Reduced IGF-1 signaling delays age-associated proteotoxicity in mice

Abstract

The insulin/insulin growth factor (IGF) signaling (IIS) pathway is a key regulator of aging of worms, flies, mice, and likely humans. Delayed aging by IIS reduction protects the nematode C. elegans from toxicity associated with the aggregation of the Alzheimer's disease-linked human peptide, Abeta. We reduced IGF signaling in Alzheimer's model mice and discovered that these animals are protected from Alzheimer's-like disease symptoms, including reduced behavioral impairment, neuroinflammation, and neuronal loss. This protection is correlated with the hyperaggregation of Abeta leading to tightly packed, ordered plaques, suggesting that one aspect of the protection conferred by reduced IGF signaling is the sequestration of soluble Abeta oligomers into dense aggregates of lower toxicity. These findings indicate that the IGF signaling-regulated mechanism that protects from Abeta toxicity is conserved from worms to mammals and point to the modulation of this signaling pathway as a promising strategy for the development of Alzheimer's disease therapy.

Figure 1

Reduction of IGF signaling protects mice from Aβ associated behavioral impairments. A. Long-lived mice carrying one Igf1r copy were crossed with transgenic Alzheimer's disease (AD) model mice harboring two AD-linked mutated genes, APP_swe (containing the human Aβ sequence) and PS1ΔE9 to obtain offspring of four genotypes: (i) Wild Type – harbor two Igf1r copies and no AD linked transgenes (WT). (ii) Long-lived mice with one Igf1r copy and no AD linked transgenes (Igf1r+/−). (iii) AD model mice with two Igf1r copies and both AD-linked transgenes (AD). (iv) Mice that harbor only one Igf1r copy and both AD-linked transgenes (AD;Igf1r+/−). B. Latency time for reaching the cued platform significantly decreased through the acquisition sessions (P= 0, F = 35.49, df = 3) without differences between the four genotypes (P>0.05, F = 1.84, df = 3, n=8, 15, 16, 18 for AD, AD;Igf1r+/−, WT and Igf1r+/− respectively), suggesting no impairment of learning. C. For the submerged platform test significant differences were observed between genotypes (P = 5E-4, 2-Way-ANOVA, F=7.71, df = 3,) and across the acquisition days (P = 0.032, F = 2.97, df = 3, n=8, 15, 16, 18 for AD, AD;Igf1r+/−, WT and Igf1r+/−, respectively). The AD mice spent a longer time (P<0.05. Fisher-LSD) after the second day in order to find the submerged platform while no difference was observed among the three other genotypes. D. Analysis of the number of crosses of the previous platform location indicated that AD;Igf1r+/− animals crossed significantly (P=0.024, Kruskal-Wallis, χ² =9.38, df = 3) more times than their AD counterparts. E. We analyzed the performance of individuals after the age of plaque formation (i.e. 12–16 month) in a Rota Rod task. Animals of the different genotypes significantly differed in their performance (P < 0.01, one-way ANOVA, df = 3, F = 4.25; n = 31, 32, 29 and 28 individuals for AD, AD;Igfr+/−, Igf1r+/− and wild type, respectively). AD mice performed worst among the four genotypes while AD;Igfr+/− mice where partially rescued from this impairment as they performed significantly better than AD animals (p < 0.05, Tuckey's LSD). Yet no statistical difference appeared between this later genotype and the two other control genotypes. In all behavioral tests, 11–15 month old mice were tested and age match controlled.

Figure 2

Reduced IGF signaling reduces Aβ associated neuroinflamation. A – H. Immuno histochemistry using GFAP antibody indicated reduced astrocytosis in brain sections of 12–13 month old AD;Igf1r+/− mice (D and H) compared to age matched AD mice (C and G). I. Image analysis confirmed the significance of the GFAP signal difference (six mice per genotype and 3 sections per animal were analyzed, P<0.05).

Figure 3

Reduced IGF signaling protects from Aβ associated neuronal and synpatic loss. A–H. Immuno histochemistry using NeuN antibody indicated that neural densities in the brains of 12–13 month old AD;Igf1r+/− (D and H), WT (A and E) and Igf1r+/− (B and F) mice were comparable, while remarkable neuronal loss was observed in brains of age matched AD animals (C and G). I. Image analysis of the NeuN signals indicated that neural density in both cortices and hippocampuses of AD animals was significantly lower compared to their age matched WT counterparts (Cortex: p<0.001, One way ANOVA, F=16.03; Hippocampus p<0.05, Kruskal-Wallis χ²=9.36, df=3). No significant difference was observed among brains of AD;Igf1r+/− and Igf1r+/− mice (six mice per genotype and 3 sections per animal were analyzed). J and K. Immuno histochemistry using synaptophysin antibody revealed that AD;Igf1r+/− mice exhibit significantly higher synaptic densities than their age matched AD counterparts in both frontal (J) and hippocampal (K) brain regions (AD n=7, AD;Igf1r+/− n=5).

Figure 4

Reduced IGF signaling facilitates Aβ hyper-aggregation. A. Thioflavin-S amyloid labeling showed similar Aβ plaque burden in brains of AD (panels III and VII) and AD;Igf1r+/− animals (panels IV and VIII). Image analysis indicated that the Thioflavin-S signals are similar in brains of AD and AD;Igf1r+/− mice, but signifcantly different from WT and Igf1r+/− mice (panel IX). Six 12–13 month old animals per genotype were analyzed. B. Aβ plaque signal density was measured using Aβ specific antibody (82E1). The signal per area ratio in brains of AD;Igf1r+/− animals (panels IV and VIII) was significantly higher (panel IX, P<0.05) compared to brains of age matched AD animals (panels III and VII) indicating higher plaque compaction in brains of AD;Igf1r+/− mice (six mice per genotype and 3 sections per animal were analyzed, DG – Dentate gyrus, NC – Neocortex).

Figure 5

Eleectron microscopy and in-vitro kinetic aggreagtion assays reveal densely packed Aβ aggregates in the brains of protected AD;Igf1r+/− mice. A. Electron micrographs of immuno-gold labelled Aβ amyloids in the cortex of AD and AD;Igf1r+/− mouse brains at different ages. Gold labeled amyloid and fibrillar Aβ structures can be observed in the higher magnification electron micrographs (right panels). The amyloid load similarly increased with age in both genotypes, but a highly ordered and condensed amyloid was present in AD;Igf1r+/− cortices (arrows) but not in the cortices of their AD counterparts (White scale bars 1 μm, black bars represent 200nm). B. Unbiased automated image processing indicates that median intensities of regions of interest (ROIs) around the gold particles labeling Aβ plaques of AD;Igf1r+/− mice (black) are significantly (P<0.04) lower than the plaque intensities of age matched AD animals (red), confirming the higher compaction state of Aβ plaques of AD;Igf1r+/− (six mice per genotype were processed, 135 images (34,087 ROIs) of AD and 101 images (26,066 ROIs) of AD;Igf1r+/− were collected and analyzed). C. Using an in-vitro kinetic aggreagtion assay to assess fibril load, 12–13 month old AD;Igf1r+/− mouse brain homogenates (blue) accelerated Thioflavin-T (ThT) monitored in-vitro kinetic aggregation significantly (P=0.035) faster than homogenates of age matched AD brains (brown) indicating more Aβ seeding competent assemblies in AD;Igf1r+/− mouse brains. Inset: statistical analysis of results obtained in C.

Figure 6

AD brains contain more soluble Aβ oligomers than brains of AD;Igf1r+/− animals. A and B. ELISA assay detected significantly higher amounts of soluble Aβ_1–40 (A) (P<0.001) and Aβ_1–42 (B) (P<0.005) in brain homogenates of 12–13 month old AD mice compared to brains of age matched AD;Igf1r+/− animals. C and D. Western blot analysis reveals no detectable difference in the amount of SDS sensitive Aβ monomers and small oligomeric assemblies between AD and AD;Igf1r+/− brain homogenates. * indicates significant difference from WT or Igfr1+/− mice. E. Native size exclusion chromatography (SEC) indicated that Aβ dimers were mainly associated with large structures in brains of 16–17 month AD;Igf1r+/− mice (panel iii) while more soluble in brains of age matched AD animals (panel ii, arrowhead)(panels represent 6 AD and 6 AD;Igf1r+/− animals that were analyzed). Loading of total samples onto the gel and subsequent western blot analysis using 6E10, confirmed equal protein loading onto the column (panel i).

Figure 7

IGF-1 signaling can play several roles in mitigating the toxicity of Aβ. The digestion of APP creates Aβ monomers that spontaneously aggregate to form toxic oligomers and possibly higher order structures in-vivo. At least two biological mechanisms can detoxifiy Aβ oligomers: (1) conversion of toxic oligomers into monomers (disaggregation) and (2) the conversion of the toxic oligomers into a much less toxic, larger structures (active aggregation). Within scenario 1, IGF-1 signaling could normally funciton to reduce protein disaggregases, possibly mediated by HSF transcription factors. Therefore, reduction of IGF-1 signaling would be predicted to result in less oligomers and more monomeric forms of Aβ due to the activation of protein disaggregases. Our resutls are inconsistent with this scenario since we find less oligomers, but equal amounts of monomeric forms of Aβ. Alternatively, within in scenario 2, IGF-1 signaling could normally function to reduce protective protein aggregases that convert toxic species into larger, less toxic forms. Therefore, reduced IGF-1 signaling results in increased aggreagase activity that in turn reduces the load of toxic oligomers and increases the compaction of less toxic fibrils. In support of scenario 2, we observed less soluble oligomers and higher compaction of the amyloid plaques in the AD animals with reduced IGF1R signaling (AD;Igf1r+/− animals). Alternatively, (3) IGF-1 signaling could promote protoeotoxicity and neuro-inflamation in response to toxic Aβ assemblies. Our results are also consistent with this proposed mechanism as we observed much less neuro-inflammation in the brains of protected AD;Igf1r+/− animals. However, this lower inflammation rate could be directly related to the reduction of Aβ oligomers in the protected animals by increased aggregases. Finally,in scenario 4, reduction of toxic secondary factors, such as Reactive Oxygen Species (ROS), might synergize with the production of toxic Aβ assemblies to promote neuronal loss. Consistent with this mechanism, Igf1r+/− mice are much more resistant to oxidative damage than wild type mice. Taken together, IGF-1 signaling could impinge at multiple steps on the path to neuronal loss and neurodegeneration in response to Aβ production and none of the interventions are mutually exclusive. Therefore, methods to reduced IGF-1 signaling provide multiple opportunities for disease intervention. Our data is most consistent with a model in which reduced IGF-1 signaling reduces the load of toxic Aβ structures, presumably dimers, that results in higher compaction of plaques, reduced neruo-inflammation and reduced neuronal loss.