Network Analysis of a Membrane-Enriched Brain Proteome across Stages of Alzheimer's Disease

Abstract

Previous systems-based proteomic approaches have characterized alterations in protein co-expression networks of unfractionated asymptomatic (AsymAD) and symptomatic Alzheimer's disease (AD) brains. However, it remains unclear how sample fractionation and sub-proteomic analysis influences the organization of these protein networks and their relationship to clinicopathological traits of disease. In this proof-of-concept study, we performed a systems-based sub-proteomic analysis of membrane-enriched post-mortem brain samples from pathology-free control, AsymAD, and AD brains (n = 6 per group). Label-free mass spectrometry based on peptide ion intensity was used to quantify the 18 membrane-enriched fractions. Differential expression and weighted protein co-expression network analysis (WPCNA) were then used to identify and characterize modules of co-expressed proteins most significantly altered between the groups. We identified a total of 27 modules of co-expressed membrane-associated proteins. In contrast to the unfractionated proteome, these networks did not map strongly to cell-type specific markers. Instead, these modules were principally organized by their associations with a wide variety of membrane-bound compartments and organelles. Of these, the mitochondrion was associated with the greatest number of modules, followed by modules linked to the cell surface compartment. In addition, we resolved networks with strong associations to the endoplasmic reticulum, Golgi apparatus, and other membrane-bound organelles. A total of 14 of the 27 modules demonstrated significant correlations with clinical and pathological AD phenotypes. These results revealed that the proteins within individual compartments feature a heterogeneous array of AD-associated expression patterns, particularly during the preclinical stages of disease. In conclusion, this systems-based analysis of the membrane-associated AsymAD brain proteome yielded a unique network organization highly linked to cellular compartmentalization. Further study of this membrane-associated proteome may reveal novel insight into the complex pathways governing the earliest stages of disease.

Keywords: biomarkers; preclinical; proteomics; synapse; vesicles.

Figure 1

Characterization of Membrane-Enriched Samples. (A) Proteins identified by gene ontologies (GOs) as uniquely cytoplasmic, membrane, or intrinsic to membranes were quantified in both membrane and soluble fractions by peptide log₂ spectral count ratio (membrane/soluble). Proteins were binned into deciles ranked by the log₂(ratio) to represent the decile-specific average degree of enrichment or depletion within the membrane fraction. Only proteins with three or more peptide spectral counts were considered. VAMP2, a representative intrinsic membrane protein, was enriched in the membrane fraction, whereas the peripheral membrane protein ROCK2 was among proteins depleted in the membrane compared to the soluble fraction. The presynaptic protein GAP43 was also enriched in membrane fraction. (B) Western blots of total and phosphorylated (pSer41) GAP43 in total brain homogenate, soluble, and membrane fractions were performed on the 6 control samples. The left panel depicts the blot from two representative cases, while the right panel shows the quantified densitometry totals for all 6 cases. Enrichment of both phosphorylated and unmodified GAP43 was observed in the membrane fraction. Abbreviations: WB, Western Blot.

Figure 2

Differential Protein Abundance Across Disease Stages. (A) The top graph depicts extracted ion chromatograms for a fully tryptic amyloid precursor protein (APP) peptide corresponding to residues 17–28 of the Aβ sequence (LVFFAEDVGSNK) in a representative control, Asymptomatic Alzheimer’s Disease (AsymAD) and Alzheimer’s disease (AD) case. Signals were normalized by setting the maximum signal intensity of the AD sample to 100%. The bottom graph demonstrates the normalized peptide intensity of this Aβ sequence in all 18 cases. As expected, this measurement increased incrementally from control to AsymAD to AD cases. (B) Venn diagram for the 530 proteins significantly altered (p < 0.05) among the three pairwise comparisons, i.e., AD vs. Control, AsymAD vs. Control, and AD vs. AsymAD. APP and the synaptic protein SNAP25 were the only two proteins to demonstrate significant changes in all pairwise comparisons. (C) Volcano plots display the log-transformed fold change (Log₂ Difference) against the log-transformed Tukey-adjusted ANOVA p value (−Log₁₀ p Value) for all proteins of each pairwise comparison. Those proteins with significantly decreased expression (p < 0.05) for each comparison are shown in blue, while the proteins with significantly increased expression (p < 0.05) are noted in red. Abbreviations: AD, Alzheimer’s disease; AsymAD, Asymptomatic Alzheimer’s Disease; Log₂Diff, Log₂ Difference (i.e., Log₂ Fold Change).

Figure 3

Network Modules Correlate to Membrane-Bound Cellular Compartments. (A) Weighted protein co-expression network analysis (WPCNA) grouped proteins (n = 1808) into distinct protein modules (M1–M27) that were then clustered to assess module relatedness based on correlation of protein co-expression eigenproteins. A hypergeometric Fisher exact test revealed only two networks with significant enrichment of cell-type specific markers (* p < 0.05; ** p < 0.01). (B) A separate Fisher exact test demonstrated strong module associations with human gene ontologies related to membrane-bound cellular compartments (* p < 0.05; ** p < 0.01). There were six modules (M2, M4, M15, M19, M20, M26) that correlated most strongly and/or specifically with gene ontologies related to the mitochondrion or mitochondrial membrane. In contrast, there were four modules (M5, M8, M9, M25) with strong correlations to synaptic/cell surface terms. Other membrane-bound compartments highly represented in this proteome included the endoplasmic reticulum (M2, M17, M21), nucleus (i.e., chromosome) (M10, M11), and Golgi apparatus (M8, M16). Finally, five modules were highly linked to the cytosol/cytoskeleton (M1, M3, M13, M22, M23). Abbreviations: M, Module; GO, Gene Ontology.

Figure 4

Unfractionated and Membrane-Associated Co-Expression Networks Demonstrate Minimal Overlap. A hypergeometric one-tailed Fisher’s exact test (FET) was used to identify modules that shared significant overlap of protein members between the membrane-fractionated (M) network and that of unfractionated control, AsymAD, and AD cases derived from the Baltimore Longitudinal Study of Aging (BLSA). The 16 modules of the unfractionated (U) BLSA network, clustered by eigenprotein relatedness, are shown on the x-axis along with their top protein ontologies. These BLSA modules were aligned to the 27 modules of the membrane-associated network (y-axis). Module gene symbol lists showed either significant overlap (red) or no significant under- or over-representation (white) in protein membership. Numbers are positive signed −Log₁₀(FDR-corrected p values) representing the degree of overlap (* p < 0.05; ** p < 0.01). Notable overlapping modules are highlighted to the right of the FET results.

Figure 5

Modules in the Membrane Proteome are Correlated to Clinical and Pathological AD Phenotypes. (A) Biweight midcorrelation (bicor) analysis of module co-expression eigenproteins to clinical and pathological disease traits, including AD diagnosis, cognitive decline as measured by Cognitive Assessment Screening Instrument (CASI) score, and cortical levels of amyloid (Consortium to Establish a Registry for Alzheimer’s Disease (CERAD) score) and tau (Braak score). There were 14 modules with significant correlations to one or more disease-associated traits (* p < 0.05; ** p < 0.01). (B–F) Module expression profiles and key hub proteins of trait-associated modules organized by compartment localization (GO terms). p values were calculated for each expression profile using Kruskal–Wallis one-way nonparametric ANOVA. Abbreviations: AD, Alzheimer’s disease; AsymAD, Asymptomatic Alzheimer’s Disease; ER, Endoplasmic Reticulum.