Brain-derived neurotrophic factor protects against tau-related neurodegeneration of Alzheimer's disease

Abstract

Reduced expression of brain-derived neurotrophic factor (BDNF) has a crucial role in the pathogenesis of Alzheimer's disease (AD), which is characterized with the formation of neuritic plaques consisting of amyloid-beta (Aβ) and neurofibrillary tangles composed of hyperphosphorylated tau protein. A growing body of evidence indicates a potential protective effect of BDNF against Aβ-induced neurotoxicity in AD mouse models. However, the direct therapeutic effect of BDNF supplement on tauopathy in AD remains to be established. Here, we found that the BDNF level was reduced in the serum and brain of AD patients and P301L transgenic mice (a mouse model of tauopathy). Intralateral ventricle injection of adeno-associated virus carrying the gene encoding human BDNF (AAV-BDNF) achieved stable expression of BDNF gene and restored the BDNF level in the brains of P301L mice. Restoration of the BDNF level attenuated behavioral deficits, prevented neuron loss, alleviated synaptic degeneration and reduced neuronal abnormality, but did not affect tau hyperphosphorylation level in the brains of P301L mice. Long-term expression of AAV-BDNF in the brain was well tolerated by the mice. These findings suggest that the gene delivery of BDNF is a promising treatment for tau-related neurodegeneration for AD and other neurodegenerative disorders with tauopathy.

Figure 1

Brain-derived neurotrophic factor (BDNF) protein levels in serum and brain are reduced in Alzheimer's disease (AD) patients and P301L transgenic mice. (a) Comparison of BDNF levels in serum between AD patients and age- and sex-matched healthy elderly controls (HE; mean±s.e.m., Student's t-test, **P<0.01). (b) Detection rate of BDNF in cerebrospinal fluid (CSF) by enzyme-linked immunosorbent assay (ELISA; %, χ²-test, *P<0.05). (c) Comparison of BDNF levels in serum between AD patients and HE controls (detected by ELISA) (mean±s.e.m., Student's t-test, *P<0.05). (d, e) Representative western blot images (d) and quantification (e) of BNDF expression in brain homogenates from pathologically confirmed AD patients and age-matched HE (mean±s.e.m., n=12, Student's t-test, *P<0.05). (f, g) BDNF levels in serum (f) and brain homogenates (g) of P301L tau transgenic pR5 mice and their wild-type (Wt) littermate controls at different ages (mean±s.e.m., n=6 per age and per type, two-way analysis of variance (ANOVA), Tukey's test, **P<0.01). (h, i) Representative western blot images (h) and quantification (i) of mouse BDNF in brain homogenates of 4-month-old mice (mean±s.e.m., n=6 per age and per type, Student's t-test, *P<0.05). 4mo P301L denotes 4-month old P301L transgenic mouse; 4mo Wt denotes 4-month-old Wt littermates. (j, k) Representative western blot images (j) and quantification (k) of mouse BDNF in brain homogenates of 12-month-old mice (mean±s.e.m., n=6 per age and per type, Student's t-test, *P<0.05, **P<0.01).

Figure 2

Wide and stable expression of human brain-derived neurotrophic factor (BDNF) after intraventricular injection of adeno-associated virus carrying BDNF (AAV-BDNF). (a) Schematics of the construction and characterization of AAV-green fluorescent protein (AAV-GFP; left panel) and AAV-BDNF (right panel) vectors. (b, c) In order to exclude endogenous mouse BDNF, we applied a specific anti-human BDNF antibody to detect the expression of human BDNF within P301L mouse brain. Human BDNF expression is detectable in multiple brain regions (b), and GFP or human BDNF was selectively expressed in neurons but not in microglia (CD45-positive) or astrocytes (glial fibrillary acidic protein (GFAP)-positive) (c). Scale bar=80μm (b) or 40μm (c). (d) Comparison of human BDNF levels in brain homogenates of P301L mice treated with AAV-BDNF or AAV-GFP at 3 weeks (3w), 6 weeks (6w), 9 weeks (9w) and 9 months (9 m) after AAV injection initiated at 3 months of age (n=4 per group for 3w, 6w and 9w; n=6 for 9 m group; mean±s.e.m.; two-way analysis of variance (ANOVA), Tukey's test, **P<0.01). The concentrations of human BDNF were determined using specific human BDNF enzyme-linked immunosorbent assay (ELISA) kits. (e) Western analysis of total BDNF expression (including endogenous mouse BDNF and exogenous human BDNF) in brain homogenates (n=5 per group; mean±s.e.m., one-way ANOVA, Tukey's test, **P<0.05; n.s. denotes no significant difference). 12mo P301L (AAV-BDNF) and 12mo P301L (AAV-GFP) denote 12-month-old P301L mice treated with AAV-BDNF and AAV-GFP, respectively; 12mo P301L (untreated) denotes untreated 12-month-old P301L mice; 12mo Wt denotes age-matched wild-type littermates.

Figure 3

Delivery of adeno-associated virus carrying brain-derived neurotrophic factor (AAV-BDNF) rescues the behavioral impairment of P301L mice. (a) Time schedule of behavioral tests. (b, c) The alternation percentage (b) and total number of entries (c) in the spontaneous alternation test (mean±s.e.m.; n=8 per group, one-way analysis of variance (ANOVA), Tukey's test, *P<0.05). (d) Comparison of the percentages of novel arm entries among different groups (mean±s.e.m.; n=8 per group, one-way ANOVA, Tukey's test, *P<0.05, **P<0.01). (e) Comparison of the preference index in the novel object recognition test among different groups (mean±s.e.m.; n=8 per group, one-way ANOVA, Tukey's test, *P<0.05). (f) Escape latency (seconds) during platform trials in Morris water maze (mean±s.e.m., n=8 per group, two-way ANOVA, Tukey's test, *P<0.05). GFP, green fluorescent protein; Wt, wild-type.

Figure 4

Neuroprotective effects of adeno-associated virus carrying brain-derived neurotrophic factor (AAV-BDNF) on neurons. (a, b) Representative images (a) and quantitative statistics (b) of neurons in the CA1 region of the hippocampus of P301L mice (untreated, AAV-BDNF treated and AAV-green fluorescent protein (AAV-GFP) treated) and wild-type (Wt) littermates, using NeuN immunofluorescence (IF) assays (six mice per group; mean±s.e.m., one-way analysis of variance (ANOVA), Tukey's test, *P<0.05, **P<0.01). Scale bar=25μm. (c) Schematic of basal forebrain in the coronal plane (upper-left panel) and representative images of choline acetyltransferase (ChAT)-positive cholinergic neurons in the medial septum (MS), intermediate part of the lateral septum (LSI) and vertical limb of the diagonal band (VDB) in AAV-BDNF and AAV-GFP groups. The inset shows the representative morphology of cholinergic neurons in MS at a higher magnification. V, ventral; D, dorsal; L, left; R, right. Scale bar=25μm. (d) Comparison of numbers of cholinergic neurons in MS, LSI and VDB between the two groups (six mice per group; mean±s.e.m., Student's t-test, *P<0.05, **P<0.01). (e) Representative electron micrographs of pyramidal neurons from the CA1 region of the hippocampus in the four groups. It is worth noticing that in 12mo P301L (AAV-GFP) group and untreated P301L group, a number of pyramidal cells shows structural changes typical of apoptosis, including an enlarged perinuclear space, chromatic agglutination and karyopyknosis, whereas apoptosis-like cells were rarely found in 12mo P301L (AAV-BDNF) group, and pyramidal cells in this group usually displayed with uniform chromatin density and a large nucleus, which was similar to 12mo Wt group. Scale bar=2μm.

Figure 5

Supportive effects of adeno-associated virus carrying brain-derived neurotrophic factor (AAV-BDNF) on neurogenesis and synapse. (a, b) Representative western blot images and quantitative analyses of the expressions of nestin, neuron-specific enolase (NSE) and doublecortin (DCX) in olfactory bulb (a) and dentate gyrus (b) of the four groups (n=6 per group, mean±s.e.m., one-way analysis of variance (ANOVA), Tukey's test, *P<0.05, **P<0.01). (c) Representative western blot images and quantitative analyses for PSD-95, PSD-93, Syn-1, SNAP-25 and VAMP-1 expression in brain homogenates of the four groups (n=6 per group, mean±s.e.m., one-way ANOVA, Tukey's test, *P<0.05, **P<0.01).

Figure 6

Genetic delivery of human brain-derived neurotrophic factor (BDNF) does not affect tau hyperphosphorylation. (a) Representative images of human pS396-tau immunostaining in the hippocampus and thalamus of different groups. Scale bar=100 μm. Insets show the representative morphology of pS396-tau-positive cells at higher magnification. (b) Comparison of the area fraction of human pS396-tau-positive immunostaining in the hippocampus (right panel) and thalamus (left panel) among the four groups (n=6 per group; mean±s.e.m., one-way analysis of variance (ANOVA), Tukey's test, **P<0.01). No differences were found among 12mo P301L (untreated), 12mo P301L (AAV-green fluorescent protein, AAV-GFP) and 12mo P301L (AAV-BDNF, adeno-associated virus carrying BDNF). (c) Representative western blot images and quantitative analyses of tau hyperphosphorylation levels for multiple sites in 12mo P301L (AAV-GFP) and 12mo P301L (AAV-BDNF) groups, including pS396, pS202/T025 and pS404 (n=6 per group, mean±s.e.m., Student's t-test, no difference was found between the two groups). (d) Quantification of human pT181 (right panel) and human pT231 (left panel) levels in Tris buffer solution (TBS) and sodium dodecyl sulfonate (SDS) extracts of P301L brain tissue homogenates from AAV-GFP and AAV-BDNF groups (n=6 per group, mean±s.e.m., Student's t-test, no difference was found between two groups). (e) Representative western blot images and quantitative analyses of the Tau phosphorylation kinase GSK3beta and the Tau phosphatase PP2A in brain tissue homogenates of P301L mice from two groups (n=6 per group, mean±s.e.m., Student's t-test, no difference was found between two groups). AAV, adeno-associated virus.