Fyn kinase modulates synaptotoxicity, but not aberrant sprouting, in human amyloid precursor protein transgenic mice

Abstract

Alzheimer's disease (AD), the most common neurodegenerative disorder, results in progressive degeneration of synapses and aberrant sprouting of axon terminals. The mechanisms underlying these seemingly opposing cellular phenomena are unclear. We hypothesized that Fyn kinase may play a role in one or both of these processes because it is increased in AD brains and because it is involved in synaptic plasticity and axonal outgrowth. We investigated the effects of Fyn on AD-related synaptotoxicity and aberrant axonal sprouting by ablating or overexpressing Fyn in human amyloid precursor protein (hAPP) transgenic mice. On the fyn+/+ background, hAPP/amyloid beta peptide (Abeta) decreased hippocampal levels of synaptophysin-immunoreactive presynaptic terminals (SIPTs), consistent with previous findings. On the fyn-/- background, hAPP/Abeta did not affect SIPTs. SIPT reductions correlated with hippocampal Abeta levels in hAPP/fyn+/+, but not hAPP/fyn-/-, mice suggesting that Fyn provides a critical link between hAPP/Abeta and SIPTs. Furthermore, overexpression of Fyn exacerbated SIPT reductions in hAPP mice. We also found that the susceptibility of mice to hAPP/Abeta-induced premature mortality was decreased by Fyn ablation and increased by Fyn overexpression. In contrast, axonal sprouting in the hippocampus of hAPP mice was unaffected. We conclude that Fyn-dependent pathways are critical in AD-related synaptotoxicity and that the pathogenesis of hAPP/Abeta-induced neuronal alterations may be mechanistically heterogenous.

Figure 1.

Genetic modulation of mouse Fyn alters the extent of hAPP_FAD/Aβ-induced reductions in SIPT and premature mortality. FYN refers to the Fyn transgene, and fyn refers to the endogenous gene. n.s., Not significant. ^*p = 0.001; ^**p < 0.0005 versus NTG wild-type littermate controls. A, Overexpressing Fyn exacerbated SIPT reductions in hAPP_low mice (n = 5-9 mice/genotype; age, 6-8 months). B, Confocal microscopic images illustrating SIPT in the OML of an NTG mouse (top) and an hAPP_low/FYN doubly TG mouse (bottom). C, Ablating Fyn reduced SIPT levels but also prevented additional SIPT reductions in hAPP_high mice (n = 5-8 mice/genotype; age, 4-5 months). D, Overexpressing Fyn did not alter Aβ levels in hAPP_low mice (n = 5-9 mice/genotype; age, 6-8 months). E, Ablating Fyn did not alter Aβ levels in hAPP_high mice (n = 5-8 mice/genotype; age, 4-5 months). F, SIPT levels correlated inversely with Aβ1-42 levels and Aβ1-42/Aβ1-x ratios in hAPP_high mice on the fyn^+/+, but not fyn^-/-, background (age, 4-5 months). G, Comparable plaque load in hAPP_high mice on fyn^+/+ and fyn^-/- backgrounds (n = 7-8 mice/genotype; age, 8-10 months). H, I, Proportion of mice that died prematurely in each of the groups indicated during the first 6 months postnatally (percentage of 38-64 mice/genotype). Husbandry conditions were similar for all groups.

Figure 2.

Aberrant axonal sprouting in hAPP_FAD mice depends on age and hAPP_FAD expression levels. A, Photomicrographs of GAP-43-IR in the hippocampus of a 6-month-old NTG mouse and an hAPP_high TG littermate. Prominent aberrant sprouting is evident in the molecular layer of the hAPP_high mouse (arrow). B, In hAPP_high mice, GAP-43-IR in this layer further increased with age. C, Photomicrographs showing normal GAP-43-IR in the hippocampus of a 6-month-old NTG mouse and an hAPP_low TG littermate. D, Quantitation of GAP-43-IR in 6-month-old hAPP_high mice revealed a 60% increase in the OML and a 13% increase in the MML (n = 10-13 mice/genotype). ^*p < 0.05; ^**p < 0.0001 versus NTG wild-type littermates. E, Quantitation of GAP-43-IR in the OML and MML revealed no aberrant sprouting in hAPP_low mice (n = 11-13 mice/genotype). mo, Months; IML, inner molecular layer.

Figure 3.

Genetic modulation of Fyn does not affect hAPP/Aβ-induced aberrant sprouting. A, Comparable increases in GAP-43-IR in the OML of hAPP_high mice on the fyn^+/+ and fyn^-/- background (n = 5-8 mice/genotype; age, 4-5 months). ^*p < 0.005. B, GAP-43-IR did not correlate with hippocampal Aβ1-42 levels or Aβ1-42/Aβ1-x ratios in hAPP_high mice on the fyn^+/+ and fyn^-/- backgrounds. C, Overexpressing Fyn did not increase GAP-43-IR levels in hAPP_low mice (n = 5-9 mice/genotype; age, 6-8 months).