On CA1 ripple oscillations: reevaluating asynchronicity evidence
Curation statements for this article:-
Curated by eLife
eLife Assessment
This valuable study provides new insights into the synchronization of ripple oscillations in the hippocampus, both within and across hemispheres. Using carefully designed statistical methods, it presents convincing evidence that synchrony is significantly higher within a hemisphere than across. However, further controlling for potential confounds related to differences in animal behavior will help clarify whether this effect is influenced by memory processing. This study will be of interest to neuroscientists studying the hippocampus and memory.
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
Abstract
Sharp-wave ripples (SWRs) are hippocampal network oscillations associated with memory consolidation. They are characterized by the co-occurrence of fast and slow field potentials across CA1 layers: the fast-frequency oscillations, known as ripples, are prominent in the pyramidal cell layer, where they coincide with increased neuronal spiking, while slower negative transients, referred to as sharp waves, occur simultaneously in the stratum radiatum . SWRs have traditionally been considered globally synchronous across the hippocampus; however, recent evidence suggests that ripples may be less synchronous than previously thought, particularly between the two hemispheres (Villalobos et al., 2017). In this study, we revisited this question using a unique dataset from probes spanning the septo-temporal axis of CA1. Our results demonstrate that ripples are phase-locked within but not between hemispheres, although their occurrence remains time-locked across both the septo-temporal axis and hemispheres. We also observed a similar synchronicity pattern for spiking activity: neurons are locally phase-coupled and globally time-coupled to ripple events. Interneurons exhibit a much stronger phase coupling to both ipsilateral and contralateral ripples than pyramidal neurons. These findings suggest that ripples are locally phase-coupled through pyramidal-interneuron interactions, with global time-locking likely driven by a common bilateral CA3 input and potentially modulated by interneuron circuits.
Article activity feed
-
-
-
eLife Assessment
This valuable study provides new insights into the synchronization of ripple oscillations in the hippocampus, both within and across hemispheres. Using carefully designed statistical methods, it presents convincing evidence that synchrony is significantly higher within a hemisphere than across. However, further controlling for potential confounds related to differences in animal behavior will help clarify whether this effect is influenced by memory processing. This study will be of interest to neuroscientists studying the hippocampus and memory.
-
Reviewer #1 (Public review):
Summary:
In this manuscript, the authors analyze electrophysiological data recorded bilaterally from the rat hippocampus to investigate the coupling of ripple oscillations across the hemispheres. Commensurate with the majority of previous research, the authors report that ripples tend to co-occur across both hemispheres. Specifically, the amplitude of ripples across hemispheres is correlated but their phase is not. These data corroborate existing models of ripple generation suggesting that CA3 inputs (coordinated across hemispheres via the commisural fibers) drive the sharp-wave component while the individual ripple waves are the result of local interactions between pyramidal cells and interneurons in CA1.
Strengths:
The manuscript is well-written, the analyses well-executed and the claims are supported by …
Reviewer #1 (Public review):
Summary:
In this manuscript, the authors analyze electrophysiological data recorded bilaterally from the rat hippocampus to investigate the coupling of ripple oscillations across the hemispheres. Commensurate with the majority of previous research, the authors report that ripples tend to co-occur across both hemispheres. Specifically, the amplitude of ripples across hemispheres is correlated but their phase is not. These data corroborate existing models of ripple generation suggesting that CA3 inputs (coordinated across hemispheres via the commisural fibers) drive the sharp-wave component while the individual ripple waves are the result of local interactions between pyramidal cells and interneurons in CA1.
Strengths:
The manuscript is well-written, the analyses well-executed and the claims are supported by the data.
Weaknesses:
One question left unanswered by this study is whether information encoded by the right and left hippocampi is correlated.
-
Reviewer #2 (Public review):
Summary:
The authors completed a statistically rigorous analysis of the synchronization of sharp-wave ripples in the hippocampal CA1 across and within hemispheres. They used a publicly available dataset (collected in the Buzsaki lab) from 4 rats (8 sessions) recorded with silicon probes in both hemispheres. Each session contained approximately 8 hours of activity recorded during rest. The authors found that the characteristics of ripples did not differ between hemispheres, and that most ripples occurred almost simultaneously on all probe shanks within a hemisphere as well as across hemispheres. The differences in amplitude and exact timing of ripples between recording sites increased slightly with the distance between recording sites. However, the phase coupling of ripples (in the 100-250 Hz range), changed …
Reviewer #2 (Public review):
Summary:
The authors completed a statistically rigorous analysis of the synchronization of sharp-wave ripples in the hippocampal CA1 across and within hemispheres. They used a publicly available dataset (collected in the Buzsaki lab) from 4 rats (8 sessions) recorded with silicon probes in both hemispheres. Each session contained approximately 8 hours of activity recorded during rest. The authors found that the characteristics of ripples did not differ between hemispheres, and that most ripples occurred almost simultaneously on all probe shanks within a hemisphere as well as across hemispheres. The differences in amplitude and exact timing of ripples between recording sites increased slightly with the distance between recording sites. However, the phase coupling of ripples (in the 100-250 Hz range), changed dramatically with the distance between recording sites. Ripples in opposite hemispheres were about 90% less coupled than ripples on nearby tetrodes in the same hemisphere. Phase coupling also decreased with distance within the hemisphere. Finally, pyramidal cell and interneuron spikes were coupled to the local ripple phase and less so to ripples at distant sites or the opposite hemisphere.
Strengths:
The analysis was well-designed and rigorous. The authors used statistical tests well suited to the hypotheses being tested, and clearly explained these tests. The paper is very clearly written, making it easy to understand and reproduce the analysis. The authors included an excellent review of the literature to explain the motivation for their study.
Weaknesses:
The authors state that their findings (highly coincident ripples between hemispheres), contradict other findings in the literature (in particular the study by Villalobos, Maldonado, and Valdes, 2017), but fail to explain why this large difference exists. They seem to imply that the previous study was flawed, without examining the differences between the studies.
The paper fails to mention the context in which the data was collected (the behavior the animals performed before and after the analyzed data), which may in fact have a large impact on the results and explain the differences between the current study and that by Villalobos et al. The Buzsaki lab data includes mice running laps in a novel environment in the middle of two rest sessions. Given that ripple occurrence is influenced by behavior, and that the neurons spiking during ripples are highly related to the prior behavioral task, it is likely that exposure to novelty changed the statistics of ripples. Thus, the authors should analyze the pre-behavior rest and post-behavior rest sessions separately. The Villalobos et al. data, in contrast, was collected without any intervening behavioral task or novelty (to my knowledge). Therefore, I predict that the opposing results are a result of the difference in recent experiences of the studied rats, and can actually give us insight into the memory function of ripples.
In one figure (5), the authors show data separated by session, rather than pooled. They should do this for other figures as well. There is a wide spread between sessions, which further suggests that the results are not as widely applicable as the authors seem to think. Do the sessions with small differences between phase coupling and amplitude coupling have low inter-hemispheric amplitude coupling, or high phase coupling? What is the difference between the sessions with low and high differences in phase vs. amplitude coupling? I noticed that the Buzsaki dataset contains data from rats running either on linear tracks (back and forth), or on circular tracks (unidirectionally). This could create a difference in inter-hemisphere coupling, because rats running on linear tracks would have the same sensory inputs to both hemispheres (when running in opposite directions), while rats running on a circular track would have different sensory inputs coming from the right and left (one side would include stimuli in the middle of the track, and the other would include closer views of the walls of the room). The synchronization between hemispheres might be impacted by how much overlap there was in sensory stimuli processed during the behavior epoch.
The paper would be a lot stronger if the authors analyzed some of the differences between datasets, sessions, and epochs based on the task design, and wrote more about these issues. There may be more publicly available bi-hemispheric datasets to validate their results.
-
