Alterations of Coherent Theta and Gamma Network Oscillations as an Early Biomarker of Temporal Lobe Epilepsy and Alzheimer's Disease

Abstract

Alzheimer's disease (AD) and temporal lobe epilepsy (TLE) are the most common forms of neurodegenerative disorders characterized by the loss of cells and progressive irreversible alteration of cognitive functions, such as attention and memory. AD may be an important cause of epilepsy in the elderly. Early diagnosis of diseases is very important for their successful treatment. Many efforts have been done for defining new biomarkers of these diseases. Significant advances have been made in the searching of some AD and TLE reliable biomarkers, including cerebrospinal fluid and plasma measurements and glucose positron emission tomography. However, there is a great need for the biomarkers that would reflect changes of brain activity within few milliseconds to obtain information about cognitive disturbances. Successful early detection of AD and TLE requires specific biomarkers capable of distinguishing individuals with the progressing disease from ones with other pathologies that affect cognition. In this article, we review recent evidence suggesting that magnetoencephalographic recordings and coherent analysis coupled with behavioral evaluation can be a promising approach to an early detection of AD and TLE. Highlights -Data reviewed include the results of clinical and experimental studies.-Theta and gamma rhythms are disturbed in epilepsy and AD.-Common and different behavioral and oscillatory features of pathologies are compared.-Coherent analysis can be useful for an early diagnostics of diseases.

Keywords: Alzheimer’s disease; coherent analysis; early diagnostics; electroencephalography; memory; oscillatory activity; temporal lobe epilepsy.

Figure 1

Theta coherence between brain areas changes during epileptogenesis. (A) Phase–phase coupling of theta oscillations between two brain areas (i) and (ii). To the left: synthetic data used for theta rhythm illustration. To the right: coherence spectrum (or phase-specific measures) between two signals can determine the strength of theta phase coupling. (B) Behavioral data for rats during the performance the episodic-like memory task. Distribution of exploratory times per object in the test phase for the control and epileptic groups; *p < 0.05, **p < 0.01, ***p < 0.005. The inset represents the object configuration in the task. (C) Representative hippocampal activity of an epileptic rat recorded in the stratum pyramidale (SP), lacunosum moleculare (SLM) and moleculare (ML) during walking. (D) Specific alterations in hippocampal theta activity in temporal lobe epilepsy (TLE) brain during object exploration in the episodic-like memory task; the time–frequency power spectrum of hippocampal field potentials in the SLM and the ML layers is shown for the 1–30 Hz frequency band. (E) Theta coherence between hippocampal SLM–ML layers during exploration of each individual object in the episodic-like memory task; the mean values of theta coherence per object within the mean (red line) and standard deviation (discontinuous line) for the whole session in the control and epileptic animals are shown. (F) Theta coherence between the hippocampus and medial prefrontal cortex (mPFC) increases pre-ictally. To the left: a mean coherogram (coherence vs. time, 0–20 Hz) of 120 s and 30 s pre-ictal local field potential (LFP) segments from the hippocampus and mPFC (30 s pre-ictal segment is designated by a black rectangle). To the right: mean ± standard error of the mean (SEM; solid ± dashed lines) coherence of 120 s (blue lines) and 30 s (green lines) recordings before seizures. (G) Representative wavelet coherograms and smoothed standard deviation of wavelet coherence of LFPs recorded in the hippocampus and medial septal-diagonal band (MSDB) in healthy (left) and epileptic animals. Adapted with permission from Buzsáki and Watson (2012) (A), Inostroza et al. (2013) (B–E), Broggini et al. (2016) (F) and Kabanova et al. (2011) (G).

Figure 2

Theta–gamma cross-frequency coupling (CFC) and its alteration in a rat model of TLE. (A) Schematic illustration of cross-frequency phase-phase coupling. Phases of theta and gamma oscillations are correlated, as shown (to the right) by the phase-phase plot of the two frequencies; (i) and (ii)—different brain areas, Hi—hippocampus. (B) A heuristic model of cross-frequency phase–amplitude coupling. Gamma oscillations are large (red line) in the excitatory phase of theta wave (black line) and small (blue line) in the inhibitory phase of theta wave. (C) The theta phase modulates the low-frequency gamma (LG) amplitude. A phase–amplitude comodulogram computed for LFP of the hippocampal CA3 field recorded at SP during execution of spatial task is shown. (D) Theta modulation of the LG amplitude in the CA3 region during context exploration increases with learning. Color scale representation of the mean LG amplitude as a function of the theta phase for each trial in the session (left). The mean LG amplitude per theta phase averaged over the first and last 20 trials is also shown (right). (E) Example of comodulation maps of hippocampal theta phase modulating mPFC gamma oscillation amplitude 120 s and 30 s before seizure onset. (F) Box plot showing mean hippocampal theta/mPFC gamma modulation index (MI), 120 s and 30 s before seizure onset; *p < 0.001. Adapted with permission from Buzsáki and Watson (2012) (A), Kirihara et al. (2012) (B), Tort et al. (2009) (C,D) and Broggini et al. (2016) (E,F).

Figure 3

Gamma amplitude modulation by theta phase is impaired in a mouse model of Alzheimer’s disease (AD; amyloid precursor protein 23, APP23 mice). (A) Raw EEG (LFP), band pass filtered signals for theta (4–12 Hz) and gamma (25–100 Hz) oscillations, gamma amplitude envelope (green) and theta phase in APP23 and non-transgenic (non-tg) mice (blue). Representative signals from five animals per genotype are shown. (B) Representative phase–amplitude comodulograms computed for hippocampal LFPs recorded in non-tg and APP23 mice. (C) Phase–amplitude plot computed for hippocampal LFPs recorded in non-tg and APP23 mice (means ± SEM). (D) MI computed for the phase–amplitude distributions shown in (C); *p < 0.05. Adapted from Ittner et (; Open Access).

Figure 4

Entorhinal tau overexpression affects phase–amplitude hippocampal—prefrontal CFC. The degree of modulation of gamma amplitude by mPFC theta phase quantified with the MI. (A) Representative examples of normalized hippocampal (to the left) and prefrontal (to the right) gamma amplitude aligned to concurrent prefrontal theta phase. (B) Changes in the MI before and after the CS (median ± 25th and 75th percentile). Compared against GFP-expressing rats (light gray), the change in MI of prefrontal theta-hippocampal gamma coupling was smaller while that of prefrontal theta-prefrontal gamma coupling was larger in tau-expressing rats (dark gray); ***p < 0.001. Abbreviations: CS, conditioned stimulus; GFP, green fluorescent protein. Adapted with permission from Tanninen et al. (2017).