Differential ripple propagation along the hippocampal longitudinal axis
Curation statements for this article:-
Curated by eLife
eLife assessment
This is an interesting study investigating the propagation dynamics of ripples recorded from the dorsal hippocampus of mice using an open-access dataset. Findings may have theoretical and practical implications for the study and manipulation of sharp-wave ripples, a main oscillatory event underlying memory consolidation. While the strength of evidence is solid and claims broadly supported, there are some points requiring additional analysis to clarify issues regarding the anatomical axes involved and to reinforce mechanistic insights.
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
- Neuroscience (eLife)
Abstract
Hippocampal ripples are highly synchronous neural events critical for memory consolidation and retrieval. A minority of strong ripples has been shown to be of particular importance in situations of increased memory demands. The propagation dynamics of strong ripples inside the hippocampal formation are, however, still opaque. We analyzed ripple propagation within the hippocampal formation in a large open-access dataset comprising 267 Neuropixel recordings in 49 awake, head-fixed mice. Surprisingly, strong ripples (top 10% in ripple strength) propagate differentially depending on their generation point along the hippocampal longitudinal axis. The septal hippocampal pole is able to generate longer ripples that engage more neurons and elicit spiking activity for an extended time even at considerable distances. Accordingly, a substantial portion of the variance in strong ripple duration ( R ² = 0.463) is explained by the ripple generation location on the longitudinal axis, in agreement with a possible distinctive role of the hippocampal septal pole in conditions of high-memory demand. Moreover, we observed that the location of the ripple generation has a significant impact on the spiking rate modulation of different hippocampal subfields, even before the onset of the ripple. This finding suggests that ripple generation location plays a crucial role in shaping the neural activity across the hippocampus.
Article activity feed
-
-
eLife assessment
This is an interesting study investigating the propagation dynamics of ripples recorded from the dorsal hippocampus of mice using an open-access dataset. Findings may have theoretical and practical implications for the study and manipulation of sharp-wave ripples, a main oscillatory event underlying memory consolidation. While the strength of evidence is solid and claims broadly supported, there are some points requiring additional analysis to clarify issues regarding the anatomical axes involved and to reinforce mechanistic insights.
-
Reviewer #1 (Public Review):
In this study, the authors use open-access datasets of Neuropixel recordings to explore the relationship between ripple strength and propagation in the septal/dorsal hippocampal pole. They found that the ripple strength correlates with the direction of propagation and that the duration of the events is dependent on the site of initiation. Medial pole ripples are longer and engage significantly more neurons than lateral ripples. These findings may have theoretical and practical implications for the study of sharp-wave ripples, a main oscillatory event underlying memory consolidation. While the approach is not entirely novel (e.g. Patel et al., JN 2013; Kumar and Deshmukh 2020), the study provides some additional insights. The strength of evidence of propagation dynamics is solid and claims are broadly …
Reviewer #1 (Public Review):
In this study, the authors use open-access datasets of Neuropixel recordings to explore the relationship between ripple strength and propagation in the septal/dorsal hippocampal pole. They found that the ripple strength correlates with the direction of propagation and that the duration of the events is dependent on the site of initiation. Medial pole ripples are longer and engage significantly more neurons than lateral ripples. These findings may have theoretical and practical implications for the study of sharp-wave ripples, a main oscillatory event underlying memory consolidation. While the approach is not entirely novel (e.g. Patel et al., JN 2013; Kumar and Deshmukh 2020), the study provides some additional insights. The strength of evidence of propagation dynamics is solid and claims are broadly supported. Some points however may require revision. In particular, issues regarding the definition of the longitudinal and transversal axes, as well as additional analysis on microcircuit interactions and neuronal dynamics per cell types and hippocampal sectors should be more thoroughly addressed in support of mechanisms.
-
Reviewer #2 (Public Review):
De Filippo et al. investigated the spatiotemporal dynamics of the ripples propagation in the hippocampus of head-fixed mice. By leveraging the LFP and the isolated units of an open dataset of 49 animals with ~6 Neuropixels probes in the longitudinal axis of the hippocampus, they found: first, that stronger ripples (>ninth decile of power) originated in the most septal pole of the hippocampus (medially, anatomically) tend to travel more (M to L) than more lateral ripples (closer to the temporal pole). Second, while strong ripples were mainly local, the authors found that they are most likely to be generated in the temporal pole of the hippocampus, from where they can travel with relatively small attenuation. Finally, they found that strong/septal ripples elicit high spiking activity along the entire …
Reviewer #2 (Public Review):
De Filippo et al. investigated the spatiotemporal dynamics of the ripples propagation in the hippocampus of head-fixed mice. By leveraging the LFP and the isolated units of an open dataset of 49 animals with ~6 Neuropixels probes in the longitudinal axis of the hippocampus, they found: first, that stronger ripples (>ninth decile of power) originated in the most septal pole of the hippocampus (medially, anatomically) tend to travel more (M to L) than more lateral ripples (closer to the temporal pole). Second, while strong ripples were mainly local, the authors found that they are most likely to be generated in the temporal pole of the hippocampus, from where they can travel with relatively small attenuation. Finally, they found that strong/septal ripples elicit high spiking activity along the entire mediolateral axis of the hippocampus. Longer/stronger ripples have been proposed to be important in situations with high memory load, and these analyses increase our understanding of their physiology and mechanisms of generation.
The conclusions of this paper are mostly well supported by data, but some aspects of interpretation and data analysis need to be clarified and extended.
High amplitude ripples preferentially occur in distal CA1, and ripples can propagate at a higher degree on the proximo-distal than in the septo-temporal axis of the hippocampus (Kumar and Deshmuckh, 2020). Therefore, a proximo-distal bias in the Neuropixel positioning could explain part of the variance the authors report. Authors should consider (or control for) the proximodistal positioning of the electrodes.
In my opinion, the dynamics of the ripple-induced spiking activity for the events generated in the medial or lateral section of the hippocampus are very striking, more even considering that only a minority of the detected ripples are strong/long events (less than 5% in a familiar environment, Fernandez-Ruiz et al, 2019), while, according to the authors, majority of the ripples (grouped as 'common' by the authors) travel on the opposite direction (from the lateral section towards the septal pole, figure 2). Moreover, in the 50-120ms window, the most lateral positions (>3500um) seem to be more influenced by the medial ripples than relatively more central electrodes (~3000um). How can the authors explain this? To understand a little bit more how ripple features relate to the spiking dynamics, authors could try to generate heatmaps of the differential spiking between medial and lateral ripples (as they did in Fig. 4D-E) for 'strong' and 'common' ripples, or for local and propagating ripples.
-
Reviewer #3 (Public Review):
Using a large Neuropixels dataset provided by the Allen Institute (https://allensdk.readthedocs.io/en/latest/visual_coding_neuropixels.html), Filippo & Schmitz examined propagation profiles of the hippocampal ripples along the longitudinal axis. In addition to the previously described correlation between the ripple strength and distance (Patel et al., 2013; Kumar et al., 2019), the authors revealed heterogeneous propagation patterns depending on the strength and the origin. Within the septal half of the hippocampus, 'strong' ripples (top 10% strength in a session) is more likely to propagate from the medial to the lateral while the other ripples move in the other direction. Interestingly, these strong ripples are unique in that they are generated locally and more in the medial part of the septal hippocampus. …
Reviewer #3 (Public Review):
Using a large Neuropixels dataset provided by the Allen Institute (https://allensdk.readthedocs.io/en/latest/visual_coding_neuropixels.html), Filippo & Schmitz examined propagation profiles of the hippocampal ripples along the longitudinal axis. In addition to the previously described correlation between the ripple strength and distance (Patel et al., 2013; Kumar et al., 2019), the authors revealed heterogeneous propagation patterns depending on the strength and the origin. Within the septal half of the hippocampus, 'strong' ripples (top 10% strength in a session) is more likely to propagate from the medial to the lateral while the other ripples move in the other direction. Interestingly, these strong ripples are unique in that they are generated locally and more in the medial part of the septal hippocampus. Finally, the authors found that more neurons, with higher firing rates, are engaged in the strong ripples generated in the medial part of the septal hippocampus.
The major strength of the present study is their finding of the unique propagation of the strong ripples across the longitudinal axis. Past studies examining ripple propagations did not have a particular focus on the strength of ripples and thus have not described this feature. On the other hand, however, I believe the manuscript would represent a higher significance if the authors provided more thoughts on physiological impacts and or particular roles of this unique propagation pattern. The authors propose 1) the integration of the different kinds of information and 2) the contribution of the septal hippocampus to higher memory demand (Lines 275-296). Although these views are interesting, the former only explains the longer propagation of the ripples but not the direction (i.e., the ripples could propagate from the lateral to the medial), and the latter idea is less convincing because the Neuropixels data is collected from the mice only passively receiving visual stimuli.
The propagation of the locally generated ripples across the septotemporal axis has been well described in past studies (Patel et al., 2013; Kumar & Deshmukh, 2019). The authors' findings about different directionalities of ripple propagation depending on the origin would provide a valuable view for the expert in the field of the hippocampal physiology.
-