Semaphorin 3A-Glycosaminoglycans Interaction as Therapeutic Target for Axonal Regeneration

Abstract

Semaphorin 3A (Sema3A) is a cell-secreted protein that participates in the axonal guidance pathways. Sema3A acts as a canonical repulsive axon guidance molecule, inhibiting CNS regenerative axonal growth and propagation. Therefore, interfering with Sema3A signaling is proposed as a therapeutic target for achieving functional recovery after CNS injuries. It has been shown that Sema3A adheres to the proteoglycan component of the extracellular matrix (ECM) and selectively binds to heparin and chondroitin sulfate-E (CS-E) glycosaminoglycans (GAGs). We hypothesize that the biologically relevant interaction between Sema3A and GAGs takes place at Sema3A C-terminal polybasic region (SCT). The aims of this study were to characterize the interaction of the whole Sema3A C-terminal polybasic region (Sema3A 725-771) with GAGs and to investigate the disruption of this interaction by small molecules. Recombinant Sema3A basic domain was produced and we used a combination of biophysical techniques (NMR, SPR, and heparin affinity chromatography) to gain insight into the interaction of the Sema3A C-terminal domain with GAGs. The results demonstrate that SCT is an intrinsically disordered region, which confirms that SCT binds to GAGs and helps to identify the specific residues involved in the interaction. NMR studies, supported by molecular dynamics simulations, show that a new peptoid molecule (CSIC02) may disrupt the interaction between SCT and heparin. Our structural study paves the way toward the design of new molecules targeting these protein-GAG interactions with potential therapeutic applications.

Keywords: NMR; glycosaminoglycan–protein interaction; peptoids; semaphorin 3A.

Figure 1

Schematic illustration of the mechanism of action proposed for our small cationic peptidomimetics through the direct inhibition of a biologically relevant Sema3A polybasic C-terminal region (SCT)–GAG interaction. Sema3A forms ternary complexes with its receptor proteins (Nrps-Plxs; Nrp1-PlxnA2 is as example of interaction partners). The figure shows the proposed inhibition mechanism of the Sema3A pathway by interfering the colocalization of secreted Sema3A with its receptor proteins (Nrp1-PlxnA2) due to CSIC02 binding to GAG sites. Long black lines represent the protein part of proteoglycans (PGs), red lines represent negatively charged sugar (GAG) chains. Sema3A/Nrp1/PllnxA2 domains: PSI (plexin-semaphorin-integrin), Ig (immunoglobulin), SCT (basic), IPT (transcription factors), GAP (GTPase-Activating Protein), a1/a2 (CUB), b1/b2 (FV/VIII), and c (MAM).

Figure 2

Structural characterization of Sema3A C-terminal domain. The scheme shows the amino acid sequence for Sema3A C-terminal region (SCT_WT) and the peptides studied in our previous work (labeled in dark rose) [40]. For these Sema3A peptides, we used the terms FS2 and (N)FS3 (for Furin processing sites 2 and 3). (a) [¹H,¹⁵N]-HSQC spectrum of ¹³C,¹⁵N-labeled Sema3A C-terminal region at pH 5 (10 mM of acetate and 50 mM of NaCl, 298 K) with the assignment for backbone amides. (b) Bioinformatics analysis of the intrinsic disorder predisposition of the Sema3A C-terminal region obtained using IUPred2A, PONDR^® VLXT, and SPOT-Disorder2. Disordered segments are indicated by values higher than the default cut-off (0.5), lower values predict structured regions. (c) Structural propensity plot using ncSPC (neighbor connected structural propensity plot). (d) Estimation of secondary structure populations using δ2D. Red bars indicate helical population estimate, grey bars indicate beta population estimates, dark rose bars indicate PPII population estimates, and remaining white is % of random coil. Chemical shift values for HN, N, C’, Cα, and Cβ nuclei were used for (c,d) graph calculations.

Figure 3

SPR sensorgrams of increasing amounts (0 to 1 μM) of the SCT_WT construct flown over (HBS-T buffer) (a) immobilized biotinylated heparin (350 RU) (b) CD Spectra of 5 μM Sema3A C-terminal region in the absence (blue) and the presence of 1 μM UFH (orange), measured at pH 5.5 in 10 mM NaP. (c) Heparin sepharose affinity chromatography profile of SCT_WT, with a single peak eluting in ~1.06 M NaCl. SCT_WT was loaded onto a HiTrap Heparin column equilibrated with 15 mM Tris.HCl (pH 7.5), and eluted with a linear gradient to 2 M NaCl. Left ordinate axis, absorbance; righ ordinate axis, % NaCl (% B).

Figure 4

(a) Overlay of [¹H-¹⁵N]-HSQC spectra of 0.2 mM Sema3A C-terminal region in the absence (orange) and the presence of 0.05 mM (or 0.35 mM in disaccharide units) dp14 heparin oligosaccharide (purple). (b) Weighted chemical shift perturbation (CSP) of C-terminal Sema3A region in the presence of dp14 heparin. The absolute values of the chemical shift differences between the presence and absence of dp14 are plotted in ppm and calculated using the weighting factors, described in [57]. Spectra were acquired in 10 mM acetate, 150 mM NaCl, pH 4.5, 10% D₂O.

Figure 5

(a) Representative snapshot of the heparin/FS2 simulation. Heparin is shown as CPK balls and FS2 as sticks. (b) Interactions fraction per FS2 residue. The stacked bar charts are normalized over the course of the trajectory, such that a value of 1.0 suggest that the interaction is maintained over 100% of the simulation time. Values over 1.0 are possible as some residues may establish multiple contacts of same subtype. (c) Time dependence of the total number of interactions and of interactions between each residue of peptide FS2 and heparin. Numbering of peptide residues according to Sema3A sequence, ACE, and NMA correspond to acetyl and N-methylamide capping groups of the N- and C-termini.

Figure 6

(a) Chemical structures of triprotonated SICHI and CSIC02–04, as the main species in aqueous medium at neutral pH, and relative increase in MB absorbance at 665 nm versus charge ratio after the addition of the peptoids to the preformed MB/heparin complex (4.71 µM Hep repeating units + 9.83 µM MB in 5 mM Tris buffer at pH = 7.5). (b) 1D ¹H WaterLOGSY NMR experiments showing that CSIC02 binds to heparin agarose resin (used as a GAG mimetic).

Figure 7

Best docked poses of (a) SICHI, (b) CSIC02, (c) CSIC03 and (d) CSIC04 to a dp8 heparin model.

Figure 8

(a) Overlay of [¹H-¹⁵N]-HSQC spectra of 0.2 mM Sema3A C-terminal region (orange) highlighting the changes of 0.05 mM (or 0.35 mM in disaccharide units) dp14 heparin oligosaccharide binding in the absence (purple) and the presence (turquoise) of 4 mM CSIC02. (b) Weighted chemical shift perturbation (CSP) of C-terminal Sema3A region in presence of both dp14 heparin and CSIC02. The absolute values of the chemical shift differences between the presence and absence of dp14 are plotted in ppm and calculated comparing with the chemical shifts of the protein alone. (c) Overlay of [¹H-¹⁵N]-HSQC spectra of 0.2 mM Sema3A C-terminal region in the absence (orange) and the presence of both 0.05 mM dp14 and 4 mM CSIC02 (turquoise). We observed two species in slow exchange for some peaks (R733, R734, and Q735 in the GAG binding region). Spectra were acquired in 10 mM acetate, 150 mM NaCl, pH 4.5, 10% D₂O.