Article Figures & Data
Figures
- Fig. 1.
Direct binding biotinylated NT3 or biotinylated DRK mutant to Trk or p75 receptors. Cells transfected with the indicated receptor were bound with the indicated concentrations of biotinylated NT3 or biotinylated DRK mutant, followed by labeled fluorescein-avidin as secondary. Cells were analyzed by flow cytometry. Shown is the ratio of the mean channel fluorescence of bell-shaped histograms, with each cell type having 50 nM biotinylated NT3 (highest concentration) normalized to 1. Average ± S.D. (n = 3 biologic replicates, 10,000 cells analyzed in each assay). Statistical analysis was done by two-way ANOVA with significance α < 0.05, followed by Bonferroni’s correction for multiple comparisons. In p75-expressing cells, mutant DRK (50 nM) has significantly reduced binding compared with NT3 (*P < 0.05). In TrkB-expressing cells, mutant DRK (2 and 10 nM) has significantly enhanced binding compared with NT3 (*P < 0.05, ****P < 0.0001, respectively). In TrkA-expressing cells, mutant DRK (2, 10 and 50 nM) has significantly enhanced binding compared with NT3 (*P < 0.05, ****P < 0.0001, respectively). In cells expressing full length TrkC (TrkC-FL) binding by mutant DRK and NT3 was not different.
- Fig. 2.
Cell survival induced by mutants. MTT assays after culture in SFM ± the indicated compounds, for 48 hours. Data are standardized to the corresponding wild-type neurotrophin (NT3 for TrkC, BDNF for TrkB, NGF for TrkA = 100% survival). Agents tested at suboptimal (0.2 nM) or optimal (2 nM) concentrations. The values from untreated controls (SFM without supplementation) are set to 0% survival. Survival was calculated using the formula [(OD test – OD SFM) × 100/(OD optimal NF – OD SFM)]. Data shown are the mean of n = 3 independent biologic replicate experiments ± S.D.; each independent experiment had n = 4 or n = 6 technical replicate wells per condition. Statistical analyses are one-way ANOVA with significance α < 0.05, followed by Bonferroni’s correction for multiple comparisons. (A) NIH-TrkC cells; the asterisk (*) indicates differences between RK and NT3 (**P < 0.05) at high concentrations. There are no significant differences between treatments at lower concentrations. (B) NIH-TrkB and control NIH-TrkA cells; only the 10 nM NF concentration is shown. Asterisks (*) indicate differences between the cognate wild-type NF for each receptor (BDNF for TrkB and NGF for TrkA) and the test agents (***P < 0.001, **P < 0.01). The number sign (#) indicates differences between DRK and NT3 in the survival of NIH-TrkB cells (###P < 0.001). The difference between D versus DRK is significant (*P < 0.05). However, in NIH-TrkB cells D versus RK is not significant, and RK versus DRK is significant (****P < 0.0001) (for clarity, this is not shown in the graph).
- Fig. 3.
Mutants induce cell survival and differentiation in cells expressing TrkC+ and p75+. (A) Survival of nnr5-TrkC cells (a PC12 cell variant expressing TrkC+ p75+). Data are the mean of n = 3 independent biologic replicate experiments ± S.D.; each independent experiment had 4–6 technical replicate wells per condition. No significant differences are seen between wild-type NT3 and the mutants when compared at their equivalent concentrations. (B) Representative assay of cell differentiation. Nnr5-TrkC cells were untreated (control; Unt.) or were treated with wild-type NT3 or mutants at 0.1 nM, and after 72 hours they were immunostained with anti–microtubule-associated protein 2. Nuclei were counterstained with DAPI. Magnification 40×. (C) Quantification of neurite outgrowth (differentiation). Data are shown as percentages of differentiated cells ± S.D., relative to NT3 (n = 3 independent experiments). Statistics were calculated by one-way ANOVA with significance α < 0.05, followed by Bonferroni’s correction for multiple comparisons; ***P < 0.001 with respect to untreated control. Treatment groups show no significant differences. DAPI, 4′,6-diamidino-2-phenylindole; MAP2, microtubule-associated protein 2.
- Fig. 4.
Mutants activate TrkC and TrkB signal transduction. Fibroblast cells transfected to express TrkC (A), TrkB (B), or TrkA (C) were serum-starved for 2 hours and then were treated with the indicated NF or mutants (2 nM) for 15 minutes, or were untreated (Unt). Positive controls are NT3 for TrkC-expressing cells, BDNF for TrkB-expressing cells, and NGF for TrkA-expressing cells. Cell lysates were analyzed by Western blots for pTrk, pAkt, or pErk1/2 by exposing membrane strips cut based on a molecular weight marker to the corresponding primary antibody. After stripping, membranes were reprobed with anti-actin, and the actin band shown is the standard loading control for that panel. The blots shown for a protein are assembled form the same exposure, and data are representative. Densitometry was done from proper linear exposures of the blots to quantify pTrkC, pTrkB, or pTrkA, the phosphotyrosine-containing bands at ∼140 kDa in each experiment. Assays were repeated at least three independent times with unique biologic samples. Quantification reports the average ± S.D. of all independent biologic experiments (n = 3), relative to wild-type NT3 (set to 1). Statistical analysis was done by one-way ANOVA with significance α < 0.05, followed by Bonferroni’s correction for multiple comparisons. *P < 0.05, **P < 0.01, ***P < 0.001.
- Fig. 5.
Trk-mediated pErk1/2 activation by mutants. Concentration-dependent pERK phosphorylation (AlphaLISA tests) in cells expressing human TrkB (hTrkB) (A) or human TrkC (hTrkC) (B). Concentration-response curves were generated after normalization of responses to 10 nM BDNF or 10 nM NT3 as respective controls. Curve fitting was performed using the log(agonist) vs. response (three parameters) function in GraphPad Prism 8. EC50 and maximal inhibition values calculated from these data are shown in Table 2. Data are mean ± S.D. values from the number of biologic replicates indicated in Table 2.
- Fig. 6.
Effects of mutants on p75-dependent TNFα expression. The levels of TNFα transcript were quantified by quantitative real-time PCR after 6 hours of the indicated treatment in HEK293-p75 cells (stably expressing p75). Each sample was assayed in triplicate technical replicate wells in at least three independent biologic replicate experiments. Data shown are the mean of the independent experiments ± S.D., with values standardized to the untreated (Unt) control. Statistical analysis was done by one-way ANOVA with significance α < 0.05, followed by Bonferroni’s correction for multiple comparisons. *P < 0.05, **P < 0.01, ***P < 0.001. LPS, lipopolysaccharide.
- Fig. 7.
Mutants delay loss of RGC somata after optic nerve (ON) axotomy. For each treatment, the surviving RGCs were quantified. Data for each mouse is the ratio of the axotomized eye ± treatments and the contralateral control healthy eye (100% RGC cell bodies). Percent data are averaged per group. Shown is the average (n = 8 mice/group) of the measurements at the 14-day endpoint. Mutant DRK affords significant RGC neuroprotection compared with untreated or PBS-treated control and compared with wild-type NGF, BDNF, or NT3. NGF-C is a mutant that does not bind to p75 but activates TrkA selectively. One-way ANOVA was used with significance α < 0.05, followed by Bonferroni’s correction for multiple comparisons; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. The differences between D and RK are not significant, RK versus DRK are significant (P < 0.0001), and D versus DRK are significant (P < 0.001).
- Fig. 8.
Mutants prevent loss of RGC nerve fibers in diabetic retinopathy. Mice were made diabetic (STZ model with verified hyperglycemia), causing loss of retinal structure and thinning of the NGI, a structure formed by RGCs and their axons and fibers. At week 2.5 of diabetes (with ongoing disease) mice were administered an intravitreal injection of the indicated agent or vehicle (Unt.). Controls (Ctl) are healthy nondiabetic retinas of normal thickness from sibling mice (n = 3–4 animals per group). (A) Representative sections of B-scan OCT images at 6 weeks of diabetes; treated with NT3 or DRK at week 2.5 of diabetes, compared with age-matched control-healthy mice. (B) NGI thickness in micrometers, and shown are the Inner Nuclear Layer (INL) and Outer Nuclear Layer (ONL). All diabetic mice are significantly worse than control nondiabetic (P < 0.001), and DRK-treated eyes are significantly better than untreated (P < 0.001) or than BDNF (P < 0.05) groups. (C) Nerve fiber loss (%) in all STZ-treated groups, reflective of the quality of the RGC fibers. Compared with untreated control (100% damage), the mutants D (P < 0.05) and DRK (P < 0.001) groups experience significantly reduced damage. Compared with BDNF, DRK is also significantly better (P < 0.01). One-way ANOVA test was used with significance α < 0.05, followed by Bonferroni’s correction for multiple comparisons. *P < 0.05, **P < 0.01, ***P < 0.001.
Tables
Biotinylated NT3 ± competitors Biotin-NT3 binding (% of untreated control) TrkC p75 TrkB No competitor 100 100 100 NT3 wild-type 60 ± 3.3 53 ± 7.4 56 ± 3.1 D 52 ± 3.8 55 ± 3.1 57 ± 3.2 RK 51 ± 5.8 83 ± 8.1 61 ± 5.3 DRK 56 ± 6.6 85 ± 6.1 64 ± 2.5 BDNF wild-type Not done Not done 58 ± 2.6 Cells transfected with the indicated receptor were preincubated for 15 minutes in binding buffer on ice without (untreated) or with competitors wild-type NF (control) or test mutants D, RK, or DRK (each at 380 nM). Then, saturating biotinylated NT3 was added for 15 minutes. The labeled secondary was fluorescein-avidin, and cells were analyzed by flowcytometry. The MCFs of bell-shaped histograms were standardized, with no competition = 100% binding. Values shown are the average ± S.D. (n = 3 biologic replicates; in each assay 5000 cells analyzed). Statistical analysis was done by one-way ANOVA with significance α < 0.05, followed by Bonferroni’s correction for multiple comparisons. In Trk-expressing cells, the three mutants, wild-type NT3, and wild-type BDNF significantly inhibited NT3-biotin binding (P < 0.001 in TrkC and TrkB). There are no significant differences in competition between NT3 wild-type versus the mutants or BDNF. In p75-expressing cells, wild-type NT3 and mutant D significantly inhibited NT3-biotin binding (P < 0.05). In p75-expressing cells, mutants RK and DRK did not significantly compete binding compared with untreated control, and binding remained significantly higher compared with control wild-type NT3 competition (P < 0.05).
BDNF NT3 D RK DRK Human TrkB EC50 nM 0.59 (0.43–0.80) 0.87 (0.57–1.32) 0.18 (0.05–0.70) 0.22 (0.08–0.61) 0.14 (0.03–0.62) % Max 96.7 ± 9.53 120 ± 7.30 91.2 ± 14.2 109 ± 16.9 98.5 ± 26.9 N 7 3 4 4 4 Human TrkC EC50 nM 10.0 (2.64–37.9) 0.58 (0.29–1.16) 0.36 (0.16–0.57) 0.11 (0.02–0.61) 0.27 (0.10–0.69) % Max 80.1 ± 21.7 101 ± 6.16 109 ± 61.9 84.9 ± 37.7 111 ± 32.6 N 5 5 3 4 4 EC50 values and maximal inhibition values (% Max) were derived from the concentration-response curves in Fig. 5. The EC50 values are the geometric means derived from negative logarithm of the EC50 (pEC50 values) with 95% confidence limits, and the % Max values are means ± S.D. for n biological replicates.
