Chronic Ca2+ imaging of cortical neurons with long-term expression of GCaMP-X

  1. Jinli Geng
  2. Yingjun Tang
  3. Zhen Yu
  4. Yunming Gao
  5. Wenxiang Li
  6. Yitong Lu
  7. Bo Wang
  8. Huiming Zhou
  9. Ping Li
  10. Nan Liu
  11. Ping Wang
  12. Yubo Fan
  13. Yaxiong Yang
  14. Zengcai V Guo
  15. Xiaodong Liu

  • Curated by eLife

    eLife logo

    Evaluation Summary:

    This paper addresses the toxicity of fluorescent calcium indicators, comparing two series of indicators (GCaMPs and GCaMP-Xs) in mouse neurons. The paper documents GCaMP toxicity during development and following prolonged strong expression, and establishes that GCaMP-X indicators are less toxic. The paper will be of interest primarily to neuroscientists who use fluorescence calcium indicators to monitor calcium dynamics during neuronal development.

    (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 agreed to share their name with the authors.)

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Dynamic Ca 2+ signals reflect acute changes in membrane excitability, and also mediate signaling cascades in chronic processes. In both cases, chronic Ca 2+ imaging is often desired, but challenged by the cytotoxicity intrinsic to calmodulin (CaM)-based GCaMP, a series of genetically-encoded Ca 2+ indicators that have been widely applied. Here, we demonstrate the performance of GCaMP-X in chronic Ca 2+ imaging of cortical neurons, where GCaMP-X by design is to eliminate the unwanted interactions between the conventional GCaMP and endogenous (apo)CaM-binding proteins. By expressing in adult mice at high levels over an extended time frame, GCaMP-X showed less damage and improved performance in two-photon imaging of sensory (whisker-deflection) responses or spontaneous Ca 2+ fluctuations, in comparison with GCaMP. Chronic Ca 2+ imaging of one month or longer was conducted for cultured cortical neurons expressing GCaMP-X, unveiling that spontaneous/local Ca 2+ transients progressively developed into autonomous/global Ca 2+ oscillations. Along with the morphological indices of neurite length and soma size, the major metrics of oscillatory Ca 2+ , including rate, amplitude and synchrony were also examined. Dysregulations of both neuritogenesis and Ca 2+ oscillations became discernible around 2–3 weeks after virus injection or drug induction to express GCaMP in newborn or mature neurons, which were exacerbated by stronger or prolonged expression of GCaMP. In contrast, neurons expressing GCaMP-X were significantly less damaged or perturbed, altogether highlighting the unique importance of oscillatory Ca 2+ to neural development and neuronal health. In summary, GCaMP-X provides a viable solution for Ca 2+ imaging applications involving long-time and/or high-level expression of Ca 2+ probes.

Article activity feed

  1. Version published to 10.7554/elife.76691 on eLife
  2. eLife

    Author Response

    Reviewer #1 (Public Review):

    GCaMP indicators have become common, almost ubiquitous tools used by many neuroscientists. As calcium buffers, calcium indicators have the potential to perturb calcium dynamics and thereby alter neuronal physiology. With so many labs using GCaMPs across a variety of applications and brain regions, it's remarkable how few have documented GCaMP-related perturbations of physiology, but there are two main contexts in which perturbations have been observed: after prolonged expression of a high GCaMP concentration (common several weeks after infection with a virus using a strong promoter); and when cytoplasmic GCaMP is present during neuronal development. As a result, GCaMP studies are often designed to avoid these two conditions.

    Here, Xiaodong Liu and colleagues ask whether GCaMP-X series indicators are less toxic that GCaMPs. GCaMP-X indicators are modified GCaMPs with an additional N-terminal calmodulin binding domain that reduces interactions of the calmodulin moiety of GCaMP with other cellular proteins. Xiaodong Liu and colleagues document effects of GCaMP expression on neuronal morphology in vitro, calcium oscillations in vitro, and sensory responses in vivo, in each case showing that GCaMP-X indicators are less toxic. Their results are compelling.

    Unfortunately, the paper suffers two main weaknesses. Firstly, the results demonstrate that GCaMP is toxic during development, after prolonged expression via viruses in vivo, and in cell culture where maturation of the culture likely recapitulates key steps in development. GCaMPs are known to be toxic in these circumstances, such toxicity is readily circumvented by driving expression in the adult, and there are countless examples of studies in which adequate GCaMP expression was achieved without toxicity. These new results are of little relevance to the majority of GCaMP experiments. That GCaMP-X indicators are less toxic during development is a new result and may be of interest to those who wish to deploy calcium indicators during development, but this is a relatively small number of neuroscientists.

    We thank the reviewer for providing valuable opinions on these critical matters. Here, we would like to clarify:

    1. In our work, the status of neurites (length, branching, etc.) is indeed one main aspect to monitor, and neuritogenesis during the early stages of development is known to have temporal trajectories with ample dynamic range thus helpful to quantitatively compare GCaMP-X versus GCaMP. However, the key factor is the actual time and level of probe expression in neurons, and the starting timepoint of expression could vary. We have conducted additional experiments using virus-infected neurons (Figure 5—figure supplement 1) and transgenic neurons with inducible expression (Figure 7—figure supplement 3), both starting to express the probes at the mature stage. Thus, GCaMP-X imaging is not necessarily limited to developing neurons. As in the original reports of GCaMP probes with toxicity, virus injection was performed for both immature (2-3 weeks, Tian 2009 PMID: 19898485) and mature mice (~2 months, Chen 2013 PMID: 23868258). According to the protocol (Huber 2012 PMID: 22538608), GCaMP virus injection was done for adult mice (>2 months), which exhibited functional and morphological deficits in nucleus-filled neurons beyond OTW (Figure 2, Figure 5 and Figure 6). Collectively, the central principles of GCaMP-X versus GCaMP are applicable to both immature and mature neurons.

    2. Chronic GCaMP-X imaging has a broad spectrum of potential applications, not limited to neural development (Resendez 2016 PMID: 26914316). As mentioned, GCaMP-X resolves the problem of longitudinal expression thus making chronic imaging more feasible. We agree with the reviewer that a large body of our data in the original version focused on the characteristics of calcium signals during the early stage of neuronal development, which served as an exemplary scenario to compare GCaMP-X with GCaMP. Indeed, the importance of Ca2+ oscillation in neural development is commonly accepted (Kamijo 2018 PMID: 29773754; Gomez 2006 PMID: 16429121). In vivo Ca2+ imaging (Figure 2 and Figure 5) and morphological analyses (e.g., Figure 6) have extended the major conclusions onto mature neurons where dysregulations of Ca2+ oscillations are also tightly coupled with neuronal health or death/damage. Importantly, GCaMP-X paves the way to unexplored directions previously impeded or discouraged due to GCaMP perturbations, e.g., chronic imaging of cultured neurons to concurrently monitor Ca2+ activities and cell morphology as in this study.

    3. To circumvent the toxicity of GCaMP is not a trivial procedure for viral infection. The expression levels need to be carefully adjusted experimentally, e.g., by dilution studies (Resendez 2016 PMID: 26914316). A delicate balance of GCaMP expression is critical: low level (or short time) of expression would result in weak signals and poor SNR whereas high level (or long time) of expression would cause nuclear filling and neural toxicity. Even for the work-around conditions of time window and dilution dosage, nucleus-filled neurons are not uncommon judged by the expression/fluorescence patterns, e.g., in the original reports of GCaMP6 (Supplementary Figure 7, Chen 2013 PMID: 23868258), and GCaMP3 (Supplementary Figure 11, Tian 2009 PMID: 19898485). Under particular conditions (subtypes of neurons, time window of imaging, dosage of virus injection, etc.), many neurons could be found without apparent perturbation/nuclear-filling to proceed with calcium imaging. Using GCaMP-X, dosage is less restricted (10fold higher concentration for GCaMP-X with improved SNR and overall performance in Figure 2, Figure 5 and Figure 6). Practically, GCaMP-X is a simple solution for the issues related to excessive/prolonged expression. Also, GCaMP-X is expected to help maintain the total number of healthy neurons and thus the general health of the brain. Reportedly, some GCaMP lines of transgenic mice exhibit epileptic activities (Steinmetz 2017 PMID: 28932809), awaiting future studies to explore whether GCaMP-X could help.

    4. As the reviewer pointed out, the key of GCaMP-X is to resolve the unwanted (apo)GCaMP binding to endogenous proteins in neurons. We agree with the reviewer that according to the empirical observations the following factors appear to increase the severity of GCaMP perturbations: prolonged time, high concentration and nuclear accumulation. GCaMP-X is able to protect GCaMP from unwanted binding and the consequent damage to neurons, validated by various tests thus far (in vitro and in vivo). In this context, the prolonged time would result in higher GCaMP concentration, meanwhile accumulating the effects due to GCaMP interactions; higher GCaMP concentration would interfere with more binding events and targets of endogenous CaM; and enhanced/prolonged expression of GCaMP is directly correlated with nuclear accumulation, a hallmark of neuronal damage.

    Secondly, the authors extend their claims to conclude that GCaMP indicators are toxic under other circumstances, claims supported by neither their results nor the literature. To provide one example, at the end of the introduction is the statement, 'chronic GCaMP-X imaging has been successfully implemented in vitro and in vivo, featured with long-term overexpression (free of CaM-interference), high spatiotemporal contents (multiple weeks and intact neuronal network) and subcellular resolution (cytosolic versus nuclear), all of which are nearly infeasible if using conventional GCaMP.' The statement's inaccurate: there are many chronic imaging studies in vitro and in vivo using GCaMP indicators without nuclear accumulation of GCaMP or perturbed sensory responses. There are more examples throughout the paper where the conclusions overreach the results and are inaccurate. The results are simply insufficient to support many of the strong statements in the paper.

    Overall, the critics and suggestions of the reviewer have been well taken and we have revised the text accordingly. For this particular paragraph here mentioned by the reviewer, we want to clarify that it was the summary of our results in the whole manuscript, where each claim referred to the data and analyses shown in corresponding figures. In details, these figures were: 'free of CaM-interference (Figure 1), multiple weeks and intact neuronal network (in vitro: Figure 3 and Figure 4; in vivo: Figure 2, Figure 5 and Figure 6; transgenic neurons: Figure 7) and cytosolic versus nuclear (Figure 1 and the previous Figure 8). The last sentence of 'all of which are nearly infeasible if using conventional GCaMP' was meant to summarize the results comparing GCaMP versus GCaMP-X in our experimental settings of chronic imaging with prolonged/excessive probe expression. Again, we agree that for particular experimental settings and purposes the toxicity of GCaMP can be circumvented empirically. To avoid miscommunications, we have revised this paragraph by moving it to the Discussion (after all the data), also ensuring that the statements on GCaMP are backed up with data or literature. Please also see Essential Revisions, Item 3.

    Reviewer #2 (Public Review):

    Geng and colleagues provide further evidence for the lower neuronal toxicity of their improved GECI, GCaMP-X, which allows improved recordings of Ca2+ signals in neurons. As reported previously and studied in more detail here, the improved properties are primarily due to a lower tendency of GCaMP-Xc (reporting cytosolic Ca2+) to enter the nucleus. They present a systematic comparison of their cytosolic or nucleus-targeted GCamP-Xc (and Xn) with the corresponding "conventional" GCaMPs (jGCaMP7b, GCaMP6m). They, again, confirm the absence of apoGCaMP-X binding to the CaM binding domain of Cav1.3 L L-type Ca2+ channels suggesting that this is the main or one of several GCaMP interactions leading to altered intracellular signaling affecting neuronal survival, development and architecture. Evidence for more (likely) physiological Ca2+ responses were obtained from a battery of experiments, including in vivo recordings of acute sensory responses after viral expression of GCaMPs, monitoring of long-term calcium oscillations in cultured neurons, correlations measured Ca2+ oscillations with hallmarks of neuronal development (soma size, neurite outgrowth/arborizations, and long-term recordings of spontaneous Ca2+ activities in vivo in S1 primary somatosensory cortex. The latter experiments also showed that much higher doses of AAV-GCaMP6m-Xc could be administered than of GCaMP6m. They also show that unfavorable effects of GCaMPs on neurons of adult GCaMP expressing transgenic mice, both in in slices and cultured neurons. While most experiments aim at demonstrating improved performance of GCaMP-X, one finding also provides potential novel insight into the role of neuronal activity patterns during neuronal development in culture. Assuming more undisturbed physiological Ca2+ signaling even through longer time periods they can follow different Ca2+ activity patterns during neuronal development. Oscillation amplitudes and the level of synchrony correlated with neurite length and frequency inversely correlated with neurite outgrowth.

    They provide convincing experimental evidence for the improvements claimed for their novel GCamP-X constructs. Some aspects should be clarified.

    A key finding explaining the construct differences is the nuclear localization. The authors should also provide numbers for the N/C ratio for Ca2+ imaging of sensoryevoked responses in vivo (Fig. 2; pg 6: nuclear accumulation was barely noticeable from GCaMP6m-Xc even beyond OTW). Also, for chronic experiments in brain slices they state for GCaMP6m-Xc in the text that (pg 12) "meanwhile the N/C ratio remained ultra-low", yet Fig. 6 shows a N/C ratio of 0.2. This does not appear to be "ultra low".

    We appreciate the reviewer for bringing up the matter of N/C ratio (indicative of nuclear accumulation). We have appended the values of N/C ratio for in vivo experiments (revised Figure 2). Following the previous report, the criteria of N/C ratio was set to 0.8 to regroup the neurons into two subpopulations. A significant fraction of GCaMP neurons were nucleus-filled (N/C ratio>0.8); meanwhile, nearly no neuron expressing GCaMP-XC was found with N/C ratio greater than 0.8 when examined 8-13 weeks post injection. Generally, due to imaging resolution, confocal microscopy provided more precise evaluation for N/C ratio than two-photon in vivo images. In Figure 6, even more clear difference in nuclear distribution was observed between GCaMP and GCaMP-X, which was described as “ultralow” (GCaMP-X). Of note, the N/C ratio of YFP itself was ~1.3. The N/C ratio for GCaMP-XC was not close to zero, consistent with the measurements from other NES-tagged peptides (Yang 2022 PMID: 35589958). GCaMP-XC was not completely excluded from cell nuclei, thus producing some fluorescence there. In light of this comment, we have revised the relevant text including the phrase of “ultralow” (Page 14, Line 393). In addition, Figure 5 was also revised accordingly.

    Along these lines, since nuclear-filled neurons were observed in their experiments with GCaMP-Xc, the authors should comment if altered Ca2+ signals were also seen for the few neurons expressing GCaMP-Xc in the nucleus.

    During 2-photon imaging experiments in vivo, occasionally GCaMP-XC neurons appeared to have some level of nuclear expression especially in those blurred images of low quality. Judged by the criteria of N/C ratio (0.8), these neurons rarely fell into the nucleus-filled group (Figure 2B and Figure 5C, also see confocal imaging Figure 1B). On the other hand, a small fraction of GCaMP-XC could be “leaked” into the nucleus. GCaMP-XN also eliminated toxic (apo)GCaMP interactions in neurons, sharing the same design principle with GCaMP-XC (Figure 1). Therefore, nuclear GCaMP-XC is expected to resemble GCaMP-XN. Experimentally, with GCaMP-XC or GCaMP-XN present in the nucleus, no significant change in neuronal Ca2+ or neurite morphology has been observed. Meanwhile, this comment has pointed out one important direction of future research, i.e., to more precisely confine GCaMP-X within the targeted organelles, e.g., by improving or replacing localization tags.

    Since they performed a systematic comparison of two constructs to demonstrate an (expected) superiority of one of them, the experiments, or at least the analysis, should ideally be performed in a blinded way. The authors should clarify how they avoided experimental bias.

    For in vitro experiments, multiple independent trials of experiments with analyses were performed by two (or more) researchers to ensure the reproducibility and to minimize any bias. And the results and conclusions have been highly consistent (among different trials/researchers). Following the suggestion, we have assured that in vivo experiments and data analyses were separately conducted by the researchers from two different labs. For long-term expression/imaging, the differences between GCaMP-X and GCaMP were often discernable directly in the images even without further calculations or statistics (e.g., Figure 3B). Related information can be found in the Methods (Page 32, Line 799).

    In their chronic Ca2+ fluorescence imaging for autonomous Ca2+ oscillations in cultured cortical neurons ultralong lasting signals (Fig. 3B, DIV 17, GCaMP6m) could be observed. It would be helpful to further describe the nature of these transients, ideally by adding it to their video collection.

    As suggested by the reviewer, the video for Figure 3B (DIV 17, GCaMP6m) has been included in this revision (Figure 3—video supplement 2). In contrast to the oscillatory signals normally observed from healthy neurons, the pronounced and sustained Ca2+ signals are associated with apoptosis and other pathological conditions in neurons (Khan 2020 PMID: 32989314; Nicotera 1998 PMID: 9601613; Harr 2010 PMID: 20826549). The Ca2+ wave with broadened width (FWHM) was indicative of damaged neurons by GCaMP (Figure 3F), rather than (altered) sensing characteristics of GCaMP. We agree that this observation is a notable and interesting phenomenon, worth to follow up in future studies.

    The discussion is very long. In my opinion it would benefit from shortening, avoid redundancies and focus only on the key findings in this paper. This includes the chapter on design and application guidelines for CaM-based GECIs. The main message what the advantage of their GCaMP-X modifications has been made before in the discussion. A more detailed discussion on this appears more suitable in a review article.

    In response to this suggestion, we have made it as concise as possible, by simplifying or removing several topics including the design and application guidelines for CaMbased GECIs.

    It may be worthwhile to include another aspect in the discussion: does the improved GCaMP-Xc cause no change in neuronal function or morphology or is it just less damaging than other GCaMPs. How can this issue be addressed experimentally.

    We have revised the discussion accordingly (Page 21, Line 588). We agree that additional experiments would help evaluate how close GCaMP-X data are to the reality, considering the Ca2+-buffering effect intrinsic to Ca2+ probes and also other factors. In light of this suggestion and also those from Reviewer #1, we have incorporated more experimental controls, including Ai140 mice (GFP, Figure 7—figure supplement 2) and Fluo-4 AM (Ca2+ dye, Figure 3—figure supplement 4). The results have been encouraging in that GCaMP-X neurons were nearly indistinguishable in the morphological and functional aspects from GFP or Fluo-4 AM controls. The incoming feedbacks from GCaMP-X users should continue to help clarify this matter, which we would like to follow up.

  3. eLife

    Evaluation Summary:

    This paper addresses the toxicity of fluorescent calcium indicators, comparing two series of indicators (GCaMPs and GCaMP-Xs) in mouse neurons. The paper documents GCaMP toxicity during development and following prolonged strong expression, and establishes that GCaMP-X indicators are less toxic. The paper will be of interest primarily to neuroscientists who use fluorescence calcium indicators to monitor calcium dynamics during neuronal development.

    (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 agreed to share their name with the authors.)

  4. eLife

    Reviewer #1 (Public Review):

    GCaMP indicators have become common, almost ubiquitous tools used by many neuroscientists. As calcium buffers, calcium indicators have the potential to perturb calcium dynamics and thereby alter neuronal physiology. With so many labs using GCaMPs across a variety of applications and brain regions, it's remarkable how few have documented GCaMP-related perturbations of physiology, but there are two main contexts in which perturbations have been observed: after prolonged expression of a high GCaMP concentration (common several weeks after infection with a virus using a strong promoter); and when cytoplasmic GCaMP is present during neuronal development. As a result, GCaMP studies are often designed to avoid these two conditions.

    Here, Xiaodong Liu and colleagues ask whether GCaMP-X series indicators are less toxic that GCaMPs. GCaMP-X indicators are modified GCaMPs with an additional N-terminal calmodulin binding domain that reduces interactions of the calmodulin moiety of GCaMP with other cellular proteins. Xiaodong Liu and colleagues document effects of GCaMP expression on neuronal morphology in vitro, calcium oscillations in vitro, and sensory responses in vivo, in each case showing that GCaMP-X indicators are less toxic. Their results are compelling.

    Unfortunately, the paper suffers two main weaknesses. Firstly, the results demonstrate that GCaMP is toxic during development, after prolonged expression via viruses in vivo, and in cell culture where maturation of the culture likely recapitulates key steps in development. GCaMPs are known to be toxic in these circumstances, such toxicity is readily circumvented by driving expression in the adult, and there are countless examples of studies in which adequate GCaMP expression was achieved without toxicity. These new results are of little relevance to the majority of GCaMP experiments. That GCaMP-X indicators are less toxic during development is a new result and may be of interest to those who wish to deploy calcium indicators during development, but this is a relatively small number of neuroscientists.

    Secondly, the authors extend their claims to conclude that GCaMP indicators are toxic under other circumstances, claims supported by neither their results nor the literature. To provide one example, at the end of the introduction is the statement, 'chronic GCaMP-X imaging has been successfully implemented in vitro and in vivo, featured with long-term overexpression (free of CaM-interference), high spatiotemporal contents (multiple weeks and intact neuronal network) and subcellular resolution (cytosolic versus nuclear), all of which are nearly infeasible if using conventional GCaMP.' The statement's inaccurate: there are many chronic imaging studies in vitro and in vivo using GCaMP indicators without nuclear accumulation of GCaMP or perturbed sensory responses. There are more examples throughout the paper where the conclusions overreach the results and are inaccurate. The results are simply insufficient to support many of the strong statements in the paper.

  5. eLife

    Reviewer #2 (Public Review):

    Geng and colleagues provide further evidence for the lower neuronal toxicity of their improved GECI, GCaMP-X, which allows improved recordings of Ca2+ signals in neurons. As reported previously and studied in more detail here, the improved properties are primarily due to a lower tendency of GCaMP-Xc (reporting cytosolic Ca2+) to enter the nucleus. They present a systematic comparison of their cytosolic or nucleus-targeted GCamP-Xc (and Xn) with the corresponding "conventional" GCaMPs (jGCaMP7b, GCaMP6m). They, again, confirm the absence of apo-GCaMP-X binding to the CaM binding domain of Cav1.3L L-type Ca2+ channels suggesting that this is the main or one of several GCaMP interactions leading to altered intracellular signaling affecting neuronal survival, development and architecture. Evidence for more (likely) physiological Ca2+ responses were obtained from a battery of experiments, including in vivo recordings of acute sensory responses after viral expression of GCaMPs, monitoring of long-term calcium oscillations in cultured neurons, correlations measured Ca2+ oscillations with hallmarks of neuronal development (soma size, neurite aoutgrowth/arborizations, and long-term recordings of spontaneous Ca2+ activities in vivo in S1 primary somatosensory cortex. The latter experiments also showed that much higher doses of AAV-GCaMP6m-Xc could be administered than of GCaMP6m. They also show that unfavorable effects of GCaMPs on neurons of adult GCaMP-expressing transgenic mice, both in in slices and cultured neurons. While most experiments aim at demonstrating improved performance of GCaMP-X, one finding also provides potential novel insight into the role of neuronal activity patterns during neuronal development in culture. Assuming more undisturbed physiological Ca2+ signaling even through longer time periods they can follow different Ca2+ activity patterns during neuronal development. Oscillation amplitudes and the level of synchrony correlated with neurite length and frequency inversely correlated with neurite outgrowth.

    They provide convincing experimental evidence for the improvements claimed for their novel GCamP-X constructs. Some aspects should be clarified.

    A key finding explaining the construct differences is the nuclear localization. The authors should also provide numbers for the N/C ratio for Ca2+ imaging of sensory-evoked responses in vivo (Fig. 2; pg 6: nuclear accumulation was barely noticeable from GCaMP6m-Xc even beyond OTW). Also, for chronic experiments in brain slices they state for GCaMP6m-Xc in the text that (pg 12) "meanwhile the N/C ratio remained ultra-low", yet Fig. 6 shows a N/C ratio of 0.2. This does not appear to be "ultra low".

    Along these lines, since nuclear-filled neurons were observed in their experiments with GCaMP-Xc, the authors should comment if altered Ca2+ singals were also seen for the few neurons expressing GCaMP-Xc in the nucleus.

    Since they performed a systematic comparison of two constructs to demonstrate an (expected) superiority of one of them, the experiments, or at least the analysis, should ideally be performed in a blinded way. The authors should clarify how they avoided experimental bias.

    In their chronic Ca2+ fluorescence imaging for autonomous Ca2+ oscillations in cultured cortical neurons ultralong lasting signals (Fig. 3B, DIV 17, GCaMP6m) could be observed. It would be helpful to further describe the nature of these transients, ideally by adding it to their video collection.

    The discussion is very long. In my opinion it would benefit from shortening, avoid redundancies and focus only on the key findings in this paper. This includes the chapter on design and application guidelines for CaM-based GECIs. The main message what the advantage of their GCaMP-X modifications has been made before in the discussion. A more detailed discussion on this appears more suitable in a review article.

    It may be worthwhile to include another aspect in the discussion: does the improved GCaMP-Xc cause no change in neuronal function or morphology or is it just less damaging than other GCaMPs. How can this issue be addressed experimentally.

  6. Version published to 10.1101/2022.01.09.475579v1 on bioRxiv