Animals and dissections
The Melbourne Health Animal Ethics Committee and the University of Adelaide Animal Ethics Committee approved procedures involved in the breeding and handling of animals. Mice were housed under a 12-hour light and 12-hour dark cycle with access to unlimited food and water. Mice were culled by CO2 inhalation and all dissections of mouse embryos, brains and organs were carried out according to the methods described previously .
Deep sequencing and analysis
Total RNA was isolated from a whole brain dissected from an E15.5 embryo of C57BL/6 background using TRIzol reagent (Invitrogen) according to the manufacturer's protocol. Small RNAs with sizes ranging from 16-30nt were isolated from 10 μg total RNA using polyacrylamide gel electrophoresis. The complementary small RNA library was constructed using the Small RNA Sample Prep Kit version 1.0 (Illumina) according to the manufacturer's protocol with 5'-GTTCAGAGTT CTACAGTCCG ACGATC-3' and 5'-TCGTATGCCG TCTTCTGCTT GT-3' adapters at the 5' and 3' ends, respectively. Sequencing was carried out using a Genome Analyzer II (Illumina). Image data was generated by the Genome Analyzer II and was processed using the Illumina pipeline software (Pipeline version 1.0 was used for the FASTQ data). This consists of an image analysis module (Firecrest), followed by basecalling using the BUSTARD module and finally production of a data file in FASTQ format using the GERALD module.
Sequence annotation pipeline
The FASTQ data was ranked according to decreasing abundance of the unique tags. This file was created using a PERL script in Linux without taking into consideration any filters (adapter sequences) or quality. A file with unique tags and their corresponding counts was generated. All unique tags (including those with a single count) were mapped to the NCBI Mouse Assembly Build 37.1 using the Bowtie program . Two sets of alignments were carried out: one stripping off 14 bases from the 5' end of unique tags and the other stripping off 14 bases from the 3' end. In both alignments, no mismatches are allowed and unique tags that hit more than one locus within the mouse genome were discarded. Unique tags with a single hit within the genome were further annotated using various databases such as RepeatMasker (analysis was performed on NCBI Mouse Assembly build 37.1 and the output was downloaded from UCSC genome browser on the 28th of November, 2008), mouse RefSeq in release 32, mouse miRNA in miRBase release 12.0 and redundant mouse EST database (downloaded from UCSC mm9 on 27th January, 2009).
Identification of candidate novel miRNAs
Unique tags that mapped to a genomic locus with a RefSeq, redundant EST or no annotations were subjected to pre-miRNA prediction using the RNA22 program . Sequences encompassing 100- to 200-nt upstream and downstream of these unique sequences were used to predict any potential pre-miRNAs with hairpin structures. The minimum number of patterns that should support a pre-miRNA before it can get reported was set to 60, and the minimum and maximum pre-miRNA lengths were set to 60nt and 150nt, respectively. All predicted pre-miRNA sequences based on these settings were used to determine the hairpin fold structure using RNAfold program . The predicted hairpin fold structure with the lowest minimum free energy (MFE) (cut off at -30 kcal/mol or lower) and conforming to the annotation criteria for pre-miRNA  was selected as the final predicted pre-miRNA. Briefly, the predicted precursor structure must be between 60-80 nt in size and must not have a large internal loop or any asymmetric bulges. The predicted pre-miRNA must contain the aligned unique sequence within one arm of the hairpin and include at least 16 bp from the 5' end of the unique sequence and the other arm of the hairpin.
Small RNA northern analysis
Eight blots were prepared from four independent E15.5 whole brains. Approximately 30 μg of total RNA was denatured in 1X Ambion Gel Loading Buffer II (Ambion®) at 85°C for 3 minutes. RNAs were electrophoresed in 15% acrylamide/urea gels (48% (w/v) urea, 15% (v/v) acrylamide, 0.05% (w/v) ammonium persulfate and 0.1% (v/v) tetramethylethylenediamine prepared in 1X TBE) in 1X TBE buffer at 300 V for 90 minutes. Separated small RNAs in the gel were then transferred onto Hybond-N+ nylon membrane (GE Healthcare) using Trans-Blot® SD Semi-Dry Electrophoretic Transfer Cell (Bio-Rad) at a constant 0.4 V for 45 minutes. The pre-hybridisation step was carried out in Amersham Rapid-hyb™ Buffer (GE Healthcare) with 100 μg/ml of herring sperm DNA (Promega) at 42°C for 1 hour and was followed by the hybridisation step. The same pre-hybridisation solution was used for hybridisation with addition of 2 × 106 dpm/ml labelled probe prepared using 20 U of T4 Polynucleotide Kinase (Promega) in 1X kinase buffer (Promega) and 50 pmol of [γ-32P]-dATP (GE Healthcare) (3000 Ci/mmol). Hybridisation was carried out for 18 hours and filters were washed in 5 × SSC with 0.1% (w/v) sodium dodecyl sulfate (SDS) (20 minutes at 37°C) followed by 1 × SSC with 0.1% (w/v) SDS and 0.2 × SSC with 0.1% (w/v) SDS (15 minutes each time at 65°C until a clean background signal was obtained). The membrane was exposed to a storage phosphor screen in a cassette at room temperature for 1 day for miR-3099 blot and 8 days for other blots before scanned using Typhoon™ 9400 (GE Healthcare).
Reverse transcription of the small RNA was performed based on modified methods [75, 76]. cDNA was synthesised from 150 ng-2.5 μg of small RNA enriched total RNA using 0.05 μM of an in-house designed stem loop primer (5'-GTTGGCTCT GGTAGGATG CCGCTCTCA GGGCATCCT ACCAGAGCCA AACTCCCCA-3', GeneWorks), and the Superscript® III Reverse Transcriptase Kit (Invitrogen) with modifications to the manufacturer's protocol. The stem loop primer was added after a denaturation step at 65°C for 5 minutes. The last 6nt at the 3' end of the stem loop primer complements the last 6nt of the 3' end of miR-3099 small RNA. The stem loop RT primer contains a target site for a universal reverse primer (5'-GTAGGATGCC GCTCTCAGG-3', GeneWorks) and a target site for UniversalProbe Library (UPL) Probe #21 (Roche Diagnostics), which were used in subsequent cDNA amplification processes together with a specific forward primer for miR-3099 (5'-CGCGTAGGCT AGAGAGAGGT-3', GeneWorks). Briefly, cDNA synthesis was performed at 16°C for 30 minutes followed by 60 cycles of 20°C for 30 seconds, 42°C for 30 seconds and 50°C for 1 second. A final incubation at 75°C for 15 minutes was performed to inactivate the reverse transcriptase enzyme.
Prior to qPCR, pre-PCR of miR-3099 was performed in a 10 μl reaction volume containing 1X LC480 Probe Master mix (Roche Diagnostics), 50 nM of each forward and universal reverse primers and 0.2X of synthesised cDNA. Pre-PCR was initially carried out at 95°C for 10 minutes, 55°C for 2 minutes and 75°C for 2 minutes and followed by 14 additional cycles of 95°C for 15 seconds and 60°C for 4 minutes. After pre-PCR, 0.01X of amplicons were used for qPCR.
QPCR was carried out in 10 μl reaction volume using 1X LightCycler 480 (LC480) Probe Master mix (Roche Diagnostics), 0.1 μM of a relevant Universal ProbeLibrary probe (Roche Diagnostics), 0.25 μM of each forward and reverse primers and 1 μl of 0.1X of synthesised cDNA. Reactions were prepared in 384-well plates and RT-qPCR was performed using a LightCycler® 480 Real Time PCR System instrument (Roche Diagnostics). QPCR was performed with an initial denaturation at 95°C for 10 minutes followed by 45 cycles at 95°C for 10 seconds, 60°C for 30 seconds and 72°C for 10 seconds, and a final step at 40°C for 1 second.
Real-Time amplification signals were acquired during the elongation step and recorded live using LightCycler® 480 Software version 1.5 (Roche Diagnostics). The cycle threshold or crossing point (Cp) from each signal was calculated based on the Second Derivative Maximum method . A 4-data point standard curve was constructed using serially diluted pooled cDNAs for each primer set used in qPCR in each run. The standard curve was used to determine the PCR efficiency and reproducibility of each PCR system. The Hmbs gene was used as reference gene normalisation according to the method as described .
Two or three independent biological replicates were used for each tissue/organ in each experiment. Two qPCR experiments were performed on the tissue of each biological replicate. The qPCR results were normalized to Hmbs, and those that were not outliers, log2 transformed and then averaged to give the expression data for the biological replicate. One-way ANOVA was used to compare the expression levels among the tissues. A P value of <0.05 was considered statistically significant. Where significant differences were detected among the tissues the least significant difference(s) (LSD) were provided with the analysis (see Additional file 14 for analysis details).
Locked Nucleic Acids - In situ hybridisation
Paraffin embedded sections (8 μm) were used for LNA-ISH. Sections were de-paraffinised with washes in xylene (3× for 5 minutes each) and hydrated in a series of ethanol concentrations into RNase-free water. Subsequently, sections were fixed in 4% (w/v) PFA (pH7.0) in 1X PBS (10 minutes) followed by Proteinase K digestion (6.7 μg/ml of Proteinase K, 50 mM of Tris HCl pH7.5, 5 mM of EDTA) for 30 minutes, re-fixed in 4% (w/v) PFA in 1X PBS for 5 minutes and acetylated (0.1 M of triethanolamine, 0.178% (v/v) of concentrated HCl and 0.25% (v/v) of acetic anhydride) for 10 minutes. Between each step, sections were washed multiple times using 1X PBS.
The pre-hybridisation step was carried out in a humidified chamber (50% (v/v) formamide, 5X sodium chloride/sodium citrate, SSC) at 60°C. Amersham Rapid-hyb™ Buffer (GE Healthcare) was used for pre-hybridisation with additional Escherichia coli tRNA (Sigma Aldrich) and Herring Sperm DNA (Promega) to a final concentration of 100 μg/ml each. After 1-2 hours of pre-hybridisation, custom-made Sox4_sir3 LNA probes (Cat. no: EQ-70537, Exiqon) were added to the buffer to give a concentration of 0.020 pmol/μl. Hybridisation was carried out in the oven for 16-20 hours.
After the hybridisation step, sections were washed in 5 × SSC (20 minutes at hybridisation temperature) followed by 0.2 × SSC (3 hours at hybridisation temperature). Sections were then rinsed in fresh 0.2 × SSC for 5 minutes and in pre-blocking buffer (0.1 M of Tris HCl pH7.5, 0.15 M of NaCl and 240 μg/ml of levamisole) for a further 5 minutes. In a humidified chamber, sections were blocked in 20% (v/v) foetal calf serum (Sigma Aldrich) and 2% (w/v) blocking powder (Roche Diagnostics) in maleate buffer for 1 hour. After blocking, sections were incubated with 0.0002X (0.00015 U) anti-DIG antibody with alkaline phosphatase, Fab fragments (Roche Diagnostics) in blocking buffer for 16 hours in the dark. Subsequently, sections were washed in NTMT buffer (3× for 10 minutes each: 0.1 M Tris HCl pH9.5, 0.1 M NaCl, 0.05 M MgCl2, 1% (v/v) Tween-20 and 240 μg/ml levamisole) and then with nitro blue tetrazolium chloride (NBT)/5-Bromo-4-chloro-3-indolyl phosphate, toluidine salt (BCIP) colour reaction (0.375 mg/ml of NBT and 0.188 mg/ml of BCIP in NTMT buffer) for 3 hours to 5 days. After the colour reaction step, sections were washed with Tris EDTA buffer pH8.0 (0.01 M of Tris HCl pH7.5 and 0.001 M EDTA pH8.0) for 10 minutes and were mounted in Entellan® media (ProSciTech).
P19 teratocarcinoma cells
Propagation and differentiation of P19 cells were carried out according to protocols previously described [18, 78].
Mouse embryonic stem (mES) cells with Dicer1c conditional allele
Mouse embryonic stem (mES) cells with Dicer1 activity were of a line heterozygous for a conditionally mutant Dicer1 allele (Dicer1
) and a null Dicer1 allele (Dicer1
-), these genetic modifications have been previously described . mES cells without Dicer1 activity were produced by transient transfection of this Dicer1
/- line with Cre recombinase to produce Dicer1
-/- subclones (JRM and DMM, unpublished data). The mES cells were propagated as previously described .
Mouse E3.5 blastocysts
C57BL/6 females of 3-4 weeks of age were superovulated using 5IU of Folligon (PMSG) followed by 5IU of Chorulon (HCG) 47.5 hours later and mated with B6D2F1 entire stud males. Microdrop culture dishes were set up to equilibrate in 37°C, 5% CO2 incubator 4 hours prior to culture. KSOM (Millipore) media was used in 20 μl droplets in a 35 mm dish, overlaid with Embryo Tested Mineral Oil (Sigma). Superovulated female mice were sacrificed after 2.5 days of superovulation induction and mating, and oviducts were collected into M2 handling media (Millipore). Oviducts were flushed using M2 media, a blunt 30G needle and a1ml syringe. Morulae were collected and cultured in pre-equilibrated KSOM. Blastocysts were collected from culture a day later under a dissecting microscope. These were considered E3.5 blastocysts.