Bridging integrator 1 (BIN1) protein expression increases in the Alzheimer's disease brain and correlates with neurofibrillary tangle pathology

Abstract

Recent genome wide association studies have implicated bridging integrator 1 (BIN1) as a late-onset Alzheimer's disease (AD) susceptibility gene. There are at least 15 different known isoforms of BIN1, with many being expressed in the brain including the longest isoform (iso1), which is brain-specific and localizes to axon initial segments and nodes of Ranvier. It is currently unknown what role BIN1 plays in AD. We analyzed BIN1 protein expression from a large number (n = 71) of AD cases and controls from five different brain regions (hippocampus, inferior parietal cortex, inferior temporal cortex, frontal cortex (BA9), and superior and middle temporal gyri). We found that the amount of the largest isoform of BIN1 was significantly reduced in the AD brain compared to age-matched controls, and smaller BIN1 isoforms were significantly increased. Further, BIN1 was significantly correlated with the amount of neurofibrillary tangle (NFT) pathology but not with either diffuse or neuritic plaques, or with the amount of amyloid-β peptide. BIN1 is known to be abnormally expressed in another human disease, myotonic dystrophy, which also features prominent NFT pathology. These data suggest that BIN1 is likely involved in AD as a modulator of NFT pathology, and that this role may extend to other human diseases that feature tau pathology.

Keywords: Alzheimer's disease; ZNF9; amyloid-β peptide; cellular nucleic acid binding protein; myotonic dystrophy; tau.

Figure 1. Identification of BIN1 Isoforms

(A) The largest isoform (iso1) is found in neurons and nerve terminals, whereas the smaller isoform (iso9) is relatively generic, and is expressed ubiquitously. The major differences between these two forms are insertions encoded by variants of exons 6 and 12. (B) Comparison with H4 cells overexpressing different isoforms of BIN1 indicated that antibody 99D detects the largest full-length isoform 1 (~80 kDa; iso1) in human brain along with at least two smaller isoforms (~60kDa), one of which is likely isoform 9 (iso9; V = empty vector), based on migration at the similar relative molecular weights.

Figure 2. Small M.W. Isoforms of BIN1 Increase in Alzheimer’s Disease Brain

(A) Representative Western blot showing that smaller BIN1 isoforms (~40-60 kDa; antibody 99D) were increased in AD cases compared to controls (2% SDS extracts from frontal cortex, area BA9, are shown); GAPDH levels are shown to confirm equal gel loading. (B) The larger BIN1 isoform (iso1) was significantly decreased in AD cases across brain regions analyzed individually, and when all regions were combined and treated as a single variable in the analysis. (C) In contrast, the smallest isoform (iso9) showed a corresponding increase. Mann-Whitney U-Test; * = p<0.05, ** = p<0.01; error bars represent the standard error of the mean (s.e.m.).

Figure 3. Confirmation of Increase in Smaller Isoforms of BIN1 and Relationship with AD Neuropathology

(A) Representative Western blot using a different antibody (Epitomics anti-BIN1; 2% SDS extracts from frontal cortex, area BA9, are shown); GAPDH levels are shown to confirm equal gel loading. Interestingly, the Epitomics antibody was considerably less effective at detecting the higher molecular weight isoform of BIN1 (iso1), indicating that the epitope for this antibody may be modified in the human brain. (B) Densitometric analysis confirms that the smaller isoforms of BIN1 were increased across brain regions in the AD cases using this antibody (similar to Figure 2, above; the difference in the SMTG was also significant, p<0.05; not shown). (C) The smaller BIN1 isoform was significantly correlated with the number of NFTs in the brain over a large number of cases (sections were taken from the same brain region as the tissue sample used for Western blotting). Mann-Whitney U-Test; * = p<0.05, ** = p<0.01, *** = p < 0.001; error bars represent the standard error of the mean (s.e.m.).