Binary and analog variation of synapses between cortical pyramidal neurons
Curation statements for this article:-
Curated by eLife
Evaluation Summary:
Synaptic connections are crucial for determining neural circuit function and for storing adaptive changes in the brain. This study performs a highly detailed quantitative analysis of certain excitatory connections in mouse neocortex, and finds that the physical size of these connections has a bimodal distribution. This is an important finding that has implications for our understanding of synaptic plasticity and neural circuit dynamics.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #2 and Reviewer #3 agreed to share their names with the authors.)
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
- Neuroscience (eLife)
Abstract
Learning from experience depends at least in part on changes in neuronal connections. We present the largest map of connectivity to date between cortical neurons of a defined type (layer 2/3 [L2/3] pyramidal cells in mouse primary visual cortex), which was enabled by automated analysis of serial section electron microscopy images with improved handling of image defects (250 × 140 × 90 μm 3 volume). We used the map to identify constraints on the learning algorithms employed by the cortex. Previous cortical studies modeled a continuum of synapse sizes by a log-normal distribution. A continuum is consistent with most neural network models of learning, in which synaptic strength is a continuously graded analog variable. Here, we show that synapse size, when restricted to synapses between L2/3 pyramidal cells, is well modeled by the sum of a binary variable and an analog variable drawn from a log-normal distribution. Two synapses sharing the same presynaptic and postsynaptic cells are known to be correlated in size. We show that the binary variables of the two synapses are highly correlated, while the analog variables are not. Binary variation could be the outcome of a Hebbian or other synaptic plasticity rule depending on activity signals that are relatively uniform across neuronal arbors, while analog variation may be dominated by other influences such as spontaneous dynamical fluctuations. We discuss the implications for the longstanding hypothesis that activity-dependent plasticity switches synapses between bistable states.
Article activity feed
-
-
Evaluation Summary:
Synaptic connections are crucial for determining neural circuit function and for storing adaptive changes in the brain. This study performs a highly detailed quantitative analysis of certain excitatory connections in mouse neocortex, and finds that the physical size of these connections has a bimodal distribution. This is an important finding that has implications for our understanding of synaptic plasticity and neural circuit dynamics.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #2 and Reviewer #3 agreed to share their names with the authors.)
-
Reviewer #1 (Public Review):
Dorkenwald & colleagues perform serial EM to fully reconstruct the neurites and synaptic connections between several hundred L2/3 pyramidal cells in the neocortex of a mouse. They find that automated methods struggle with the large volume of tissue they sample and devise methods for augmenting training data for human-assisted convolutional network-based segmentation. The data and code are available in various forms on a dedicated website. The scientific findings of the study are that the distribution of synapse spine volumes is best fit by a two-component log-normal mixture model, and that synapse sizes are coarsely correlated when multiple shared connections are considered, with the binary state accounting for much of the shared variation. This is interesting because multimodality in synaptic strength is an …
Reviewer #1 (Public Review):
Dorkenwald & colleagues perform serial EM to fully reconstruct the neurites and synaptic connections between several hundred L2/3 pyramidal cells in the neocortex of a mouse. They find that automated methods struggle with the large volume of tissue they sample and devise methods for augmenting training data for human-assisted convolutional network-based segmentation. The data and code are available in various forms on a dedicated website. The scientific findings of the study are that the distribution of synapse spine volumes is best fit by a two-component log-normal mixture model, and that synapse sizes are coarsely correlated when multiple shared connections are considered, with the binary state accounting for much of the shared variation. This is interesting because multimodality in synaptic strength is an outcome that constrains models of synaptic learning and suggests additional structure in the connections that may form functional assemblies. The paper is well presented and well written and should be of interest to neurophysiologists, theoretical neuroscientists, cell biologists and neuroanatomists.
-
Reviewer #2 (Public Review):
The authors demonstrated the existence of binary and analogue forms of variabilities in dendritic spines using the double synaptic contacts between two neurons within the volume of 3D EM reconstructions (250*140*90um). The 160 double synaptic connections (EFig. 3) are stunning and exceptionally valuable, which provide a novel line of evidence for two modes of spine structural changes. The major weakness is the absence of behavioural manipulations to confirm the learning dependence of the results. However, the new ways of interpreting the EM data will be surprising to the neuroscience community and have a significant impact.
-
Reviewer #3 (Public Review):
The authors present a large scale study of connectomics in mouse primary visual cortex. Their study focuses on correlations among the spine head volumes of pairs of dual connections, and specifically those formed between Layer 2/3 pyramidal neurons. They define a "dual connection" as a pair of synapses connecting the same pre- and post-synaptic cells. They report on a novel and very interesting finding that the distribution of sizes among these pairs is bimodal can be fit by a mixture of a binary variable and a log-normal distribution. They draw the conclusion that the binary component could be due to synaptic plasticity rules and that the log-normal component is due to other factors. The authors also relate their novel findings to previous studies of correlations among dual connections.
The authors make the …
Reviewer #3 (Public Review):
The authors present a large scale study of connectomics in mouse primary visual cortex. Their study focuses on correlations among the spine head volumes of pairs of dual connections, and specifically those formed between Layer 2/3 pyramidal neurons. They define a "dual connection" as a pair of synapses connecting the same pre- and post-synaptic cells. They report on a novel and very interesting finding that the distribution of sizes among these pairs is bimodal can be fit by a mixture of a binary variable and a log-normal distribution. They draw the conclusion that the binary component could be due to synaptic plasticity rules and that the log-normal component is due to other factors. The authors also relate their novel findings to previous studies of correlations among dual connections.
The authors make the claim that the bimodality of the size distribution of "dual connections" is due to plasticity rules while the continuous gradation of the synapse size is dominated by "other influences" besides plasticity rules. But couldn't it be the case that precise synaptic plasticity rules are always at play and that the bimodality is due to the impact of "other influences" on these rules. These influences could be dendrite diameter, synapse density along the dendrite, distance from the soma, branch-point failure of propagation of the pre-synaptic AP or post-synaptic bAP, to name just a few. The continuous log-normal component of the gradation of synapse sizes is not at all inconsistent with the outcome of highly precise synaptic plasticity rules. Indeed, Bartol et al., 2015 showed that synaptic plasticity in rat CA1 stratum radiatum is as precise for small synapses as it is for large synapses over a broad, continuous, log-normal range of synapse sizes. Each individual synapse has its own unique activation history and a resulting, precise synaptic weight. And across the population of synapses there is a continuous distribution of activation histories.
The bimodality is a new and very interesting observation in this work and deserves much closer scrutiny to reveal the root cause behind it. The authors have shown that it exists in connections between layer 2/3 PyCs but is absent in other connections traced in their reconstruction of mouse neocortex. Perhaps this reveals something special about information processing in the neural subcircuit of layer 2/3 PyCs. With their highly detailed large-scale reconstruction, the authors have the opportunity to probe this question deeper.
-
-