Presence of miR-134 in Xenopus neural tissues. (A) Taqman stem-loop real-time PCR assays on the expression of three miRNAs, miR-103, miR-134, and miR-191 in rat brain, whole Xenopus embryos, and Xenopus neural tube tissues. Please note that both Xenopus samples were total RNAs, whereas the rat brain sample was processed for microRNA enrichment using a mirVana miRNA isolation kit (see Methods). As a result, the microRNA level in the rat brain sample was much higher and merely serves as a positive control. The relative miRNA levels are plotted in logarithmic scale. The error bars represent the standard error of the mean. Asterisk: p < 0.01 Student's t-test. (B) Whole mount in situ analysis of miR-134 expression in Xenopus embryos (stage 24) using a LNA probe against miR-134 or a scrambled probe. Cross sections of the spinal cord are shown on the right side. Dotted red lines depict the boundary of miR-134 positive and negative regions.

Blockade of BDNF-induced growth cone turning by miR-134 mimic and antisense oligonucleotides. (A) Representative images of growth cones at the beginning and the end of 30 min exposure to a BDNF gradient. Dashed lines indicate the original direction of growth cone extension. Dotted lines indicate the position of the growth cone at the onset of the turning assay. Insets are fluorescent images of the same growth cones, showing the presence of the fluorescent tracer FITC-dextran co-injected with miR-134 mimic or antisense. (B) Trajectory tracings of all the growth cones subjected to 30 min turning assay in a BDNF gradient. Neurons injected with control oligonucleotides (Ctrl), miR-134 antisense inhibitors or mimics are shown here. The origin is the center of the growth cone at the onset of the BDNF gradient and the original direction of growth cone extension is vertical. Scale bars: 20 μm. Arrows indicate the direction of the BDNF gradient. (C) Average turning angles (top) and net extension (bottom) of different groups of all the growth cones examined. Numbers indicate the total number of growth cones examined for each group. Each group is labeled at the bottom: None: neurons not treated; Cyclo: bath application of cycloheximide; Ctrl: neurons injected with miR-134 control oligonucleotides; AS: injected with miR-134 antisense inhibitors; MM: injected with miR-134 mimics. Error bars represent the standard error of the mean. Asterisks depict the statistical significance (p < 0.01, Mann-Whitney test).

Detection of Xenopus limk1 in Xenopus neurons. (A) RT-PCR detection of Xlimk1 mRNA from RNA samples extracted from Stage 20-22 Xenopus neural tube tissues using specific primers. RNA samples were processed without (-RT) and with reverse transcriptase (RT). (B) Representative fluorescence images of cultured Xenopus growth cones labeled using a specific antibody against LIMK1. (C) Detection of Xlimk1 mRNA in Xenopus growth cones by fluorescence in situ hybridization. Top panels are the differential interference contrast (DIC) images of the growth cones. Bottom panels are the FISH images of Xenopus growth cones labeled with digoxigenin-conjugated probes (three probes, ~50 nt each) that are specifically complementary to different parts of the coding region of Xlimk1 mRNA. The reverse probes were used as the control. Scale: 10 μm.

The schematic diagram shows a proposed model on how different cues might act through PS-dependent and -independent pathways to regulate growth cone steering. BMP7 acts as bidirectional guidance molecular through phosphorylation regulation of actin depolymerizing factor (ADF)/cofilin (AC). We propose that BDNF gradients release the inhibition of translation by miR-134 and induce the local translation of Xlimk1 in an asymmetric way, leading to the asymmetric modification of the actin dynamics for growth cone steering. BDNF also acts to elicit asymmetric β-actin translation, which could plausibly operate in parallel to or in concert with the miR-134 regulation of LIMK1 translation for growth cone turning. Future experiments are required to test this model. For simplicity, the mTOR pathway is not depicted in the model.

Enrichment of miR-134 in Xenopus growth cones. Fluorescence in situ hybridization was used to detect miR-134 in cultured Xenopus spinal neurons using a LNA probe (A) or scrambled probe (B). Phase contrast images of the growth cones are shown as insets. Arrows indicate the miR-134 puncta in the lamellipodia and filopodia. Scale bars: 10 μm.

Quantitative analysis of BDNF-induced p44/42 MAPK activation by immunofluorescence of phospho-p44/42 levels in Xenopus growth cones. Neurons injected with control, miR-134 mimics and antisense inhibitors were exposed to control saline or BDNF (50 ng/ml) for 30 min. The immunofluorescence intensity of BDNF-exposed growth cones was normalized to that of the corresponding group without BDNF exposure. Numbers indicate the number of growth cones examined. Error bars: the standard error of the mean. Double asterisks: p < 0.001 (Student's t-test).

Xenopus limk1 mRNA as a potential target of miR-134. (A-B) Representative confocal double FISH images showing growth cones subjected to either reverse Xlimk1 probes and scrambled miRNA probes (A) or Xlimk1 probes and LNA-miR-134 probes (B). Color panels are the merged channels of both Xlimk1 (green) and miR-134 (red) signals, of which an intensity threshold was applied to each channel to highlight the FISH signals. Yellow colors indicate co-localization of Xlimk1 and miR-134. Scale bars: 10 μm. (C) Quantification of the percentage of Xlimk1 puncta co-localized with miR-134 puncta. Numbers indicate the number of growth cones examined. (D) The predicted duplex formation between miR-134 (red) and Xlimk1 (green) 3'UTR is shown on the left (optimal binding energy: -28.2 kcal/ml). The results of luciferase assays using a Xlimk1 3'UTR-luciferase reporter are shown in the bar graph on the right. Numbers indicate the total numbers of samples from three repeated trials. Error bars depict the standard error of the mean. Asterisks in (C & D): p < 0.01 Student's t-test.