Basic Science and Pathogenesis

Background: Apolipoprotein E4 (APOE4), a common variant of APOE, is a major genetic risk factor for Alzheimer's disease (AD), but APOE4 carriers do not always develop AD. Several large-scale genetic studies have identified a common haplotype of the aging factor klotho that modify disease risk in APOE4 carriers. In humans, klothoharbors two common missense variants (rs9536314, p.F352V; rs9527025, p.C370S) that define the klotho V/S (KL-V/S) haplotype, which is protective against AD in APOE4 carriers, and the klotho F/C (KL-F/C) haplotype, which is not. We sought to understand this genetic interaction using novel mouse models.

Method: Mouse klotho differs at the key amino acid position p.C370S rendering the native mouse haplotype as "KL-F/S", matching neither of the observed human haplotypes. We have replaced this with either the homozygous protective KL-V/S or common KL-F/C haplotypes using CRISPR/Cas9. We have tested the effects of these klotho haplotypes on klotho expression in serum, brain, and kidney relative to B6 controls and to the amyloidogenic App^SAA AD model, which exhibit significant amyloid neuropathology. We have also run RNA-seq on whole brain homogenates at 12 months of age.

Result: Secreted p-KL was decreased in KL-F/C mice harboring the novel mouse risk haplotype relative to age matched B6 controls and to KL-V/S mice. In contrast, KL-V/S mice showed no significant differences in soluble p-KL serum levels when compared to B6 controls. Secreted p-KL levels in KL-V/S mice were significantly elevated when compared to the APP^SAA mice, whereas young KL-F/C mice showed similar levels to the aged APP^SAA model. Transcriptional analysis indicated that the KL-V/S haplotype was anti-correlated with AMP-AD human modules, while the KL-F/C risk haplotype was significantly correlated to AD signatures in modules enriched for neuronal system, cell cycle, myelination, and cellular stress response pathways.

Conclusion: These data suggest the native mouse klotho allele may act similarly to the protective KL-V/S haplotype, while the KL-F/C haplotype leads to lower secreted klotho and potentially greater disease risk. These models are therefore suitable for experiments to elucidate how klotho variants protect from AD pathologies, and will enable detailed studies of how this protection is specific to the APOE4 genotype.