Complex and multidimensional lipid raft alterations in a murine model of Alzheimer's disease

Abstract

Various animal models of Alzheimer's disease (AD) have been created to assist our appreciation of AD pathophysiology, as well as aid development of novel therapeutic strategies. Despite the discovery of mutated proteins that predict the development of AD, there are likely to be many other proteins also involved in this disorder. Complex physiological processes are mediated by coherent interactions of clusters of functionally related proteins. Synaptic dysfunction is one of the hallmarks of AD. Synaptic proteins are organized into multiprotein complexes in high-density membrane structures, known as lipid rafts. These microdomains enable coherent clustering of synergistic signaling proteins. We have used mass analytical techniques and multiple bioinformatic approaches to better appreciate the intricate interactions of these multifunctional proteins in the 3xTgAD murine model of AD. Our results show that there are significant alterations in numerous receptor/cell signaling proteins in cortical lipid rafts isolated from 3xTgAD mice.

Figure 1

Morris Water Maze testing of control and 3xTgAD mice. (a) Distance travelled (cm) in the nonvisible probe target results for control (n = 10, blue bars) and 3xTgAD mice (n = 10, red bars) for 8 days of training. (b) Water maze escape latency (s) for days 1 to 8 of training in the nonvisible probe target. (c) Swim speed (cm/s) assessment of control and 3xTgAD animals during days 1–8 of training. *P < .05; **P < .01.

Figure 2

Quantification of detergent-resistant lipid rafts. (a) The pictorial panel depicts an image of iodixanol-separated detergent resistant membrane fractions (centered on red arrow), captured using a Nikon 3200 digital camera. The line diagram indicates the direction of collection of centrifugal fractions 1–10, and the associated Western blot for flotillin-1 demonstrates its enrichment in the raft fractions. (b) Captured Joint Photographic Expert Group (JPG) images were converted to a Tagged Image File Format (TIFF) version and imported to Image Gauge (v4.2) software and the specific area of interest (red box), that is, the detergent-resistant, flotillin-1-rich lipid raft band was quantified into absorbance units minus background absorbance per square pixel area ((AU-B)/px²) values. (c) Representative set of male 8-month-old control (1) and 3xTgAD (2) mice detergent-resistant membranes isolated from plasma membrane fractions separated using an Iodixanol gradient. The associated histogram depicts mean ± s.e. (standard error) mean detergent-resistant membrane intensity ((AU-B)/px²) from at least three separate control and 3xTgAD mice (P = .033, nonpaired, two-tailed t-test). (d) Representative set of male 16-month-old control (1) and 3xTgAD (2) mice detergent resistant membranes isolated from plasma membrane fractions separated using an Iodixanol gradient. The associated histogram depicts mean ± s.e. (standard error) mean detergent resistant membrane intensity ((AU-B)/px²) from at least three separate C57-BL6 and 3xTgAD mice (P = .0224, non-paired, two-tailed t-test). (e) Representative flotillin-1 Western blot for the lipid raft fraction series for 3xTgAD (red outline) or control (C57-BL6: blue outline) mice. The associated histogram depicts mean ± s.e. (standard error) mean fraction 2 flotillin expression intensity ((AU-B)/px²) from at least three separate C57-BL6 and 3xTgAD mice (P = .015, non-paired, two-tailed t-test).

Figure 3

Differential protein expression in control versus 3xTgAD lipid raft extracts. (a) Proportionately drawn Venn diagram analysis of reliably identified proteins from control lipid rafts (blue line) and 3xTgAD rafts (red line). (b) Total protein loading control for centrifugal fraction 2. A total of 50 mg of fraction 2 protein was loaded and stained with SYPRO Ruby and scanned using a phosphorimager. (c)–(n). Representative western blots from multiple expression analysis experiments for differential presentation of proteins in fraction 2 extracts from control (blue) or 3xTgAD mice (red). Associated with each panel (c)–(n) the associated histograms represent the mean ± s.e. mean of protein expression intensity (measured in ((AU-B)/px₂)) from at least three separate experiments. *P < .05; **P < .01; ***P < .001. Protein abbreviations are as follows. FAK: focal adhesion kinase; Pyk2: proline-rich tyrosine kinase 2; GIT-1: GRK interactor-1; Jak2: Janus kinase 2; Crk: v-Crk avian sarcoma virus CT10 oncogene homolog; Fyn: Fyn tyrosine kinase; IRS-1: insulin receptor substrate-1; casp 7: caspase 7; mTOR: mammalian target of rapamycin; Grin2: G protein-regulated inducer of neurite outgrowth 2; IGF-1R: insulin-like growth factor-1 receptor.

Figure 4

Functional pathway informatic clustering of control and 3xTgAD lipid raft proteins. (a) Subtractive representation of hybrid score generation after clustering of lipid raft proteins from control and 3xTgAD animal rafts into cellular signaling pathways. The hybrid scores were generated by multiplication of the protein enrichment ratio of the specific pathway with the negative log₁₀ (−log₁₀) of the probability of that enrichment (see Section 2). The data is presented as a numerical value of the control pathway hybrid score subtracted from the 3xTgAD pathway hybrid score. Pathways in which the score in 3xTgAD was greater than the control are denoted in red; pathways in which the control hybrid score is greater than the 3xTgAD hybrid score are denoted in blue. A similar depiction format is employed for differential analysis of control versus 3xTgAD Neuronal Function pathways (b), Energy Regulation/Metabolism pathways (c), and Stress Response pathways (d).

Figure 5

Functional receptor signaling pathway informatic clustering of control and 3xTgAD lipid raft proteins. Subtractive representation of hybrid score generation after clustering of raft proteins from control and 3xTgAD animal rafts into receptor signaling pathways. The data is presented as a numerical value of the control pathway hybrid score subtracted from the 3xTgAD pathway hybrid score. Pathways in which the score in 3xTgAD was greater than the control are denoted in red; pathways in which the control hybrid score was greater than the 3xTgAD hybrid score are denoted in blue.

Figure 6

Multidimensional protein latent semantic indexing (LSI) analysis of proteins extracted from control and 3xTgAD lipid rafts. Proteins from the control or 3xTgAD extracted datasets that possessed an explicit latent semantic indexing (LSI, GeneIndexer, Computable Genomix) score in at least two of the multiple GeneIndexer interrogation terms (Alzheimer's, oxidation, neurodegeneration, synaptic transmission, neurogenesis, scaffolding, and GPCR) are represented in a heatmap format. Proteins are identified on the left side of the heatmap as an individual number (see Appendix H for key). The presence of a colored panel (3xTgAD, red: control, blue: 3xTgAD and control: yellow) on the same lateral as the numbered protein denotes explicit textual correlation of that protein with the specific vertical interrogation term.

Figure 7

Receptor-restricted control and 3xTgAD network interaction analysis. (a) Proportionally drawn Venn diagram depicting the relative distribution between control or 3xTgAD raft samples of receptor-specific proteins filtered using IPA version 8.5 (control filtered protein list, Table 1; 3xTgAD filtered protein list, Table 2). (b) The highest scoring protein interaction network generated from IPA Network analysis (network scores and focus molecules are listed in Appendix I) of the receptor-specific control dataset. (c) The highest scoring protein interaction network generated from IPA Network analysis (network scores and focus molecules are listed in Appendix I) of the receptor-specific 3xTgAD dataset (Appendix J). A full description of the nature of interactions based on the connecting lines can be found at the following webpage linked to the IPA analysis module (https://analysis.ingenuity.com/pa/info/help/help.htm#ipa_help.htm). Dashed lines represent indirect gene interactions while solid lines represent empirically measured direct interactions. The two highest significantly scoring networks (B—control, C—3xTgAD) are based on the highest percentage of the network occupation by specific proteins (focus molecules) from the input receptor-specific datasets.

Figure 8

Alzheimer's disease correlation of control or 3xTgAD receptor-specific lipid raft proteins. (a) The histogram represents the number of receptor-specific raft proteins (Table 1: control, Table 2: 3xTgAD) that demonstrate an explicit LSI correlation to the term “Alzheimer's” for control (blue) or 3xTgAD (red) extracts. (b) The histogram depicts the cumulated LSI correlation scores for the Alzheimer's-related receptor-specific proteins identified in (a). panels (c) and (d) represent the phylogenetic dendrograms for the receptor-specific raft proteins linked to the interrogation term “Alzheimer's” from control (c) or 3xTgAD (d) datasets. The proteins highlighted in red or blue and indicated by an arrow were specifically clustered into the highest scoring protein interaction networks for control (Figure 7(b)) or 3xTgAD (Figure 7(c)) samples.