1. A Splicing-Independent Function of SF2/ASF in MicroRNA Processing
Han Wu, Shuying Sun, Kang Tu, Yuan Gao, Bin Xie, Adrian R. Krainer, Jun Zhu 
Molecular Cell 
Volume 38, Issue 1, Pages (April 2010)
DOI: /j.molcel
Copyright © 2010 Elsevier Inc. Terms and Conditions

2. Figure 1 Differentially Expressed miRNAs upon SF2/ASF Induction
(A) Normalized counts of individual miRNAs at 24, 48, or 72 hr after SF2/ASF induction were compared to that of cells without induction. Differentially expressed miRNAs (>1.5-fold change and q < 0.01) are shown, and z scores are plotted in the heat map.
(B) Northern blotting analysis of mature miR-7 in the stable HeLa cell line with inducible SF2/ASF expression. U6 snRNA was used as an internal control. The changes in relative miR-7 expression levels are shown at the bottom of the panel.
(C) HeLa cells were transfected with small interfering RNAs (siRNAs) against either luciferase or SFRS1. Two days after transfection, radiolabeled RT-PCR and western blotting were used to monitor the SFRS1 mRNA level (top two panels; GAPDH as an internal control) and protein level (middle two panels; α-tubulin as an internal control), respectively. Northern blotting analysis of endogenous miR-7 level with U6 as an internal control (bottom two panels). Results from triplicate experiments are plotted in the right panel. The ∗ indicates p < 0.05 (t test, n = 3). Error bar represents SEM.
Molecular Cell  , 67-77DOI: ( /j.molcel )
Copyright © 2010 Elsevier Inc. Terms and Conditions

3. Figure 2 SFRS1 Is a Physiological Target of miR-7
(A) Base-pairing between mature miR-7 and its putative target site in the 3′UTR of SFRS1.
(B) Schematic diagram of the luciferase reporter constructs. The putative miR-7 target site is shown.
(C) Dual luciferase assays were performed in triplicate. For each construct, the relative luciferase activity was plotted by normalizing between cells transfected with control RNAs (dark gray) and miR-7 precursors (light gray).
(D) HeLa cells were transfected with either control RNAs or synthetic miR-7 precursors. Two days after transfection, the levels of mature miR-7 and endogenous SF2/ASF protein were quantified by northern (U6 as internal control; top two panels) and western blotting (α-tubulin as an internal control; bottom two panels), respectively. The relative levels of SF2/ASF protein from three separate transfections are plotted (t test, p < 0.05; middle panel). Quantitative RT-PCR results of endogenous SFRS1 mRNA in the transfected cells are shown in the right panel; GAPDH mRNA was used as an internal control. Error bars represent SEM.
Molecular Cell  , 67-77DOI: ( /j.molcel )
Copyright © 2010 Elsevier Inc. Terms and Conditions

4. Figure 3
The Domain Requirement of SF2/ASF for Promoting miRNA Production and Alternative Splicing
(A) Diagram of the hnRNPK minigene reporter. Proximal/distal splice sites and miR-7-1 precursor (dashed box) are shown. (B) The hnRNPK minigene was cotransfected with control vector (ctrl), wild-type/mutant SF2/ASF, SC35, or 9G8 cDNAs into HeLa cells. Cells were harvested 2 days after transfection. The overall level of hnRNPK (top panel) and its alternative splice-site usage (bottom panel) were monitored by radiolabeled RT-PCR. The mature miR-7 level was determined by northern blotting (middle panel). The relative miR-7 levels (normalized to total hnRNPK levels) and the alternative splicing ratios (proximal/distal) are plotted in (C) and (D), respectively (t test, p < 0.05; n = 3). Error bars represent SEM.
Molecular Cell  , 67-77DOI: ( /j.molcel )
Copyright © 2010 Elsevier Inc. Terms and Conditions

5. Figure 4
A Splicing-Independent Function of SF2/ASF in Promoting miR-7 Expression
(A and B) (A) Wild-type hnRNPK minigene, the proximal 3′ splice-site mutant (PM) or the distal 3′ splice-site mutant (DM), was transfected into HeLa cells together with a control vector or a cDNA expressing SF2/ASF. Alternative splicing of hnRNPK and miR-7 expression were monitored by radiolabeled RT-PCR and northern blotting, respectively. The levels of mature miR-7 were normalized to total hnRNPK levels. Results from triplicate experiments were plotted in (B) (t test, p < 0.05; n = 3).
(C) Intronless minigene expressing miR-7 (pCG-miR-7) was transfected into HeLa cells together with a control vector or SF2/ASF cDNA. The levels of EGFP and miR-7 were monitored by radiolabeled RT-PCR and northern blotting, respectively.
(D) The miR-7 levels were normalized to EGFP, and the results from triplicate experiments are plotted (t test, p < 0.05; n = 3). Error bars represent SEM.
Molecular Cell  , 67-77DOI: ( /j.molcel )
Copyright © 2010 Elsevier Inc. Terms and Conditions

6. Figure 5
Direct Interaction between SF2/ASF and the Stem-Loop Region of miR-7-1
(A) The stem-loop region of hsa-miR-7-1: the high-score SF2/ASF motif is shown in red (dashed box); mature miR-7 and miR-7∗ are shown in pink. The SF2/ASF motif was disrupted by mutations in four nucleotides (upper case, blue) with compensatory mutations in the opposite arm (upper case, green).
(B) Wild-type or mutant pCG-miR-7 was transiently transfected into HeLa cells. RNA samples obtained from input and SF2/ASF-specific CLIP were analyzed by radiolabeled RT-PCR. Two primer pairs to specifically amplify the EGFP and the miR-7 stem-loop regions are shown.
(C) The relative enrichments of the miR-7 stem-loop region before and after SF2/ASF CLIP are plotted for wild-type and mutant pCG-miR-7 transcripts (t test, p < 0.05; n = 3). Error bars represent SEM.
(D) In vitro-transcribed pri-miR-7-1 RNA or its mutant was coupled to agarose beads and incubated with HeLa extract. The bound proteins were analyzed by western blotting with α-tubulin as a negative control. Approximately 1/200 of the input extract was loaded as positive control (lanes 1 and 3). The relative SF2/ASF protein levels are shown at the bottom of the panel.
Molecular Cell  , 67-77DOI: ( /j.molcel )
Copyright © 2010 Elsevier Inc. Terms and Conditions

7. Figure 6 Sequence-Dependent Promotion of miR-7 Maturation by SF2/ASF
(A) Wild-type or mutant miR-7 construct was cotransfected into HeLa cells with a control vector or SF2/ASF-expressing plasmid. Transfection efficiencies were normalized to the EGFP levels (radiolabeled RT-PCR). Precursor and mature miR-7 levels were monitored by northern blotting. The relative activities of SF2/ASF in promoting miR-7 expression are shown in the right panel (t test, p < 0.05; n = 3). Error bars represent SEM. (B) HeLa cells were transfected with either luciferase or SFRS1-specific siRNA. Two days after transfection, cells were harvested to prepare whole-cell extracts, for which the levels of SF2/ASF protein were analyzed by western blotting. In vitro pri-miRNA processing assay was performed with wild-type or mutant pri-miR-7-1 substrates (Michlewski et al., 2008); the results are shown in (C) and (D), respectively.
Molecular Cell  , 67-77DOI: ( /j.molcel )
Copyright © 2010 Elsevier Inc. Terms and Conditions

8. Figure 7 SF2/ASF Is Broadly Involved in miRNA Biogenesis
(A) HeLa cells were transfected with intronless miR-29b-1/2, miR-221, or miR-222 constructs, together with an empty or SF2/ASF expression vector. The levels of EGFP and miR-7 were monitored by radiolabeled RT-PCR and northern blotting, respectively.
(B) Schematic diagram of the SFRS1/miR-7 negative feedback loop. SF2/ASF directly binds to pri-miR-7 to promote its nuclear cropping, although effects on later steps (pre-miRNA export and/or Dicer cleavage; dashed lines) cannot be ruled out. Additional factors (circle with a question mark) may also be involved in efficient pri-miRNA maturation, possibly through the interaction with the RS domain of SF2/ASF. In the cytoplasm, mature miR-7 (yellow line) binds to the 3′UTR of the SFRS1 mRNA (rectangle box) and represses the production of SF2/ASF protein via translational inhibition.
Molecular Cell  , 67-77DOI: ( /j.molcel )
Copyright © 2010 Elsevier Inc. Terms and Conditions