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
. 2013 Jul 10;79(1):39-53.
doi: 10.1016/j.neuron.2013.05.017. Epub 2013 Jun 20.

SAD kinases sculpt axonal arbors of sensory neurons through long- and short-term responses to neurotrophin signals

Brendan N Lilley  1 , Y Albert Pan, Joshua R Sanes
Affiliations

SAD kinases sculpt axonal arbors of sensory neurons through long- and short-term responses to neurotrophin signals

Brendan N Lilley et al. Neuron. .

Abstract

Extrinsic cues activate intrinsic signaling mechanisms to pattern neuronal shape and connectivity. We showed previously that three cytoplasmic Ser/Thr kinases, LKB1, SAD-A, and SAD-B, control early axon-dendrite polarization in forebrain neurons. Here, we assess their role in other neuronal types. We found that all three kinases are dispensable for axon formation outside of the cortex but that SAD kinases are required for formation of central axonal arbors by subsets of sensory neurons. The requirement for SAD kinases is most prominent in NT-3 dependent neurons. SAD kinases transduce NT-3 signals in two ways through distinct pathways. First, sustained NT-3/TrkC signaling increases SAD protein levels. Second, short-duration NT-3/TrkC signals transiently activate SADs by inducing dephosphorylation of C-terminal domains, thereby allowing activating phosphorylation of the kinase domain. We propose that SAD kinases integrate long- and short-duration signals from extrinsic cues to sculpt axon arbors within the CNS.

PubMed Disclaimer

Figures

Figure 1
Figure 1. LKB1 and SAD kinases are dispensable for axon formation by subcortical neurons
(A–F) Neurofilament immunohistochemistry of hindbrain (A,B), spinal cord (C,D) and retina (E,F) from E18.5 control and LKB1Nestin-cre animals reveals that LKB1 deletion in neural progenitors does not affect axon formation throughout the neuraxis. (G–L) Immunohistochemistry for NF-H and Isl1/2 in E13.5 control (G,I,K) or SAD-A/B−/− (H,J,L) hindbrain (G,H), spinal cord (I,J) and retina (K,L) shows that SADs are also dispensable for axon formation in these regions. Scale bars, 100 µm. See also Figure S1.
Figure 2
Figure 2. Ia proprioceptive sensory neurons require SADs, but not LKB1, for central axon projections
(A–E) Immunohistochemistry reveals SAD-A/B in DRG at P0 (A), absent in SADIsl1-cre mutants (B); in spinal cord at E13.5 (C); in intramuscular axons (D; co-labeled with antibodies to neurofilaments and SytII, D’) and in whisker pad at E15.5 (E; co-labeled with TUJ1, E’). (F–I) E15.5 transverse spinal cord sections from animals labeled with DiI (F,H) or anti-Parvalbumin (G,I) show loss of ventral horn projections in SADIsl1-cre mutants compared to controls. (J–P) Ventral horn projections remain truncated in SADIsl1-cre mutants at E18.5 (J,K) and P8 (M,N), whereas deletion of LKB1 has no effect (L,O,P). Scale bars, 50 µm. See also Figure S2.
Figure 3
Figure 3. SAD kinases regulate central, but not peripheral axon differentiation in NT-3 dependent sensory neurons
(A,B) Unilateral labeling of T12 DRG shows loss of midline crossing projections both medially and laterally in the dorsal horn of SADIsl1-cre mutants. Arrows indicates midline. (C–F) Immunohistochemistry of dorsal horn from lumbar spinal cord of P0 (C,E) or P8 (D,F) animals. Loss of SAD kinases in sensory neurons does not effect central projection of axons labeled with TrkA (C,E), CGRP (D,F, red) or VGLUT1 (D,F, green). (G–H’’) Immunohistochemistry using antibodies to PV and NF-H (G’,H’) shows that SAD-deficient axons are prominent in the cuneate fascicle (CF), but fail to grow into the cuneate nucleus (outlined, CN). (I,J) Transganglionic DiI labeling of the B2 whisker follicle shows defects in central axon projections of whisker afferents. DiI labeled axons on the left are in the spinal trigeminal tract adjacent to the SpVc, collaterals branch off and grow into the SpVc and form dense arbors. SADIsl1-cre mutant axons fail to elaborate axon arbors that enter into the correct target (outlined); growth cones tip many of the labeled axons (arrows). (K–R) Parvalbumin immunohistochemistry of E17.5 triceps muscle showing that annulospiral endings of IaPSNs (K,M) and flame-shaped Golgi tendon organs of IbPSNs (L,N) differentiate normally in SADIsl1-cre animals. (O,Q) Immunohistochemistry for TUJ1 (axons) and TROMA-I (Merkel cells) of P0 skin at the dorsal midline shows normal formation of Merkel cell-neurite complexes in SADIsl1-cre mutants. (P,R) TUJ1 immunohistochemistry of P0 mystacial pad shows that whisker follicles (wf) and epidermis (epi) are innervated normally in the absence of SADs. Scale bars, 50 µm. See also Figure S3.
Figure 4
Figure 4. SAD kinases act downstream of NT-3 to induce axon outgrowth
(A–E) Sections of lumbar DRG stained for PV at E15.5 (A,B) or for ER81 at E13.5 (C,D) and quantification of PV+ soma size (E, mean ± SEM, n≥60 cells per genotype). SADs are not required for transducing signals from peripheral NT-3 sources. (F–J) Axon outgrowth of DRG explants from control (F,H) and SADIsl1-cre mutants (G,I) grown for 2DIV with the indicated neurotrophin shows that SADs are required for NT-3 induced growth in vitro. Scale bars in A, C and F, 100 µm. Quantification (J) shows large decreases in NT-3-induced axon outgrowth and modest effects on NGF-induced growth (mean ± SEM from three experiments; NT-3: n ≥40 ganglia per genotype, p<.0001; NGF: n ≥36 ganglia per genotype; p=.005, unpaired t-test). (K) Immunoblot from control and SADIsl1-cre E13.5 lumbar DRGs shows loss of SAD pALT reactivity in double KO SADIsl1-cre animals. TUJ1 is a loading control. (L) E13.5 + 2DIV DRG cultures grown in NT-3 were starved of NT-3 for 5 hours, then NT-3 was added for the indicated times before cells lysis. (M) Growth of Bax−/− DRG neurons in the absence of neurotrophin for 3 days, followed by stimulation with NT-3 for 15 minutes induces SAD ALT phosphorylation. (N) DRG neurons were grown in the presence of NT-3 for 2DIV then were either grown with NT-3 (+) or without NT-3 for 18 hours (−), or deprived of NT-3 for 18 hours followed by growth with NT-3 for 24 hours (- -> +). See also Figure S4.
Figure 5
Figure 5. NT-3 post-translationally regulates SAD protein expression via the Raf/MEK/ERK pathway
(A) qRT-PCR of SAD-A and SAD-B mRNA shows that SAD mRNA levels are the same in NT-3 starved and NT-3 treated cultures (mean ± SEM, three experiments; 1.0 equals levels in NT-3 treated samples). (B) DRG neurons infected with lentiviral vectors expressing GFP-tagged versions of wild type (WT) or D-box mutant (DBM) SAD-A were grown with NT-3 for 2DIV followed by additional NT-3 treatment (+) or starvation (−) for 18 hours. Quantification of GFP immunoblot signals of (−) relative to (+) NT-3 (mean ± SEM, N=3, p=.025) is shown below representative blot. (C) DRG neurons cultured for 2 DIV with NT-3 were starved for 18 hours (−), followed by NT-3 re-addition for 24 hours in the presence of vehicle, MEK1/2 inhibitor (PD-325901, 10 µM), PI3K inhibitor (LY294002, 50 µM) or mTORC1 inhibitor (rapamycin, 100 nM). (D) DRG neurons were grown for 2DIV in either NT-3 and were treated or starved as in (B) or were treated with NT-3 and MEK1/2 inhibitor. (E) DRG neurons infected with B-RAF V600E-expressing lentivirus were cultured as in (B). (F) Bax-deficient DRG neurons not treated with neurotrophic factors were infected as in (D) and were lysed at 3DIV. All samples were analyzed by immunoblotting with the indicated antibodies. (G) Summary of pathway regulating SAD protein levels.
Figure 6
Figure 6. Inhibitory phosphorylation in SAD C-terminal domains blocks kinase activation
(A) Immunoblots from P0 cortices of animals of the indicated SAD genotype shows that SADs migrate heterogeneously in SDS-PAGE and that the predominant SAD-A pALT species migrates at 76 kDa. (B) Two color immunoblot of lysates from HeLa cells transfected with LKB1 and SAD-A shows that SAD-A is present in 85 and 76 kDa forms, but only the 76 kDa form is ALT phosphorylated. (C) Serum starved HeLa cells transfected with SAD-A without or with TrkC were treated with 50 ng/ml NT-3 for the indicated times followed by cell lysis and immunoblotting. SAD-A shifts from a predominantly 85 kDa form to the 76 kDa form upon NT-3 treatment only in TrkC+ cells. (D) Treatment of lysates of DRG cultures or HeLa cell transfectants with λ protein phosphatase shows that phosphorylation accounts for the difference in molecular weight. (E) Domain structure of SAD-A and Phosphomouse display of identified sites of phosphorylation in mouse brain (yellow lines). (F) Analysis of lysates and anti-HA (SAD-A) immunoprecipitates from HeLa cells expressing WT or 18A SAD-A shows that WT, but not 18A is reactive with anti-p[S/T]P. The anti-p[S/T]P signal in lysate lanes is from non-SAD-A species. (G) Immunoprecipitation of SAD-A followed by kinase assay with purified LKB1/STRAD/MO25. Only dephospho-CTD SAD-A is ALT phosphorylated. (H) Co-expression of Tau and SAD-A WT or 18A in 293T cells followed by immunoblotting. SAD-A18A shows increased kinase activity towards Tau (S262) and increased pALT reactivity. (I) Model for how SAD-A CTD phosphorylation inhibits ALT phosphorylation and activity. See also Figures S5 and S6.
Figure 7
Figure 7. Multiple pathways regulate the phosphorylation state of the SAD-A CTD
(A) Treatment of SAD-AWT expressing HeLa cells or DRG neurons with 10 µM Roscovitine for 30 minutes causes SAD-A to shift mobility to the dephospho-CTD form. (B) Co-expression of SAD-A (WT or 18A) with CDK5 (WT or Kinase dead: D145N) and p35 in 293T cells followed by immunoblotting. Active CDK5 causes CTD phosphorylation, mobility shift and loss of pALT reactivity of SAD-AWT; SAD-A18A is not affected. (C) Immunoprecipitation for SAD-A followed by immunoblotting shows that NT-3 treatment of TrkC/SAD-A expressing HeLa cells causes dephosphorylation of CTD [S/T]P sites. (D) Treatment of TrkC/SAD-A expressing HeLa cells with MEK1/2 or PLCγ inhibitors partially blocks SAD-A CTD dephosphorylation in response to NT-3. (E) Inhibition of NT-3 induced SAD ALT phosphorylation in DRG neurons by blocking of PLCγ (U73122), but not MEK1/2 (U0126). Ca2+ ionophore A23187 induces SAD ALT phosphorylation independently of NT-3. (F) Treatment of serum starved HeLa cells with calcium ionophore A23187 (5 µM) for 5 minutes induces SAD-A CTD dephosphorylation independently of MEK1/2 activity. (G) Model for NT-3 dependent regulation of SADs. See also Figure S6.
Figure 8
Figure 8. Increasing SAD activation induces DRG axon branching in vitro
(A–D) Representative images of neurons expressing empty vector control or the indicated SAD-A cDNA constructs. Scale bar equals 100 µm. (E,F) Quantification of neurite length and branch points in cultured neurons. Differences among means in (E) did not reach statistical significance (p>.01 using one way ANOVA with Tukey’s multiple comparisons test; graph shows mean ± SEM; Vector: n=76; WT: n=87; T175A: n=101; 18A: n=93) whereas the difference in branch numbers (F) between SAD-A18A and all others were highly significant (mean ± SEM, p<.0001, One way ANOVA with Tukey’s multiple comparisons test). (G) Model for how NT-3 signaling controls SAD kinase activity by regulating SAD levels and the fraction of SAD that is ALT phosphorylated.
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. Airaksinen MS, Koltzenburg M, Lewin GR, Masu Y, Helbig C, Wolf E, Brem G, Toyka KV, Thoenen H, Meyer M. Specific subtypes of cutaneous mechanoreceptors require neurotrophin-3 following peripheral target innervation. Neuron. 1996;16:287–295. - PubMed
    1. Albers KM, Perrone TN, Goodness TP, Jones ME, Green MA, Davis BM. Cutaneous overexpression of NT-3 increases sensory and sympathetic neuron number and enhances touch dome and hair follicle innervation. J Cell Biol. 1996;134:487–497. - PMC - PubMed
    1. Alessi DR, Sakamoto K, Bayascas JR. LKB1-dependent signaling pathways. Annu Rev Biochem. 2006;75:137–163. - PubMed
    1. Alvarado-Kristensson M, Rodríguez MJ, Silió V, Valpuesta JM, Carrera AC. SADB phosphorylation of gamma-tubulin regulates centrosome duplication. Nat Cell Biol. 2009;11:1081–1092. - PubMed
    1. Arber S, Ladle DR, Lin JH, Frank E, Jessell TM. ETS gene Er81 controls the formation of functional connections between group Ia sensory afferents and motor neurons. Cell. 2000;101:485–498. - PubMed

Publication types