Inhibition of soluble TNF signaling in a mouse model of Alzheimer's disease prevents pre-plaque amyloid-associated neuropathology

Abstract

Microglial activation and overproduction of inflammatory mediators in the central nervous system (CNS) have been implicated in Alzheimer's disease (AD). Elevated levels of the pro-inflammatory cytokine tumor necrosis factor (TNF) have been reported in serum and post-mortem brains of patients with AD, but its role in progression of AD is unclear. Using novel engineered dominant negative TNF inhibitors (DN-TNFs) selective for soluble TNF (solTNF), we investigated whether blocking TNF signaling with chronic infusion of the recombinant DN-TNF XENP345 or a single injection of a lentivirus encoding DN-TNF prevented the acceleration of AD-like pathology induced by chronic systemic inflammation in 3xTgAD mice. We found that chronic inhibition of solTNF signaling with either approach decreased the LPS-induced accumulation of 6E10-immunoreactive protein in hippocampus, cortex, and amygdala. Immunohistological and biochemical approaches using a C-terminal APP antibody indicated that a major fraction of the accumulated protein was likely to be C-terminal APP fragments (beta-CTF) while a minor fraction consisted of Av40 and 42. Genetic inactivation of TNFR1-mediated TNF signaling in 3xTgAD mice yielded similar results. Taken together, our studies indicate that soluble TNF is a critical mediator of the effects of neuroinflammation on early (pre-plaque) pathology in 3xTgAD mice. Targeted inhibition of solTNF in the CNS may slow the appearance of amyloid-associated pathology, cognitive deficits, and potentially the progressive loss of neurons in AD.

Figures

Figure 1. Hippocampal infusion of XENP345 attenuates appearance of intraneuronal 6E10-immunoreactive protein induced by chronic systemic inflammation in 3xTgAD mice

(A) Schematic of experimental design: 4.5 month old mice were cannulated in the CA1 region of the hippocampus in the right hemisphere of the brain and fitted with osmotic pumps that delivered 0.1 mg/kg/day XENP345 or an equivalent volume of vehicle for 4 weeks. During these 4 weeks, animals were given 7.5 × 105 E.U./kg of LPS twice a week. (B) Immunohistological analysis of 6E10 immunoreactivity in hippocampal brain sections reveals that 3xTgAD mice that received intrahippocampal saline and i.p. LPS displayed significant accumulation of intraneuronal 6E10-imunoreactive protein (green). Inset shows 6E10-positive neurons in the hilus, magnified. 3xTgAD mice that received intrahippocampal XENP345 and i.p LPS displayed reduced intraneuronal 6E10-immunoreactivity compared to those that received intrahippocamapal saline and i.p. LPS. Inset shows 6E10-negative neurons in the hilus, magnified. Non-Tg animals of the same genetic background that received intrahippocampal saline or XENP345 and i.p LPS displayed no intraneuronal 6E10-immunoreactive protein. Nuclei in (B) are counterstained with bisbenzimide (blue). Scale bar = 200 μm. (C) Quantification of 6E10-positive neurons in brain sections from animals that received intrahippocampal saline/LPS i.p. or intrahippocampal XENP345/LPS i.p. reveals a statistically significant difference in 6E10-immunoreactive protein accumulation between 3xTgAD saline/LPS and 3xTgAD XENP345/LPS groups. Values shown are group means ± SEM for n = 4 animals per group, * denotes p < 0.05 by one-tailed Student's t test. (D) Quantification of total cells (nuclei counterstained with bisbenzimide) in the hilus of 3xTgAD animals infused with intrahippocampal XENP345 or saline reveals no significant difference between groups by one-tailed Student's t test, confirming that reduction in Aβ is not due to loss of a TNF-dependent subpopulation of cells in the hilus. Values shown are group means ± SEM for n = 4 animals per group.

Figure 2. Hippocampal infusion of XENP345 reduces appearance of intraneuronal 6E10-immunoreactive protein

(A) Schematic of experimental design: 4.5 month old mice were cannulated in the CA1 region of the hippocampus in the right hemisphere of the brain and fitted with osmotic pumps that delivered 0.1 mg/kg/day XENP345 or an equivalent volume of vehicle for 4 weeks. During these 4 weeks, animals were given 7.5 × 105 E.U./kg of LPS or an equivalent volume of saline twice a week. (B–C) Animals that were injected with i.p. low-dose LPS show increased 6E10-immunoreactive protein in the hippocampus; this increase is inhibited by XENP345. (A). Images were taken at 4× magnification. CA1 is surrounded by curly brackets, scale bar = 500 μm. (B) Images were taken at 40× magnification, scale bar = 50 μm. (D) Stereological analysis of 6E10-positive cell number in the hippocampus ipsilateral and contralateral to the cannula in 3xTgAD mice that received intrahippocampal saline or XENP345 and i.p. saline or LPS. Values shown are group means ± SEM for n = 4 animals per group. * denotes significantly different from saline pump/saline injections at p < 0.05) by one-factor ANOVA followed by Tukey's post hoc test.

Figure 3. In vitro validation of the biological activity of lentiviral-derived DN-TNF in primary glial cells

(A) Transduction of SS01 murine astrocytes blocks TNF-induced NFκB pathway activation as measured by cytoplasmic-to-nuclear enrichment of the p65RelA subunit. SS01 cells were mock-transduced or transduced with lenti-GFP or lenti-DN-TNF 24 hours after plating. Forty-eight hours later, cells were stimulated with saline vehicle (left panels) or 2ng/mL TNF (right panels) for 15 minutes. Cells were fixed and immunostained with an antibody specific to the p65RelA subunit of NFκB. All cells display detectable cytoplasmic localization of p65 prior to stimulation. Upon stimulation with TNF, mock- and lenti-GFP-transduced cells display enhanced nuclear localization of p65 (white box), whereas lenti-DN-TNF-transduced cells still display a high degree of cytoplasmic p65 staining (white box). Scale bar = 50 μm. (B, C) Transduction of primary microglia with lenti-DN-TNF blocks TNF-induced microglia activation. Primary microglia cultures from 3xTgAD mice were mock-transduced or transduced with lenti-DN-TNF, or lenti-GFP. After 48 hrs, cells were treated with saline vehicle, 10 ng/mL TNF alone or 10 ng/mL TNF + 200 ng/mL XENP345 as a positive control for TNF inhibition. Cells were fixed 24 hr later and immunostained with the microglial activation marker CD45. (B) Images of primary microglia stained with the activation marker CD45 show an increase in CD45-immunoreactive microglia in lenti-GFP infected cells treated with TNF but not in lenti-DN-TNF infected cells treated with TNF. Scale bar = 50 μm. (C) Quantification of TNF--induced microglia activation in lenti-DN-TNF versus lenti-GFP or mock-transduced 3xTgAD primary microglia cultures. Total microglia (all CD45low + high positive cells) and activated microglia (CD45high cells with nuclear enrichment of CD45) were counted. Values shown are percent activated microglia (CD45high / CD45low + high) × 100 ± S.E.M., * denotes significantly different from mock-infected/vehicle-treated at p < 0.05 by one-way ANOVA followed by Tukey's post hoc test.

Figure 4. ICV administration of Lenti-DN-TNF results in detectable in vivo expression in 3xTgAD mice

(A) Schematic of experimental design: 3 month-old 3xTgAD mice received an ICV injection of lenti-DN-TNF or lenti-GFP. At 4.5 months old, mice received a bi-weekly regimen of intraperitoneal (i.p.) LPS (7.5 × 105 E.U./kg) injections for 6 weeks. Tissue was harvested at the time mice were 6 months old. (B) Immunohistological analysis of anti-GFP immunoreactivity in the hippocampus of lenti-DN-TNF-IRES-GFP transduced mouse brain, but not in non-transduced mouse brains, confirms expression of the lentiviral-encoded genes. Scale bar = 100 μm.

Figure 5. ICV administration of Lenti-DN-TNF attenuates accumulation of intraneuronal 6E10-immunoreactive protein(s) in 3xTgAD mice

(A) Brains of 3xTgAD mice that received lenti-GFP and chronic LPS exposure display accumulation of intraneuronal 6E10-immunoreactive amyloid-associated protein(s) in layer V of parietal cortex (enclosed by black curly brackets) and in the hippocampus, CA1 region (enclosed by white curly brackets) shown on the left at 4X magnification, scale bar = 500 μm; and on the right at 40X magnification, scale bar = 50 μm. Brains of 3xTgAD mice that received lenti-DN-TNF and chronic LPS exposure display reduced accumulation of intraneuronal 6E10-immunoreactive amyloid-associated protein(s) in both parietal cortex and hippocampus. (B) Brains of 3xTgAD mice that received lenti-DN-TNF and chronic LPS exposure display marked reduction of intraneuronal 6E10-immunoreactive protein(s) in the amygdala compared to animals that received lenti-GFP ICV and chronic LPS i.p. injections. On the left 4X magnification, medial amygdala enclosed in a white rectangle, scale bar = 500 μ; on the right 40X magnification, scale bar = 50 μm. (C). Stereological analysis of the number of 6E10-immunoreactive neurons in hippocampus and cortex of 3xTgAD mice that received lenti-GFP/LPS or lenti-DN-TNF/LPS. Values shown are group means ± SEM. for n = 3 animals, * denotes p = 0.026 by one-tailed t-test.

Figure 6. Relative abundance of APP, C-terminal APP fragments (β-CTF and α-CTF), and Aβ peptides and quantification of Aβ40 and Aβ42 by densitometry and ELISA in LPS-trated 3xTgAD mice transduced with lenti-GFP or lenti-DN-TNF

(A) Immunofluorescent confocal Z-slice images of hippocampal sections double-labeled with 6E10 (red) and a C-terminal fragment (CTF) specific APP antibody (green) were taken to construct a Z-stack to establish the relative abundance of Aβ (6E10+/CTF−, red only), APP/β-CTF (6E10+/CTF+, yellow) and α-CTF (6E10−/CTF+, green only) in 3 equal size square fields per animal which were averaged and compared by t-test in 3xTgAD mice from various treatment groups. Results indicate lower abundance of Aβ and higher abundance of APP/β-CTF species (6E10+/CTF+, yellow). Densitometric analysis of Aβ only signal in brain sections of (B) animals infused with XENP345 (or saline as negative control) or (C) injected with lenti-DN-TNF (or lenti-GFP as negative control) and subjected to chronic systemic inflammation indicate TNF signaling inhibition has minimal effect on Aβ levels in (B) mice fitted with osmotic pumps (p = 0.34, Kruskal-Wallis) or (C) in lenti-injected mice (p = 0.15, Kruskall-Wallis). Quantitative ELISAs for (D)Aβ40 and (E) Aβ42 revealed low but detectable levels of both peptides and no significant difference between lenti-DN-TNF and lenti-GFP-treated 3xTgAD mice. One-way ANOVA, n = 4 for lenti-DN, n = 3 for lenti-GFP, Aβ40 p = 0.12; Aβ40 p = 0.10.

Figure 7. solTNF inhibition reduces accumulation of β-CTF and/or APP

(A) Brains of 3xTgAD mice that received lenti-GFP and chronic systemic LPS injections display accumulation of intraneuronal anti-APP C9-immunoreactive material in layer V of parietal cortex (surrounded by black curly brackets) and in the CA1 region of the hippocampus (surrounded by white curly brackets); C9 immunoreactivity is attenuated by lenti-DN-TNF ICV transduction. On the left,4X magnification: scale bar = 500 μm; On the right 40X magnification: scale bar = 50 μm. (B) 3xTgAD mice that were injected with low-dose LPS i.p. show increased 6E10-immunoreactive protein in the CA1 region of the hippocampus surrounded by curly brackets; this increase is inhibited by chronic infusion of the solTNF-selective inhibitor XENP345. On the left, 4X magnification, scale bar = 500 μm; on the right, 40X magnification, scale bar = 50 μm.

Figure 8. Genetic deletion of TNFR1 in 3xTgAD transgenic mice prevents accumulation of intraneuronal 6E10 immunoreactivity in response to chronic systemic inflammation

Compound transgenic mice were generated by crossing 3xTgAD mice with TNFR1-deficient (TNFR1−/−) mice or with WT mice of the same genetic background (C57B6/J, TNFR1+/+). Immunohistological analysis revealed intraneuronal accumulation of (A) 6E10 and (B) C9-immunoreactive amyloid-associated proteins in 3xTgAD;TNFR1+/+ mice that received chronic systemic LPS injections but not in 3xTgAD;TNFR1−/− mice which lack functional solTNF signaling. 40X magnification, scale bar = 50 μm.