Fig 2: Analysis of SCG2 concentration in clinical serum samples. SCG2 concentration in serum samples of patients and control groups was analyzed with enhanced nanoplasmonic immunosensor (E-NPIS) (A) and ELISA (B). The serum samples were 1:10 diluted (E-NPIS) or 1:5 diluted (ELISA) with PBS buffer. Data represent the means ± s.d. of three replicates from 5 controls and 8 patients for E-NPIS analysis, 2 controls and 6 patients for ELISA analysis. SCG2 concentration was below the LOD of ELISA for some clinical samples. P = 0.089 for E-NPIS and p = 0.383 for ELISA (Student’s t-test). Basic information of clinical serum samples was described in Table S1 (Supplementary information).

Fig 3: Enhanced nanoplasmonic immunosensor based on tyramide amplification strategy. (A) Schematic illustration of the enhanced nanoplasmonic immunosensor using enzyme precipitation reaction combined with tyramide signal amplification for signal enhancement. After immunoreaction is completed, the tyramide-biotin conjugates were deposited by HRP (horseradish peroxidase) catalyzed reaction. A subsequent reaction with STA-AP results in the localization of the enhancement of the AP signal at the site of tyramide deposition. (B) Standard curve was prepared by plotting the ?? measured with varying concentration of SCG2 (from 0 to 30 ng/mL). The inset plot shows the assay response in the low concentration area at below 1.0 ng/mL. The regression equation was y = 0.7947x + 1.5427 (R2 = 0.9983) in the linear range. (C) The correlation between enhanced nanoplasmonic immunosensor (E-NPIS) and conventional ELISA for the detection of SCG2 (2 R2 = 0.9844). Each data point is the average of N = 3 individual measurements, and the error bars indicate standard deviation.

Fig 4: Dendritic arborization and synaptic formation attenuated by SCG2 overexpression. (A) Attenuated dendritic arborization by SCG2 overexpression. Rat hippocampal neurons were transfected with human SCG2 clone (hSCG2) (Fig. S1B). The numbers of dendrites and arborization significantly decreased after SCG2 overexpression (SCG2 OE) as compared with that in the control group that expressed EGFP only (Cont). Means \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\pm$$\end{document}± s.e.m. of data from 10 control neurons (Cont) and 10 hSCG2 neurons (SCG2 OE) are shown. For primary dendrites: ***p < 0.001; for secondary dendrites: ***p < 0.001; for Sholl analysis of arborization: *p < 0.05, **p < 0.01, and ***p < 0.001 (Student’s t-test). Scale bar: 50 µm. (B–D) Decreased synaptic formation by overexpression of SCG2. The density of dendritic spines and the numbers of excitatory and inhibitory synapses significantly decreased after SCG2 overexpression (SCG2 OE) as compared with that in the control group (Cont). Means \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\pm$$\end{document}± s.e.m. of data from 28 dendrites of 7 control neurons (Cont) and 28 dendrites of 7 hSCG2 neurons (SCG2 OE). (B) The density of dendritic spines: *p < 0.05. (C) The numbers of excitatory synapses: *p < 0.05. (D) The numbers of inhibitory synapses: **p < 0.01 (Student’s t-test). Scale bar: 10 µm.

Fig 5: Dendritic arborization and synaptic formation attenuated by SCG2 knockdown. (A) Decreased SCG2 in the brain tissues of PTPRT-/- null mice. SCG2 expression was examined in the cortical and hippocampal tissues of PTPRT-/- null mice. Western blotting displayed that SCG2 levels relative to ß-actin decreased in PTPRT-/- null mice as compared with wild-type mice (n = 7, 6). **p < 0.01 (Student’s t-test). The original images are presented in Supplementary Fig. S6. (B) Attenuated dendritic arborization by SCG2 knockdown. Rat hippocampal neurons were transfected with SCG2-shRNA to knock down SCG2 expression (Fig. S7A). The numbers of dendrites and arborization significantly decreased by SCG2 knockdown (SCG2 KD) as compared with the control group (Cont), and were rescued by the addition of Res SCG2 resistant to SCG2-shRNA to this context (SCG2 res) (Fig. S7B). Means \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\pm$$\end{document}± s.e.m. of data from 12 control neurons (Cont), 12 SCG2-shRNA neurons (SCG2 KD), and 14 SCG2-shRNA + Res SCG2 neurons (SCG2 res) are shown. By the Turkey post hoc test after application of one-way ANOVA, primary dendrites: *p < 0.05 and **p < 0.01, F = 6.887, p = 0.003; secondary dendrites: **p < 0.01 and ***p < 0.001, F = 45.80, p < 0.0001. Sholl analysis of arborization: *p < 0.05, **p < 0.01, and ***p < 0.001 (Student’s t-test). Scale bar: 50 µm. (C–E) Decreased synaptic formation by SCG2 knockdown. The density of dendritic spines and the numbers of excitatory and inhibitory synapses significantly decreased by SCG2 knockdown (SCG2 KD) as compared with the control group (Cont), and were rescued by the addition of Res SCG2 to this context (SCG2 res). Means \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\pm$$\end{document}± s.e.m. of data from 28 dendrites of 7 control neurons (Cont), 26 dendrites of 7 SCG2-shRNA neurons (SCG2 KD), and 26 dendrites of 7 SCG2-shRNA + Res SCG2 neurons (SCG2 res) are shown. By the Turkey post hoc test after application of one-way ANOVA, the density of dendritic spines: ***p < 0.001, F = 66.67, p < 0.0001; the numbers of excitatory synapses: *p < 0.05 and ***p < 0.001, F = 24.67, p < 0.0001; the numbers of inhibitory synapses: *p < 0.05 and ***p < 0.001; F = 37.65, p < 0.0001. Scale bar: 10 µm.