Complex network analysis of CA3 transcriptome reveals pathogenic and compensatory pathways in refractory temporal lobe epilepsy

Abstract

We previously described - studying transcriptional signatures of hippocampal CA3 explants - that febrile (FS) and afebrile (NFS) forms of refractory mesial temporal lobe epilepsy constitute two distinct genomic phenotypes. That network analysis was based on a limited number (hundreds) of differentially expressed genes (DE networks) among a large set of valid transcripts (close to two tens of thousands). Here we developed a methodology for complex network visualization (3D) and analysis that allows the categorization of network nodes according to distinct hierarchical levels of gene-gene connections (node degree) and of interconnection between node neighbors (concentric node degree). Hubs are highly connected nodes, VIPs have low node degree but connect only with hubs, and high-hubs have VIP status and high overall number of connections. Studying the whole set of CA3 valid transcripts we: i) obtained complete transcriptional networks (CO) for FS and NFS phenotypic groups; ii) examined how CO and DE networks are related; iii) characterized genomic and molecular mechanisms underlying FS and NFS phenotypes, identifying potential novel targets for therapeutic interventions. We found that: i) DE hubs and VIPs are evenly distributed inside the CO networks; ii) most DE hubs and VIPs are related to synaptic transmission and neuronal excitability whereas most CO hubs, VIPs and high hubs are related to neuronal differentiation, homeostasis and neuroprotection, indicating compensatory mechanisms. Complex network visualization and analysis is a useful tool for systems biology approaches to multifactorial diseases. Network centrality observed for hubs, VIPs and high hubs of CO networks, is consistent with the network disease model, where a group of nodes whose perturbation leads to a disease phenotype occupies a central position in the network. Conceivably, the chance for exerting therapeutic effects through the modulation of particular genes will be higher if these genes are highly interconnected in transcriptional networks.

Figure 1. Concentric levels.

Example of concentric levels of a network for node as reference (i.e. centered at node ). Each concentric level is represented by rings , namely , and , with concentric node degrees and .

Figure 2. Node categories.

Networks illustrating the three proposed categories of nodes with VIPs presenting low node degree but high concentric node degree at first level, hubs with lower concentric node degree and higher node degree, and high-hubs presenting high values of both properties.

Figure 3. Network projection and visualization.

On left side of the figure the network is randomly placed over the desired metric space (in this case, the 2D plane) and the forces described by equations (1) and (2) acting on each node are shown as arrows. Nodes without connection between them (displayed as fading blue arrows) interact only by repulsive forces while connected nodes (red arrows) present both attractive and repulsive interactions. Through the molecular dynamics simulation the system attains a lower energy state with connected nodes being close together if an edge is present and further away otherwise, as shown on the figure's right side. This methodology preserves the network topological structure. Final visualization is obtained by vertex color mapping according to a property or measurements, e.g. node degree.

Figure 4. Transcriptional interaction network for FS group.

FS transcriptional interaction network based on Pearson's correlation of 307 differentially expressed GO annotated genes (FS-DE). Hubs (red) and VIPs (blue) are identified by their gene symbols.

Figure 5. Transcriptional interaction network for NFS group.

NFS transcriptional interaction network based on Pearson's correlation of 307 differentially expressed GO annotated genes (NFS-DE). Hubs (red) and VIPs (blue) genes are identified by their gene symbols.

Figure 6. Node distribution and categorization for differentially expressed FS and NFS networks.

Normalized degree distribution log-log plot (A) for FS and NFS differentially expressed networks, alongside their respective best fitted power law (lines), corresponding to the function . Scatter plot (B,C) of node degree () vs concentric node degree () measures of GO annotated genes obtained in the networks for differentially expressed genes (DE). Hubs (red) and VIPs (blue) genes are identified by their gene symbols. 240 nodes for the FS network (B) and 246 nodes for the NFS network (C).

Figure 7. Node distribution and categorization for complete FS and NFS networks.

Normalized degree distribution log-log plot (A) for FS and NFS complete networks, alongside their respective best fitted power law (lines), corresponding to the function . Scatter plot (B,C) of node degree () vs concentric node degree () measures of GO annotated genes obtained in the complete transcriptional interaction networks (CO). High-Hubs (green), Hubs (red) and Vips (blue) genes are identified by their gene symbols.15,585 nodes for the FS network (B) and 11,233 nodes for the NFS network (C).

Figure 8. Interactome networks.

Interactome networks for DE (A, B) and CO (C, D) hubs/high-hubs (depicted in red) and VIPs (depicted in blue) using MINT, HPRD and IntAct databases for FS (A, C) and NFS (B, D) groups. Node shapes and colors represent biological processes, as follows: diamond for actin/cytoskeleton; hexagon for signaling (blue or pink borders stands for neurotransmission or inflammation); octagon for calcium ion binding/transport/homeostasis; octagon with blue or pink border stand for cation channel or cation transport respectively; parallelogram for cell-cell interaction/adhesion; rectangle for cell processes (blue or green borders stand for apoptosis or proliferation); bordered triangles stand for synaptic transmission/SNARE complex (light blue), voltage-gated ion channel activity (pink), GABAergic synaptic transmission (red), synaptic transmission (green) and neuronal differentiation/motility (purple); vee for cholesterol homeostasis; ellipse for other processes. Nodes in red represent high-hubs or hubs in CO or DE transcriptional networks, respectively; Nodes in blue represent VIPs in CO and DE transcriptional networks; Node size is related to node degree (number of links). Links in red represent the first and second hub connections, centered in DE or CO VIPs.

Figure 9. Histopathological results.

Histopathological findings in sclerotic hippocampi from patients with epilepsy and semiquantitative analysis of granule cell loss (GCL) in patients with (FS) or without (NFS) febrile seizures as initial precipitant insult. A: semiquantitative analysis of granule cell loss (GCL) in patients with (FS) or without (NFS) febrile seizures as initial precipitant insult (*: p<0.05). B: overview of a histological slice. C: preserved granule cell layer. D: granule cell loss. E: granule cell dispersion. F: granule cell bilamination. G: stargazin-positive cells (arrows). Calibration bars: B 4 mm; C–F 200 um; G 50 um. GCL was graded into the following categories: grade zero (no GCL), grade 1 (mild reduction of neuronal density), grade 2 (moderate reduction of neuronal density), grade 3 (severe reduction of neuronal density), and grade 4 (disruption of neuronal cell layer).