the increase in number of DCs in BM culture became conspicuous (data not shown for BM cultures in media with 25% or 50% of the Flt3L conditioned medium). As estimated by Western blot analysis above, 10% content of the Flt3L conditioned medium corresponds to 1 µg/ml of Flt3L in culture medium. The efficacy to induce DCs in vivo by this Flt3L protein has also been demonstrated previously (2526).

To identify miRNAs possibly involved in DC development, a Web-based miRNA-target interaction database miRWalk2.0 (27), which amalgamates data from a collection of miRNA-target prediction programs, was used. Ten gene targets that are prominent transcription factors known to have a pivotal role in DC development – TCF4, Bcl6, Irf2, Irf4, Irf8, Id2, SpiB, Batf3, Notch2, Zbtb46 – were picked for the prediction software to identify miRNAs that possibly regulate these genes by binding to their 3'UTR. From the provided list of miRNAs that have a high potential to bind to the 3'UTR of these transcription factors, candidate miRNAs were selected dependent upon availability of detection reagents in our in-house miRNA probe collection. Overall, 20 candidate miRNAs were specified according to the analyses of 3'UTR sequences from the 10 transcription factors (Supplemental Fig. 1).

In preliminary efforts to reveal miRNAs that may play a role in DC development, expression profiles of the candidate miRNAs were assessed in Flt3L culture system of mouse BM cells. Using our in-house probe collection, the expressions of 20 candidate miRNAs were screened in CD11c⁺ DCs derived from BM culture with Flt3L, which were further assorted into B220⁻ cDCs and B220⁺ pDCs (Fig. 2A). Total RNAs isolated from these two subsets of cultured DCs were subjected to analysis of miRNA expression profile by real-time RT-PCR. When the expression of candidate miRNAs was normalized and their relative levels were compared, an exceedingly high expression of miR-124 was observed in the B220⁻ cDC population from BM culture with Flt3L (Fig. 2B). In the case of each individual candidate miRNA profile, all were expressed more in B220⁻ cDCs than in B220⁺ pDCs except for miR-17 (Fig. 2C). Exceptionally high expression of miR-124 in cDCs and its contrasting expression between cDCs and pDCs in BM-derived DCs cultured with Flt3L suggest that miR-124 may also be differentially expressed during ontogeny of DCs in vivo and may play a role in their development.

According to the analysis of miRNA-target interactions above, miR-124 is predicted to bind to the 3'UTR of TCF4 (Supplemental Fig. 1A). miR-124 is also expected to bind to Zbtb46 although the probability is low (i.e., the sum of prediction algorithms with significant scores is 1; Supplemental Fig. 1J). To see whether miR-124 actually binds to these genes and regulates their expression, a dual luciferase reporter assay was performed using a pmirGLO vector (Fig. 3A). A section, 342 bps, of the 3'UTR of TCF4 containing the predicted binding site of miR-124 was inserted into the multiple cloning site (MCS) of a pmirGLO vector (Fig. 3B). A 480 bp section of the 3'UTR of Zbtb46 was cloned likewise into a pmirGLO vector (Fig. 3C). The cloned vectors were co-transfected with miR-124 mimic or the negative control into HeLa cells and the luciferase activity was measured. The assay showed that the overexpression of miR-124 was able to down-regulate luciferase activity of the pmirGLO-TCF4 vector by ~30% while not those of the pmirGLO-control or pmirGLO-Zbtb46 (Fig. 3D). Besides, the sequences of miR-124 binding site found in 3'UTRs of both mouse and human TCF4 transcripts are highly conserved (Fig. 3E). Therefore, it appears that miR-124 might directly bind to both 3'UTRs of mouse and human TCF4 and thus posttranscriptionally regulate their activity.

TCF4 is a critical transcription factor in the development and homeostasis of pDCs and is expressed at higher levels in pDCs and pDC precursors than in other DCs and progenitors (528). Therefore, it seems developmentally relevant that lower expression of miR-124 in pDCs than in cDCs was observed amongst BM-derived DCs cultured with Flt3L in our preliminary screening (Fig. 2). As previously demonstrated by others (72930), we were able to culture BM cells in vitro with Flt3L for 6 to 12 days to efficiently produce three major DC subsets that respectively correspond to pDC, cDC1, and cDC2 lineage cells in lymphoid tissues in vivo. These DC populations from Flt3L-cultured BM were identified and isolated as per their surface markers of CD11c⁺ B220⁺ for pDCs, CD11c⁺ B220⁻ CD24⁺ CD172α⁻ for cDC1, and CD11c⁺ B220⁻ CD172α⁺ CD24^int for cDC2 (Fig. 4A). Then, these three DC populations purified from BM culture with Flt3L by flow cytometric sorting were subjected to RNA extraction and real-time RT-PCR. As mentioned above, the dynamic and differential expression of various transcription factors across the DC subsets is critical for DC development (10). Expressions of TCF4, Batf3, and Irf8 in the DC subsets isolated from BM culture with Flt3L (Fig. 4B) were parallel to those previously reported in the DC subsets isolated from lymphoid tissues (9), indicating that the isolated DC subsets were classified appropriately. Expression of miR-124, similar to the results of the preliminary screen (Fig. 2), was lower in pDCs than in cDCs, i.e., cDC1 and cDC2 cells (Fig. 4C). Moreover, cDC1 cells were shown to express significantly higher levels of miR-124 than both pDCs and cDC2 cells.

To further verify the expression profiles of miR-124 amongst DC subsets present in BM culture with Flt3L, we examined the expression of miR-124 from DC subsets and progenitors present in steady-state BM. MDP, CDP, pDC, cDC1, and cDC2 cells present in mouse BM were respectively purified ex vivo by flow cytometric sorting according to their surface makers (Fig. 5A-C). When the extracted RNAs of the respective DC subsets and progenitors isolated from BM were analyzed by real-time RT-PCR and compared, miR-124 expression was prominently found in cDC1 cells at a level much higher than those of MDP, CDP, and other DC subsets (Fig. 5D). In addition, we also had several BM populations of pre-DCs, including precDCs and pre-pDCs (6731), purified and analyzed for their expressions of miR-124, which were lower than that of cDC1 cells but similar to those of other DC subsets and progenitors (data not shown). Therefore, higher expression of miR-124 is consistently found in the cDC1 lineage from both BM cells and BM-derived cell cultures with Flt3L. Then, to observe the patterns of miR-124 expression in steady-state spleen, DC subsets and pre-DCs were sorted ex vivo from freshly prepared splenocytes according to their surface makers (Fig. 6A-C). Real-time RT-PCR of the extracted RNA from the sorted splenic cells also revealed a relatively high expression of miR-124 in cDC1 cells compared to other DC subsets and pre-DCs in spleen (Fig. 6D), therefore suggesting that the abundant expression of miR-124 might be important to the development of all cDC1 lineage cells in general.

As depicted in Fig. 7A, there are three primary miRNA genes for miR-124: pri-miR-124-1, pri-miR-124-2, and pri-miR-124-3 (3233). Three transcripts originate from these three different genes on separate chromosomes but all convert into the same mature miR-124 sequence. With oligonucleotide probes to distinguish and detect the three pri-miR-124 transcripts (24), real-time RT-PCR was performed to determine which miR-124 precursors were expressed more in the cDC1 lineage. Expression profiling of pri-miR-124 in the cDC1 cells from steady-state BM (Fig. 7B) and spleen (Fig. 7C) showed similar patterns of miR-124 precursor expression. In both tissues, cDC1 cells showed that pri-miR-124-1 was expressed the least and pri-miR-124-3 was expressed the most but all three precursors were expressed within the ranges of no significant statistical difference. In addition, expression of the three pri-miR-124 genes was measured in other DC subsets where transcripts of all three precursor genes were also detected significantly (data not shown). In other words, the definitive expression of all three pri-miR-124 transcripts indicates that they all contribute significantly to the generation of mature miR-124 in cDC1 lineage cells.