Yang JJ et al. (2019), The Expression of Key Guidance Genes at a Foreb...

XB-ART-55877

eNeuro March 1, 2019; 6 (2):

The Expression of Key Guidance Genes at a Forebrain Axon Turning Point Is Maintained by Distinct Fgfr Isoforms but a Common Downstream Signal Transduction Mechanism.

Yang JJ , Bertolesi GE , Dueck S , Hehr CL , McFarlane S .

Abstract
During development the axons of neurons grow toward and locate their synaptic partners to form functional neural circuits. Axons do so by reading a map of guidance cues expressed by surrounding tissues. Guidance cues are expressed at a precise space and time, but how guidance cue expression is regulated, and in a coordinated manner, is poorly understood. Semaphorins (Semas) and Slits are families of molecular ligands that guide axons. We showed previously that fibroblast growth factor (Fgf) signaling maintains sema3a and slit1 forebrain expression in Xenopus laevis, and these two repellents cooperate to guide retinal ganglion cell (RGC) axons away from the mid-diencephalon and on towards the optic tectum. Here, we investigate whether there are common features of the regulatory pathways that control the expression of these two guidance cues at this single axon guidance decision point. We isolated the sema3a proximal promoter and confirmed its responsiveness to Fgf signaling. Through misexpression of truncated Fgf receptors (Fgfrs), we found that sema3a forebrain expression is dependent on Fgfr2-4 but not Fgfr1. This is in contrast to slit1, whose expression we showed previously depends on Fgfr1 but not Fgfr2-4. Using pharmacological inhibitors and misexpression of constitutively active (CA) and dominant negative (DN) signaling intermediates, we find that while distinct Fgfrs regulate these two guidance genes, intracellular signaling downstream of Fgfrs appears to converge along the phosphoinositol 3-kinase (PI3K)-Akt signaling pathway. A common PI3K-Akt signaling pathway may allow for the coordinated expression of guidance cues that cooperate to direct axons at a guidance choice point.

PubMed ID: 30993182
PMC ID: PMC6464512
Article link: eNeuro

Genes referenced: actb akt1 dynll1 fgfr1 fgfr2 fgfr3 fgfr4 gab1 grb2 hpse map3k1 mapk1 myc pik3ca pik3cg sdc2 sema3a slit1 slit2 spry1 tec tubb
GO keywords: axon guidance [+]
Antibodies: Akt Ab10 Gapdh Ab4 Mapk1 Ab7 Sema3a Ab1

Article Images: [+] show captions

	Figure 1. sema3a is co-expressed with fgfrs in the embryonic forebrain, but in complementary domains to slit1. A, dFISH on transverse sections through the forebrain using specific antisense riboprobes against fgfr1-4 (green) and sema3a (red). There is co-expression (yellow) of sema3a with all fgfrs in the ventricular zone of the forebrain. The insets reveal co-localization of dFISH signal. The rightmost column shows cartoons of the fgfr (green) and sema3a (red) domains, with co-expression in yellow. The fgfr2 region of co-expression is restricted to the dorsal ventricular zone (unfilled arrowhead). B, dFISH on transverse sections through the forebrain by using specific antisense riboprobes against slit1 (green) and sema3a (red). sema3a is expressed by cells around the forebrain ventricle, whereas slit1 is localized to the pial cells and the floor plate of the neural tube. There is some limited co-expression (yellow) of sema3a and slit1 at the interface (asterisk) of the two expression domains. The rightmost column shows cartoons of the slit1 (green) and sema3a (red) domains, with co-expression in yellow. Scale bars, 50 µm. fp, floor plate; ve, ventricle.
	Figure 2. Characterization of the sema3a promoter. A, Schematic of the sema3a deletion fragments inserted upstream of the firefly luciferase (luc) gene in the pGL3 vector. B, XTC cells express endogenous Sema3a protein by Western blotting. C, The reporter constructs of the sema3a promoter fragments and the Renilla luciferase construct were co-transfected into XTC cells. Luminescence corresponding to each deletion fragment was normalized to Renilla to account for transfection efficiency and expressed as relative light units compared to the promoterless pGL3 basic vector. Each bar represents mean ± SEM from n = 9 wells, N = 3. Statistical significance was determined by one-way ANOVA (α = 0.05) and the following post hoc tests: ♦p < 0.05 and p < 0.01 by Dunnett’s multiple comparison test versus the control (pGL3 vector) and #p < 0.05 by Bonferroni’s multiple comparison test between selected deletion fragments. D, Expression of fgfr genes in XTC cells, shown by RT-PCR. XTC cells express all the fgfr genes (as verified by sequencing of amplicons) except for fgfr4a/b. Of note, the fgfr1b amplicons smaller than 900 bp may be alternatively spliced variants (Friesel and Dawid, 1991). E, F, The heparanase promoter construct (E, n = 12 wells, N = 2 for both bars), –2119 –55 hpse::luc, and the –2930 +63 sema3a::luc construct (F, n = 12 wells, N = 3 for both bars) were co-transfected into XTC cells in 96-well plates with the Renilla luciferase plasmid and treated with 100 µM SU5402. Bars reflect mean ± SEM; p < 0.05, compared to DMSO control, two-tailed Student’s t test. G, –2930 +63 sema3a::luc was co-transfected with truncated fgfrs in XTC cells (N = 7 for all bars; control n = 30 wells, DN fgfr1 n = 16 wells, DN fgfr2 n = 12 wells, sfgfr3 n = 14 wells). Bars reflect mean ± SEM; *p < 0.05 by ANOVA and Bonferroni post hoc test compared to pCS2-GFP transfection as control.
	Figure 3. Fgfr2-4 inhibition downregulates sema3a in the forebrain. A, A cartoon of stage 27/28 forebrain electroporation. B, C, Embryos were electroporated with pCS2-DNfgfr1 (B) and pCS108-DNfgfr2 (C) and processed in transverse cryostat sections by FISH for fgfr1/2 at stage 32 to confirm transgene expression. D–F, Embryos were electroporated with control pCS2-GFP (D, n = 37 brains, N = 3), pCS2-DNfgfr1 (E, n = 20 brains, N = 3), or pCS108-DNfgfr2 (F, n = 16/24 brains had decreased expression, N = 3) and processed for sema3a expression by wholemount ISH at stage 32. G, sema3a mRNA levels in brains electroporated with truncated fgfrs was measured by RT-qPCR. spry1 was a readout of Fgfr inhibition. Bars represent mean ± SEM for control (n = 32, N = 4), DN fgfr1 (n = 24, N = 4), DN fgfr2 (n = 20, N = 3), sfgfr3 (n = 30, N = 4), and DN fgfr4 (n = 31, N = 4); *p < 0.05 statistical significance versus the control was determined using the REST algorithm. Scale bars, 50 µm. di, diencephalon; fp, floor plate; h, hypothalamus; pi, pineal gland; tec, optic tectum; tel, telencephalon; ve, ventricle.
	Figure 4. PI3K inhibition decreases both sema3a and slit1 expression in the forebrain. A, The canonical sequence of Fgf signaling intermediates is indicated by solid arrows. Dotted arrows show further signaling through downstream factors (Rodriguez-Viciana et al., 1994; Zimmermann and Moelling, 1999; Jiminez et al., 2002; Nakayama et al., 2008; Aksamitiene et al., 2012; Mendoza et al., 2011; Chen et al., 2012; Selvaraj et al., 2014). The activated complex of HSPG-FGF-FGFR dimerizes to phosphorylate the FGFR intracellular tails and recruits FRS2α, GRB2, GAB1, SOS, and RAS (Rubinfeld and Seger, 2004; Lemmon and Schlessinger, 2010). GAB1 activates the PI3K-AKT relay (orange). RAS activates the RAF-MEK-ERK cascade, i.e., MAPK signaling (green). PLCγ can directly bind to the phosphorylated FGFR to hydrolyze PIP2 to IP3 and DAG (magenta). The inhibitors against FGFRs, MEK, PI3K, and PLCγ are SU5402, U0126, LY294002, and U73122, respectively. CA RAF and CA MEKK1 were used to overactivate the ERK MAPK pathway. DN MEKK1 was used to inhibit the MAPK pathway. The CA and DN PKB (AKT) constructs were used for gain and loss of function of AKT signaling, respectively. Wild-type PLCγ was overexpressed to increase PLCγ signaling, and the PLCγ-deficient DN FGFR was used to inhibit PLCγ activation. B–G, Stage 32 brains were treated with control (B, n = 46 brains, N = 4; C, n = 39 brains, N = 3) or 10 µM U73122 solution and processed by wholemount ISH for sema3a (D, n = 12 brains, N = 3) and slit1 (E, n = 11 brains, N = 2). Black arrowheads indicate the sema3a and slit1 domains of interest. Mek inhibition with 100 µM U0126 treatment did not visibly decrease sema3a (F, n = 11 brains, N = 2) or slit1 (G, n = 11 brains, N = 2) expression by wholemount ISH. H, Phospho-Erk (pErk) knockdown in U0126-treated forebrains confirmed by Western blot analysis (n = 18 brains and N = 3 for both treated and control). I, J, RT-qPCR of sema3a and slit1 forebrain mRNA following treatment with U0126 (I, n = 23 brains and N = 4 for both U0126 and control) and 100 µM SU5402 (J, n = 17 brains and N = 3 for both SU5402 and control). K–M, PI3K inhibition with 25 µM LY294002 treatment decreased sema3a (K, n = 15/20 LY294002 brains had decreased expression vs the control in B, N = 3) and slit1 (L, n = 11/15 LY294002 brains had decreased expression vs the control in C, N = 2) expression by ISH. slit1 expression in the floor plate (L, unfilled arrowhead) was unaffected by LY294002 treatment. M, RT-qPCR for sema3a and slit1 mRNA with LY294002 treatment (n = 11 brains and N = 2 for both LY294002 and control). In all RT-qPCR data, bars represent the mean ± SEM; *p < 0.05 using the REST algorithm for statistical significance. Scale bar, 50 µm. chi, optic chiasm; di, diencephalon; fp, floor plate; h, hypothalamus; pi, pineal gland; tec, optic tectum; tel, telencephalon.
	Figure 5. Activation of the MAPK pathway is sufficient to inhibit both sema3a and slit1 expression in the forebrain. Constructs for Fgfr downstream signaling intermediates were electroporated into the forebrain of stage 27/28 embryos, and spry1, sema3a, and slit1 expression in the forebrain were quantified by RT-qPCR 24 h after electroporation (stage 32). A, B, RT-qPCR when PLCγ signaling was augmented by pEYFP-PLCγ (A, PLCγ n = 32 brains, N = 4 vs control n = 31, N = 4), and inhibited with the PLCγ-deficient FGFR1, phRluc-N1-FGFR1Y766F (B, FGFR1Y766F n = 22 brains, N = 3 vs control n = 24, N = 3). C–G, The MAPK pathway was inhibited with a DN MEKK1, pCS2-MEKK1-KM, and activated with either pCS2-MEKK1+ or pCS108-BRAFV600E. Immunohistochemistry on transverse sections of the forebrain against myc (C, E) and GFP (G) tags confirm successful electroporations. Changes in gene expression was quantified by RT-qPCR after electroporation with MEKK1-KM (D, n = 30, N = 4), MEKK1+ (F, n = 28, N = 4), and BRAFV600E (H, n = 32, N = 4) and compared to control pCS2-GFP electroporation (D, F, H, n = 31, N = 4). Bars represent the mean ± SEM; *p < 0.05 using the REST algorithm for statistical significance. Scale bar, 50 µm. ve, ventricle.
	Figure 6. Akt signaling positively regulates sema3a and slit1 in the forebrain. Plasmids encoding DN AKT (A, B) T308A/S473APKB, and CA AKT (C, D) T308D/S473DPKB, and (E, F) AKT-myr were electroporated into the forebrain of stage 27/28 embryos. A, C, E, Forebrain transgene expression was assessed by immunohistochemistry for HA-tagged AKT mutants. B, D, F, spry1, sema3a, and slit1 expression in the forebrain 24 h after electroporation was quantified by RT-qPCR for (B) T308A/S473APKB (n = 21, N = 3 vs control n = 23, N = 3), (D) T308D/S473DPKB (n = 40, N = 5 vs control n = 39, N = 5), and (F) AKT-myr (n = 22, N = 3 vs control n = 24, N = 3), where the control electroporation was pCS2-GFP. Bars represent the mean ± SEM; *p < 0.05 using the REST algorithm for statistical significance. Scale bar, 50 µm. ve, ventricle.
	Figure 7. PI3K signaling regulates the behavior of RGC axons at the mid-diencephalic turn. Lateral views of stage 40 wholemount brains with HRP-labeled optic tracts that were exposed at stage 33/34 either to (A) control DMSO solution (n = 16 brains, N = 3) or (B, C) 20 µM LY294002 (n = 33 brains, N = 3). Black line represents the approximate anterior border of the optic tectum, and the asterisk the location of the mid-diencephalic turn guidance choice point. D, The RGC axon stall phenotype was quantified by representing the width of the optic tract post-mid-diencephalic turn (W2) as a ratio (W2/W1) to the width of the optic tract (W1) at the mid-diencephalic turn. On each bar is the number of individual brains pooled from 3 independent experiments. Bars represent the mean ± SEM; ***p < 0.001 using the unpaired Student’s t test. Scale bar, 50 µm. Hb, hindbrain; Pi, pineal gland; Tec, tectum; Tel, telencephalon; v.di, ventral diencephalon.
	slit1 (slit guidance ligand 1) gene expression in brain of Xenopus laevis embryo, assayed via in situ hybridization, NF stage 32, lateral view, anterior left, dorsal up.
	sema3a (semaphorin 3A) gene expression in brain of Xenopus laevis embryo, assayed via in situ hybridization, NF stage 32, lateral view, anterior left, dorsal up.

References [+] :

Aksamitiene, Cross-talk between mitogenic Ras/MAPK and survival PI3K/Akt pathways: a fine balance. 2012, Pubmed