Sensory gating endophenotype based on its neural oscillatory pattern and heritability estimate

Abstract

Context: The auditory sensory gating deficit has been considered a leading endophenotype in schizophrenia. However, the commonly used index of sensory gating, P50, has low heritability in families of people with schizophrenia, raising questions about its utility in genetic studies. We hypothesized that the sensory gating deficit may occur in a specific neuronal oscillatory frequency that reflects the underlying biological process of sensory gating. Frequency-specific sensory gating may be less complex than the P50 response, and therefore closer to the direct genetic effects, and thus a more valid endophenotype.

Objectives: To compare the gating of frequency-specific oscillatory responses with the gating of P50 and to compare their heritabilities.

Design: We explored single trial-based oscillatory gating responses in people with schizophrenia, their relatives, and control participants from the community.

Setting: Outpatient clinics.

Participants: Persons with schizophrenia (n = 102), their first-degree relatives (n = 74), and control participants from the community (n = 70).

Main outcome measures: Gating of frequency-specific oscillatory responses, gating of the P50 wave, and their heritability estimates.

Results: Gating of the theta-alpha-band responses of the control participants were significantly different from those with schizophrenia (P < .001) and their first-degree relatives (P = .04 to .009). The heritability of theta-alpha-band gating was estimated to be between 0.49 and 0.83 and was at least 4-fold higher than the P50 heritability estimate.

Conclusions: Gating of the theta-alpha-frequency oscillatory signal in the paired-click paradigm is more strongly associated with schizophrenia and has significantly higher heritability compared with the traditional P50 gating. This measure may be better suited for genetic studies of the gating deficit in schizophrenia.

Figure 1

An illustration of 8-level bio5.5 discrete wavelet decomposition on a single trial recording in responses to S1 and S2. Time 0 indicates the onset of stimuli. Each 500 ms epoch was divided into 4 125-ms epochs (T0–T3) for the purpose of extracting the temporal evolution of the PSD. The y-axis of each detail ranges from −12 to +12 on wavelet scale, marked by the lower and upper lines that separate the details. There were no discernable signals that could be easily observed from the original single trial recording (top). With the wavelet transform, it appeared that very small but discernable energy changes might have appeared at D6 (12–20 Hz) and D7 (5–12 Hz) in response to S1 (marked by ovals), which appeared attenuated in response to S2. In comparison, for instance at D5, the energy in response to S2 appeared the same or even slightly increased compared to the energy in response to S1 in this particular trial.

Figure 2

Oscillatory patterns in response to S1 (left) and S2 (right panel) in community controls, schizophrenia patients, and their relatives. Error bars are standard errors.

Figure 3

Gating of beta and alpha-theta oscillatory components. A ratio of 1 (gray line) indicates no gating; less than 1 indicates a gated response to S2; above 1 indicates facilitated response to S2. Error bars are standard errors. * Both subjects with schizophrenia and their first-degree relatives showed significantly reduced sensory gating in this time-frequency component compared to controls.

Figure 4

Heritability estimates of different sensory gating components. Error bars are standard errors. If h² = 0, no standard errors are calculated. Note that the most heritable component resides at the T1 (26–150 ms) epoch across different groupings. * p<0.05 ; ** p<0.01.