ZFIN is incorporating published figure images and captions as part of an ongoing project. Figures from some publications have not yet been curated, or are not available for display because of copyright restrictions.

ZFIN is incorporating published figure images and captions as part of an ongoing project. Figures from some publications have not yet been curated, or are not available for display because of copyright restrictions.

Schematic of experimental design and workflow. (A) Bioinformatic analysis was performed on p53 network genes listed in the Ingenuity Pathways Analysis database and p53 Knowledgebase, and miR-125b binding sites predicted by the TargetScan and MicroCosm databases. (B) p53 network genes were screened for miR-125b targets by using gain- (GOF) and loss-of-function (LOF) of miR-125b in human cells, mouse cells and zebrafish embryos, as indicated by effects on gene expression using qRT-PCR. (C) p53 network genes that were positive in either the GOF or LOF screen were assayed for direct binding to miR-125b using a biotinylated microRNA pull-down method. (D) p53 network genes that were also positive in the miR-125b pull-down were finally validated as miR-125b targets by 3′ UTR luciferase reporter assays and Western blots for protein expression. (E) A model of how miR-125b regulates the p53 network across vertebrates was constructed using our combined datasets for human, mouse and zebrafish cells.

(A) Loss-of-function (LOF) screens were performed in human primary lung fibroblasts (hLF) or mouse 3T3 fibroblasts by transfecting an antisense RNA against both miR-125a and miR-125b (miR-125a/b-AS), or by microinjecting morpholinos (MO) against pre-mir-125b hairpin precursors (all 3 isoforms) into zebrafish embryos. Gain-of-function (GOF) screens were performed in human SH-SY5Y and mouse N2A neuroblastoma by transfecting the miR-125b duplex into cells in culture, or by coinjecting the miR-125b duplex with the morpholinos against pre-mir-125b into zebrafish embryos. Fold changes in gene expression were measured by qRT-PCR twenty-four hours after transfection or injection, relative to the mock and negative control miRNA or morpholino, and shown as log₂(fold change) using a heat-map. (B) Human: 13 genes were significantly derepressed by a loss of miR-125b, while 20 genes were significantly repressed by a gain of miR-125b, making a total of 22 genes that passed the screen (P<0.05, fold change > 1.3, relative to mock control). (C) Mouse: 11 genes were significantly derepressed by a loss of miR-125b, while 12 genes were significantly repressed by a loss of miR-125b, making a total of 13 genes that passed the screen (P<0.05, fold change > 1.3, relative to mock control). (D) Zebrafish: 13 genes were significantly derepressed by a loss of pre-mir-125b (P<0.05, fold change > 1.3, relative to control MO), while 12 genes were significantly repressed/rescued by a gain of miR-125b (P<0.05, fold change > 1.3, relative to pre-mir-125b MO), making a total of 14 genes that passed the screen. All experiments were performed with at least three biological replicates.

Biotinylated miR-125b was used as bait to pull-down mRNAs bound to miR-125b, using streptavidin-conjugated magnetic beads. The mRNAs were quantified by qRT-PCR, normalized to Gapdh, and then compared relative to the same mRNA species pulled down by a biotinylated C. elegans negative control microRNA. The enrichment of mRNAs bound to miR-125b is presented as mean log₂ fold change ± s.e.m. (n≥3 biological replicates). (A) Human: 13 out of 22 candidate targets were significantly enriched by miR-125b pull-down in human primary lung fibroblasts (hLF) 24 hours after transfection. (B) Mouse: 11 out of 13 candidate targets were significantly enriched by miR-125b pull-down in mouse 3T3 fibroblasts 24 hours after transfection. (C) Zebrafish: 8 of 14 candidate targets were significantly enriched by miR-125b pull-down in zebrafish embryos 24 hours after injection. Dashed line: cutoff for genes that were significantly enriched (Log₂ Fold change > 0.5, P<0.05).

Candidate p53 network genes that were positive in both the GOF/LOF screen and miR-125b pull-down were validated for targeting by miR-125b using the 3′ UTR luciferase reporter assay and Western blots for protein expression. (A-C), Reporter genes containing the full-length 3′ UTRs of each selected target gene were co-transfected with miR-125b duplex into 293T cells. Luciferase readings were obtained 48 hours after transfection and presented here as the average percentage of luciferase activity ± s.e.m. (n≥3) relative to a scrambled duplex co-transfected control (100%). A reporter containing a 23-nucleotide-binding-site with perfect complementarity to miR-125b was used as the perfect match positive control, while the unmodified luciferase reporter was used as the empty negative control. (A) Human: 10 out of 13 candidate genes' 3′ UTRs showed significant repression by miR-125b relative to the control (p<0.01). (B) Mouse: 9 out of 11 candidate genes' 3′ UTRs showed significant repression by miR-125b relative to the control (p<0.01). (C) Zebrafish: 7 out of 8 candidate genes' 3′ UTRs showed significant repression by miR-125b relative to the control (p<0.01). (D) Alignment of predicted miR-125b binding sites in the 3′UTRs of Ppp1ca, Prkra and Tp53 across three species. Seed-binding sequences are underlined. Bases conserved in two (blue) or three (black) species are highlighted. (E) The 3′UTR seed-binding sequences of 7 target mRNAs were mutated and assayed for direct binding to miR-125b using the luciferase reporter assay, relative to wild-type 3′UTR sequences. (E) The seed-binding sequences in the 3′UTR of 7 predicted target mRNAs were mutated and compared to wild-type sequences for binding to miR-125b using luciferase reporter assay. (F-G) Western blot analysis of protein expression of selected target genes two days after a transfection of miR-125b duplex, miR-125b antisense (AS) or negative control duplex or negative control antisense. (F) Western blots showed that miR-125b repressed BAK1, PPP1CA, TP53, and PPP2CA levels in human SH-SY5Y neuroblastoma cells, while the antisense RNA miR-125b-AS derepressed expression of these proteins in human ReNcell VM neural progenitor cells. (G) Western blots showed that miR-125b repressed BAK1, PPP1CA, PUMA, and ITCH levels in mouse N2A neuroblastoma cells. Abbreviations: h, human; m, mouse; z, zebrafish.

Only targets that passed ≥ 2 validation assays, in at least one species, are shown. Red: predicted targets validated by 3 assays; Orange: predicted targets validated by 2 assays; Yellow: predicted targets validated by 1 assay; Pink: predicted targets not validated by any assay, but validated by 3 assays in another species.

(A) Human p53 network. (B) Mouse p53 network. (C) Zebrafish p53 network. Models were constructed by Ingenuity Pathway Analysis. Red: predicted targets validated by 3 assays; Orange: predicted targets validated by 2 assays; Yellow: predicted targets validated by 1 assay; Pink: predicted targets not validated by any assay, but validated by 3 assays in another species. (D) Incoherent feedforward loop (FFL) motifs characterize miR-125b regulation of p53 network genes that mediate apoptosis or cell cycle arrest.