PREreview of Neuronal activity inhibits axonal mitochondrial transport in a region-specific manner

In this study Tom and Vanden Berghe provide a pipeline to quantify mitochondrial axonal transport and simultaneously measure Calcium (Ca) concentration fluctuations with high spatial resolution during physiologic neuronal stimulations. Previously it has been proposed that Ca levels regulate mitochondrial transport. The authors add interesting data showing that mitochondrial transport is activity-dependent in axons with functional synapses but the spatial differences on mitochondrial transport along the axon might not correlate with spatially differential Ca levels.

Positive aspects of the study: 

• The authors provide a strong framework to simultaneously record mitochondrial transport and calcium dynamics with high spatial resolution in single axons of primary hippocampal neurons.

• The authors are able to reproduce previous work showing how strong and sustained (KCl) depolarization inhibits mitochondrial axonal transport.  

• The authors provide data proving that mitochondrial transport along axons is activity-dependent in connecting segments under physiologically relevant stimulations, and suggest that it does not depend on Ca levels, which has never been suggested before.

Major comments:

• The authors conclude “In contrast to KCl-induced depolarization (fig. 2), activity-induced calcium elevations did not inhibit axonal mitochondrial transport”. But later in this study they do observe inhibited axonal mitochondrial transport in connecting axons. We would suggest to clarify that no inhibition of transport was observed in non-connecting axons.

• We believe that the paper can gain in clarity if some figures are reorganized. We would suggest to:

  • Switch the order of figure 1 and 2 to present first the reproduced results or combine them in one figure.

  • Combine figures 3, 4 and 5 in one small figure consisting of a schematic on how the method works and its main readouts plus one example for the readouts (e.g. comparing calibration value and example) and include the rest of the panels in the supplementary figures section as they provide method optimization information rather than further data on the role of calcium levels on mitochondrial transport. In particular, we would suggest reducing the number of panels of figure 5. A graphical abstract or scheme of the segmentation pipeline in combination with a few exemplary images would make this easier to grasp. Reducing the number of  images would make the flow of information easier to follow.

  • Include figure 7 in the supplementary figures section, as it helps to strengthen the author's hypothesis but doesn't provide direct data on influence of Ca levels on mitochondrial transport along microtubules.

  • Skip figure 8. We consider that the data on Ca responses in dendritic segments is not relevant unless mitochondrial transport is also assessed in this neuronal compartment.

• It is known that cell density influences maturation of neuronal cultures in vitro and neuronal culture activity. Although the authors name the advantages of using sparse cultures for their approach, they should also assess the maturation state of their (sparse and mixed) cultures not only by post-hoc IHC but also functionally and show the capacity to generate spontaneous activity. We would recommend comparing spontaneous activity after 7 DIV and 14 DIV.  If they are not spontaneously active, they should comment on the functional limitations for the study.

• We would like to encourage the authors to discuss in depth the physiological relevance of investigating activity-dependent mitochondrial axonal transport in non-connecting axons (no functional synapses). Why are these experiments important and what can we learn from them?

• It is not clear how the analysis of the spatial pattern is achieved. Is mitochondrial transport measured within the same axon in regions lacking functional synapses and in regions containing synapses?

• We would like the authors to follow on the discussion on mitochondrial transport along axons being activity-dependent in connecting segments under physiologically relevant stimulations, but not dependent on Ca levels. We find this a very exciting potential contribution of this study and have some suggestions on how to expand on it: 

  • It would help to have a figure where the extent of calcium responses is compared side-by-side between non-connecting and connecting axons

  • Is mitochondrial transport affected to the same extent along the axon? We would appreciate it if the authors would investigate and show if the position of halted mitochondria spatially overlaps with localization of synapses along the axon. Is the transport unaffected in functionally connected axons but within axonal regions lacking synaptic contacts?  

  • It would be interesting to show the differential localization of syntaphilin in sparse vs connected cultures. This could potentially provide more mechanistic insights on the differences observed between sparse and connected cultures.

Minor comments:

• We find that the subheadings within the result section could describe better the results explained in the following subsection. We would suggest reformulating them:

  • The header of the first subsection is confusing as the authors show later on how neuronal activity indeed can regulate mitochondrial transport in functional axons. 

  • In the second subsection, the method is center work. Perhaps one could include the development of the quantitative method in the title.

• We consider that Figure 1A can gain some clarity on illustrating how simultaneous Ca imaging and mitochondria tracking was achieved

• The authors explain that “KCl-induced depolarization has been used previously to demonstrate the calcium-dependent inhibition of mitochondrial transport”. A reference should be included to support this statement.

• We would appreciate a discussion about the choice of specific activity patterns as "physiological conditions". We would like to encourage the authors to give a reason why these stimulation frequencies were chosen. Also, including supporting references would help.

• In figure 1 and 2, the temporal scale should be included in the kymographs. In figure 1, 3, 5 and 6, the scale bars of images are missing.

• When introducing fura-2 measurements, elaborate on the origin of the 2 different emission signals and their correlation, how does this dye work?

• In the second method subsection, we would recommend to elaborate more on why using fura-2 signal helps on defining a mask of active neurons

• In figure 6, it is difficult to visualize the axons that are indicated by arrows. To more clearly depict this negative synapsin 1 staining, we would suggest to show it hand to hand with a positive example and use zoomed in images to better appreciate the punctate localization of synapsis. In panel b, to which staining is the white signal referring to? In panels h, i and j, what experimental condition is “control” referring to? No stimulation or non-connecting axonal regions?

Comments on reporting:

• We consider that some formal aspects of the manuscript could be improved to adjust to the conventional guidelines of original research articles. We would recommend the authors to:

  • Avoid repeating exact same sentences in the abstract and other sections of the article.

  • Focus the introduction on the relevant information necessary to understand the following experiments. Avoid going into detail on mechanisms (ex: molecular motors that transport mitochondrial along microtubules) that will not be addressed in the results and the discussion.

  • Use the last paragraph of the introduction to state not only the conclusions of your research but first a brief sentence of what they did to arrive at these conclusions and summarize how they studied the connection (triage) between neuronal activity, Ca2+ and mito redistribution.

  • Important information to understand the experiment should be in the main results text, avoid placing it in the figure legend. Use the figure legend only to describe in detail the content of the figures.

  • We recommend using same color-scheme across the figures

  • In the figure legends, we suggest that while describing images, include in brackets the color used for each staining and highlight the axonal segment with the ROI that has been used to obtain the kymograph

  • A thorough description of the analysis pipeline (ex: image analysis pipeline and mathematical formulas applied) is best included in the methods section rather than in the results

  • Avoid commenting on possible artifacts on the last paragraph of the discussion, especially if they refer to artifacts that don’t directly affect relevant measurements of this study (ex: cytosolic Ca measurements). If considered necessary, we would recommend to address the topic earlier

• In the figure legends, authors should report not only in the number of repetitions of each experiment, but also in the number of different animal that has been used

Suggestions for future studies:

• We would like to suggest to explore automated/unbiased ways to quantify the mitochondrial trajectories

• It would be interesting to show whether mito distribution/dynamics is altered by calcium fluctuations induced by different mechanisms than (enforced) neuronal activity. e.g. altering  calcium buffering at interorganelle contact sites. This would specify the regulatory effect of neuronal activity induced mito redistribution.

Conflicts of interest of reviewers:

• None declared

Competing interests

The authors declare that they have no competing interests.