Fig 1.
Sema6A exerts constitutive and PlxnA2-dependent cell-autonomous functions.
(A-F) NIH3T3 cells expressing GFP alone (A and B), myc-Sema6A-FL (Sema6A; B and E), or myc-Sema6A-K393D (Sema6A-K393D; C and F) were treated with purified AP-Fc (AP; A-C) or PlxnA2-EC-Fc (PlxnA2-EC; D-F). Scale bar = 20 μm. (G) Graph represents the cell area in cells transfected with GFP, Sema6A or Sema6A-K393D and treated with AP or PlxnA2-EC; n = 100–200 cells per experimental condition. Data are expressed as mean ± s.e.m; ***P <0.001; Student’s t-test.
Fig 2.
Forward and reverse Sema6A-PlxnA2 interactions have different effects on the migration of cerebellar neurons.
(A-L) EGL explants of WT (A-D), PlxnA2-/- (E-H), and Sema6A-/- (I-L) mice cultured on a monolayer of NIH3T3 cells expressing dTomato alone (Control cells), or together with Sema6A-FL (Sema6A Cells), PlxnA2-FL (PlxnA2 Cells) and PlxnA2-A396E (PlxnA2-A396E Cells). Migrating granular neurons are visualised in green and NIH3T3 cells in red. Scale bar = 100 μm. (M) Graph summarising the migration of different EGL explants cultured together with different NIH3T3 layers; n = 50–100 explants per experimental condition. Data are expressed as mean ± s.e.m; *P < 0.05, **P ≤ 0.005 and ***P < 0.001; ***P < 0.001; one-way ANOVA followed by Bonferroni multiple comparison test.
Fig 3.
Sema6A signaling reduces the axonal length of granular neurons.
(A-C) GFP-expressing granular neurons were cultured on Control cells (A), Sema6A Cells (B), PlxnA2 Cells (C) and PlxnA2-A396E Cells (A396E Cells, D). Scale bar = 20 μm. (E) Graph represents the axonal length of granular neurons grown on different NIH3T3 layers; n = 100–200 cells per experimental condition. Data are expressed as mean ± s.e.m; ***P < 0.001; one-way ANOVA followed by Bonferroni multiple comparison test. (F) Graph represents the axonal length of PlxnA2+/+ (dark columns) and PlxnA2-/- (light columns) granular neurons grown on NIH3T3 cells. Neurons were transfected with GFP alone or together with PlxnA2-FL, and cultured on control (Ctrl) or on Sema6A cell layers; n = 100–200 cells per experimental condition. Data are expressed as mean ± s.e.m; ***P < 0.001; Student’s t-test. (G) Graph represents the axonal length of Sema6A+/- (dark columns) and Sema6A-/- (light columns) granular neurons grown on NIH3T3 cells. Neurons were transfected with GFP alone or together with Sema6A-FL, and cultured on control (Ctrl) or on PlxnA2 cell layers; n = 100–200 cells per experimental condition. Data are expressed as mean ± s.e.m; ***P < 0.001; Student’s t-test. (H-J) Granular neurons transfected with GFP alone (H and I) or together with Sema6A-∆cyt (J), and grown on control or on PlxnA2 cells. Scale bar = 20 μm. (K) Graph represents the axonal length of Sema6A+/- (dark columns) and Sema6A-/- (light columns) granular neurons. Neurons were transfected with GFP alone or together with Sema6A-∆cyt, and cultured on control (Ctrl) or on PlxnA2 cell layers; n = 100–200 cells per experimental condition. Data are expressed as mean ± s.e.m; ***P < 0.001; Student’s t-test.
Fig 4.
PlxnA2 interacts with Sema6A in cis and in trans.
(A) Scheme summarising Sema6A-AP bindings. (B-F): COS-7 cells were transfected with the empty vector (mock; B), Sema6A-FL (C), PlxnA2-FL (D), PlxnA2-FL together with Sema6A-FL (E) or PlxnA2-A396E (F); and treated with Sema6A-AP. (G) Scheme summarising PlxnA2-AP bindings. (H-K): COS-7 cells were transfected with the empty vector (mock; H), Sema6A-FL (I), PlxnA2-FL (J) or PlxnA2-FL together with Sema6A-FL (K) and treated with PlxnA2-AP. (L) PlxnA2 immunoprecipitation (IP) from protein samples of COS-7 cells transfected with Sema6A-FL, PlxnA2-FL or Sema6A-FL together PlxnA2-FL. Antibodies against PlxnA2 and Sema6A were used in the immunoblots. (M) Sema6A IP from protein samples of COS-7 cells transfected with Sema6A-FL, PlxnA2-FL or Sema6A-FL together with PlxnA2-FL. (N) PlxnA2 IP from protein samples of mouse cerebellums or EGL explants. (O) PlxnA2 IP from protein samples of COS-7 cells transfected with PlxnA2-FL, Sema6A-∆cyt or Sema6A-∆cyt together with PlxnA2-FL. (P) PlxnA2 IP from protein samples of COS-7 cells transfected with PlxnA2-A396D, Sema6A-FL or Sema6A-FL together with PlxnA2-A396D. (Q) PlxnA2 IP from protein samples of COS-7 cells transfected with PlxnA2-FL, Sema6A-K393D or Sema6A-K393D together with PlxnA2-FL.
Fig 5.
Sema6A cytoplasmic domain interacts with Abl and Mena.
(A) Immunoprecipitations from untreated or PlxnA2-EC-treated COS-7 cells previously transfected with different combinations of myc-Sema6A, Abl-V5 and PlxnA2. Immunoprecipitations (Ip) were performed employing an anti-V5 antibody. Antibodies against V5, myc and PlxnA2 were used in the immunoblots (Ib). (B) Immunoprecipitations from untreated or PlxnA2-EC treated COS-7 cells previously transfected with different combinations of myc-Sema6A, Mena-V5 and PlxnA2. Immunoprecipitations (Ip) were performed employing an anti-myc antibody. Antibodies against V5, myc and PlxnA2 were used in the immunoblots (Ib). (C) and (D) Interactions were quantified by immunoblotting and densitometric analysis. Histograms represent quantification of interaction normalized to the amount of immunoprecipitated protein (Abelson-V5 for Sema6A-Abelson or Sema6A-myc for Sema6A-Mena) from three independent experiments. Students t-test *p<0,05 indicating signifiance. S6A-Abl/S6A-Abl+PlA2 p = 0.013, S6A-Abl-PlA2/S6A-Abl-PlA2+PlA2 p = 0.034. Errors bars represent standard error.
Fig 6.
Sema6A reverse signaling is stimulated by multimerisation.
(A) Scheme representing how the induced multi-aggregation of the cytosolic domain of Sema6A activates signaling pathways. A chimeric construct was engineered by putting in frame the entire cytosolic domain of Sema6A (Sema6A-cyt) together with 3 self-aggregation FKBP domains, and GFP. The chimeric peptide targets the plasma membrane by the myristoylating sequence located at the N-terminal. The application of the drug FK1012 triggers the chemically induced multimerisation (CIM) of Sema6A-cyt via the self-aggregation of FKBP. (B) Cerebellar neurons transfected with MF3-GFP or MF3-Sema6A-cyt-GFP (MF3-Sema6A-cyt), and treated with 500nM FK1012 (“CIM”). GFP-positive aggregates are indicated (arrowheads). Scale bar = 20μm. (C) Axonal shaft of cerebellar neurons transfected with myc-Sema6A-FL or MF3-Sema6A-cyt-GFP (MF3-Sema6A-cyt), and treated with PlxnA2-EC or CIM respectively. Scale bar = 5 μm. (D) and (E) Graphs summarising the myc-Sema6A-FL aggregation in myc-Sema6A-FL-expressing neurons treated with PlxnA2-EC at different time points. The aggregation is represented as the average distance between Sema6A clusters, or the average size of Sema6A clusters; n = 50–100 cells per time point. Data are expressed as mean ± s.e.m; ***P < 0.001; one-way ANOVA followed by Bonferroni multiple comparison test. (F) MF3-Sema6A-cyt-expressing granular neurons were cultured on NIH3T3 cell layers, and treated with CIM. Scale bar = 50 μm. (G) Graph representing the axonal length of MF3-Sema6A-cyt-expressing cerebellar neurons treated with CIM for 24 and 48h; n = 50–100 cells per experimental condition. Data are expressed as mean ± s.e.m; ***P < 0.001; one-way ANOVA followed by Bonferroni multiple comparison test.
Fig 7.
Cell-contraction induced by PlxnA2-Sema6A interaction depends on Abl signaling.
(A-F) NIH3T3 cells expressing GFP (A and D), Sema6A-FL (B and E), and Sema6A-∆Abl (C and F) were treated with purified AP-Fc (AP; a-c) or PlxnA2-EC-Fc (PlxnA2-EC; D-F). Scale bar = 20 μm. (G) Scheme illustrates that the cell contraction is via Abl signaling. (H) Graph represents the contraction of the cell area in NIH3T3 transfected with GFP, Sema6A or Sema6A-∆Abl and treated with AP or PlxnA2-EC; n = 100–200 cells per experimental condition. Data are expressed as mean ± s.e.m; ***P < 0.001; Student’s t-test.
Fig 8.
Schematic of diverse Sema6A-PlxnA2 interactions.
Neighbouring cells are shown expressing either Sema6A (blue), PlxnA2 (red) or both molecules. (A) Signaling between cells expressing Sema6A alone and PlxnA2 alone can proceed in both directions (blue and red arrows). (B) Expression of Sema6A in cis blocks receptor function of PlxnA2 in the top cell but PlxnA2 can still act as a ligand to stimulate Sema6A receptor function in the bottom cell (blue arrow). (C) Sema6A-PlxnA2 complexes can act as receptors for PlxnA2 in trans (top cell), but not for Sema6A in trans (top cell in B). In some cases in vivo [12], Sema6A can still act as a ligand when co-expressed with PlxnA2 (red arrow, bottom cell). In others [10, 11, 51], PlxnA2 may directly occlude or indirectly antagonise this function (scenario not shown). (D) In populations of cells expressing both Sema6A and PlxnA2, the complexes should act as receptors for PlxnA2, through reverse signaling by Sema6A (blue arrows). The balance between these various interactions will likely depend on the precise stoichiometry of the proteins and may also be dynamically altered by interactions with additional proteins or newly encountered cells (e.g., ref. [53]).