Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2012 Apr 12;74(1):95-107.
doi: 10.1016/j.neuron.2012.02.022.

Intra-axonal translation of SMAD1/5/8 mediates retrograde regulation of trigeminal ganglia subtype specification

Sheng-Jian Ji  1 , Samie R Jaffrey
Affiliations

Intra-axonal translation of SMAD1/5/8 mediates retrograde regulation of trigeminal ganglia subtype specification

Sheng-Jian Ji et al. Neuron. .

Abstract

In many cases, neurons acquire distinct identities as their axons navigate toward target cells and encounter target-derived signaling molecules. These molecules generate retrograde signals that activate subtype-specific gene transcription. Mechanisms by which axons convert the complex milieu of signaling molecules into retrograde signals are not fully understood. Here, we examine retrograde signaling mechanisms that specify neuronal identity in the trigeminal ganglia, which relays sensory information from the face to the brain. We find that neuron specification requires the sequential action of two target-derived factors, BDNF and BMP4. BDNF induces the translation of axonally localized SMAD1/5/8 transcripts. Axon-derived SMAD1/5/8 is translocated to the cell body, where it is phosphorylated to a transcriptionally active form by BMP4-induced signaling endosomes and mediates the transcriptional effects of target-derived BDNF and BMP4. Thus, local translation functions as a mechanism by which coincident signals are converted into a retrograde signal that elicits a specific transcriptional response.

PubMed Disclaimer

Figures

Figure 1
Figure 1. A Signaling Endosome Mechanism Mediates Retrograde BMP4 Signaling
(A) Axonal application of BMP4 induces increases in nuclear pSMAD1/5/8 and Tbx3 levels. Dissociated E13.5 trigeminal ganglia neurons were cultured in microfluidic chambers for 2 DIV, to allow axons to cross to the axonal compartment. Application of BMP4 (20 ng/ml) to the axonal compartment leads to an increase in pSMAD1/5/8 and Tbx3 levels in the nuclei of trigeminal neurons. Shown are anti-pSMAD1/5/8 and anti-Tbx3 immunostaining of neurons following axonal application of BMP4 for 1 hr (pSMAD1/5/8) and 4 hr (Tbx3), respectively. (B and C) The BMP4 retrograde signal is conveyed by molecular motors and requires BMP4 receptor activity in the cell body. BMP4 treatment in axons induced a 2.2-fold increase in pSMAD1/5/8 (B) and a 2.0-fold increase in Tbx3 (C) immunofluorescence intensity in cell bodies. This effect was blocked by co-application of the retrograde motor inhibitor EHNA (1 mM) to the axons. Retrograde BMP4 signaling was also blocked by application of the BMP receptor inhibitor dorsomorphin (DMP, 10 μM) to the cell body compartment, but not by application of the inhibitor to the axonal compartment. (D) Overexpression of dynamitin impairs BMP4 retrograde signaling. Axonal BMP4 treatment induced a 2.5-fold increase of pSMAD (red, arrows) only in the nuclei of neurons not expressing dynamitin (arrows). However, the increase in pSMAD was not seen in neurons overexpressing dynamitin (indicated by nGFP expression, green, arrowheads). The graph shows the quantification. (E) Retrograde trafficking of biotinylated BMP4 from axons to the soma. Application of biotinylated BMP4 (20 ng/ml) to axons for 1 hr resulted in prominent anti-biotin immunofluorescence signals in the cell body (red). As controls, application of biotinylated BSA (20 ng/ml) or vehicle (PBS/borate/glycine) resulted in only baseline labeling in cell bodies. For quantification, all data are mean ± SEM, n = 3 replicates. Between 100–200 cells were analyzed for each condition in each experiment. ns, not significant, **p < 0.01, *p < 0.05, compared with axonal BMP4 treatment group as indicated (B and C) or vehicle treatment group (D), post hoc Tukey test after one-way ANOVA. Scale bars represent 20 μm (A); 10 μm (D and E).
Figure 2
Figure 2. Retrograde BMP4 Signaling Requires Intra-Axonal Protein Synthesis
(A and B) Retrograde BMP4 signaling requires intra-axonal protein synthesis. The increase in nuclear pSMAD1/5/8 (A) and Tbx3 (B) induced by axonal BMP4 application were inhibited by inclusion of protein synthesis inhibitors anisomycin (Aniso, 200 μM) or cycloheximide (CHX, 10 μM) in the axonal compartment. Protein synthesis inhibitors and BMP4 were added to the axonal compartment at the same time. (C–E) Repression of OC1, OC2 and Hmx1 retrograde BMP4 signaling requires intra-axonal protein synthesis. Application of BMP4 to the axonal compartment for 4 hr leads to a decrease in OC1 (C), OC2 (D) and Hmx1 (E) levels in the cell body by 44.3% (OC1), 60.0% (OC2) and 29.4% (Hmx1), respectively. The decrease in OC1, OC2 and Hmx1 levels in the cell body were inhibited by inclusion of protein synthesis inhibitor anisomycin (Aniso) in the axonal compartment. For quantification, all data are mean ± SEM, n = 3 replicates. Between 100–200 cells were analyzed for each condition in each experiment. **p < 0.01, *p < 0.05, compared with axonal BMP4 treatment group as indicated, post hoc Tukey test after one-way ANOVA.
Figure 3
Figure 3. SMAD Protein and mRNA Are Localized to Axons of Trigeminal Ganglia Neurons
(A) SMAD1/5/8 immunohistochemistry (red) reveals high levels in the maxillary (MX) projections (arrows) of trigeminal ganglia (TG), but not in the mandibular (MD) projections (outlined by white dotted line, arrowhead). Axon bundles were visualized by Tau1 immunofluorescence (green). (B) Expression of SMAD1, 5, and 8 in cell bodies and axons of trigeminal ganglia (TG) neurons. Immunohistochemistry of E13.5 rat embryo sagittal sections using specific antibodies against individual SMAD1, 5 and 8 isoforms shows SMAD protein in trigeminal ganglia cell bodies (upper panels). Each SMAD isoform was also detected in the maxillary axon bundles (lower panels, arrows). DAPI signal is excluded from axon bundles. (C) SMAD1, 5, and 8 transcripts are detected in the distal axons of trigeminal neurons. SMAD1, 5 and 8 mRNA were detected by RT-PCR using RNA harvested from the distal axon and neuronal soma, respectively. BMPR1a, BMPR1b, BMPR2 and Tbx3 transcripts were not detected in axons. β-actin is the positive control for axonal localization, and γ-actin was used as a negative control to exclude the possibility of contamination of the axonal fraction with cell body–derived material. All transcripts were detected in the trigeminal cell bodies. (D) Localization of SMAD1, 5, and 8 and 8 transcripts to axons by fluorescence in situ hybridization (FISH). Dissociated E13.5 rat trigeminal ganglia neurons were cultured for 2 DIV. SMAD1, 5, 8 and β-actin transcripts were detected in trigeminal axons in a characteristic punctate localization (red) typically seen with axonal transcripts (Lin and Holt, 2008). The control transcript, γ-actin, as well as Tbx3, exhibited similar minimal reactivity as observed with a sense SMAD1 probe. GAP43 immunostaining (green) was used to visualize axonal borders. Insets, FISH labeling in the cell bodies of trigeminal neurons. The graph shows the quantification. Bars indicate mean ± SEM, n = 3 replicates. A minimum of 15 axons for each probe was analyzed. ns, not significant, *p < 0.05 (compared with sense probe group, post hoc Tukey's test after one-way ANOVA). Scale bars represent 100 μm (A and B); 10 μm (D).
Figure 4
Figure 4. SMAD1/5/8 Is Locally Synthesized in Axons
(A) Axonal levels of SMAD1/5/8 are dependent on axonal protein synthesis. Application of anisomycin (Aniso, 200 μM) to the axonal compartment for 4 hr resulted in a 62.8% reduction of axonal SMAD1/5/8 (red) protein levels. SMAD1/5/8 fluorescence intensity was normalized to the area defined by Tau1 fluorescence in the distal 50 μm of the axon, including the growth cone. (B) FRAP (fluorescence recovery after photobleaching) analysis of Dendra2-SMAD1 in axons. pDendra2-SMAD1-3'UTRSMAD1 was transfected into E13.5 trigeminal neurons and axons were severed from the cell body after 2 DIV. Expression of the fusion protein in axons was detected by its green fluorescence. Photobleaching was carried out in the distal axon (> 500 μm from soma), which resulted in a near complete elimination of green signals. The recovery of green Dendra2-SMAD1 signal after photobeaching was monitored by live-cell imaging. A gradual appearance of green signal was observed, which increased 3.2-fold by 30 min. This recovery was blocked by treatment with anisomycin. The graph shows the quantification of green signal recovery. For quantifications, bars indicate mean ± SEM, n = 3 replicates. More than 20 axons for each treatment were analyzed. ns, not significant; *p < 0.05 (unpaired, two-tailed t test; panel A); **p < 0.01 (post hoc Tukey's test after one-way ANOVA as indicated; panel B). Scale bars represent 10 μm.
Figure 5
Figure 5. SMAD1/5/8 Retrogradely Traffics from Axon to Soma
(A) Time-dependent depletion of SMAD1/5/8 from axons is mediated by molecular motors. E13.5 rat trigeminal explants were cultured and axons were severed from the cell body after 2 DIV. A 55.4% loss of SMAD1/5/8 immunofluorescence intensity in the severed axons was observed after 1 hr treatment of anisomycin compared with vehicle. EHNA, but not LLnL, blocked the reduction of axonal SMAD1/5/8 levels. All experimental conditions included BMP4. Scale bar represents 10 μm. (B) Locally synthesized SMAD1/5/8 protein is retrogradely trafficked back to cell body. L-azidohomoalanine (AHA) was added to the axonal compartment to label the axonally synthesized proteins. Axons were treated with anisomycin or BMP4, as indicated. Lysates were collected from cell body compartments and immunoprecipitation with anti-pSMAD1/5/8 antibody was carried out, followed by Click-iT reaction to biotinylate AHA-labeled proteins. Anti-biotin western blotting detected AHA-labeled pSMAD1/5/8 in the cell body compartment. Levels of axonally derived pSMAD1/5/8 were increased upon axonal application of BMP4, while reduced levels were detected upon axonal application of anisomycin. The quantification is shown on the right. (C) Photoconversion of Dendra2-SMAD1 in axons. pDendra2-SMAD1-3'UTRSMAD1 was transfected into E13.5 trigeminal neurons. Expression of the fusion protein was detected by its green fluorescence. After photoconversion of a 1-μm segment (> 500 μm from soma) in the axon by 0.1 sec irradiation with a 406 nm laser, the photoconverted signal was detected (red). Scale bar represents 10 μm. (D) Translocation of Dendra2-SMAD1 protein from the axon to the nucleus. pDendra2-SMAD1-3'UTRSMAD1 and pDendra2 expression constructs were transfected into E13.5 trigeminal neurons respectively. After photoconverting ~20 1-μm segments (> 50 μm from soma) in axons on DIV2, the accumulation of red signal in nucleus was measured 2 min later. To control for nonspecific photoconversion during imaging of Dendra2 or Dendra2-SMAD1 in neuronal nuclei, non-photoconverted neurons in the same microscopic field as the photoconverted neurons were used as controls. Photoconversion of Dendra2-SMAD1 in axons resulted in a significant increase in photoconverted Dendra2-SMAD1 in the nucleus, compared with Dendra2. For quantifications, bars indicate mean ± SEM, n = 3 replicates. More than 20 axons (A) and neurons (D) for each treatment were analyzed. ns, not significant, *p < 0.05, **p < 0.01 (post hoc Tukey's test after one-way ANOVA compared with the vehicle treatment group in panels A and B, and 0 min timepoint group in panel D).
Figure 6
Figure 6. Retrograde BMP4 Signaling Requires Axonally Synthesized SMAD1/5/8
(A) Selective depletion of SMAD1/5/8 from distal axons. Both siRNA cocktails (siRNA 2-1-1 and 3-3-3), containing siRNAs specific to individual SMAD1, 5 and 8 mRNAs, lead to a significant reduction (by 67–75%) in axonal SMAD1/5/8 (red) protein levels compared with a non-targeting siRNA control. Tau1 staining (green) was used to define the borders of axons and growth cones for fluorescence measurements. Scale bar represents 10 μm. (B and C) Retrograde BMP4 signaling is impaired following axonal knockdown of SMAD1, 5, and 8. Knockdown of axonal SMAD1, 5, and 8 inhibited BMP4-mediated retrograde induction of pSMAD1/5/8 by 68–76% (B) and Tbx3 by 51–62% (C). For quantification, all data are mean ± SEM, n = 3 replicates. Between 100–200 cells (B and C) and more than 15 axons (A) were analyzed for each condition. ns, not significant, **p < 0.01, *p < 0.05 (compared with control siRNA group, post hoc Tukey's test after one-way ANOVA).
Figure 7
Figure 7. BDNF Acts Locally to Regulate SMAD1/5/8 Translation and Retrograde BMP4 Signaling
(A) BDNF induces intra-axonal synthesis of SMAD1/5/8. E13.5 trigeminal explants were grown in microfluidic chambers for 2 days, and then neurotrophin-free media was applied to the axonal compartment for 24 hr. Axons were then severed to ensure that any increase in SMAD1/5/8 levels do not derive from transport from the soma. Axons were treated with BMP4 (20 ng/ml), BDNF (25 ng/ml), anisomycin (Aniso, 200 μM), NT-3 (50 ng/ml) or NGF (50 ng/ml) for 30 min. In untreated axons, SMAD1/5/8 (red) levels are nearly undetectable. However, SMAD1/5/8 levels increase significantly upon application of BDNF. This increase is blocked by anisomycin, indicating that the increase in SMAD1/5/8 levels derives from an increase in local translation. Substantially less or no induction was seen following treatment with other neurotrophins or BMP4. Tau1 staining (green) was used to define the borders of axons and growth cones for fluorescence measurements. Scale bar represents 10 μm. (B) Quantification of results in (A). BDNF induced a 1.7-fold increase in axonal SMAD1/5/8 levels, which was blocked with anisomycin. NT-3 induced a substantially smaller, although statistically significant 0.7-fold increase in axonal SMAD1/5/8 levels. (C) Retrograde BMP4 signaling requires both BDNF and BMP4. Dissociated E13.5 trigeminal neurons were cultured in microfluidic chambers for 2 days, and then the media was switched to neurotrophin-free media for 4 hr. Axons were treated with NGF/NT-3, BDNF, or BMP4, alone and in combination for 1 hr, and nuclear pSMAD1/5/8 was measured by immunofluorescence. Retrograde induction of pSMAD1/5/8 was not elicited by axonal application of NGF/NT-3, BDNF or BMP4 alone, but was only observed when both neurotrophins and BMP4 were applied to axons. Nuclear pSMAD1/5/8 levels increased by 96.1% following axonal application of both BDNF and BMP4, and by 36.6% after axonal application of both NGF/NT-3 and BMP4. (D) BDNF functions locally to regulate retrograde BMP4 signaling. Retrograde BMP4 signaling was blocked following axonal application of K252a (2 μM), but not application of K252a to the cell body. BDNF was present in the culturing media. For quantification, all data are mean ± SEM, n = 3 replicates. Between 100–200 cells (C and D) and more than 15 axons (B) were analyzed for each condition. ns, not significant, *p < 0.05, **p < 0.01, compared with vehicle (B, C) or axonal BMP4 treatment (D) group as indicated, post hoc Tukey's test after one-way ANOVA.
Figure 8
Figure 8. Loss of Axonal SMAD1/5/8 Protein and Trigeminal Ganglia Positional Identity Markers in BDNF−/− Embryos
(A) Loss of SMAD1/5/8 in maxillary (MX) axons in BDNF−/− embryos. BDNF is expressed in the maxillary and ophthalmic target fields of the trigeminal ganglia. SMAD1/5/8 (red) is readily detectable in the maxillary axonal projections (arrows) of the trigeminal ganglia in E12.5 BDNF+/− mouse embryos. However, axonal SMAD1/5/8 staining is significantly reduced in BDNF−/− embryos (arrowheads), indicating a role for BDNF in regulating axonal SMAD1/5/8 levels. Axon projections were visualized by Tau1 immunofluorescence (green). The quantification shows that SMAD1/5/8 labeling was decreased by 60.0% in the axon projections of BDNF−/− embryos compared to BDNF+/− littermates. (B) Loss of pSMAD1/5/8 and Tbx3 expression in the trigeminal ganglia in BDNF−/− embryos. Expression of both markers is readily detected in sections of the trigeminal ganglia from E12.5 BDNF+/− embryos. However, expression of both markers is essentially abolished in BDNF−/− embryos. Induction of Tbx3 mRNA was also assessed at E13.5 by in situ hybridization. Tbx3 transcripts were readily detected in BDNF+/− embryos but not in BDNF−/− embryos. Trigeminal neuron cell bodies were visualized by Isl1 immunofluorescence (green). Isl1 staining was used to identify neuronal nuclei for quantification of pSMAD1/5/8 and Tbx3 (C and D) and to confirm normal trigeminal ganglia formation in BDNF−/− embryos. A portion of the trigeminal ganglia is shown at higher power to demonstrate changes in staining intensity in the nucleus. The red dotted lines indicate the boundary between the maxillary/ophthalmic (MX/OP) and mandibular (MD) regions of the trigeminal ganglia. (C and D) Quantification of loss of pSMAD and Tbx3 expression seen in (B). Nuclear pSMAD1/5/8 and Tbx3 signals were quantified in maxillary and ophthalmic regions (MX/OP) and mandibular (MD) regions of trigeminal ganglia. Isl1 labeling was used as a mask to demarcate the borders of neuronal nuclei for quantification. Nuclear pSMAD (C) and Tbx3 (D) expression in MX/OP of BDNF−/− embryos was reduced by 46.2% and Tbx3 by 62.4%, to the low baseline levels seen in the MD region of the ganglia. The expression of these markers in the MD portion of the trigeminal ganglia was not affected in the BDNF−/− embryos. For quantifications (A, C and D), all data are mean ± SEM, n = 3 replicates. Between 4–5 sections were analyzed for each genotype. ns, not significant, *p < 0.05, **p < 0.01, unpaired, two-tailed t test. Scale bars represent 50 μm (A); 100 μm (B).
See this image and copyright information in PMC

Comment in

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Arumae U, Pirvola U, Palgi J, Kiema TR, Palm K, Moshnyakov M, Ylikoski J, Saarma M. Neurotrophins and their receptors in rat peripheral trigeminal system during maxillary nerve growth. J Cell Biol. 1993;122:1053–1065. - PMC - PubMed
    1. Bassell GJ, Zhang H, Byrd AL, Femino AM, Singer RH, Taneja KL, Lifshitz LM, Herman IM, Kosik KS. Sorting of ß-actin mRNA and protein to neurites and growth cones in culture. Journal of Neuroscience. 1998;18:251–265. - PMC - PubMed
    1. Burkhardt JK, Echeverri CJ, Nilsson T, Vallee RB. Overexpression of the dynamitin (p50) subunit of the dynactin complex disrupts dynein-dependent maintenance of membrane organelle distribution. J Cell Biol. 1997;139:469–484. - PMC - PubMed
    1. Chao DL, Ma L, Shen K. Transient cell-cell interactions in neural circuit formation. Nat Rev Neurosci. 2009;10:262–271. - PMC - PubMed
    1. Chen X, Xu H, Yuan P, Fang F, Huss M, Vega VB, Wong E, Orlov YL, Zhang W, Jiang J, et al. Integration of external signaling pathways with the core transcriptional network in embryonic stem cells. Cell. 2008;133:1106–1117. - PubMed

Publication types

MeSH terms

LinkOut - more resources