Cholecystokinin modulates age-dependent Thalamocortical Neuroplasticity

  1. Xiao Li
  2. Jingyu Feng
  3. Xiaohan Hu
  4. Peipei Zhou
  5. Tao Chen
  6. Xuejiao Zheng
  7. Peter Jendrichovsky
  8. Xue Wang
  9. Mengying Chen
  10. Hao Li
  11. Xi Chen
  12. Dingxuan Zeng
  13. Mengfan Zhang
  14. Zhoujian Xiao
  15. Ling He
  16. Stephen Temitayo Bello
  17. Jufang He

  • Curated by eLife

    eLife logo

    eLife Assessment

    This is an important study demonstrating that cholecystokinin is a key modulator of auditory thalamocortical plasticity during development and in young adult but not aged mice, though cortical application of this neuropeptide in older animals appears to go some way to restoring this age-dependent loss in plasticity. A strength of this work is the use of multiple experimental approaches, which together provide convincing support for the proposed involvement of cholecystokinin. This work is likely to be influential in opening up a new avenue of investigation into the roles of neuropeptides in sensory plasticity.

Reviewed by

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

Log in to save this article

Abstract

Abstract

The thalamocortical pathways exhibit neuroplasticity not only during the critical period but also in adulthood. In this study, we investigated how cholecystokinin (CCK) modulates age-dependent thalamocortical plasticity. Our findings demonstrated that CCK is expressed in thalamocortical neurons and that high-frequency stimulation (HFS) of the thalamocortical pathway triggers the release of CCK in auditory cortex (ACx), as detected by a CCK sensor. HFS of the medial geniculate body (MGB) induced thalamocortical long-term potentiation (LTP) in wild-type young adult mice. However, knockdown of Cck expression in MGB neurons or blockade of the CCK-B receptor (CCKBR) in the ACx abolished HFS-induced LTP. Interestingly, this LTP could not be elicited in juvenile (3-week-old) or aged mice (over 18-month-old) due to distinct mechanisms: the absence of CCKBR in juveniles and the inability to release CCK in aged mice. Notably, exogenous administration of CCK into the ACx rescued LTP in aged mice and significantly improved frequency discrimination. These findings highlight the potential of CCK as a therapeutic intervention for ameliorating neuroplasticity deficits associated with thalamocortical connectivity.

Article activity feed

  1. eLife

    eLife Assessment

    This is an important study demonstrating that cholecystokinin is a key modulator of auditory thalamocortical plasticity during development and in young adult but not aged mice, though cortical application of this neuropeptide in older animals appears to go some way to restoring this age-dependent loss in plasticity. A strength of this work is the use of multiple experimental approaches, which together provide convincing support for the proposed involvement of cholecystokinin. This work is likely to be influential in opening up a new avenue of investigation into the roles of neuropeptides in sensory plasticity.

  2. eLife

    Reviewer #1 (Public review):

    This report addresses a compelling topic. The authors demonstrate that tetanic stimulation of the auditory thalamus induces cortical long-term potentiation (LTP), which can be elicited by either electrical or optical stimulation of the thalamus or by noise bursts. They further show that thalamocortical LTP is abolished when thalamic CCK is knocked down or when cortical CCK receptors are blocked. Notably, in 18-month-old mice, thalamocortical LTP was largely absent but could be restored by cortical application of CCK. The authors conclude that CCK is a critical contributor to thalamocortical plasticity and may enhance this form of plasticity in aged subjects.

    The findings presented in this report are valuable and advance our understanding of thalamocortical plasticity.

  3. eLife

    Reviewer #2 (Public review):

    Summary:

    This work used multiple approaches to show that CCK is critical for long-term potentiation (LTP) in the auditory thalamocortical pathway. They also showed that the CCK mediation of LTP is age-dependent and supports frequency discrimination. This work is important because is opens up a new avenue of investigation of the roles of neuropeptides in sensory plasticity.

    Strengths:

    The main strength is the multiple approaches used to comprehensively examine the role of CCK in auditory thalamocortical LTP. Thus, the authors do provide a compelling set of data that CCK mediates thalamocortical LTP in an age-dependent manner.

    Weaknesses:

    The behavioral assessment is relatively limited, but may be fleshed out in future work.

  4. eLife

    Reviewer #3 (Public review):

    Summary:

    Cholecystokinin (CCK) is highly expressed in auditory thalamocortical (MGB) neurons and CCK has been found to shape cortical plasticity dynamics. In order to understand how CCK shapes synaptic plasticity in the auditory thalamocortical pathway, they assessed the role of CCK signaling across multiple mechanisms of LTP induction with the auditory thalamocortical (MGB - layer IV Auditory Cortex) circuit in mice. In these physiology experiments that leverage multiple mechanisms of LTP induction and a rigorous manipulation of CCK and CCK-dependent signaling, they establish an essential role of auditory thalamocortical LTP on the co-release of CCK from auditory thalamic neurons. By carefully assessing the development of this plasticity over time and CCK expression, they go on to identify a window of time that CCK is produced throughout early and middle adulthood in auditory thalamocortical neurons to establish a window for plasticity from 3 weeks to 1.5 years in mice, with limited LTP occurring outside of this window. The authors go on to show that CCK signaling and its effect on LTP in the auditory cortex is also capable of modifying frequency discrimination accuracy in an auditory PPI task. In evaluating the impact of CCK on modulating PPI task performance, it also seems that in mice <1.5 years old CCK-dependent effects on cortical plasticity is almost saturated. While exogenous CCK can modestly improve discrimination of only very similar tones, exogenous focal delivery of CCK in older mice can significantly improve learning in a PPI task to bring their discrimination ability in line with those from young adult mice.

    Strengths:

    (1) The clarity of the results along with the rigor multi-angled approach provide significant support for the claim that CCK is essential for auditory thalamocortical synaptic LTP. This approach uses a combination of electrical, acoustic, and optogenetic pathway stimulation alongside conditional expression approaches, germline knockout, viral RNA downregulation and pharmacological blockade. Through the combination of these experimental configures the authors demonstrate that high-frequency stimulation-induced LTP is reliant on co-release of CCK from glutamatergic MGB terminals projecting to the auditory cortex.

    (2) The careful analysis of the CCK, CCKB receptor, and LTP expression is also a strength that puts the finding into the context of mechanistic causes and potential therapies for age-dependent sensory/auditory processing changes. Similarly, not only do these data identify a fundamental biological mechanism, but they also provide support for the idea that exogenous asynchronous stimulation of the CCKBR is capable of restoring an age-dependent loss in plasticity.

    (3) Although experiments to simultaneously relate LTP and behavioral change or identify a causal relationship between LTP and frequency discrimination are not made, there is convincing evidence that CCK signaling in the auditory cortex (known to determine synaptic LTP) is important for auditory processing/frequency discrimination. These experiments are key for establishing the relevance of this mechanism.

    Weaknesses:

    The following are weaknesses or limitations of the study that may also fall outside of the scope of this work, but which could be addressed in the future.

    (1) Given the magnitude of the evoked responses, one expects that pyramidal neurons in layer IV are primarily those that undergo CCK-dependent plasticity, but the degree to which PV-interneurons and pyramidal neurons participate in this process differently is unclear.

    (2) While these data support an important role for CCK in synaptic LTP in the auditory thalamocortical pathway, perhaps temporal processing of acoustic stimuli is as or more important than frequency discrimination. Given the enhanced responsivity of the system, it is unclear whether this mechanism would improve or reduce the fidelity of temporal processing in this circuit. Understanding this dynamic may also require consideration of cell type as raised in weakness #1.

    (3) In Figure 1, an example of increased spontaneous and evoked firing activity of single neurons after HFS is provided. Yet it is surprising that the group data are analyzed only for the fEPSP. It seems that single neuron data would also be useful at this point to provide insight into how CCK and HFS affect temporal processing and spontaneous activity/excitability.

  5. eLife

    Author response:

    The following is the authors’ response to the previous reviews

    Reviewer #1 (Public review):

    This report addresses a compelling topic. However, I have significant concerns, which necessitate a reassessment of the report's overall value.

    Anatomical Specificity and Stimulation Site:

    While the authors clarify that the ventral MGB (MGv) was the intended stimulation target, the electrode track (Fig. 1A) and viral spread (Fig. 2E) suggest possible involvement of the dorsal MGB (MGd) and broader area. Given that MGv-AI and MGd-AC pathways have distinct-and sometimes opposing-effects on plasticity, the reported LTP values (with unusually small standard deviations) raise concerns about the specificity of the findings. Additional anatomical verification would help resolve this issue.

    We thank the reviewer for highlighting the importance of anatomical specificity in MGv targeting. In the revised manuscript, we have taken several steps to address these issues:

    (1) Higher-magnification histology has been added to Figure 1A, clearly identifying the electrode tip localized within the MGv.

    (2) Figure 2E has been replaced with a new image showing viral expression largely confined to MGB, with minimal spread to surrounding structures.

    (3) In the Discussion, we explicitly acknowledge that although targeting was guided by stereotaxic coordinates and histological confirmation, some viral spread throughout the MGB occurred. We also discuss the possibility that both MGv-A1 and MGd-AC pathways may contribute to the recorded responses, which could influence the observed plasticity, as previously suggested by the reviewer.

    These additions and acknowledgments are now incorporated to ensure the reader can interpret the data with full consideration of anatomical targeting limitations.

    Results section:

    “Higher-magnification histology confirmed accurate MGv targeting (Figure 1A, lower-middle panel)’”

    Discussion section:

    “Although our experiment targeting the MGv was guided by stereotaxic coordinates and verified post hoc, we acknowledge potential contributions from non-lemniscal medial geniculate nucleus dorsal (MGd) projections. Anatomical and physiological evidence indicates that MGv-AC projections provide rapid, frequency‑specific, tonotopically organized excitation, whereas MGd pathways target higher‑order auditory cortex with broader tuning, less precise tonotopy, longer response latencies, and greater context‑dependence, features that can differentially shape cortical sensory integration and plasticity (Lee and Sherman, 2010; Smith et al., 2012; Ohga et al., 2018; Lee, 2015; Hu, 2003). While the co-recruitment of lemniscal and non-lemniscal inputs may enhance the generality of our CCK-dependent mechanism, the differing response characteristics of these pathways suggest subtle differences in their relative engagement in the observed plasticity. Future pathway-specific manipulations will help clarify their respective contributions”

    Lee, C.C., and Sherman, S.M. (2010). Topography and physiology of ascending streams in the auditory tectothalamic pathway. Proceedings of the National Academy of Sciences 107, 372-377. doi:10.1073/pnas.0907873107.

    Smith, P.H., Uhlrich, D.J., Manning, K.A., and Banks, M.I. (2012). Thalamocortical projections to rat auditory cortex from the ventral and dorsal divisions of the medial geniculate nucleus. Journal of Comparative Neurology 520, 34-51.

    Ohga, S., Tsukano, H., Horie, M., Terashima, H., Nishio, N., Kubota, Y., Takahashi, K., Hishida, R., Takebayashi, H., and Shibuki, K. (2018). Direct Relay Pathways from Lemniscal Auditory Thalamus to Secondary Auditory Field in Mice. Cerebral Cortex 28, 4424-4439. 10.1093/cercor/bhy234.

    Lee, C.C. (2015). Exploring functions for the non-lemniscal auditory thalamus. Frontiers in Neural Circuits 9, 69.

    Hu, B. (2003). Functional organization of lemniscal and nonlemniscal auditory thalamus. Experimental Brain Research 153, 543-549. 10.1007/s00221-003-1611-5.

    Figure legend section:

    “Post-hoc histology at higher magnification (lower-middle) shows the electrode tip confined within the MGv. White lines delineate the MGv/MGd border based on cytoarchitectonic landmarks.”

    Statistical Rigor and Data Variability:

    The remarkably low standard deviations in LTP measurements are unexpected based on established variability in thalamocortical plasticity. The authors' response confirms these values are accurate, but further justification, such as methodological controls or replication-would bolster confidence in these results. Additionally, the comparison of in vivo vs. in vitro LTP variability requires more substantive support.

    We appreciate the reviewer's concern regarding the unusually small variability. We would like to clarify that the error bars in our figures represent Standard Error of the Mean (SEM) rather than Standard Deviations (SD). As SEM is derived from the SD while incorporating sample size, it is inherently smaller than SD, which may have led to the impression of unrealistically low variability. This has now been explicitly clarified in the figure legends and Methods.

    To illustrate the raw variability, we have added Supplementary Figure S1E showing unaveraged fEPSP slopes compare to SEM, corresponding to Figure S1C. This addition ensures transparency and allows readers to directly assess the quality and consistency of our recordings.

    Regarding the comparison between in vivo and in vitro LTP variability:

    We agree that clarifying the basis of our in vivo vs. in vitro variability comparison is important. For example, in Chen et al., 2019, using identical LTP induction protocols (Fig. J), the SED of in vitro slice measurements (Fig. K) was substantially larger than that of in vivo recordings (Fig. L).

    This difference likely reflects:

    (1) In vitro: neighboring data points within a single experiment are highly correlated; variability across experiments is large due to heterogeneous sensitivity to LTP induction (10–200% increasement).

    (2) In vivo: lower correlation between neighboring data points, but each is averaged from 12 recordings over 2 min, reducing cross-trial variability; sensitivity to LTP induction is less variable across experiments (5–60% changes).

    We hope that these clarifications and additional data address the reviewer’s concerns regarding statistical rigor and data variability.

    Methods section:

    “The slopes of the evoked fEPSPs were calculated and normalized using a customized MATLAB script, and the group data were plotted as mean ± Standard Error of the Mean (SEM).”

    “All data are presented as mean ± SEM. Error bars and shaded areas represent SEM. Here, n represents the number of stimulation-recording sites or and N represents the number of animals in each experiment. At each time point, fEPSPs were averaged across 12 consecutive trials (2 min) to reduce within-experiment fluctuation. Normalized time courses were then used for repeated-measures analyses.”

    Figure legend section:

    “Data are mean ± SEM; error bars indicate SEM.”

    “(E) Unaveraged fEPSP slopes are shown for each time point, with individual data points corresponding to all sites included in Fig. 1C; mean ± SEM overlays are shown in black. Note that all individual data points are displayed in this figure, whereas in Figure S1C, only the averaged values are shown.”

    Viral Targeting and Specificity:

    The manuscript does not clearly address whether cortical neurons were inadvertently infected by AAV9. Given the potential for off-target effects, explicit confirmation (e.g., microphotograph of stimulation site) would strengthen the study's conclusions.

    We appreciate the request for quantitative confirmation of off-target cortical infection. We clarify that our histological verification was conducted by systematic sampling rather than exhaustive quantification. Under the same sampling procedure, we did not detect tdTomato-positive cortical somata after AAV9‑Syn‑ChrimsonR‑tdTomato injections into the MGB, whereas we observed rare EYFP-positive cortical somata after AAV9‑EF1a‑DIO‑ChETA‑EYFP (median < 1 cell per 0.4 × 0.4 mm² section, Supplementary Figure S1E). Although these observations do not constitute a formal statistical estimate, they were consistent across sampled sections and are in line with the low-level trans-synaptic transfer reported for AAV9. We have discussed their potential implications for data interpretation in the Discussion.

    We hope these clarifications and the newly presented histological evidence address the reviewer’s concerns and further strengthen the rigor of our study.

    Discussion section:

    “Another potential limitation of our study is the trans-synaptic transfer property of AAV9 (Figure S1F). To mitigate this risk, we carefully control the injection volume, rate, and viral expression time, while also verifying expression post-hoc. Systematic sampling histological analysis detected no tdTomato-positive cortical somata in the ACx (Figure 2E lower panel), whereas rare EYFP-positive cortical somata were observed after AAV9-EF1a-DIO-ChETA-EYFP injections (median < 1 cell in 0.4 × 0.4 mm2 section, Figure S1F, corresponds to Figure 2A upper-middle panel). These construct‑dependent observations align with occasional low‑level trans‑synaptic transfer reported for AAV9 (Zingg et al., 2017) and indicate that off‑target cortical infection was negligible for ChrimsonR and exceedingly rare for ChETA under our experimental conditions.”

    Zingg, B., Chou, X.L., Zhang, Z.G., Mesik, L., Liang, F., Tao, H.W., and Zhang, L.I. (2017). AAV-Mediated Anterograde Transsynaptic Tagging: Mapping Corticocollicular Input-Defined Neural Pathways for Defense Behaviors. Neuron 93, 33-47. 10.1016/j.neuron.2016.11.045.

    Figure legend:

    “Representative histological images demonstrating low-level transsynaptic spread following AAV9-EF1a-DIO-ChETA-EYFP injection into the MGv. Rare EYFP-positive cortical neurons were observed (median < 1 cell per 0.4 × 0.4 mm² section). Scale bar: 100 µm.”

    Integration of Prior Literature:

    The discussion of existing work is adequate but could be more comprehensive. A deeper engagement with contrasting findings would provide better context for the study's contributions.

    We appreciate the reviewer’s suggestion to engage more deeply with contrasting findings. In the revised Introduction and Discussion, we have:

    (1) Refocused the historical context toward adult auditory thalamocortical plasticity and explicitly contrasted it with visual and somatosensory cortices, while adult ACx exhibits weaker and more gated NMDAR dependence.

    (2) Positioned CCK–CCKBR signaling as a permissive/gating mechanism that can complement or partially compensate for postsynaptic NMDAR signaling, potentially reconciling variability across cortical areas and life stages.

    (3) Clarified the potential differential contributions of lemniscal (MGv) and non‑lemniscal (MGd) streams to plasticity expression and variability, acknowledging pathway-specific response properties.

    These additions are now integrated in the Introduction (paragraphs 2–3) and Discussion (sections “CCK Dependence of Thalamocortical Neuroplasticity in the ACx” and “Developmental and Age‑Dependent CCK‑Mediated Plasticity”), providing a more comprehensive and balanced context for our findings.

    Introduction section:

    “However, converging evidence shows that thalamocortical inputs retain a capacity for experience-dependent modification in adulthood. Sensory enrichment or deprivation can gate or reinstate thalamocortical plasticity. In the adult ACx, pairing sounds with neuromodulatory drive can reshape cortical representations. In vivo high-frequency stimulation (HFS) of dorsal lateral geniculate nucleus (LGN) or medial geniculate body (MGB) induces LTP in sensory cortices and has been linked to perceptual learning beyond the critical period. Notably, auditory thalamocortical plasticity appears less dependent on NMDA receptors compared to other cortical regions. The mechanisms underlying thalamocortical plasticity in the mature brain remain poorly understood.

    Cholecystokinin (CCK) and its receptor CCK-B receptor (CCKBR) are well positioned to influence thalamocortical transmission: Cck mRNA is abundant in MGB neurons and CCKBR is enriched in layer IV of ACx, the principal thalamorecipient layer.”

    Discussion section:

    “These findings suggest a potential involvement of CCK in thalamocortical plasticity. Our data extend this framework by identifying CCK–CCKBR signaling as a permissive modulator of adult thalamocortical LTP.”

    “We propose that CCKBR activation may trigger intracellular calcium release and AMPAR recruitment in parallel to, or partially compensating for,independently of postsynaptic NMDAR signaling, while the complementarity of CCKBR and NMDARs may contribute to robust thalamocortical plasticity. This complementary arrangement may reconcile differences across developmental stages and cortical areas, and highlights neuropeptidergic signaling as a lever to re-enable adult thalamocortical plasticity.

    Notably, exogenous CCK alone failed to induce LTP in the absence of accompanying stimulation (Figure S2A and S2B), emphasizing that CCK function as a modulator rather than a direct initiator of LTP. Activation of the thalamocortical pathway is also essential for LTP induction. Although our experiment targeting the MGv was guided by stereotaxic coordinates and verified post hoc, we acknowledge potential contributions from non-lemniscal medial geniculate nucleus dorsal (MGd) projections. Anatomical and physiological evidence indicates that MGv-AC projections provide rapid, frequency‑specific, tonotopically organized excitation, whereas MGd pathways target higher‑order auditory cortex with broader tuning, less precise tonotopy, longer response latencies, and greater context‑dependence, features that can differentially shape cortical sensory integration and plasticity. While the co-recruitment of lemniscal and non-lemniscal inputs may enhance the generality of our CCK-dependent mechanism, the differing response characteristics of these pathways suggest subtle differences in their relative engagement in the observed plasticity. Future pathway-specific manipulations will help clarify their respective contributions. Another potential limitation of our study is the trans-synaptic transfer property of AAV9 (Figure S1F). To mitigate this, we carefully controlled the injection volume, rate, and viral expression time, and conducted post-hoc histological analyses to minimize off-target effects, thereby reducing the likelihood of trans-synaptic transfer confounding the interpretation of our findings.”

    Therapeutic Implications:

    The authors' discussion of therapeutic potential is now appropriately cautious and well-reasoned.

    Conclusion:

    While the study presents intriguing findings, the concerns outlined above must be addressed to fully establish the validity and impact of the results. I appreciate the authors' efforts thus far and hope they can provide additional data or clarification to resolve these issues. With these revisions, the manuscript could make a valuable contribution to the field.

    Reviewer #2 (Public review):

    Summary:

    This work used multiple approaches to show that CCK is critical for long-term potentiation (LTP) in the auditory thalamocortical pathway. They also showed that the CCK mediation of LTP is age-dependent and supports frequency discrimination. This work is important because is opens up a new avenue of investigation of the roles of neuropeptides in sensory plasticity.

    Strengths:

    The main strength is the multiple approaches used to comprehensively examine the role of CCK in auditory thalamocortical LTP. Thus, the authors do provide a compelling set of data that CCK mediates thalamocortical LTP in an age-dependent manner.

    Weaknesses:

    There are some details that should be addressed, primarily regarding potential baseline differences in comparison groups. The behavioral assessment is relatively limited, but may be fleshed out in future work.

    We appreciate the reviewer’s suggestion regarding potential baseline differences. In our study, all groups underwent harmonized procedures, including identical exposure, timing, and acquisition parameters. Group allocation and data collection were performed under standardized conditions. For electrophysiology, baseline fEPSP measures and stimulation intensities were calibrated per site using consistent input-output procedures, with analyses based on normalized slopes relative to each site’s own baseline. For behavior, animals from the same litter served as both experimental and control groups, matched for handling conditions; startle/PPI data were acquired using identical hardware and timing settings. While no additional post hoc re-processing was performed, we have clarified these controls in the Methods to enhance transparency.

    We agree that the behavioral assessment is intentionally focused and does not encompass broader auditory perceptual functions (e.g., temporal processing). We now explicitly state this limitation and propose future studies to examine temporal acuity and cell-type-specific manipulations. These experiments will clarify how CCK-dependent thalamocortical plasticity generalizes to other perceptual domains.

    Reviewer #3 (Public review):

    Summary:

    Cholecystokinin (CCK) is highly expressed in auditory thalamocortical (MGB) neurons and CCK has been found to shape cortical plasticity dynamics. In order to understand how CCK shapes synaptic plasticity in the auditory thalamocortical pathway, they assessed the role of CCK signaling across multiple mechanisms of LTP induction with the auditory thalamocortical (MGB - layer IV Auditory Cortex) circuit in mice. In these physiology experiments that leverage multiple mechanisms of LTP induction and a rigorous manipulation of CCK and CCK-dependent signaling, they establish an essential role of auditory thalamocortical LTP on the co-release of CCK from auditory thalamic neurons. By carefully assessing the development of this plasticity over time and CCK expression, they go on to identify a window of time that CCK is produced throughout early and middle adulthood in auditory thalamocortical neurons to establish a window for plasticity from 3 weeks to 1.5 years in mice, with limited LTP occurring outside of this window. The authors go on to show that CCK signaling and its effect on LTP in the auditory cortex is also capable of modifying frequency discrimination accuracy in an auditory PPI task. In evaluating the impact of CCK on modulating PPI task performance, it also seems that in mice <1.5 years old CCK-dependent effects on cortical plasticity is almost saturated. While exogenous CCK can modestly improve discrimination of only very similar tones, exogenous focal delivery of CCK in older mice can significantly improve learning in a PPI task to bring their discrimination ability in line with those from young adult mice.

    Strengths:

    (1) The clarity of the results, along with the rigor multi-angled approach, provide significant support for the claim that CCK is essential for auditory thalamocortical synaptic LTP. This approach uses a combination of electrical, acoustic, and optogenetic pathway stimulation alongside conditional expression approaches, germline knockout, viral RNA downregulation and pharmacological blockade. Through the combination of these experimental configures the authors demonstrate that high-frequency stimulation-induced LTP is reliant on co-release of CCK from glutamatergic MGB terminals projecting to the auditory cortex.

    (2) The careful analysis of the CCK, CCKB receptor, and LTP expression is also a strength that puts the finding into the context of mechanistic causes and potential therapies for age-dependent sensory/auditory processing changes. Similarly, not only do these data identify a fundamental biological mechanism, but they also provide support for the idea that exogenous asynchronous stimulation of the CCKBR is capable of restoring an age-dependent loss in plasticity.

    (3) Although experiments to simultaneously relate LTP and behavioral change or identify a causal relationship between LTP and frequency discrimination are not made, there is still convincing evidence that CCK signaling in the auditory cortex (known to determine synaptic LTP) is important for auditory processing/frequency discrimination. These experiments are key for establishing the relevance of this mechanism.

    Weaknesses:

    (1) Given the magnitude of the evoked responses, one expects that pyramidal neurons in layer IV are primarily those that undergo CCK-dependent plasticity, but the degree to which PV-interneurons and pyramidal neurons participate in this process differently is unclear.

    We agree with the reviewer that the relative contributions of pyramidal neurons and PV-interneurons to CCK-dependent thalamocortical plasticity remain to be determined. Our recordings primarily reflected excitatory postsynaptic activity from layer IV pyramidal neurons, given the fEPSP metrics used. As PV-interneurons are essential in shaping cortical inhibition and temporal precision, they may also be modulated by CCK release from thalamocortical inputs. We have explicitly acknowledged this limitation in the Discussion section of the manuscript and propose that future studies should employ cell-type-specific recording or manipulation approaches to dissect the respective roles of inhibitory and excitatory neuronal populations in CCK-dependent thalamocortical plasticity. We appreciate the reviewer’s suggestion and believe this is a valuable direction for ongoing research.

    (2) While these data support an important role for CCK in synaptic LTP in the auditory thalamocortical pathway, perhaps temporal processing of acoustic stimuli is as or more important than frequency discrimination. Given the enhanced responsivity of the system, it is unclear whether this mechanism would improve or reduce the fidelity of temporal processing in this circuit. Understanding this dynamic may also require consideration of cell type as raised in weakness #1.

    We acknowledge that the current study primarily examined frequency discrimination and did not directly assess temporal processing. Enhanced network responsivity could have variable effects on temporal precision, depending on the balance between excitation and inhibition. PV-interneurons, in particular, are known to support temporal fidelity in auditory processing (Nocon et al., 2023; Cai et al., 2018). We discussion that future work should investigate how CCK modulation influences temporal coding at both the circuit and single-cell level, and whether such changes align with or diverge from the mechanisms underlying frequency discrimination improvements.

    (3) In Figure 1, an example of increased spontaneous and evoked firing activity of single neurons after HFS is provided. Yet it is surprising that the group data are analyzed only for the fEPSP. It seems that single neuron data would also be useful at this point to provide insight into how CCK and HFS affect temporal processing and spontaneous activity/excitability, especially given the example in 1F.

    We appreciate the reviewer’s suggestion. While we recorded single-unit activity during HFS protocols, long-term stability over >1.5 hours was less consistent compared to fEPSP measurements, leading to higher variability in spike-based metrics. We therefore used fEPSPs as our primary quantitative measure for robustness. We agree, however, that single-neuron data could yield valuable complementary insights. In future experiments combining stable single-unit recording with synaptic measurements will be conducted to better link cellular excitability and network plasticity.

    (4) The circuitry that determines PPI requires multiple brain areas, including the auditory cortex. Given the complicated dynamics of this process, it may be helpful to consider what, if anything, is known specifically about how layer IV synaptic plasticity in the auditory cortex may shape this behavior.

    We agree that PPI involves multiple cortical and subcortical nodes. In our paradigm, layer IV neurons receive segregated MGv inputs, high-frequency activation of thalamocortical projections induces robust synaptic plasticity in layer IV. The potentiation at these synapses could amplify the cortical representation of weak prepulses, facilitating their detection and enhancing PPI performance. This interpretation is consistent with prior work showing that local CCK infusion combined with auditory stimuli can augment cortical responses (Li et al., 2014). We have expanded the Discussion to highlight that in aged animals, where baseline PPI performance is often reduced due to degraded auditory inputs (Ouagazzal et al., 2006; Young et al., 2010), restoring thalamocortical plasticity via CCK may partially compensate for sensory gating deficits. We further note that the exact contribution of layer IV to PPI circuitry warrants future investigation using pathway-specific perturbations.

    Comments on revisions:

    The manuscript is much improved and many of the issues or questions have been addressed. Ideally, evidence for the degree of transsynaptic spread for AAV9-Syn-ChrimsonR-tdTomato would also be provided in some form since in the authors' response in sounds like some was observed, as expected.

    We thank the reviewer for this important point and for the opportunity to clarify. As requested, we have carefully examined the possibility of transsynaptic spread in our experiments:

    We clarify that our histological verification was conducted by systematic sampling rather than exhaustive quantification. Under the same sampling procedure, we did not detect tdTomato-positive cortical somata after AAV9‑Syn‑ChrimsonR‑tdTomato injections into the MGB, whereas we observed rare EYFP-positive cortical somata after AAV9‑EF1a‑DIO‑ChETA‑EYFP (median < 1 cell per 0.4 × 0.4 mm² section, see Figure 2A and Figure S1F), consistent with occasional low-level transsynaptic spread reported in the literature.

    We have updated the Discussion sections to clearly report these findings, and to emphasize the potential for vector- and construct-dependent variability in transsynaptic spread. We also explicitly acknowledge this technical limitation and discuss its implications for data interpretation.

    We hope these clarifications and additions address the reviewer’s concern regarding viral specificity and transsynaptic spread.

    Discussion section:

    “Another potential limitation of our study is the trans-synaptic transfer property of AAV9 (Figure S1F). To mitigate this risk, we carefully control the injection volume, rate, and viral expression time, while also verifying expression post-hoc. Systematic sampling histological analysis detected no tdTomato-positive cortical somata in the ACx (Figure 2E lower panel), whereas rare EYFP-positive cortical somata were observed after AAV9-EF1a-DIO-ChETA-EYFP injections (median < 1 cell in 0.4 × 0.4 mm2 section, Figure S1F, corresponds to Figure 2A upper-middle panel). These construct‑dependent observations align with occasional low‑level trans‑synaptic transfer reported for AAV9 (Zingg et al., 2017) and indicate that off‑target cortical infection was negligible for ChrimsonR and exceedingly rare for ChETA under our experimental conditions.”

    Zingg, B., Chou, X.L., Zhang, Z.G., Mesik, L., Liang, F., Tao, H.W., and Zhang, L.I. (2017). AAV-Mediated Anterograde Transsynaptic Tagging: Mapping Corticocollicular Input-Defined Neural Pathways for Defense Behaviors. Neuron 93, 33-47. 10.1016/j.neuron.2016.11.045.

    Figure legend:

    " Representative histological images demonstrating low-level transsynaptic spread following AAV9-EF1a-DIO-ChETA-EYFP injection into the MGv. Rare EYFP-positive cortical neurons were observed (median < 1 cell per 0.4 × 0.4 mm² section). Scale bar: 100 µm."

    Reviewer #1 (Recommendations for the authors):

    Thank you for your efforts in revising the manuscript. While progress has been made, I have a few remaining concerns that I hope you can address to further strengthen the study.

    Focus of the Introduction:

    Auditory thalamocortical plasticity is known to be NMDA-dependent, albeit with weaker dependence during early development. Given that this work examines thalamocortical LTP in young adult and aged mice, I recommend refining the Introduction to place greater emphasis on auditory thalamocortical plasticity in the adult brain. The current discussion of somatosensory plasticity during early development, while interesting, seems less directly relevant to the present study. A sharper focus on the auditory system would better frame your research questions.

    We thank the reviewer for this constructive suggestion. We have revised the Introduction to emphasize adult auditory thalamocortical plasticity and to streamline content less directly related to our study. Specifically:

    (1) We now foreground evidence that thalamocortical inputs retain experience-dependent plasticity beyond the critical period in adult ACx, including neuromodulatory pairing, HFS-induced LTP, and experience-dependent reinstatement.

    (2) We explicitly note that adult auditory thalamocortical plasticity is more weakly NMDAR-dependent than in other cortices, thereby motivating our focus on CCK–CCKBR signaling as a permissive mechanism for adult LTP.

    (3) We have condensed the discussion of somatosensory plasticity during early development to a brief background and shifted the focus to adult auditory mechanisms and knowledge gaps that directly frame our research questions.

    These changes appear in the revised Introduction (paragraphs 2–3), which now provide a sharper rationale for investigating CCK‑dependent thalamocortical LTP in young adult and aged mice.

    Introduction section:

    “However, converging evidence shows that thalamocortical inputs retain a capacity for experience-dependent modification in adulthood. Sensory enrichment or deprivation can gate or reinstate thalamocortical plasticity. In the adult ACx, pairing sounds with neuromodulatory drive can reshape cortical representations. In vivo high-frequency stimulation (HFS) of dorsal lateral geniculate nucleus (LGN) or medial geniculate body (MGB) induces LTP in sensory cortices and has been linked to perceptual learning beyond the critical period. Notably, auditory thalamocortical plasticity appears less dependent on NMDA receptors compared to other cortical regions. The mechanisms underlying thalamocortical plasticity in the mature brain remain poorly understood.

    Cholecystokinin (CCK) and its receptor CCK-B receptor (CCKBR) are well positioned to influence thalamocortical transmission: Cck mRNA is abundant in MGB neurons and CCKBR is enriched in layer IV of ACx, the principal thalamorecipient layer.”

    Anatomical Specificity of MGv Targeting:

    The mouse MGv is a small and deep structure, and precise targeting is critical given the functional differences between MGv and MGd pathways. In the current figures:

    Fig. 1A suggests the electrode track may have approached the MGd.

    Fig. 2E indicates some viral spread beyond the MGB.

    Since MGv-AI and MGd-AC pathways exhibit distinct (and sometimes opposing) effects on plasticity, I encourage you to provide additional clarification or verification of the stimulated/infected regions. This would greatly enhance the interpretability of your LTP data.

    Please see above.

    Data Variability and Transparency:

    The reported thalamocortical LTP values exhibit remarkably small standard deviations, which is somewhat unexpected given typical experimental variability in such measurements. To address this concern, it would be helpful to include example raw traces of the recorded LTP (e.g., in a supplementary figure). This would allow readers to better evaluate the data quality and consistency.

    Please see above.

    Reviewer #2 (Recommendations for the authors):

    Overall, the authors did an excellent job of responding to our critiques, both in their direct responses and in the modified text. The modified text is also more readable than before. Two issues that the authors should consider addressing;

    (1) Unless I missed it, there is no commentary stated about the impact of using aged C57 mice, which lose their hearing, such that the effects seen in the older mice could be related to hearing loss rather than aging alone. Some discussion of this point should be made.

    We thank the reviewer for raising this important point. C57BL/6 mice are known to develop age-related hearing loss, which could potentially affect PPI performance in older animals. We note that in our internal screening we observed markedly reduced startle amplitudes and frequent negative PPI values in many mice >20 months, indicating severe auditory impairment. To minimize this confound a priori, we excluded mice older than 20 months and restricted the aged cohort to 17–19 months, which consistently exhibited robust startle responses and reliable PPI. While some degree of presbycusis may still be present in this age range in C57BL/6 mice, the improvement of PPI following CCK administration combined with acoustic exposure indicates that the auditory pathways remained sufficiently functional to support sensorimotor gating. In fact, the presence of partial hearing loss in these aged mice may have allowed us to better detect the beneficial effects of CCK, further highlighting its therapeutic potential for age-related deficits. The greater improvement in PPI observed in older mice —as compared to younger mice, whose PPI in control group is already high—likely reflect the combined effects of age-related hearing loss and CCK deficiency, with CCK-induced restoration of thalamocortical plasticity being the primary focus of our study. We have now added a discussion of this point in the revised manuscript.

    Discussion section:

    “In aged mice, PPI deficits are commonly observed due to impaired auditory processing. Notably, C57BL/6 mice exhibit age-related hearing loss (Johnson et al., 1997). Both age-associated changes in auditory function and CCK deficiency contribute to impaired sensory gating. The presence of partial hearing loss in aged mice may have facilitated the detection of CCK’s beneficial effects, further highlighting its therapeutic potential for age-related deficits. Our results suggest that enhanced thalamocortical plasticity mediated by CCK might partially compensate for these deficits by amplifying residual auditory signals in aged mice.”

    Johnson, K.R., Erway, L.C., Cook, S.A., Willott, J.F., and Zheng, Q.Y. (1997). A major gene affecting age-related hearing loss in C57BL/6J mice. Hearing Research 114, 83-92. https://doi.org/10.1016/S0378-5955(97)00155-X.

    (2) Minor point - I do not agree with the use of the term "ventral to bregma" to describe where the craniotomies were placed (e.g., line 599). The direction being described is more typically referred to as "lateral." If the authors prefer to use the term "ventral," perhaps additional clarification can be added.

    We thank the reviewer for pointing out this issue and apologize for any confusion. We agree that “ventral to bregma” is not the standard terminology and have revised the Methods section to use “below the temporal ridge”. We have also clarified that the craniotomy for accessing the auditory cortex was performed on the lateral aspect of the skull in rodents, just below the temporal ridge. We hope this revision resolves the ambiguity.

    Method section:

    “A craniotomy was performed over the temporal bone, as the auditory cortex is located on the lateral surface of the brain (coordinates: 1.5 to 3.0 mm below the temporal ridge and 2.0 to 4.0 mm posterior to bregma for mice; 2.5 to 6.5 mm below the temporal ridge and 3.0 to 5.0 mm posterior to bregma for rats) to access the auditory cortex.”

    “Six-week after CCK-sensor virus injection, a craniotomy was performed to access the auditory cortex at the temporal bone (1.5 to 3.0 mm below the temporal ridge and 2.0 to 4.0 mm posterior to bregma), and the dura mater was opened.”

  6. Version published to 10.7554/elife.101513.3 on eLife
  7. Version published to 10.7554/elife.101513 on eLife
  8. Version published to 10.7554/elife.101513.2 on eLife
  9. eLife

    eLife Assessment

    This is an important study demonstrating that cholecystokinin is a key modulator of auditory thalamocortical plasticity during development and in young adult but not aged mice, though cortical application of this neuropeptide in older animals appears to go some way to restoring this age-dependent loss in plasticity. A strength of this work is the use of multiple experimental approaches, which together provide convincing support for the proposed involvement of cholecystokinin. While there are some limitations in the electrophysiological recordings and areas where the manuscript would benefit from further clarification, this work is likely to be influential in opening up a new avenue of investigation into the roles of neuropeptides in sensory plasticity.

  10. eLife

    Reviewer #1 (Public review):

    This report addresses a compelling topic. However, I have significant concerns, which necessitate a reassessment of the report's overall value.

    Anatomical Specificity and Stimulation Site:
    While the authors clarify that the ventral MGB (MGv) was the intended stimulation target, the electrode track (Fig. 1A) and viral spread (Fig. 2E) suggest possible involvement of the dorsal MGB (MGd) and broader area. Given that MGv-AI and MGd-AC pathways have distinct-and sometimes opposing-effects on plasticity, the reported LTP values (with unusually small standard deviations) raise concerns about the specificity of the findings. Additional anatomical verification would help resolve this issue.

    Statistical Rigor and Data Variability:
    The remarkably low standard deviations in LTP measurements are unexpected based on established variability in thalamocortical plasticity. The authors' response confirms these values are accurate, but further justification, such as methodological controls or replication-would bolster confidence in these results. Additionally, the comparison of in vivo vs. in vitro LTP variability requires more substantive support.

    Viral Targeting and Specificity:
    The manuscript does not clearly address whether cortical neurons were inadvertently infected by AAV9. Given the potential for off-target effects, explicit confirmation (e.g., microphotograph of stimulation site) would strengthen the study's conclusions.

    Integration of Prior Literature:
    The discussion of existing work is adequate but could be more comprehensive. A deeper engagement with contrasting findings would provide better context for the study's contributions.

    Therapeutic Implications:
    The authors' discussion of therapeutic potential is now appropriately cautious and well-reasoned.

    Conclusion:
    While the study presents intriguing findings, the concerns outlined above must be addressed to fully establish the validity and impact of the results. I appreciate the authors' efforts thus far and hope they can provide additional data or clarification to resolve these issues. With these revisions, the manuscript could make a valuable contribution to the field.

  11. eLife

    Reviewer #2 (Public review):

    Summary:

    This work used multiple approaches to show that CCK is critical for long-term potentiation (LTP) in the auditory thalamocortical pathway. They also showed that the CCK mediation of LTP is age-dependent and supports frequency discrimination. This work is important because is opens up a new avenue of investigation of the roles of neuropeptides in sensory plasticity.

    Strengths:

    The main strength is the multiple approaches used to comprehensively examine the role of CCK in auditory thalamocortical LTP. Thus, the authors do provide a compelling set of data that CCK mediates thalamocortical LTP in an age-dependent manner.

    Weaknesses:

    There are some details that should be addressed, primarily regarding potential baseline differences in comparison groups. The behavioral assessment is relatively limited, but may be fleshed out in future work.

  12. eLife

    Reviewer #3 (Public review):

    Summary:

    Cholecystokinin (CCK) is highly expressed in auditory thalamocortical (MGB) neurons and CCK has been found to shape cortical plasticity dynamics. In order to understand how CCK shapes synaptic plasticity in the auditory thalamocortical pathway, they assessed the role of CCK signaling across multiple mechanisms of LTP induction with the auditory thalamocortical (MGB - layer IV Auditory Cortex) circuit in mice. In these physiology experiments that leverage multiple mechanisms of LTP induction and a rigorous manipulation of CCK and CCK-dependent signaling, they establish an essential role of auditory thalamocortical LTP on the co-release of CCK from auditory thalamic neurons. By carefully assessing the development of this plasticity over time and CCK expression, they go on to identify a window of time that CCK is produced throughout early and middle adulthood in auditory thalamocortical neurons to establish a window for plasticity from 3 weeks to 1.5 years in mice, with limited LTP occurring outside of this window. The authors go on to show that CCK signaling and its effect on LTP in the auditory cortex is also capable of modifying frequency discrimination accuracy in an auditory PPI task. In evaluating the impact of CCK on modulating PPI task performance, it also seems that in mice <1.5 years old CCK-dependent effects on cortical plasticity is almost saturated. While exogenous CCK can modestly improve discrimination of only very similar tones, exogenous focal delivery of CCK in older mice can significantly improve learning in a PPI task to bring their discrimination ability in line with those from young adult mice.

    Strengths:

    (1) The clarity of the results, along with the rigor multi-angled approach, provide significant support for the claim that CCK is essential for auditory thalamocortical synaptic LTP. This approach uses a combination of electrical, acoustic, and optogenetic pathway stimulation alongside conditional expression approaches, germline knockout, viral RNA downregulation and pharmacological blockade. Through the combination of these experimental configures the authors demonstrate that high-frequency stimulation-induced LTP is reliant on co-release of CCK from glutamatergic MGB terminals projecting to the auditory cortex.

    (2) The careful analysis of the CCK, CCKB receptor, and LTP expression is also a strength that puts the finding into the context of mechanistic causes and potential therapies for age-dependent sensory/auditory processing changes. Similarly, not only do these data identify a fundamental biological mechanism, but they also provide support for the idea that exogenous asynchronous stimulation of the CCKBR is capable of restoring an age-dependent loss in plasticity.

    (3) Although experiments to simultaneously relate LTP and behavioral change or identify a causal relationship between LTP and frequency discrimination are not made, there is still convincing evidence that CCK signaling in the auditory cortex (known to determine synaptic LTP) is important for auditory processing/frequency discrimination. These experiments are key for establishing the relevance of this mechanism.

    Weaknesses:

    (1) Given the magnitude of the evoked responses, one expects that pyramidal neurons in layer IV are primarily those that undergo CCK-dependent plasticity, but the degree to which PV-interneurons and pyramidal neurons participate in this process differently is unclear.

    (2) While these data support an important role for CCK in synaptic LTP in the auditory thalamocortical pathway, perhaps temporal processing of acoustic stimuli is as or more important than frequency discrimination. Given the enhanced responsivity of the system, it is unclear whether this mechanism would improve or reduce the fidelity of temporal processing in this circuit. Understanding this dynamic may also require consideration of cell type as raised in weakness #1.

    (3) In Figure 1, an example of increased spontaneous and evoked firing activity of single neurons after HFS is provided. Yet it is surprising that the group data are analyzed only for the fEPSP. It seems that single neuron data would also be useful at this point to provide insight into how CCK and HFS affect temporal processing and spontaneous activity/excitability, especially given the example in 1F.

    (4) The circuitry that determines PPI requires multiple brain areas, including the auditory cortex. Given the complicated dynamics of this process, it may be helpful to consider what, if anything, is known specifically about how layer IV synaptic plasticity in the auditory cortex may shape this behavior.

    Comments on revisions:

    The manuscript is much improved and many of the issues or questions have been addressed. Ideally, evidence for the degree of transsynaptic spread for AAV9-Syn-ChrimsonR-tdTomato would also be provided in some form since in the authors' response in sounds like some was observed, as expected.

  13. eLife

    Author response:

    The following is the authors’ response to the original reviews

    Public Reviews:

    Reviewer #1 (Public review):

    This study offers a valuable investigation into the role of cholecystokinin (CCK) in thalamocortical plasticity during early development and adulthood, employing a range of experimental techniques. The authors demonstrate that tetanic stimulation of the auditory thalamus induces cortical long-term potentiation (LTP), which can be evoked through either electrical or optical stimulation of the thalamus or by noise bursts. They further show that thalamocortical LTP is abolished when thalamic CCK is knocked down or when cortical CCK receptors are blocked. Interestingly, in 18-month-old mice, thalamocortical LTP was largely absent but could be restored through the cortical application of CCK. The authors conclude that CCK contributes to thalamocortical plasticity and may enhance thalamocortical plasticity in aged subjects.

    While the study presents compelling evidence, I would like to offer several suggestions for the authors' consideration:

    (1) Thalamocortical LTP and NMDA-Dependence:

    It is well established that thalamocortical LTP is NMDA receptor-dependent, and blocking cortical NMDA receptors can abolish LTP. This raises the question of why thalamocortical LTP is eliminated when thalamic CCK is knocked down or when cortical CCK receptors are blocked. If I correctly understand the authors' hypothesis - that CCK promotes LTP through CCKR-intracellular Ca2+-AMPAR. This pathway should not directly interfere with the NMDA-dependent mechanism. A clearer explanation of this interaction would be beneficial.

    Thank you for your question regarding the role of CCK and NMDA receptors (NMDARs) in thalamocortical LTP. We propose that CCK receptor (CCKR) activation enhances intracellular calcium levels, which are crucial for thalamocortical LTP induction. Calcium influx through NMDARs is also essential to reach the threshold required for activating downstream signaling pathways that promote LTP (Heynen and Bear, 2001). Thus, CCKRs and NMDARs may function in a complementary manner to facilitate LTP, with both contributing to the elevation of intracellular calcium.

    However, it is important to note that the postsynaptic mechanisms of thalamocortical LTP in the auditory cortex (ACx) differ from those in other sensory cortices. Studies have shown that thalamocortical LTP in the ACx appears to be less dependent on NMDARs (Chun et al., 2013), which is distinct from somatosensory or visual cortices. Our previous studies also found that while NMDAR antagonists can block HFS-induced LTP in the inner ACx, LTP can still be induced in the presence of CCK even after the NMDARs blockade (Chen et al. 2019). These findings suggest that CCK may act through an alternative mechanism involving CCKR-mediated calcium signaling and AMPAR modulation, which partially compensates for the loss of NMDAR signaling. This distinction may reflect functional differences between the ACx and other sensory cortices, as highlighted in previous studies (King and Nelken, 2009).

    While our current study focuses on the role of CCKR-mediated plasticity in the auditory system, further investigations are needed to elucidate how CCKRs and NMDARs interact within the broader framework of thalamocortical neuroplasticity across different cortical regions. Understanding whether similar mechanisms operate in other sensory systems, such as the visual cortex, will be an important direction for future research.

    Heynen, A.J., and Bear, M.F. (2001). Long-term potentiation of thalamocortical transmission in the adult visual cortex in vivo. J Neurosci 21, 9801-9813. 10.1523/jneurosci.21-24-09801.2001.

    Chun, S., Bayazitov, I.T., Blundon, J.A., and Zakharenko, S.S. (2013). Thalamocortical Long-Term Potentiation Becomes Gated after the Early Critical Period in the Auditory Cortex. The Journal of Neuroscience 33, 7345-7357. 10.1523/jneurosci.4500-12.2013.

    Chen, X., Li, X., Wong, Y.T., Zheng, X., Wang, H., Peng, Y., Feng, H., Feng, J., Baibado, J.T., Jesky, R., et al. (2019). Cholecystokinin release triggered by NMDA receptors produces LTP and sound-sound associative memory. Proc Natl Acad Sci U S A 116, 6397-6406. 10.1073/pnas.1816833116.

    King, A. J., & Nelken, I. (2009). Unraveling the principles of auditory cortical processing: can we learn from the visual system? Nature neuroscience, 12(6), 698-701.

    (2) Complexity of the Thalamocortical System:

    The thalamocortical system is intricate, with different cortical and thalamic subdivisions serving distinct functions. In this study, it is not fully clear which subdivisions were targeted for stimulation and recording, which could significantly influence the interpretation of the findings. Clarifying this aspect would enhance the study's robustness.

    Thank you for your valuable feedback. We would like to clarify that stimulation was conducted in the medial geniculate nucleus ventral (MGv), and recording was performed in layer IV of the ACx. Targeting the MGv allows us to investigate the influence of thalamic inputs on auditory cortical responses. Layer IV of the ACx is known to receive direct thalamic projections, making it an ideal site for assessing how thalamic activity influences cortical processing. We will incorporate this clarification into the revised manuscript to enhance the robustness of our study.

    Results section:

    “Stimulation electrodes were placed in the MGB (specifically in the medial geniculate nucleus ventral subdivision, MGv), and recording electrodes were inserted into layer IV of ACx”

    “The recording electrodes were lowered into layer IV of ACx, while the stimulation electrodes were lowered into MGB (MGv subdivision). The final stimulating and recording positions were determined by maximizing the cortical fEPSP amplitude triggered by the ES in the MGB. The accuracy of electrode placement was verified through post-hoc histological examination and electrophysiological responses.”

    (3) Statistical Variability:

    Biological data, including field excitatory postsynaptic potentials (fEPSPs) and LTP, often exhibit significant variability between samples, sometimes resulting in a standard deviation that exceeds 50% of the mean value. The reported standard deviation of LTP in this study, however, appears unusually small, particularly given the relatively limited sample size. Further discussion of this observation might be warranted.

    Thank you for your question. In our experiments, the sample size N represents the number of animals used, while n refers to the number of recordings, with each recording corresponding to a distinct stimulation and recording sites. To adhere to ethical guidelines and minimize animal usage, we often perform multiple recordings within a single animal, such as from different hemispheres of the brain. Although N may appear small, our statistical analyses are based on n, ensuring sufficient data points for reliable conclusions.

    Furthermore, as our experiments are conducted in vivo, we observe lower variability in the increase of fEPSP slopes following LTP induction compared to brain slice preparations, where standard deviations exceeding 50% of the mean are common. This reduced variability likely reflects the robustness of the physiologically intact conditions in the in vivo setup.

    (4) EYFP Expression and Virus Targeting:

    The authors indicate that AAV9-EFIa-ChETA-EYFP was injected into the medial geniculate body (MGB) and subsequently expressed in both the MGB and cortex. If I understand correctly, the authors assume that cortical expression represents thalamocortical terminals rather than cortical neurons. However, co-expression of CCK receptors does not necessarily imply that the virus selectively infected thalamocortical terminals. The physiological data regarding cortical activation of thalamocortical terminals could be questioned if the cortical expression represents cortical neurons or both cortical neurons and thalamocortical terminals.

    Thank you for your question. In Figure 2A, EYFP expression indicates thalamocortical projections, while the co-expression of EYFP with PSD95 confirms the identity of thalamocortical terminals. The CCK-B receptors (CCKBR) are located on postsynaptic cortical neurons. The observed co-labeling of thalamocortical terminals and postsynaptic CCKBR suggests that CCK-expressing neurons in the medial geniculate body (MGB) can release CCK, which subsequently acts on the postsynaptic CCKBR. This evidence supports our interpretation of the functional role of CCK modulating neural plasticity between thalamocortical inputs and cortical neurons. As shown in Figure 2A, we aim to demonstrate that the co-labeling of thalamocortical terminals with CCK receptors accounts for a substantial proportion of the thalamocortical terminals. We will ensure that this clarification is emphasized in the revised manuscript to address your concerns.

    Results section:

    “Cre-dependent AAV9-EFIa-DIO-ChETA-EYFP was injected into the MGB of CCK-Cre mice. EYFP labeling marked CCK-positive neurons in the MGB. The co-expression of EYFP thalamocortical projections with PSD95 confirms the identity of thalamocortical terminals (yellow), which primarily targeted layer IV of the ACx (Figure 2A, upper panel). Immunohistochemistry revealed that a substantial proportion (15 out of 19, Figure 2A lower right panel) of thalamocortical terminals (arrows) colocalize with CCK receptors (CCKBR) on postsynaptic cortical neurons in the ACx (Figure 2A lower panel), supporting the functional role of CCK in modulating thalamocortical plasticity.”

    (5) Consideration of Previous Literature:

    A number of studies have thoroughly characterized auditory thalamocortical LTP during early development and adulthood. It may be beneficial for the authors to integrate insights from this body of work, as reliance on data from the somatosensory thalamocortical system might not fully capture the nuances of the auditory pathway. A more comprehensive discussion of the relevant literature could enhance the study's context and impact.

    Thank you for your valuable feedback. We will enhance our discussion on auditory thalamocortical LTP during early development and adulthood to provide a more comprehensive context for our study.

    (6) Therapeutic Implications:

    While the authors suggest potential therapeutic applications of their findings, it may be somewhat premature to draw such conclusions based on the current evidence. Although speculative discussion is not harmful, it may not significantly add to the study's conclusions at this stage.

    Thank you for your thoughtful feedback. We agree that the therapeutic applications mentioned in our study are speculative at this stage and should be regarded as a forward-looking perspective rather than definitive conclusions. Our intention was to highlight the broader potential of our findings to inspire further research, rather than to propose immediate clinical applications.

    In light of your feedback, we have adjusted the language in the manuscript to reflect a more cautious interpretation. Speculative discussions are now explicitly framed as hypotheses or possibilities for future exploration. We emphasize that our findings provide a foundation for further investigations into CCK-based plasticity and its implications.

    We believe that appropriately framed forward-thinking discussions are valuable in guiding the direction of future research. We sincerely hope that our current and future work will contribute to a deeper understanding of thalamocortical plasticity and, over time, potentially lead to advancements in human health.

    Reviewer #2 (Public review):

    Summary:

    This work used multiple approaches to show that CCK is critical for long-term potentiation (LTP) in the auditory thalamocortical pathway. They also showed that the CCK mediation of LTP is age-dependent and supports frequency discrimination. This work is important because it opens up a new avenue of investigation of the roles of neuropeptides in sensory plasticity.

    Strengths:

    The main strength is the multiple approaches used to comprehensively examine the role of CCK in auditory thalamocortical LTP. Thus, the authors do provide a compelling set of data that CCK mediates thalamocortical LTP in an age-dependent manner.

    Weaknesses:

    The behavioral assessment is relatively limited but may be fleshed out in future work.

    Reviewer #3 (Public review):

    Summary:

    Cholecystokinin (CCK) is highly expressed in auditory thalamocortical (MGB) neurons and CCK has been found to shape cortical plasticity dynamics. In order to understand how CCK shapes synaptic plasticity in the auditory thalamocortical pathway, they assessed the role of CCK signaling across multiple mechanisms of LTP induction with the auditory thalamocortical (MGB - layer IV Auditory Cortex) circuit in mice. In these physiology experiments that leverage multiple mechanisms of LTP induction and a rigorous manipulation of CCK and CCK-dependent signaling, they establish an essential role of auditory thalamocortical LTP on the co-release of CCK from auditory thalamic neurons. By carefully assessing the development of this plasticity over time and CCK expression, they go on to identify a window of time that CCK is produced throughout early and middle adulthood in auditory thalamocortical neurons to establish a window for plasticity from 3 weeks to 1.5 years in mice, with limited LTP occurring outside of this window. The authors go on to show that CCK signaling and its effect on LTP in the auditory cortex is also capable of modifying frequency discrimination accuracy in an auditory PPI task. In evaluating the impact of CCK on modulating PPI task performance, it also seems that in mice <1.5 years old CCK-dependent effects on cortical plasticity are almost saturated. While exogenous CCK can modestly improve discrimination of only very similar tones, exogenous focal delivery of CCK in older mice can significantly improve learning in a PPI task to bring their discrimination ability in line with those from young adult mice.

    Strengths:

    (1) The clarity of the results along with the rigor multi-angled approach provide significant support for the claim that CCK is essential for auditory thalamocortical synaptic LTP. This approach uses a combination of electrical, acoustic, and optogenetic pathway stimulation alongside conditional expression approaches, germline knockout, viral RNA downregulation, and pharmacological blockade. Through the combination of these experimental configures the authors demonstrate that high-frequency stimulation-induced LTP is reliant on co-release of CCK from glutamatergic MGB terminals projecting to the auditory cortex.

    (2) The careful analysis of the CCK, CCKB receptor, and LTP expression is also a strength that puts the finding into the context of mechanistic causes and potential therapies for age-dependent sensory/auditory processing changes. Similarly, not only do these data identify a fundamental biological mechanism, but they also provide support for the idea that exogenous asynchronous stimulation of the CCKBR is capable of restoring an age-dependent loss in plasticity.

    (3) Although experiments to simultaneously relate LTP and behavioral change or identify a causal relationship between LTP and frequency discrimination are not made, there is still convincing evidence that CCK signaling in the auditory cortex (known to determine synaptic LTP) is important for auditory processing/frequency discrimination. These experiments are key for establishing the relevance of this mechanism.

    Weaknesses:

    (1) Given the magnitude of the evoked responses, one expects that pyramidal neurons in layer IV are primarily those that undergo CCK-dependent plasticity, but the degree to which PV-interneurons and pyramidal neurons participate in this process differently is unclear.

    Thank you for this insightful comment. We agree that the differential roles of PV-interneurons and pyramidal neurons in CCK-dependent thalamocortical plasticity remain unclear and acknowledge this as an important limitation of our study. Our primary focus was on pyramidal neurons, as our in vivo electrophysiological recordings measured the fEPSP slope in layer IV of the auditory cortex, which primarily reflects excitatory synaptic activity. However, we recognize the critical role of the excitatory-inhibitory balance in cortical function and the potential contribution of PV-interneurons to this process. In future studies, we plan to utilize techniques such as optogenetics, two-photon calcium imaging and cell-type-specific recordings to investigate the distinct contributions of PV-interneurons and pyramidal neurons to CCK-dependent thalamocortical plasticity, thereby providing a more comprehensive understanding of how CCK modulates thalamocortical circuits.

    (2) While these data support an important role for CCK in synaptic LTP in the auditory thalamocortical pathway, perhaps temporal processing of acoustic stimuli is as or more important than frequency discrimination. Given the enhanced responsivity of the system, it is unclear whether this mechanism would improve or reduce the fidelity of temporal processing in this circuit. Understanding this dynamic may also require consideration of cell type as raised in weakness #1.

    Thank you for this thoughtful comment. We acknowledge that our study did not directly address the fidelity of temporal processing, which is indeed a critical aspect of auditory function. Our behavioral experiments primarily focused on linking frequency discrimination to the role of CCK in synaptic strengthening within the auditory thalamocortical pathway. However, we agree that enhanced responsivity of the system could also impact temporal processing dynamics, such as the precise timing of auditory responses. Whether this modulation improves or reduces the fidelity of temporal processing remains an open and important question.

    As you noted, understanding these dynamics will require a deeper investigation into the interactions between different cell types, particularly the balance between excitatory and inhibitory neurons. Exploring how CCK modulation affects both the circuit and cellular levels in temporal processing is an important direction for future research, which we plan to pursue. Thank you again for raising this important point.

    Disscusion section:

    “While we focused on homosynaptic plasticity at thalamocortical synapses by recording only fEPSPs in layer IV of ACx, it is essential to further explore heterosynaptic effects of CCK released from thalamocortical synapses on intracortical circuits, particularly its role in modulating the excitatory-inhibitory balance. PV-interneurons, as key regulators of cortical inhibition, may contribute to the temporal fidelity of sensory processing, which is critical for auditory perception (Nocon et al., 2023; Cai et al., 2018). Additionally, CCK may facilitate cross-modal plasticity by modulating heterosynaptic plasticity in interconnected cortical areas. Future studies would provide valuable insights into the broader role of CCK in shaping sensory processing and cortical network dynamics.”

    Nocon, J.C., Gritton, H.J., James, N.M., Mount, R.A., Qu, Z., Han, X., and Sen, K. (2023). Parvalbumin neurons enhance temporal coding and reduce cortical noise in complex auditory scenes. Communications Biology 6, 751. 10.1038/s42003-023-05126-0.

    Cai, D., Han, R., Liu, M., Xie, F., You, L., Zheng, Y., Zhao, L., Yao, J., Wang, Y., Yue, Y., et al. (2018). A Critical Role of Inhibition in Temporal Processing Maturation in the Primary Auditory Cortex. Cereb Cortex 28, 1610-1624. 10.1093/cercor/bhx057.

    (3) In Figure 1, an example of increased spontaneous and evoked firing activity of single neurons after HFS is provided. Yet it is surprising that the group data are analyzed only for the fEPSP. It seems that single-neuron data would also be useful at this point to provide insight into how CCK and HFS affect temporal processing and spontaneous activity/excitability, especially given the example in 1F.

    Thank you for your insightful comment. In our in vivo electrophysiological experiments on LTP induction, we recorded neural activity for over 1.5 hours to assess changes in neuronal responses over time, both prior to and following the induction. While single neuron firing data can provide valuable insights, such measurements are inherently more variable due to factors like cortical state fluctuations and the condition of nearby neurons, which makes them less reliable for long-term analysis. For this reason, we focused on fEPSP, as it offers a more stable and robust readout of synaptic activity over extended periods.

    We appreciate your suggestion and recognize the value of single-neuron data in understanding how CCK and HFS affect temporal processing and excitability. In future studies, we will consider to incorporate single-neuron analyses to complement our synaptic-level findings and provide a more comprehensive understanding of these mechanisms.

    (4) The authors mention that CCK mRNA was absent in CCK-KO mice, but the data are not provided.

    Thank you for your comment. Data from the CCK-KO mice are presented in Figure 3A (far right) and in the upper panel of Figure 3B (far right). In the lower panel of Figure 3B, data from the CCK-KO group are not shown because the normalized values for this group were essentially zero, as expected due to the absence of CCK mRNA.

    (5) The circuitry that determines PPI requires multiple brain areas, including the auditory cortex. Given the complicated dynamics of this process, it may be helpful to consider what, if anything, is known specifically about how layer IV synaptic plasticity in the auditory cortex may shape this behavior.

    Thank you for raising this important point. Pre-pulse inhibition (PPI) of the acoustic startle response indeed involves multiple brain regions, with the ascending auditory pathway playing a key role (Gómez-Nieto et al., 2020). Within the auditory cortex, layer IV neurons receive tonotopically organized inputs from the medial geniculate nucleus and are critical for integrating thalamic inputs and shaping auditory processing.

    In our behavioral experiments, mice were required to discriminate pre-pulses of varying frequencies against a continuous background sound. Given the role of auditory cortical neurons in integrating thalamic inputs and shaping auditory processing, it is likely that synaptic plasticity in these neurons contributes to the enhanced discrimination of pre-pulses. Supporting this idea, our previous work demonstrated that local infusion of CCK, paired with weak acoustic stimuli, significantly increased auditory responses in the auditory cortex (Li et al., 2014). In the current study, we further showed that CCK release during high-frequency stimulation of the thalamocortical pathway induced LTP in layer IV of the auditory cortex. Together, these findings suggest that CCK-dependent synaptic plasticity in layer IV may amplify the cortical representation of weak auditory inputs, thereby improving pre-pulses detection and enhancing PPI performance.

    It is also worth noting that aged mice with hearing loss typically exhibit PPI deficits due to impaired auditory processing (Ouagazzal et al., 2006 and Young et al., 2010). We propose that enhanced plasticity in the thalamocortical pathway, mediated by CCK, might partially compensate for these deficits by amplifying residual auditory signals in aged mice. However, the precise mechanisms by which layer IV synaptic plasticity modulates PPI behavior remain to be fully understood. Given the complex dynamics of sensory processing, future studies could explore how layer IV neurons interact with other cortical and subcortical circuits involved in PPI, as well as the specific contributions of excitatory and inhibitory cell types. These investigations will help provide a more comprehensive understanding of the role of CCK in modulating sensory gating and auditory processing.

    Gómez-Nieto, R., Hormigo, S., & López, D. E. (2020). Prepulse inhibition of the auditory startle reflex assessment as a hallmark of brainstem sensorimotor gating mechanisms. Brain sciences, 10(9), 639.

    Li, X., Yu, K., Zhang, Z., Sun, W., Yang, Z., Feng, J., Chen, X., Liu, C.-H., Wang, H., Guo, Y.P., and He, J. (2014). Cholecystokinin from the entorhinal cortex enables neural plasticity in the auditory cortex. Cell Research 24, 307-330. 10.1038/cr.2013.164.

    Ouagazzal, A. M., Reiss, D., & Romand, R. (2006). Effects of age-related hearing loss on startle reflex and prepulse inhibition in mice on pure and mixed C57BL and 129 genetic background. Behavioural brain research, 172(2), 307-315.

    Young, J. W., Wallace, C. K., Geyer, M. A., & Risbrough, V. B. (2010). Age-associated improvements in cross-modal prepulse inhibition in mice. Behavioral neuroscience, 124(1), 133.

    Recommendations for the authors:

    Reviewer #2 (Recommendations for the authors):

    Major concerns:

    (1) In Figure 1, the authors used different metrics for fEPSP strength. In Figure 1D, the authors used the slope, while they used the amplitude in Figure 1G. It is known that the two metrics are different from each other. While the slope is calculated from the linear regression between the voltage change per time of the rising phase of the fEPSP, the amplitude represents the voltage value of the fEPSP's peak. Please clarify here and in the method what metric you used, because the two terms are not interchangeable.

    Thank you for pointing out this oversight in our manuscript. We confirm that we used the slope of the fEPSP as the metric for assessing synaptic strength throughout the study, including both Figure 1D and Figure 1G. We will make the necessary corrections to ensure clarity and consistency. Thank you for bringing this to our attention.

    (2) It is not mentioned in the details of the methods about the CCK-KO mice. Please give such details. Although the authors used the CCK-KO mouse model as a control, I think that it is not a good choice to test the hypothesis mentioned in lines 165 and 166. The experiment was supposed to monitor the CCK-BR activity after HFS of the MGB and answer whether the CCK-BR will get activated by thalamic stimulation, but the CCK-KO mouse does not have CCK to be released after the optogenetic activation of the Chrimson probe. Therefore, it is expected to give nothing as if the experimenter runs an experiment without intervention. I think that the appropriate way to examine the hypothesis is to compare mice that were either injected with AAV9-Syn-FLEX-ChrimsonR-tdTomato or AAV9-Syn-FLEX-tdTomato. However, CCK-OK would be a perfect model to confirm that LTP can be only generated dependently on CCK, by simply running the HFS of the MGB that would be associated with the cortical recording of the fEPSP. This also will rule out the assumption that the authors mentioned in lines 191 and 192.

    Thank you for your valuable feedback. The rationale behind our experimental design was to validate the newly developed CCK sensor and confirm its specificity. We aimed to verify CCK release post-HFS by comparing the responses of the CCK sensor in CCK-KO mice and CCK-Cre mice. This comparison allowed us to determine that the observed increase in fluorescence intensity post-HFS was specifically due to CCK release, rather than other neurotransmitters induced by HFS.

    We appreciate your suggestion to compare mice injected with AAV9-Syn-FLEX-ChrimsonR-tdTomato and AAV9-Syn-FLEX-tdTomato, as it is indeed a valuable approach for directly testing the hypothesis regarding CCK-BR activation. However, we prioritized using the CCK-KO model to validate the CCK sensor's efficacy and specificity. The validation can be inferred by comparing the CCK sensor activity before and after HFS.

    Regarding concerns mentioned in lines 191 and 192 about potential CCK release from other projections via indirect polysynaptic activation, CCK-KO mice were not suitable for this aspect due to their global knockout of CCK. To address this limitation, we utilized shRNA to specifically down-regulate Cck expression in MGB neurons. This approach focused on the necessity of CCK released from thalamocortical projections for the observed LTP and effectively ruled out the possibility of indirect polysynaptic activation.

    We also acknowledge that the methods section lacked sufficient details about the CCK-KO mice, which may have caused confusion. In the revised methods section, we will add the following details:

    (1) The genotype of the CCK-KO mice used in this study (CCK-ires-CreERT2, Jax#012710).

    (2) A brief description of the CCK-KO validation, emphasizing the absence of CCK mRNA in these mice (as shown in Figure 3A and 3B).

    (3) The experimental purpose of using CCK-KO mice to validate the specificity of the CCK sensor.

    We believe these additions will clarify the rationale for using CCK-KO mice and their role in this study. Thank you again for highlighting these important points.

    (3) Figure 3C: The authors should examine if there is a difference in the baseline of fEPSPs across different age groups as the dependence on the normalization in the analysis within each group would hide if there were any difference of the baseline slope of fEPSP between groups which could be related to any misleading difference after HFS. Also, I wonder about the absence of LTP in P20, which is a closer age to the critical period. Could the authors discuss that, please?

    Thank you for your insightful feedback. To address your concern regarding baseline differences in fEPSP slopes across age groups, we conducted additional analysis. Baseline fEPSP across the three groups (P20, 8w, 18m), normalized to the 8w group, were 64.8± 13.1%, 100.0 ± 20.4%, and 58.8± 10.3%, respectively. While there was a trend suggesting smaller fEPSP slopes in the P20 and 18m groups compared to the young adult group, these differences were not statistically significant due to data variability (P20 vs. 8w, P = 0.319; 8w vs. 18m, P=0.147; P20 vs. 18m, P = 1.0, one-way ANOVA). These results suggest that baseline variability is unlikely to confound the observed differences in LTP after HFS. Furthermore, we ensured that normalization minimized any potential baseline effects.

    Regarding the absence of LTP in P20, this likely reflects developmental regulation of CCKBR expression in the auditory cortex (ACx). The HFS-induced thalamocortical LTP observed in our study is CCK-dependent and mechanistically distinct from the NMDA-dependent thalamocortical LTP during the critical period. Specifically, correlated pre- and postsynaptic activity can induce NMDA-dependent thalamocortical LTP only during an early critical period corresponding to the first several postnatal days, after which this pairing becomes ineffective starting from the second postnatal week (Crair and Malenka, 1995; Isaac et al., 1997; Chun et al., 2013). In contrast, the CCK-dependent Thalamocortical LTP induced by HFS is robust in adult mice but appears absent in P20, likely due to the lack of postsynaptic CCKBR expression in the ACx at this developmental stage.

    We will include these clarifications in the revised manuscript, particularly in the Discussion section, to provide a more comprehensive explanation of our findings. Thank you for your valuable comments and suggestions.

    Crair, M.C., and Malenka, R.C. (1995). A critical period for long-term potentiation at thalamocortical synapses. Nature 375, 325-328. 10.1038/375325a0.

    Isaac, J.T.R., Crair, M.C., Nicoll, R.A., and Malenka, R.C. (1997). Silent Synapses during Development of Thalamocortical Inputs. Neuron 18, 269-280. https://doi.org/10.1016/S0896-6273(00)80267-6.

    Chun, S., Bayazitov, I.T., Blundon, J.A., and Zakharenko, S.S. (2013). Thalamocortical Long-Term Potentiation Becomes Gated after the Early Critical Period in the Auditory Cortex. The Journal of Neuroscience 33, 7345-7357. 10.1523/jneurosci.4500-12.2013.

    (4) Figure 4F: It is noticed that the baseline fEPSP of the CCK group and ACSF groups were different, which raises a concern about the baseline differences between treatment groups.

    Thank you for your valuable feedback and for pointing out this important detail. We apologize for any confusion caused by the presentation of the data. As noted in the figure legend, the scale bars for the fEPSPs were different between the left (0.1 mV) and right panels (20 µV). This difference in scale may have created the perception of baseline differences between the CCK and ACSF groups. To enhance clarity and avoid potential misunderstanding, we will unify the scale bar values in the revised figure. This adjustment will provide a clearer and more accurate comparison of fEPSPs between groups. Thank you again for bringing this issue to our attention.

    (5) From Figure S2D, it seems that different animals were injected with the drug and ACSF. Therefore, how the authors validate the position of the recording electrode to the cortical area of certain CF and relative EF. Also, there is not enough information about the basis of the selection of the EF. Should it be lower than the CF with a certain value? Was the EF determined after the initial tuning curve in each case? To mitigate this difference, it would be appropriate if the authors examined the presence of a significant difference in the tuning width and CFs between animals exposed to ACSF and CCK-4. This will give some validation of a balanced experiment between ACSF and CCK-4. I wonder also why the authors used rats here not mice, as it will be easier to interpret the results came from the same species.

    Thank you for your thoughtful comments. The effective frequency (EF) was determined after measuring the initial tuning curve for each case. The EF was selected to elicit a clear sound response while maintaining a sufficient distance from the characteristic frequency (CF) to allow measurable increases in response intensity. Specifically, EF was selected based on the starting point of the tuning peak, which corresponds to the onset of its fastest rising phase. From this point, EF was determined by moving 0.2 or 0.4 octaves toward the CF. While there were individual differences in EF selection among animals, the methodology for determining EF was standardized and applied consistently across both the ACSF and CCK-4 groups.

    Regarding the use of rats in these experiments, these studies were conducted prior to our current work with mice. The findings in rat provide valuable insights that support our current results in mice. Since the rat data are supplementary to the primary findings, we included them as supplementary material to provide additional context and validation. Furthermore, in consideration of animal welfare, we chose not to replicate these experiments in mice, as the findings from rats were sufficient to support our conclusions.

    Methods section:

    “The tuning curve was determined by plotting the lowest intensity at which the neuron responded to different tones. The characteristic frequency (CF) is defined as the frequency corresponding to the lowest point on this curve. The effective frequency (EF) was determined to elicit a clear sound response while maintaining a sufficient distance from the CF to allow measurable increases in response intensity. Specifically, EF was selected based on the starting point of the tuning peak, which corresponds to the onset of its fastest rising phase. From this point, EF was determined by moving 0.2 or 0.4 octaves toward the CF.”

    (6) Lines 384-386: There are no figures named 5H and I.

    Thank you for pointing this out. The references to Figures 5H and 5I were incorrect and should have referred to Figures 5C and 5D. We sincerely apologize for this oversight and will correct these errors in the revised manuscript to ensure clarity and accuracy. Thank you again for bringing this to our attention.

    (7) The authors should mention the sex of the animals used.

    Thank you for your comment and for highlighting this important detail. The sex of the animals used in this study is specified in the Animals section of the Methods: "In the present study, male mice and rats were used to investigate thalamocortical LTP." We appreciate your careful attention to this point and will ensure that this detail remains clearly stated in the manuscript.

    (8) Lines 534 and 648: These coordinates are difficult to understand. Since the experiment was done on both mice and rats, we need a clear description of the coordinates in both. Also, I think that you should mention the lateral distance from the sagittal suture as the ventral coordinates should be calculated from the surface of the skull above the AC and not from the sagittal suture.

    Thank you for your valuable feedback and for pointing out this important issue. We apologize for any confusion caused by our description of the coordinates. The term “ventral” was deliberately used because the auditory cortex is located on the lateral side of the skull, which may have caused some misunderstanding.

    To provide a clearer and more accurate descriptions of the coordinates, we will revise the text in the manuscript as follows: “A craniotomy was performed at the temporal bone (-2 to -4 mm posterior and -1.5 to -3 mm ventral to bregma for mice; -3.0 to -5.0 mm posterior and -2.5 to -6.5 mm ventral to bregma for rats) to access the auditory cortex.'

    We appreciate your attention to these details and will ensure that the revised manuscript includes this clarification to improve accuracy and eliminate potential confusion. Thank you again for bringing this to our attention.

    (9) Line 536: The author should specify that these coordinates are for the experiment done on mice.

    Thank you for your valuable feedback. We will revise the manuscript to explicitly specify that these coordinates refer to the experiments conducted on mice. This clarification will help improve the clarity and precision of the manuscript. We greatly appreciate your attention to this point and your effort to enhance the quality of our work.

    Methods section:

    “and a hole was drilled in the skull according to the coordinates of the ventral division of the MGB (MGv, AP: -3.2 mm, ML: 2.1 mm, DV: 3.0 mm) for experiments conducted on mice.”

    (10) Line 590: Please add the specifications of the stimulating electrode. Is it unipolar or bipolar? What is the cat.# provided by FHC?

    Thank you for your valuable feedback. The electrodes used in the experiments are unipolar. We will include the catalog number provided by FHC in the revised manuscript for clarity. The revised text will be updated as follows:

    “In HFS-induced thalamocortical LTP experiments, two customized microelectrode arrays with four tungsten unipolar electrodes each, impedance: 0.5-1.0 MΩ (recording: CAT.# UEWSFGSECNND, FHC, U.S.), and 200-500 kΩ (stimulating: CAT.# UEWSDGSEBNND, FHC, U.S.), were used for the auditory cortical neuronal activity recording and MGB ES, respectively.”

    We appreciate your attention to this detail, and we will ensure that the revised manuscript reflects this clarification accurately.

    (11) Lines 612-614: There are no details of how the optic fiber was inserted or post-examined. If there is a word limitation, the authors may reference another study showing these procedures.

    Thank you for your insightful comment and for highlighting this important aspect of the methodology. To address this, we will reference the study by Sun et al. (2024) in the revised manuscript, which provides detailed procedures for optic fiber insertion and post-examination. We believe that this reference will help enhance the clarity and completeness of the methods section.

    Sun, W., Wu, H., Peng, Y., Zheng, X., Li, J., Zeng, D., Tang, P., Zhao, M., Feng, H., Li, H., et al. (2024). Heterosynaptic plasticity of the visuo-auditory projection requires cholecystokinin released from entorhinal cortex afferents. eLife 13, e83356. 10.7554/eLife.83356.

    We appreciate your valuable suggestion, which will contribute to improving the quality of the manuscript.

    Minor concerns:

    (1) The definition of HFS was repeated many times throughout the manuscript. Please mention the defined name for the first time in the manuscript only followed by its abbreviation (HFS).

    Thank you for your suggestion and for pointing out this important detail. We will revise the manuscript to ensure that all abbreviations are defined only upon their first mention in the manuscript, with subsequent mentions using the abbreviations consistently. We appreciate your careful attention to detail and your effort to help improve the manuscript.

    (2) Line 173: There is a difference between here and the methods section (620 nm here and 635 nm there) please correct which wavelength the authors used.

    Thank you for your careful review and for bringing this discrepancy to our attention. We have corrected the inconsistency, and the wavelength has been unified throughout the manuscript to ensure accuracy and clarity. The revised text now reads as follows:

    “The fluorescent signal was monitored for 25s before and 60s after the HFLS (5~10 mW, 620 nm) or HFS application.”

    We appreciate your valuable feedback, which has helped us improve the precision and consistency of the manuscript.

    (3) Line 185: I think the authors should refer to Figure 2G before mentioning the statistical results.

    Thank you for your careful review and for pointing out this oversight. We have now added a reference to Figure 2G at the appropriate location to ensure clarity and logical flow in the manuscript, as recommended..

    (4) Line 202: I think the authors should refer to Figure 2J before mentioning the statistical results.

    Thank you again for your careful review and for highlighting this point. We have revised the manuscript to include a reference to Figure 2J before mentioning the statistical results.

    We appreciate your valuable feedback, which has helped us improve the accuracy and presentation of the results.

    (5) Line 260: Please add appropriate references at the end of the sentence to support the argument.

    Thank you for your valuable suggestion. To address this, we have add appropriate references to support the statement regarding the multiple steps involved between mRNA expression and neuropeptide release. Additionally, we have revised the statement to adopt a more cautious interpretation. The revised text is as follows:

    “It is widely recognized that mRNA levels do not always directly correlate with peptide levels due to multiple steps involved in peptide synthesis and processing, including translation, post-translational modifications, packaging, transportation, and proteolytic cleavage, all of which require various enzymes and regulatory mechanisms (38-41). A disruption at any stage in this process could lead to impaired CCK release, even when Cck mRNA is present.”

    We have included the following references to support this statement:

    38. Mierke, C.T. (2020). Translation and Post-translational Modifications in Protein Biosynthesis. In Cellular Mechanics and Biophysics: Structure and Function of Basic Cellular Components Regulating Cell Mechanics, C.T. Mierke, ed. (Springer International Publishing), pp. 595-665. 10.1007/978-3-030-58532-7_14.

    39. Gualillo, O., Lago, F., Casanueva, F.F., and Dieguez, C. (2006). One ancestor, several peptides post-translational modifications of preproghrelin generate several peptides with antithetical effects. Mol Cell Endocrinol 256, 1-8. 10.1016/j.mce.2006.05.007.

    40. Sossin, W.S., Fisher, J.M., and Scheller, R.H. (1989). Cellular and molecular biology of neuropeptide processing and packaging. Neuron 2, 1407-1417. https://doi.org/10.1016/0896-6273(89)90186-4.

    41. Hook, V., Funkelstein, L., Lu, D., Bark, S., Wegrzyn, J., and Hwang, S.R. (2008). Proteases for processing proneuropeptides into peptide neurotransmitters and hormones. Annu Rev Pharmacol Toxicol 48, 393-423. 10.1146/annurev.pharmtox.48.113006.094812.

    We greatly appreciate your helpful feedback, which has allowed us to improve both the accuracy and the depth of discussion in the manuscript.

    (6) Line 278: The authors mentioned "due to the absence of CCK in aged animals", which was not an appropriate description. It should be a reduction of CCK gene expression or a possible deficient CCK release.

    Thank you for your careful review and for pointing out the inaccuracy in our description. We agree with your suggestion and have revised the statement to more appropriately reflect the findings.

    “Our findings revealed that thalamocortical LTP cannot be induced in aged mice, likely due to insufficient CCK release, despite intact CCKBR expression.”

    This revision ensures a more accurate and precise description of the potential mechanisms underlying the observed phenomenon. We greatly appreciate your valuable feedback, which has helped us improve the clarity and accuracy of the manuscript.

    (7) Line 291: The authors mentioned that "without MGB stimulation", which is confusing. The MGB was stimulated with a single electrical pulse to evoke cortical fEPSPs. Therefore it should be "without HFS of MGB".

    Thank you for pointing this out and for highlighting the potential confusion caused by our original phrasing. Upon review, we recognize that our original phrasing "without MGB stimulation" may have been unclear and could have led to misinterpretation. To clarify, our intention was to describe the period during which CCK was present without any stimulation of the MGB.

    It is important to note that, in the presence of CCK, LTP can be induced even with low-frequency stimulation, including in aged mice. This observation underscores the potent effect of CCK in facilitating thalamocortical LTP, regardless of the specific stimulation protocol used.

    To address this issue, we have revised the sentence for improved clarity as follows::

    " To investigate whether CCK alone is sufficient to induce thalamocortical LTP without activating thalamocortical projections, we infused CCK-4 into the ACx of young adult mice immediately after baseline fEPSPs recording. Stimulation was then paused for 15 min to allow for CCK degradation, after which recording resumed."

    We believe this revision resolves the misunderstanding and provides a clearer and more accurate description of the experimental context. We greatly appreciate your insightful feedback, which has helped us refine the manuscript for clarity and precision.

    Reviewer #3 (Recommendations for the authors):

    Minor comments:

    (1) Line 99, 134, possibly other locations: "site" to "sites".

    Thank you for your careful review. We appreciate your attention to detail and have made the necessary corrections in the manuscript.

    (2) Throughout the manuscript there are some minor issues with language choice and subtle phrasing errors and I suggest English language editing.

    Thank you for your suggestion. In response, we have thoroughly reviewed the manuscript and addressed issues related to language choice and phrasing. The text has been carefully edited to ensure clarity, precision, and consistency. We believe these revisions have significantly enhanced the overall quality of the manuscript. We greatly appreciate your feedback, which has been invaluable in improving the presentation of our work.

    (3) Based on the experimental configurations, I do not think it is a problematic caveat, but authors should be aware of the high likelihood of AAV9 jumping synapses relative to other AAV serotypes.

    Thank you for bringing up the potential of AAV9 crossing synapses, a recognized characteristic of this serotype. We appreciate your observation regarding its relevance to our experimental design. In our study, we carefully considered the possibility of trans-synaptic transfer during both the experimental design and data interpretation phases. To minimize the likelihood of significant trans-synaptic spread, we implemented several measures, including controlling the injection volume, using a slow injection rate, and limiting the viral expression time. Post-hoc histological analyses confirmed that the expression of AAV9 was largely confined to the intended regions, with limited evidence of synaptic jumping under our experimental conditions.

    While we acknowledge the inherent potential for AAV9 to cross synapses, we believe this effect does not substantially confound the interpretation of our findings in the current study. To address this concern, we have added a brief discussion on this point in the revised manuscript to enhance clarity. We greatly appreciate your insightful comment, which has helped us further refine our work.

    Discussion section：

    “ One potential limitation of our study is the trans-synaptic transfer property of AAV9. To mitigate this, we carefully controlled the injection volume, rate, and viral expression time, and conducted post-hoc histological analyses to minimize off-target effects, thereby reducing the likelihood of trans-synaptic transfer confounding the interpretation of our findings.”

    (4) The trace identifiers (1-4) do not seem correctly placed/colored in Figure S1D. Please check others carefully.

    Thank you for your careful review and for bringing this issue to our attention. We have corrected the trace identifiers in Figure S1D. Additionally, we have carefully reviewed all other figures to ensure their accuracy and consistency. We greatly appreciate your attention to detail, which has helped improve the overall quality of the manuscript.

    (5) Please provide a value of the laser power range based on calibrated values.

    Thank you for your suggestion. We have included the calibrated laser power range in the revised manuscript as follows:

    “The laser stimulation was produced by a laser generator (5-20 mW(30), Wavelength: 473 nm, 620 nm; CNI laser, China) controlled by an RX6 system and delivered to the brain via an optic fiber (Thorlabs, U.S.) connected to the generator.”

    We appreciate your feedback, which has helped improve the clarity and precision of our methodological description.

    (6) It would be useful to annotate figures in a way that identifies in which transgenic mice experiments are being performed.

    Thank you for your valuable suggestion. We will add annotations to the figures to explicitly identify the type of mice used in each experiment. We believe this enhancement will improve the clarity and accessibility of our results. We greatly appreciate your input in making our manuscript more informative.

    (7) Please comment on the rigor you use to address the accuracy of viral injections. How often did they spread outside of the MGB/AC?

    Thank you for raising this important question regarding the accuracy of viral injections and the potential spread outside the MGB or AC. Below, we provide details for each set of experiments:

    shRNA Experiments:

    For the shRNA experiments targeting the MGB, our primary goal was to achieve comprehensive coverage of the entire MGB. To this end, we used larger injection volumes and multiple injection sites, which inevitably resulted in some viral spread beyond the MGB. However, this approach was necessary to ensure robust knockdown effects that were representative of the entire MGB. While strict confinement to specific subregions could not be guaranteed, this strategy allowed us to prioritize the effectiveness of the knockdown within the target region.

    Fiber photometry Experiments:

    For the fiber photometry experiments targeting the auditory cortex (AC), we used larger injection volumes and multiple injection sites to cover its relatively large size. Although this approach might have resulted in some CCK-sensor virus spread outside the AC, the placement of the optic fiber was guided by the location of the auditory cortex. Consequently, any minor viral expression outside the AC would not affect the experimental results, as recordings were confined to the intended area through precise fiber placement.

    Optogenetic Experiments:

    For the optogenetic experiments targeting the MGB, we specifically injected virus into the MGv subregion. To minimize viral spread, we employed several strategies, including the used fine injection needles, waiting for tissue stabilization (7 minutes post-needle insertion), delivering small volumes at a slow rate to prevent backflow, aspirating 5 nL of the solution post-injection, and raising the needle by 100 μm before waiting an additional 5 minutes prior to full retraction. These measures significantly reduced the risk of viral leakage to adjacent regions.

    Histological Validation:

    After the electrophysiological experiments, we systematically verified the accuracy of viral expression by examining histological sections to ensure that the expression was primarily localized within the intended regions.

    Terminology in the Manuscript:

    In the manuscript, we deliberately used the term "MGB" in the manuscript rather than specifically "MGv" to transparently acknowledge the potential for viral spread in some experiments.

    We hope this explanation clarifies the strategies we employed to address the accuracy of viral injections, as well as how we managed potential viral spread. We have also added a brief information in the revised manuscript to reflect these points and acknowledge the inherent variability in viral delivery.

  14. eLife

    eLife Assessment

    This is an important study demonstrating that cholecystokinin is a key modulator of auditory thalamocortical plasticity during development and in young adults but not aged mice, though the cortical application of this neuropeptide in older animals appears to go some way to restoring this age-dependent loss in plasticity. A strength of this work is the use of multiple experimental approaches, which together provide convincing support for the proposed involvement of cholecystokinin. Nevertheless, the specificity of the electrical and optical stimulation experiments requires further validation and some key details are missing in the presentation and discussion of these findings. This work is likely to be influential in opening up a new avenue of investigation into the roles of neuropeptides in sensory plasticity.

  15. eLife

    Reviewer #1 (Public review):

    This study offers a valuable investigation into the role of cholecystokinin (CCK) in thalamocortical plasticity during early development and adulthood, employing a range of experimental techniques. The authors demonstrate that tetanic stimulation of the auditory thalamus induces cortical long-term potentiation (LTP), which can be evoked through either electrical or optical stimulation of the thalamus or by noise bursts. They further show that thalamocortical LTP is abolished when thalamic CCK is knocked down or when cortical CCK receptors are blocked. Interestingly, in 18-month-old mice, thalamocortical LTP was largely absent but could be restored through the cortical application of CCK. The authors conclude that CCK contributes to thalamocortical plasticity and may enhance thalamocortical plasticity in aged subjects.

    While the study presents compelling evidence, I would like to offer several suggestions for the authors' consideration:

    (1) Thalamocortical LTP and NMDA-Dependence:
    It is well established that thalamocortical LTP is NMDA receptor-dependent, and blocking cortical NMDA receptors can abolish LTP. This raises the question of why thalamocortical LTP is eliminated when thalamic CCK is knocked down or when cortical CCK receptors are blocked. If I correctly understand the authors' hypothesis - that CCK promotes LTP through CCKR-intracellular Ca2+-AMPAR. This pathway should not directly interfere with the NMDA-dependent mechanism. A clearer explanation of this interaction would be beneficial.

    (2) Complexity of the Thalamocortical System:
    The thalamocortical system is intricate, with different cortical and thalamic subdivisions serving distinct functions. In this study, it is not fully clear which subdivisions were targeted for stimulation and recording, which could significantly influence the interpretation of the findings. Clarifying this aspect would enhance the study's robustness.

    (3) Statistical Variability:
    Biological data, including field excitatory postsynaptic potentials (fEPSPs) and LTP, often exhibit significant variability between samples, sometimes resulting in a standard deviation that exceeds 50% of the mean value. The reported standard deviation of LTP in this study, however, appears unusually small, particularly given the relatively limited sample size. Further discussion of this observation might be warranted.

    (4) EYFP Expression and Virus Targeting:
    The authors indicate that AAV9-EFIa-ChETA-EYFP was injected into the medial geniculate body (MGB) and subsequently expressed in both the MGB and cortex. If I understand correctly, the authors assume that cortical expression represents thalamocortical terminals rather than cortical neurons. However, co-expression of CCK receptors does not necessarily imply that the virus selectively infected thalamocortical terminals. The physiological data regarding cortical activation of thalamocortical terminals could be questioned if the cortical expression represents cortical neurons or both cortical neurons and thalamocortical terminals.

    (5) Consideration of Previous Literature:
    A number of studies have thoroughly characterized auditory thalamocortical LTP during early development and adulthood. It may be beneficial for the authors to integrate insights from this body of work, as reliance on data from the somatosensory thalamocortical system might not fully capture the nuances of the auditory pathway. A more comprehensive discussion of the relevant literature could enhance the study's context and impact.

    (6) Therapeutic Implications:
    While the authors suggest potential therapeutic applications of their findings, it may be somewhat premature to draw such conclusions based on the current evidence. Although speculative discussion is not harmful, it may not significantly add to the study's conclusions at this stage.

  16. eLife

    Reviewer #2 (Public review):

    Summary:

    This work used multiple approaches to show that CCK is critical for long-term potentiation (LTP) in the auditory thalamocortical pathway. They also showed that the CCK mediation of LTP is age-dependent and supports frequency discrimination. This work is important because it opens up a new avenue of investigation of the roles of neuropeptides in sensory plasticity.

    Strengths:

    The main strength is the multiple approaches used to comprehensively examine the role of CCK in auditory thalamocortical LTP. Thus, the authors do provide a compelling set of data that CCK mediates thalamocortical LTP in an age-dependent manner.

    Weaknesses:

    The behavioral assessment is relatively limited but may be fleshed out in future work.

  17. eLife

    Reviewer #3 (Public review):

    Summary:

    Cholecystokinin (CCK) is highly expressed in auditory thalamocortical (MGB) neurons and CCK has been found to shape cortical plasticity dynamics. In order to understand how CCK shapes synaptic plasticity in the auditory thalamocortical pathway, they assessed the role of CCK signaling across multiple mechanisms of LTP induction with the auditory thalamocortical (MGB - layer IV Auditory Cortex) circuit in mice. In these physiology experiments that leverage multiple mechanisms of LTP induction and a rigorous manipulation of CCK and CCK-dependent signaling, they establish an essential role of auditory thalamocortical LTP on the co-release of CCK from auditory thalamic neurons. By carefully assessing the development of this plasticity over time and CCK expression, they go on to identify a window of time that CCK is produced throughout early and middle adulthood in auditory thalamocortical neurons to establish a window for plasticity from 3 weeks to 1.5 years in mice, with limited LTP occurring outside of this window. The authors go on to show that CCK signaling and its effect on LTP in the auditory cortex is also capable of modifying frequency discrimination accuracy in an auditory PPI task. In evaluating the impact of CCK on modulating PPI task performance, it also seems that in mice <1.5 years old CCK-dependent effects on cortical plasticity are almost saturated. While exogenous CCK can modestly improve discrimination of only very similar tones, exogenous focal delivery of CCK in older mice can significantly improve learning in a PPI task to bring their discrimination ability in line with those from young adult mice.

    Strengths:

    (1) The clarity of the results along with the rigor multi-angled approach provide significant support for the claim that CCK is essential for auditory thalamocortical synaptic LTP. This approach uses a combination of electrical, acoustic, and optogenetic pathway stimulation alongside conditional expression approaches, germline knockout, viral RNA downregulation, and pharmacological blockade. Through the combination of these experimental configures the authors demonstrate that high-frequency stimulation-induced LTP is reliant on co-release of CCK from glutamatergic MGB terminals projecting to the auditory cortex.

    (2) The careful analysis of the CCK, CCKB receptor, and LTP expression is also a strength that puts the finding into the context of mechanistic causes and potential therapies for age-dependent sensory/auditory processing changes. Similarly, not only do these data identify a fundamental biological mechanism, but they also provide support for the idea that exogenous asynchronous stimulation of the CCKBR is capable of restoring an age-dependent loss in plasticity.

    (3) Although experiments to simultaneously relate LTP and behavioral change or identify a causal relationship between LTP and frequency discrimination are not made, there is still convincing evidence that CCK signaling in the auditory cortex (known to determine synaptic LTP) is important for auditory processing/frequency discrimination. These experiments are key for establishing the relevance of this mechanism.

    Weaknesses:

    (1) Given the magnitude of the evoked responses, one expects that pyramidal neurons in layer IV are primarily those that undergo CCK-dependent plasticity, but the degree to which PV-interneurons and pyramidal neurons participate in this process differently is unclear.

    (2) While these data support an important role for CCK in synaptic LTP in the auditory thalamocortical pathway, perhaps temporal processing of acoustic stimuli is as or more important than frequency discrimination. Given the enhanced responsivity of the system, it is unclear whether this mechanism would improve or reduce the fidelity of temporal processing in this circuit. Understanding this dynamic may also require consideration of cell type as raised in weakness #1.

    (3) In Figure 1, an example of increased spontaneous and evoked firing activity of single neurons after HFS is provided. Yet it is surprising that the group data are analyzed only for the fEPSP. It seems that single-neuron data would also be useful at this point to provide insight into how CCK and HFS affect temporal processing and spontaneous activity/excitability, especially given the example in 1F.

    (4) The authors mention that CCK mRNA was absent in CCK-KO mice, but the data are not provided.

    (5) The circuitry that determines PPI requires multiple brain areas, including the auditory cortex. Given the complicated dynamics of this process, it may be helpful to consider what, if anything, is known specifically about how layer IV synaptic plasticity in the auditory cortex may shape this behavior.

  18. Version published to 10.7554/elife.101513.1 on eLife
  19. Version published to 10.1101/2024.07.31.605964 on bioRxiv