Author response:
The following is the authors’ response to the original reviews
eLife Assessment
This study presents potentially valuable insights into the role of climbing fibers in cerebellar learning. The main claim is that climbing fiber activity is necessary for optokinetic reflex adaptation, but is dispensable for its long-term consolidation. There is evidence to support the first part of this claim, though it requires a clearer demonstration of the penetrance and selectivity of the manipulation. However, support for the latter part of the claim is incomplete owing to methodological concerns, including unclear efficacy of longer-duration climbing fiber activity suppression.
We sincerely appreciate the thoughtful feedback provided by the reviewer regarding our study on the role of climbing fibers in cerebellar learning. Each point raised has been carefully considered, and we are committed to addressing them comprehensively. We acknowledge the importance of addressing methodological concerns, particularly regarding the efficacy of long-term suppression of CF activity, as well as ensuring clarity regarding the penetrance and selectivity of our manipulation. To this end, we have outlined plans for substantial revisions to the manuscript to adequately address these issues.
Public Reviews:
Reviewer #1 (Public Review):
Summary:
The study by Seo et al highlights knowledge gaps regarding the role of cerebellar complex spike (CS) activity during different phases of learning related to optokinetic reflex (OKR) in mice. The novelty of the approach is twofold: first, specifically perturbing the activity of climbing fibers (CFs) in the flocculus (as opposed to disrupting communication between the inferior olive (IO) and its cerebellar targets globally); and second, examining whether disruption of the CS activity during the putative "consolidation phase" following training affects OKR performance.
The first part of the results provides adequate evidence supporting the notion that optogenetic disruption of normal CF-Purkinje neuron (PN) signaling results in the degradation of OKR performance. As no effects are seen in OKR performance in animals subjected to optogenetic irradiation during the memory consolidation or retrieval phases, the authors conclude that CF function is not essential beyond memory acquisition. However, the manuscript does not provide a sufficiently solid demonstration that their longterm activity manipulation of CF activity is effective, thus undermining the confidence of the conclusions.
Strengths:
The main strength of the work is the aim to examine the specific involvement of the CF activity in the flocculus during distinct phases of learning. This is a challenging goal, due to the technical challenges related to the anatomical location of the flocculus as well as the IO. These obstacles are counterbalanced by the use of a well-established and easy-to-analyse behavioral model (OKR), that can lead to fundamental insights regarding the long-term cerebellar learning process.
Weaknesses:
The impact of the work is diminshed by several methodological shortcomings.
Most importantly, the key finding that prolonged optogenetic inhibition of CFs (for 30 min to 6 hours after the training period) must be complemented by the demonstration that the manipulation maintains its efficacy. In its current form, the authors only show inhibition by short-term optogenetic irradiation in the context of electrical-stimulation-evoked CSs in an ex vivo preparation. As the inhibitory effect of even the eNpHR3.0 is greatly diminished during seconds-long stimulations (especially when using the yellow laser as is done in this work (see Zhang, Chuanqiang, et al. "Optimized photo-stimulation of halorhodopsin for long-term neuronal inhibition." BMC biology 17.1 (2019): 1-17. ), we remain skeptical of the extent of inhibition during the long manipulations. In short, without a demonstration of effective inhibition throughout the putative consolidation phase (for example by showing a significant decrease in CS frequency throughout the irradiation period), the main claim of the manuscript of phase-specific involvement of CF activity in OKR learning cannot be considered to be based on evidence.
Second, the choice of viral targeting strategy leaves gaps in the argument for CF-specific mechanisms. CaMKII promoters are not selective for the IO neurons, and even the most precise viral injections always lead to the transfection of neurons in the surrounding brainstem, many of which project to the cerebellar cortex in the form of mossy fibers (MF). Figure 1Bii shows sparsely-labelled CFs in the flocculus, but possibly also MFs. While obtaining homogenous and strong labeling in all floccular CFs might be impossible, at the very least the authors should demonstrate that their optogenetic manipulation does not affect simple spiking in PNs.
Finally, while the paper explicitly focuses on the effects of CF-evoked complex spikes in the PNs and not, for example, on those mediated by molecular layer interneurons or via direct interaction of the CF with vestibular nuclear neurons, it would be best if these other dimensions of CF involvement in cerebellar learning were candidly discussed.
We appreciate the reviewer’s thorough evaluation, which thoughtfully highlights the strengths and areas for improvement in our study.
We agree with the reviewer’s recognition of the novelty of our approach, particularly in specifically perturbing climbing fiber (CF) activity in the flocculus and examining its effects across distinct phases of learning. Additionally, our use of the well-established OKR behavior paradigm provides a robust framework for investigating cerebellar learning processes, further strengthening our study.
To address concerns regarding the efficacy of long-term optogenetic inhibition and the specificity of viral targeting, we conducted additional experiments. These include in vivo monitoring of CF activity during the irradiation period, confirming sustained inhibition of complex spikes throughout the consolidation phase. To ensure precise targeting and mitigate potential side effects, such as unintended modification of Purkinje cell (PC) simple spike activity, we demonstrated that optogenetic suppression of CF transmission did not affect simple spike firing. Furthermore, we made additional characterizations to confirm the specificity of viral targeting.
Lastly, we recognize the importance of exploring alternative mechanisms underlying CF involvement in cerebellar learning. Accordingly, we expanded the manuscript to provide a more comprehensive discussion of these mechanisms, offering a clearer perspective on the broader implications of our findings.
Reviewer #2 (Public Review):
Summary:
The authors aimed to explore the role of climbing fibers (CFs) in cerebellar learning, with a focus on optokinetic reflex (OKR) adaptation. Their goal was to understand how CF activity influences memory acquisition, memory consolidation, and memory retrieval by optogenetically suppressing CF inputs at various stages of the learning process.
Strengths:
The study addresses a significant question in the cerebellar field by focusing on the specific role of CFs in adaptive learning. The authors use optogenetic tools to manipulate CF activity. This provides a direct method to test the causal relationship between CF activity and learning outcomes.
Weaknesses:
Despite shedding light on the potential role of CFs in cerebellar learning, the study is hampered by significant methodological issues that question the validity of its conclusions. The absence of detailed evidence on the effectiveness of CF suppression and concerns over tissue damage from optogenetic stimulation weakens the argument that CFs are not essential for memory consolidation. These challenges make it difficult to confirm whether the study's objectives were fully met or if the findings conclusively support the authors' claims. The research commendably attempts to unravel the temporal involvement of CFs in learning but also underscores the difficulties in pinpointing specific neural mechanisms that underlie the phases of learning. Addressing these methodological issues, investigating other signals that might instruct consolidation, and understanding CFs' broader impact on various learning behaviors are crucial steps for future studies.
We appreciate the reviewer’s recognition of the significance of our study in addressing the fundamental question of the role of CF in adaptive learning within the cerebellar field. The use of optogenetic tools indeed provides a direct means to investigate the causal relationship between CF activity and learning outcomes.
To address concerns regarding the effectiveness of CF suppression during consolidation, we plan to conduct further in-vivo recordings. These will demonstrate how reliably CF transmission can be suppressed through optogenetic manipulation over an extended period.
In response to the concern about potential tissue damage from laser stimulation, we believe that our optogenetic manipulation was not strong enough to induce significant heat-induced tissue damage in the flocculus. According to Cardin et al. (2010), light applied through an optic fiber may cause critical damage if the intensity exceeds 100 mW, which is eight times stronger than the intensity we used in our OKR experiment. Furthermore, if there had been tissue damage from chronic laser stimulation, we would expect to see impaired long-term memory reflected in abnormal gain retrieval results tested the following day. However, as shown in Figures 2 and 3, there were no significant abnormalities in consolidation percentages even after the optogenetic manipulation.
Finally, we appreciate the reviewer’s recognition of the challenges involved in pinpointing specific neural mechanisms. We plan to expand the discussion to address these complexities and outline future research directions.
Recommendations for the authors:
Reviewer #1 (Recommendations For The Authors):
Inhibitory optogenetic actuators are generally problematic, especially in time frames longer than seconds. If the authors wish to be able to inhibit activity in the flocculus-targeting CFs for a long time, maybe it would make sense to try to retrogradely transfect the IO neurons from the flocculus (using a cre-lox approach) with inhibitory DREADDs. This approach is also full of problems, so the absence or significant decrease in CS activity throughout the period of manipulation must be demonstrated.
In addition to re-examining the strength of the evidence regarding the role of CFs in the consolidation and retrival phases, the manuscript would benefit from significant reworking of the details in the manuscript and figures. Below is a possibly incomplete list of things we would want to highlight:
(1) While the text states the authors "... verified the potential reduction of Cs firing rate in PCs of awake mice in vivo by inhibiting CF signals", the data nor a figure are shown. This is of critical importance when judging the reliability of the following results. The data presented in panels Figure 1D-E should also be improved to be more informative, specifically, the waveforms of EPSCs should be shown in higher resolution. We are not informed about how many cells/slices/animals the results are obtained from, nor how many trials were done per condition. Finally, the in vitro data is from vermal Purkinje neurons, while the focus of the work is in the flocculus. Please provide these verifications for the flocculus.
To verify the suppression of complex spike (Cs) activity, we conducted additional in-vivo experiments and added Figure 2, which presents recordings of Cs firing rates from Purkinje cells (PCs) during optogenetic suppression of climbing fiber (CF) activity. These data demonstrate that the suppression specifically and robustly targets Cs activity without affecting simple spike firing, as shown in Figure 2C. The results presented in Figure 2 were acquired at 40 minutes of optostimulation, consistently showing effective suppression of Cs activity throughout this period. While continuous recordings over several hours were not performed, the stability and sustained suppression observed at the 40-minute mark strongly suggest that the manipulation remains effective during the extended durations required for the behavioral tests.
Additionally, we have improved Figure 1D by enhancing the resolution of EPSC waveforms and including more detailed information in the figure legend regarding the number of cells and animals analyzed. For the current-clamp mode data (Figures 1E and F), we clarified the experimental conditions to provide additional context. While the in vitro data were collected from vermal PCs, these experiments were intended to illustrate the fundamental properties of CF-PC transmission.
(2) It is challenging to get a homogenous transfection of all CFs in a given region. To be able to judge the significance of the results, the readers should be provided with material allowing assessing the transfection quality. The images shown in panels Bi-ii are spatially restricted and of too low quality to make judgements. Also, it is not stated whether the images shown are from GFP or NpHR-transfected animals. These different payloads are delivered using different viral capsids (AAV1 vs. AAV9) that have significantly different transfection capacities and results from AAV9-CamKIIGFP cannot be generalized to AAV1-CamKII-NpHR. Please show the expression for the capsid used with NpHR.
To clarify, the images in Figure Bi-ii are representative of GFP expression in animals transfected using AAV1-CamKII-EGFP. The purpose of these panels is to confirm the successful targeting of the region of interest rather than to evaluate viral tropism or capsid-specific transfection efficiency. Moreover, while the transfection characteristics of AAV1 and AAV9 may differ, the key experimental parameter of effective CF suppression was validated through in-vivo electrophysiological recordings, which robustly confirm the efficacy of NpHR expression.
(3) Finally, please show the location of the optic fiber implant in the flocculus from post-mortem images.
In Figure 3a of our revised manuscript, we added post-mortem histological images showing the exact location of the optic fiber implants in the flocculus. These images provided clear confirmation that the optogenetic stimulation was targeted to the correct anatomical region, ensuring that the observed effects are attributable to CF manipulation in the flocculus.
Reviewer #2 (Recommendations For The Authors):
(1) The efficacy of CF suppression is questionable. The histology in Figure 1 shows that only a handful of CFs are transduced in their approach. This observation casts doubt on the claimed complete suppression of CF-evoked EPSCs in every recorded PC in the same figure. This necessitates a more detailed explanation for this apparent discrepancy. Also, the absence of current-clamp recordings to measure the effect on CF-evoked complex spiking in PCs and the lack of detail regarding the timing of optogenetic actuation (continuous or pulsed) during these slice experiments are also significant omissions.
We are providing additional in vivo electrophysiological recordings showing sustained CF suppression in awake animals (Figure 2). These recordings will directly demonstrate the extent of CFevoked complex spike (Cs) suppression.
Moreover, we have included additional data of current-clamp recordings to measure the impact of CF suppression on Cs activity (Figures 1E and 1F). Regarding the timing of the optogenetic actuation, the stimulation was applied continuously in the slice experiments.
(2) The authors claim that their method effectively suppresses CF activity in vivo, yet they do not present any supporting data. Given the histological evidence provided, it's questionable whether their approach truly impacts the CF population broadly, casting doubts on the efficacy of their suppression approach to identify the role of CFs during behavior. To address these concerns, further experiments and detailed quantification are essential to validate the extent and uniformity of CF suppression achieved.
As we responded earlier, we conducted additional in-vivo experiments with continuous recordings of CF-evoked complex spike (Cs) activity during optogenetic suppression (Figure 2). These data directly demonstrate effective and sustained inhibition of CF transmission throughout the behavioral experiments. Quantification of CF suppression revealed consistent inhibition across the manipulation period, with no observable alterations in Purkinje cell simple spike firing rates, confirming that our intervention specifically targeted CF activity without off-target effects. In addition to the in-vivo data, the in-vitro data presented in Figure 1 (lines 107~116) further validate the efficacy of our optogenetic manipulation, showing consistent suppression of CF transmission without any failures. These findings collectively confirm the reliability and specificity of our suppression approach for studying CF contributions to behavior.
(3) To optogenetically test the role of CFs in memory consolidation, the authors deliver continuous, high-power light to the flocculus (13 mW for 6 hrs). This extends well beyond typical experimental conditions. The sustained nature of the light exposure thus brings into question the consistency and reliability of CF suppression over time. Firstly, it is imperative to determine whether CF activity is suppressed throughout this extended period. Secondly, the intensity and duration of light exposure carry a significant risk of causing extensive damage to the surrounding tissue. Given these concerns, a thorough histological examination is warranted to assess the potential adverse effects on tissue integrity. Such an analysis is crucial not only for validating the experimental outcomes but also for ensuring that the observed effects are not confounded by light-induced tissue damage.
To address whether CF activity is suppressed throughout the extended period, we included new in-vivo recordings demonstrating robust suppression of CF transmission, as evidenced by inhibited complex spikes sustained at 40 minutes of optostimulation. Regarding potential tissue damage, our optogenetic protocol used a light intensity (13 mW), which is much lower than the 75 mW threshold reported by Cardin et al. (2010) as sufficient to maintain normal neuronal activity. Moreover, critical damage typically requires intensities exceeding 100 mW for several hours (Cardin, Jessica A., et al. "Targeted optogenetic stimulation and recording of neurons in vivo using cell-type-specific expression of Channelrhodopsin-2." Nature protocols 5.2 (2010): 247-254.). Finally, we observed no abnormalities in long-term memory consolidation or gain retrieval (Figures 3C, 4C, 4F), further supporting that our light stimulation did not induce tissue damage.
(4) The generalizability of their findings to various learning behaviors remains uncertain. Given that the flocculus plays a role in vestibulo-ocular reflex (VOR) adaptation, which encompasses both CFdependent and CF-independent learning types (gain increase and gain decrease, respectively), this system could offer a more feasible approach for investigating hypotheses about the role of CFs in guiding distinct learning processes.
In response to the reviewer’s comment on the generalizability of our findings to learning behaviors involving both CF-dependent and CF-independent mechanisms, we acknowledge the importance of examining these dynamics in cerebellar motor adaptation systems, such as the OKR. Although our study used an OKR task, findings from VOR studies apply here. Ke et al. (2009) demonstrated that VOR gain increases (CF-dependent) and gain decreases (CF-independent) involve distinct plasticity processes (Ke, Michael C., Cong C. Guo, and Jennifer L. Raymond. "Elimination of climbing fiber instructive signals during motor learning." Nature neuroscience 12.9 (2009): 1171-1179), suggesting that CF engagement is task-dependent, particularly for larger error signals that require CF-guided adaptation.
Similarly, our OKR findings suggest that CF-dependent pathways are likely used for large, persistent errors, whereas CF-independent mechanisms may drive more gradual adjustments. This alignment between OKR and VOR systems supports the generalizability of CF-selective adaptation across cerebellar learning tasks. We have elaborated on this point in our revised manuscript (lines 219~237), clarifying how CF-dependent and CF-independent mechanisms can generalize across motor learning contexts in the cerebellum.
(5) The acute effect of CF suppression on OKR eye movements warrants investigation. If OKR eye movements are altered by their method, this could complicate the interpretation of their results.
During our experiments, we monitored ocular movements during CF optogenetic manipulation and found no aberrant effects, such as nystagmus. As shown in Figures 4G and 4H, disrupting CF signaling during gain retrieval did not alter the gain, confirming that our manipulation neither acutely affects ocular reflexes nor induces abnormal eye movement. Therefore, it leads to the conclusion that the observed effects are specific to learning and memory processes.
(6) The authors raise the potential issue of inducing presynaptic LTD in CFs. Can they be sure that their manipulation doesn't generate a similar effect? Additional controls or techniques to accurately interpret the results are needed considering this concern.
However, our discussion does not claim that optogenetic suppression directly induces CF-LTD. Instead, we posit that CF suppression may have mimicked the functional consequences of CFLTD, such as reduced complex spike (Cs) activity and associated calcium signaling. This, in turn, may have indirectly interfered with the induction of parallel fiber-Purkinje cell (PF-PC) LTD, thereby preventing gain enhancement during learning.
This hypothesis is consistent with previous studies highlighting the interplay between CF and PF synaptic plasticity in cerebellar motor learning. For example, Hansel and Linden (2000) and Weber et al. (2003) discuss how changes at CF synapses can modulate Cs waveforms and calcium dynamics, which are critical for PF-PC LTD. Coesmans et al. (2004) and Han et al. (2007) further elaborate on the necessity of CF input for effective PF-PC LTD induction during learning tasks such as retinal slip correction.
While our experiments were not designed to directly measure CF-LTD, the observed prevention of gain enhancement aligns with the hypothesis that CF suppression functionally disrupted downstream PF-PC LTD. We have clarified these points in our revised manuscript (lines 250~258) to avoid misunderstanding.
(7) The specific timeframe for OKR consolidation remains uncertain, with evidence from numerous studies indicating that cerebellar memory consolidation unfolds over several days. Therefore, a more thorough investigation into these extended durations, supported by control experiments to validate the outcomes, would significantly strengthen the study's conclusions, and provide clearer insights into the consolidation process of OKR learning.
Our current study specifically focused on the early phase of the post-learning period, as supported by findings from several studies: Cooke et al., (2004); Titley et al., (2007); Steinmetz et al., (2016); Seo et al., (2024)
These studies collectively indicate that cerebellar-dependent memory consolidation—including OKR—can occur rapidly during the early consolidation phase. While the specific mechanisms examined in these studies vary (e.g., synaptic plasticity, intrinsic plasticity, or circuit-level changes), they consistently demonstrate that modifications in the cerebellum after the early consolidation period no longer influence memory storage or performance. This evidence strongly supports the relevance of our experimental focus and the timing of our interventions.
We acknowledge the importance of investigating extended consolidation periods, which could indeed provide additional insights. However, given our current aims, the rapid consolidation dynamics observed in the early phase are most relevant to the questions addressed in this study. We have elaborated on these matter in our revised manuscript (lines 273~283).
(8) Issues around whether the authors have control over CF activity with their optogenetic intervention raise questions of whether learning can be recovered during the training procedure if the optogenetic stimuli are halted. Specifically, if suppression is applied for three blocks (what the authors refer to as "sessions") during the training procedure and then ceases, does learning rapidly recover in the immediately following blocks?
While we did not directly examine the restoration of learning capability within the same training session following the cessation of optogenetic inhibition, we believe several aspects of our experimental design and insights from prior studies support our interpretation.
Our optogenetic intervention specifically targeted Purkinje cells (PCs) in the flocculus and was applied continuously during designated training sessions to modulate cerebellar activity. Notably, Medina et al. (2001) demonstrated that transient inactivation of the cerebellar cortex impairs the expression of learned responses but does not disrupt the underlying plasticity mechanisms (Medina, Javier F., Keith S. Garcia, and Michael D. Mauk. "A mechanism for savings in the cerebellum." Journal of Neuroscience 21.11 (2001): 4081-4089.). This finding suggests that cerebellar plasticity remains intact and functional even after transient perturbations.
Therefore, it is plausible that once optogenetic inhibition is lifted, the cerebellar network regains its capacity for learning and adaptation, as the intrinsic plasticity and memory encoding processes remain preserved. While we acknowledge that direct experimental confirmation of rapid recovery in our setup was not performed, this interpretation is consistent with our experimental framework and the broader literature.
(9) The study does not fully explore the instructive signals/mechanisms underlying the memory consolidation process. A detailed investigation into potential instructive signals for consolidation beyond CF-induced signaling, like the simple spiking of PCs, could significantly enhance the study's conclusions. Indeed, there is currently no evidence to suggest that CFs play a role in the consolidation phase anyway so testing their role seems a bit of a strawman argument.
While our study primarily focused on characterizing CF-dependent pathways, we acknowledge that memory consolidation is likely driven by a multifaceted interplay of instructive signals beyond CF-induced mechanisms. In particular, Purkinje cell (PC) simple spiking may act as a critical signal during the consolidation phase, either complementing or functioning independently of CF input. Emerging evidence suggests that simple spiking can modulate downstream circuitry in ways that stabilize and strengthen memory traces.
To address this, we have expanded the discussion in the revised manuscript to explore potential instructive signals for consolidation, including PC simple spiking, local circuit plasticity within the cerebellar cortex, and its interaction with the cerebellar nuclei. We propose that these mechanisms collectively contribute to the transfer and stabilization of motor memory, offering a more comprehensive framework for understanding consolidation. We have elaborated on these matter in our revised manuscript (lines 238~250).
(10) Previous reports have highlighted the necessity of CF activity for extinction/memory maintenance (Medina et al. 2002; Kim et al. 2020). That is, the absence of CF activity is consequential for cerebellar function. These results present a potential contrast to the findings reported in this current study. This discrepancy raises important questions about the experimental conditions, methodologies, and interpretations of CF function across different studies. A thorough discussion comparing these divergent outcomes is essential, as it could elucidate the specific contexts or conditions under which CF activity influences memory processes.
We acknowledge that previous studies (Medina et al., 2002; Kim et al., 2020) have suggested a role for climbing fiber (CF) activity in extinction. However, our study specifically focuses on the acquisition phase of motor learning and does not extend to extinction or maintenance. As such, we have revised our discussion to limit interpretations strictly to the scope of our findings and removed references to extinction.
The discrepancies between our results and prior work may arise from differences in methodologies and behavioral paradigms. For instance, we utilized optogenetic inhibition to achieve precise temporal and spatial control of CF activity, whereas previous studies employed pharmacological or lesion methods that may have broader effects on the cerebellar circuitry. Additionally, differences in behavioral paradigms, such as the optokinetic reflex (OKR) task used in our study compared to the eye-blink conditioning tasks in prior studies, may demand distinct roles for CF signaling depending on the specific requirements for error correction and adaptation.
This clarification is now incorporated into our revised manuscript, and the discussion has been streamlined to focus on the phase-specific role of CF activity during acquisition without extending to extinction or maintenance (lines 259~270).
