Homeostatic regulation of neuronal firing levels is a critical feature of neural circuits. It allows for prolonged stable function during changing activity levels associated with ongoing Hebbian plasticity, changes in sensory drive or input loss (Keck et al., 2017). The impairment of homeostatic regulation is thought to result in prolonged periods of excessive neuronal hyper- or hypoactivity, which may impair network function (Frere and Slutsky, 2018; Litwin-Kumar and Doiron, 2014; Marder and Prinz, 2002; Tetzlaff et al., 2011; Wu et al., 2020; Zenke et al., 2013). Homeostatic regulation occurs through plastic changes to a broad range of synaptic, intrinsic, cellular, and network-level properties (Barnes et al., 2015; Bridi et al., 2020; Bridi et al., 2018; Gainey and Feldman, 2017; Joseph and Turrigiano, 2017; Keck et al., 2017; Lee and Kirkwood, 2019; Ma et al., 2019; Turrigiano, 2011). These changes must facilitate rebalancing of activity while maintaining the underlying circuits established for ongoing sensory processing (Chen et al., 2013a; Iacaruso et al., 2017), as well as for learned (Jurjut et al., 2017; Lee et al., 2020; Poort et al., 2015) and innate behaviors (Branco and Redgrave, 2020; Evans et al., 2018). Maintaining this balance is particularly challenging with synaptic forms of homeostatic plasticity, where synaptic connections form the basis for circuitry and memory, but may also undergo homeostatic changes in strength that is inversely proportional to changes in activity levels. Previous work in adult animals has shown that synaptic homeostatic changes occur in only a subset of synapses (Barnes et al., 2017) or via additive changes (Goel and Lee, 2007), as opposed to multiplicative changes across the entire population described for some homeostatic mechanisms during development and the critical period (Desai et al., 2002; Kaneko et al., 2008; Turrigiano, 2017; Turrigiano et al., 1998). Identifying the functional properties of the subset of synapses that are homeostatically strengthened in adulthood is essential for understanding how sensory processing and homeostatic regulation are coordinated after the closure of the critical period, when sensory plasticity is reduced (Hensch, 2005; Levelt and Hübener, 2012).

Sensory cortices in adult mice are known to be multimodal and receive multiple sensory inputs (Ibrahim et al., 2016; Iurilli et al., 2012; Knöpfel et al., 2019; Todd et al., 2016; Vann et al., 2009; Yoshitake et al., 2013), as well as intrinsic contextual and intracortical inputs (Fiser et al., 2016; Khan and Hofer, 2018; Miller et al., 2014; Okun et al., 2015; Pakan et al., 2016; Pakan et al., 2018b; Petrus et al., 2015; Roth et al., 2016; Saleem et al., 2018). Large populations of neurons in visual cortex have a strong correlation or coupling with the average network activity, which is thought to be largely non-sensory and represent contextual input that may be associated with behavioral state (Okun et al., 2015; Stringer et al., 2019). High levels of network coupling are correlated with a higher probability of local intracortical connections between neurons (Okun et al., 2015) and could also reflect the presence of other common non-sensory inputs (Yoshimura et al., 2005). Following sensory deprivation, functional strengthening could occur in functional subsets of inputs, such as the remaining sensory inputs, other non-deprived sensory inputs, or network-correlated inputs, either independently or in a combination of subsets; however, it is currently unclear whether the synapses that undergo homeostatic strengthening encode a specific functional input.

Here, we examine the properties of inputs that undergo functional changes following sensory deprivation using repeated two-photon functional imaging of dendritic spines in awake mice (Chen et al., 2013b; Iacaruso et al., 2017). We find that, following deprivation, inputs with activity that is correlated with the network activity undergo functional strengthening, while inputs that have identified sensory-evoked responses do not strengthen, independent of whether they encode a deprived or non-deprived sensory input. Despite the fact that sensory-evoked spine responses are not strengthened, we find that global sensory-evoked responses homeostatically increase following deprivation and are correlated with increases in responses of network-correlated spines on their dendrites. Together, our work suggests that homeostatic synaptic regulation occurs through changes to network-correlated inputs, rather than the sensory responsive inputs that drive ongoing sensory processing.

We used chronic two-photon imaging in awake adult mice expressing the genetically encoded calcium indicator GCaMP6s (see Methods) to examine the functional properties of dendritic spines that undergo functional plasticity following sensory deprivation. Over a span of 4 days, we repeatedly measured functional calcium signals in dendritic spines on the distal tufts (in layer 2/3) of layer 5 pyramidal cells in primary monocular visual cortex before and after sensory deprivation via enucleation of the contralateral eye (Figure 1A–C; Barnes et al., 2015). We extracted spine responses using previously published methods (Chen et al., 2013b; Iacaruso et al., 2017; Wilson et al., 2016) and were able to measure the functional response properties, response amplitude, and response frequency of the same spines over the course of the imaging time series (Figure 1B–C, Figure 1—figure supplement 1A–D).

Figure 1 with 1 supplement see all

Download asset Open asset

Figure 1—source data 1

Statistical comparisons for Figure 1.

https://cdn.elifesciences.org/articles/81958/elife-81958-fig1-data1-v2.docx

Download elife-81958-fig1-data1-v2.docx

Identifying functional subsets of dendritic spines in visual cortex

We have previously used structural imaging of spine size as an in vivo proxy for synaptic strength and shown that not all spines undergo a TNF-α-dependent strengthening following deprivation (Barnes et al., 2017). One possibility is that there could be common functional properties of spines that undergo homeostatic strengthening. We therefore characterized the response properties of individual spines prior to any form of deprivation during the baseline time points (–24 and –1 hr). We first measured the responses of spines to visual stimuli: bilateral drifting gratings (Barnes et al., 2015) and a sparse noise stimulus (de Vries et al., 2020; Figure 2A). Spines that had time-locked responses to the visual stimulus at levels above chance (see Methods) (Barnes et al., 2015; Ko et al., 2011) were categorized as ‘visually responsive’ spines (Figure 2B). We found that a majority of spines that were responsive to the sparse noise stimulus also responded to drifting gratings (97%). Therefore, in a subset of experiments, we characterized visually responsive spines using only the drifting gratings stimulus to allow sufficient imaging time to test other stimuli. When we examined the functional strengthening of all of the spines, we found an inverse relationship, such that the spines with the lowest visual responsiveness prior to deprivation underwent the strongest functional strengthening after deprivation (Figure 2C). Consistent with our previous work (Barnes et al., 2017), we observed that only a subset of spines increased their functional responses (Figure 2C).

Figure 2

Download asset Open asset

Figure 2—source data 1

Statistical comparisons for Figure 2.

https://cdn.elifesciences.org/articles/81958/elife-81958-fig2-data1-v2.docx

Download elife-81958-fig2-data1-v2.docx

We next characterized other functional inputs. Within mouse visual cortex, there are high levels of correlated network activity, which may reflect the intrinsic dynamics (Okun et al., 2015; Stringer et al., 2019). We therefore hypothesized that activity in some spines may reflect this network activity, potentially via local intracortical connections (Okun et al., 2015). We measured the correlation of individual spines’ activity with the ‘network signal’, which we measured as the average activity during all experimental conditions (visual stimulation and darkness) of all the other spines in the same imaging region (Okun et al., 2015; Figure 2A, see Methods). The network signal does not reflect locomotion-related activity, as the mice were stationary during our measurements.

In our population of spines, we found that most spines (89%) had a significant correlation with the network signal. This included the visually responsive spines, all of which had a significant correlation with the network signal (see Methods). As described above, all spines with time-locked responses to visual stimuli at above chance levels (see Methods) were classified as visually responsive spines (24%; Figure 2B), even though their activity also correlated with the network signal. Spines with activity that correlated with the network signal, but did not have visually-evoked responses, were classified as network-correlated (65%). The remainder of spines neither responded to the visual stimuli nor correlated with the network activity, which we refer to as unclassified (11%; Figure 2A). We tested whether these unclassified spines were correlated with other behavioral measures, such as pupil movement, pupil dilation (diameter), and whisker movement, and found no evidence of significant correlations above chance (see Methods). There were no significant differences in the average activity in the three groups of spines prior to deprivation (measured as the area under the curve of the ΔF/F₀ fluorescence signal; comparison between network-correlated, visually responsive and unclassified spine activity, one-way ANOVA, p=0.232).

We next examined if input types were clustered on the same dendritic branches (Hedrick et al., 2022). All of the branches that we measured had all three types of spines (visually responsive, network-correlated, unclassified). Within individual dendritic branches, we found that the local spatial organization of functional response properties (visually responsive, network-correlated, unclassified) of the dendritic spines were not different from chance (Figure 2D–F, see Methods).

Functional strengthening occurs in network-correlated spines

We next examined if the spines that were functionally strengthened had common functional properties (Figure 3). Consistent with our earlier finding (Figure 2C), we found no evidence of an increase in the amplitude of the persistent spines identified as visually responsive (Figure 3A), or of the persistent unclassified spines (Figure 3B). However, following deprivation there was an increase in the response amplitude of persistent network-correlated spines (Figure 3C). Overall activity depends on both the amplitude and frequency of responses. While the average frequency of responses did not change following deprivation (Figure 1—figure supplement 1E), we examined if functional groups of spines had changes in response frequency and observed a decrease in the frequency of responses of visually responsive spines over time following deprivation (Figure 3—figure supplement 1A–C). When we examined the integral of the activity trace (which combines amplitude and frequency) over time following deprivation, we observed a decrease for visually responsive spines (Figure 3—figure supplement 1D), no change for unclassified spines (Figure 3—figure supplement 1E), and an increase for network-correlated spines (Figure 3—figure supplement 1F). We also found a strong increase in the fraction of inactive visually responsive spines after deprivation. While a larger fraction of visually responsive spines become inactive, the absolute number of network-correlated spines that became inactive was higher, since network-correlated spines reflect a larger proportion of the total spine population. These results suggest that the increase in spine loss previously reported following enucleation (Barnes et al., 2017) may include both spines that were visually responsive and network-correlated prior to deprivation (Figure 3D–F). A sizeable fraction of visually responsive spines remains active after visual deprivation (Figure 3D). It is important to note that following sensory deprivation, we no longer observed responses to visual stimuli in the visually responsive spines, since the feedforward input had been permanently removed. The responses of the previously categorized visually responsive spines consisted of network responses following deprivation.

Figure 3 with 1 supplement see all

Download asset Open asset

Figure 3—source data 1

Statistical comparisons for Figure 3.

https://cdn.elifesciences.org/articles/81958/elife-81958-fig3-data1-v2.docx

Download elife-81958-fig3-data1-v2.docx

There was no correlation between amplitude and frequency within individual spines in any of the groups (Figure 3G–I), again suggesting that frequency decrease does not regulate changes in amplitude within individual spines in this paradigm. Overall, these results suggest the network-correlated inputs, and not the identified visually responsive inputs, are homeostatically strengthened following sensory input loss.

Identified sensory responsive spines are not functionally strengthened

Having found that the network-correlated spines increase their activity, we wanted to determine whether this more generally reflected the functional strengthening of all non-deprived inputs. To investigate this issue, we used a multimodal brain area – the retrosplenial cortex (RSC) – which receives measurable auditory responsive and visually responsive inputs (Figure 4A). We again measured responses of dendritic spines on layer 5 cells in the distal tufts in layer 2/3. Using the same classification approach as before (Figure 2, see Methods), we found that the activity of 92% of the spines was correlated with the network signal. Within this group, a subset of spines had time-locked responses to visual (18.7%) or auditory (15.9%) stimuli (Figure 4A) and the rest correlated only with the network signal (57.4%, Figure 4A). The remaining 8% of spines were unclassified (no correlation with the network signal and no sensory-evoked responses). Following reversible visual deprivation via dark exposure for 48 hr (Figure 4B), we observed an increase in the average spine response amplitude across the population. This response was TNF-α-dependent (Figure 4C), similar to what we observed in visual cortex following enucleation (Figure 1). When we examined the changes in the responses of functionally identified spines, we observed an increase in the response amplitude of the network-correlated spines, but not in the visually responsive or unclassified spines (Figure 4D, Figure 4—figure supplement 1A–C), consistent with our results in visual cortex (Figure 2C, Figure 3A–C). We then examined the responses of the auditory responsive spines, which represent an identified non-deprived sensory input. After visual deprivation, the auditory responsive spines did not undergo functional strengthening but maintained their baseline amplitude following visual deprivation (Figure 4E).

Figure 4 with 1 supplement see all

Download asset Open asset

Figure 4—source data 1

Statistical comparisons for Figure 4.

https://cdn.elifesciences.org/articles/81958/elife-81958-fig4-data1-v2.docx

Download elife-81958-fig4-data1-v2.docx

To ensure that the strengthening effects that we observed were not exclusive to visual deprivation, we performed a complementary, but separate, experiment where we deprived mice of auditory input through reversible ear plugging for 48 hr and again measured post-deprivation spine responses in the RSC (Figure 4F). We observed a TNF-α-dependent population increase in spine responses (Figure 4G). Similar to visual deprivation, network-correlated spines increased their responses following auditory deprivation (Figure 4H–I, Figure 4—figure supplement 1D–F), but there was no change in the response strength of either (non-deprived) visually responsive (Figure 4H) or (deprived) auditory responsive spines (Figure 4I). Together, these data suggest that increases in spine response amplitude predominantly occur in the spines whose activity correlated with the network signal and not in the spines which are responsive to the non-deprived or deprived sensory stimulation.

Here, we report a TNF-α-dependent increase in the functional responses of dendritic spines in awake mice following sensory deprivation. Surprisingly, when we examine the functional properties of spines that undergo strengthening, we find dendritic spines whose activity is correlated with the network signal are strengthened, but spines identified as sensory responsive are not. This strengthening of network-correlated spines occurs over a similar time course and magnitude in both visual cortex and RSC and in response to different deprivation paradigms. Despite the fact that we do not see an increase in responses of sensory responsive spines, we do observe a homeostatic increase in global sensory-evoked responses following sensory deprivation. This global increase is correlated with an increase in the response strength of network-correlated spines on those branches, but not of identified sensory responsive spines. Our findings suggest that in vivo synaptic homeostasis during adulthood occurs in a functional subset of inputs, specifically those correlated only with network activity, rather than the identified sensory responsive inputs that are thought to drive sensory processing.

The pre-processed raw data can be accessed at https://doi.org/10.5281/zenodo.7399601. Data that have not been pre-processed are available upon request to any interested party, due to size constraints, by emailing georg.keller@fmi.ch, who will provide temporary transfer access for downloading the data. No proposal is required to access the data and there are not restrictions on who can access the data. Software for controlling the two-photon microscope and pre-processing of the calcium imaging data is available on https://sourceforge.net/projects/iris-scanning/ (copies archived at swh:1:rev:1f1b5616aaaf6cf862d2f4b467b65b479a2b0416, swh:1:rev:48e0b534668ef3ecb7c2580c00cdbcf650d69043, and swh:1:rev:d3b0d1a0482b5b929621f7e6e99c0e9ff2eee11f).

1. 1. Barnes SJ
   2. Keller GB
   3. Keck T

   (2022) Zenodo

   Homeostatic regulation through strengthening of neuronal network correlated synaptic inputs.

   https://doi.org/10.5281/zenodo.7399601

1. 1. Barnes SJ
   2. Sammons RP
   3. Jacobsen RI
   4. Mackie J
   5. Keller GB
   6. Keck T

   (2015) Subnetwork-specific homeostatic plasticity in mouse visual cortex in vivo

   Neuron 86:1290–1303.

   https://doi.org/10.1016/j.neuron.2015.05.010

   • PubMed
   • Google Scholar
2. 1. Barnes SJ
   2. Franzoni E
   3. Jacobsen RI
   4. Erdelyi F
   5. Szabo G
   6. Clopath C
   7. Keller GB
   8. Keck T

   (2017) Deprivation-Induced homeostatic spine scaling in vivo is localized to dendritic branches that have undergone recent spine loss

   Neuron 96:871–882.

   https://doi.org/10.1016/j.neuron.2017.09.052

   • PubMed
   • Google Scholar
3. 1. Beattie EC
   2. Stellwagen D
   3. Morishita W
   4. Bresnahan JC
   5. Ha BK
   6. Von Zastrow M
   7. Beattie MS
   8. Malenka RC

   (2002) Control of synaptic strength by glial TNFalpha

   Science 295:2282–2285.

   https://doi.org/10.1126/science.1067859

   • PubMed
   • Google Scholar
4. 1. Beaulieu-Laroche L
   2. Toloza EHS
   3. Brown NJ
   4. Harnett MT

   (2019) Widespread and highly correlated somato-dendritic activity in cortical layer 5 neurons

   Neuron 103:235–241.

   https://doi.org/10.1016/j.neuron.2019.05.014

   • PubMed
   • Google Scholar
5. 1. Branco T
   2. Häusser M

   (2010) The single dendritic branch as a fundamental functional unit in the nervous system

   Current Opinion in Neurobiology 20:494–502.

   https://doi.org/10.1016/j.conb.2010.07.009

   • PubMed
   • Google Scholar
6. 1. Branco T
   2. Häusser M

   (2011) Synaptic integration gradients in single cortical pyramidal cell dendrites

   Neuron 69:885–892.

   https://doi.org/10.1016/j.neuron.2011.02.006

   • PubMed
   • Google Scholar
7. 1. Branco T
   2. Redgrave P

   (2020) The neural basis of escape behavior in vertebrates

   Annual Review of Neuroscience 43:417–439.

   https://doi.org/10.1146/annurev-neuro-100219-122527

   • PubMed
   • Google Scholar
8. 1. Bridi MCD
   2. de Pasquale R
   3. Lantz CL
   4. Gu Y
   5. Borrell A
   6. Choi SY
   7. He K
   8. Tran T
   9. Hong SZ
   10. Dykman A
   11. Lee HK
   12. Quinlan EM
   13. Kirkwood A

   (2018) Two distinct mechanisms for experience-dependent homeostasis

   Nat Neurosci 21:843–850.

   https://doi.org/10.1038/s41593-018-0150-0

   • Google Scholar
9. 1. Bridi MCD
   2. Zong FJ
   3. Min X
   4. Luo N
   5. Tran T
   6. Qiu J
   7. Severin D
   8. Zhang XT
   9. Wang G
   10. Zhu ZJ
   11. He KW
   12. Kirkwood A

   (2020) Daily oscillation of the excitation-inhibition balance in visual cortical circuits

   Neuron 105:621–629.

   https://doi.org/10.1016/j.neuron.2019.11.011

   • Google Scholar
10. 1. Chen JL
    2. Villa KL
    3. Cha JW
    4. So PTC
    5. Kubota Y
    6. Nedivi E

    (2012) Clustered dynamics of inhibitory synapses and dendritic spines in the adult neocortex

    Neuron 74:361–373.

    https://doi.org/10.1016/j.neuron.2012.02.030

    • PubMed
    • Google Scholar
11. 1. Chen X
    2. Rochefort NL
    3. Sakmann B
    4. Konnerth A

    (2013a) Reactivation of the same synapses during spontaneous up states and sensory stimuli

    Cell Reports 4:31–39.

    https://doi.org/10.1016/j.celrep.2013.05.042

    • PubMed
    • Google Scholar
12. 1. Chen TW
    2. Wardill TJ
    3. Sun Y
    4. Pulver SR
    5. Renninger SL
    6. Baohan A
    7. Schreiter ER
    8. Kerr RA
    9. Orger MB
    10. Jayaraman V
    11. Looger LL
    12. Svoboda K
    13. Kim DS

    (2013b) Ultrasensitive fluorescent proteins for imaging neuronal activity

    Nature 499:295–300.

    https://doi.org/10.1038/nature12354

    • PubMed
    • Google Scholar
13. 1. Desai NS
    2. Cudmore RH
    3. Nelson SB
    4. Turrigiano GG

    (2002) Critical periods for experience-dependent synaptic scaling in visual cortex

    Nature Neuroscience 5:783–789.

    https://doi.org/10.1038/nn878

    • PubMed
    • Google Scholar
14. 1. de Vries SEJ
    2. Lecoq JA
    3. Buice MA
    4. Groblewski PA
    5. Ocker GK
    6. Oliver M
    7. Feng D
    8. Cain N
    9. Ledochowitsch P
    10. Millman D
    11. Roll K
    12. Garrett M
    13. Keenan T
    14. Kuan L
    15. Mihalas S
    16. Olsen S
    17. Thompson C
    18. Wakeman W
    19. Waters J
    20. Williams D
    21. Barber C
    22. Berbesque N
    23. Blanchard B
    24. Bowles N
    25. Caldejon SD
    26. Casal L
    27. Cho A
    28. Cross S
    29. Dang C
    30. Dolbeare T
    31. Edwards M
    32. Galbraith J
    33. Gaudreault N
    34. Gilbert TL
    35. Griffin F
    36. Hargrave P
    37. Howard R
    38. Huang L
    39. Jewell S
    40. Keller N
    41. Knoblich U
    42. Larkin JD
    43. Larsen R
    44. Lau C
    45. Lee E
    46. Lee F
    47. Leon A
    48. Li L
    49. Long F
    50. Luviano J
    51. Mace K
    52. Nguyen T
    53. Perkins J
    54. Robertson M
    55. Seid S
    56. Shea-Brown E
    57. Shi J
    58. Sjoquist N
    59. Slaughterbeck C
    60. Sullivan D
    61. Valenza R
    62. White C
    63. Williford A
    64. Witten DM
    65. Zhuang J
    66. Zeng H
    67. Farrell C
    68. Ng L
    69. Bernard A
    70. Phillips JW
    71. Reid RC
    72. Koch C

    (2020) A large-scale standardized physiological survey reveals functional organization of the mouse visual cortex

    Nature Neuroscience 23:138–151.

    https://doi.org/10.1038/s41593-019-0550-9

    • PubMed
    • Google Scholar
15. 1. Dombeck DA
    2. Khabbaz AN
    3. Collman F
    4. Adelman TL
    5. Tank DW

    (2007) Imaging large-scale neural activity with cellular resolution in awake, mobile mice

    Neuron 56:43–57.

    https://doi.org/10.1016/j.neuron.2007.08.003

    • PubMed
    • Google Scholar
16. 1. Dubbs A
    2. Guevara J
    3. Yuste R

    (2016) Moco: fast motion correction for calcium imaging

    Front Neuroinform 10:6.

    https://doi.org/10.3389/fninf.2016.00006

    • Google Scholar
17. 1. Evans DA
    2. Stempel AV
    3. Vale R
    4. Ruehle S
    5. Lefler Y
    6. Branco T

    (2018) A synaptic threshold mechanism for computing escape decisions

    Nature 558:590–594.

    https://doi.org/10.1038/s41586-018-0244-6

    • PubMed
    • Google Scholar
18. 1. Fiser A
    2. Mahringer D
    3. Oyibo HK
    4. Petersen AV
    5. Leinweber M
    6. Keller GB

    (2016) Experience-dependent spatial expectations in mouse visual cortex

    Nature Neuroscience 19:1658–1664.

    https://doi.org/10.1038/nn.4385

    • PubMed
    • Google Scholar
19. 1. Francioni V
    2. Padamsey Z
    3. Rochefort NL

    (2019) High and asymmetric somato-dendritic coupling of V1 layer 5 neurons independent of visual stimulation and locomotion

    eLife 8:e49145.

    https://doi.org/10.7554/eLife.49145

    • Google Scholar
20. 1. Frere S
    2. Slutsky I

    (2018) Alzheimer’s disease: from firing instability to homeostasis network collapse

    Neuron 97:32–58.

    https://doi.org/10.1016/j.neuron.2017.11.028

    • PubMed
    • Google Scholar
21. 1. Gainey MA
    2. Hurvitz-Wolff JR
    3. Lambo ME
    4. Turrigiano GG

    (2009) Synaptic scaling requires the glur2 subunit of the AMPA receptor

    The Journal of Neuroscience 29:6479–6489.

    https://doi.org/10.1523/JNEUROSCI.3753-08.2009

    • PubMed
    • Google Scholar
22. 1. Gainey M.A
    2. Feldman DE

    (2017) Multiple shared mechanisms for homeostatic plasticity in rodent somatosensory and visual cortex

    Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 372:20160157.

    https://doi.org/10.1098/rstb.2016.0157

    • PubMed
    • Google Scholar
23. 1. Goel A
    2. Lee HK

    (2007) Persistence of experience-induced homeostatic synaptic plasticity through adulthood in superficial layers of mouse visual cortex

    The Journal of Neuroscience 27:6692–6700.

    https://doi.org/10.1523/JNEUROSCI.5038-06.2007

    • PubMed
    • Google Scholar
24. 1. Grell M
    2. Douni E
    3. Wajant H
    4. Löhden M
    5. Clauss M
    6. Maxeiner B
    7. Georgopoulos S
    8. Lesslauer W
    9. Kollias G
    10. Pfizenmaier K
    11. Scheurich P

    (1995) The transmembrane form of tumor necrosis factor is the prime activating ligand of the 80 kDa tumor necrosis factor receptor

    Cell 83:793–802.

    https://doi.org/10.1016/0092-8674(95)90192-2

    • PubMed
    • Google Scholar
25. 1. He P
    2. Liu Q
    3. Wu J
    4. Shen Y

    (2012) Genetic deletion of TNF receptor suppresses excitatory synaptic transmission via reducing AMPA receptor synaptic localization in cortical neurons

    FASEB Journal 26:334–345.

    https://doi.org/10.1096/fj.11-192716

    • PubMed
    • Google Scholar
26. 1. Hedrick NG
    2. Lu Z
    3. Bushong E
    4. Singhi S
    5. Nguyen P
    6. Magaña Y
    7. Jilani S
    8. Lim BK
    9. Ellisman M
    10. Komiyama T

    (2022) Learning binds new inputs into functional synaptic clusters via spinogenesis

    Nature Neuroscience 25:726–737.

    https://doi.org/10.1038/s41593-022-01086-6

    • PubMed
    • Google Scholar
27. 1. Hengen KB
    2. Lambo ME
    3. Van Hooser SD
    4. Katz DB
    5. Turrigiano GG

    (2013) Firing rate homeostasis in visual cortex of freely behaving rodents

    Neuron 80:335–342.

    https://doi.org/10.1016/j.neuron.2013.08.038

    • PubMed
    • Google Scholar
28. 1. Hengen KB
    2. Torrado Pacheco A
    3. McGregor JN
    4. Van Hooser SD
    5. Turrigiano GG

    (2016) Neuronal firing rate homeostasis is inhibited by sleep and promoted by wake

    Cell 165:180–191.

    https://doi.org/10.1016/j.cell.2016.01.046

    • PubMed
    • Google Scholar
29. 1. Hensch TK

    (2005) Critical period plasticity in local cortical circuits

    Nature Reviews. Neuroscience 6:877–888.

    https://doi.org/10.1038/nrn1787

    • PubMed
    • Google Scholar
30. 1. Holtmaat A
    2. Bonhoeffer T
    3. Chow DK
    4. Chuckowree J
    5. De Paola V
    6. Hofer SB
    7. Hübener M
    8. Keck T
    9. Knott G
    10. Lee WCA
    11. Mostany R
    12. Mrsic-Flogel TD
    13. Nedivi E
    14. Portera-Cailliau C
    15. Svoboda K
    16. Trachtenberg JT
    17. Wilbrecht L

    (2009) Long-term, high-resolution imaging in the mouse neocortex through a chronic cranial window

    Nature Protocols 4:1128–1144.

    https://doi.org/10.1038/nprot.2009.89

    • PubMed
    • Google Scholar
31. 1. Iacaruso MF
    2. Gasler IT
    3. Hofer SB

    (2017) Synaptic organization of visual space in primary visual cortex

    Nature 547:449–452.

    https://doi.org/10.1038/nature23019

    • PubMed
    • Google Scholar
32. 1. Ibrahim LA
    2. Mesik L
    3. Ji XY
    4. Fang Q
    5. Li HF
    6. Li YT
    7. Zingg B
    8. Zhang LI
    9. Tao HW

    (2016) Cross-Modality sharpening of visual cortical processing through layer-1-mediated inhibition and disinhibition

    Neuron 89:1031–1045.

    https://doi.org/10.1016/j.neuron.2016.01.027

    • PubMed
    • Google Scholar
33. 1. Iurilli G
    2. Ghezzi D
    3. Olcese U
    4. Lassi G
    5. Nazzaro C
    6. Tonini R
    7. Tucci V
    8. Benfenati F
    9. Medini P

    (2012) Sound-driven synaptic inhibition in primary visual cortex

    Neuron 73:814–828.

    https://doi.org/10.1016/j.neuron.2011.12.026

    • PubMed
    • Google Scholar
34. 1. Joseph A
    2. Turrigiano GG

    (2017) All for one but not one for all: excitatory synaptic scaling and intrinsic excitability are coregulated by camkiv, whereas inhibitory synaptic scaling is under independent control

    The Journal of Neuroscience 37:6778–6785.

    https://doi.org/10.1523/JNEUROSCI.0618-17.2017

    • PubMed
    • Google Scholar
35. 1. Jurjut O
    2. Georgieva P
    3. Busse L
    4. Katzner S

    (2017) Learning enhances sensory processing in mouse V1 before improving behavior

    The Journal of Neuroscience 37:6460–6474.

    https://doi.org/10.1523/JNEUROSCI.3485-16.2017

    • PubMed
    • Google Scholar
36. 1. Kaneko M
    2. Stellwagen D
    3. Malenka RC
    4. Stryker MP

    (2008) Tumor necrosis factor-alpha mediates one component of competitive, experience-dependent plasticity in developing visual cortex

    Neuron 58:673–680.

    https://doi.org/10.1016/j.neuron.2008.04.023

    • PubMed
    • Google Scholar
37. 1. Keck T
    2. Keller GB
    3. Jacobsen RI
    4. Eysel UT
    5. Bonhoeffer T
    6. Hübener M

    (2013) Synaptic scaling and homeostatic plasticity in the mouse visual cortex in vivo

    Neuron 80:327–334.

    https://doi.org/10.1016/j.neuron.2013.08.018

    • PubMed
    • Google Scholar
38. 1. Keck T
    2. Toyoizumi T
    3. Chen L
    4. Doiron B
    5. Feldman DE
    6. Fox K
    7. Gerstner W
    8. Haydon PG
    9. Hübener M
    10. Lee HK
    11. Lisman JE
    12. Rose T
    13. Sengpiel F
    14. Stellwagen D
    15. Stryker MP
    16. Turrigiano GG
    17. van Rossum MC

    (2017) Integrating hebbian and homeostatic plasticity: the current state of the field and future research directions

    Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 372:20160158.

    https://doi.org/10.1098/rstb.2016.0158

    • PubMed
    • Google Scholar
39. 1. Khan AG
    2. Hofer SB

    (2018) Contextual signals in visual cortex

    Current Opinion in Neurobiology 52:131–138.

    https://doi.org/10.1016/j.conb.2018.05.003

    • PubMed
    • Google Scholar
40. 1. Knöpfel T
    2. Sweeney Y
    3. Radulescu CI
    4. Zabouri N
    5. Doostdar N
    6. Clopath C
    7. Barnes SJ

    (2019) Audio-visual experience strengthens multisensory assemblies in adult mouse visual cortex

    Nature Communications 10:5684.

    https://doi.org/10.1038/s41467-019-13607-2

    • PubMed
    • Google Scholar
41. 1. Ko H
    2. Hofer SB
    3. Pichler B
    4. Buchanan KA
    5. Sjöström PJ
    6. Mrsic-Flogel TD

    (2011) Functional specificity of local synaptic connections in neocortical networks

    Nature 473:87–91.

    https://doi.org/10.1038/nature09880

    • PubMed
    • Google Scholar
42. 1. Lee MC
    2. Yasuda R
    3. Ehlers MD

    (2010) Metaplasticity at single glutamatergic synapses

    Neuron 66:859–870.

    https://doi.org/10.1016/j.neuron.2010.05.015

    • PubMed
    • Google Scholar
43. 1. Lee HK
    2. Kirkwood A

    (2019) Mechanisms of homeostatic synaptic plasticity in vivo

    Frontiers in Cellular Neuroscience 13:520.

    https://doi.org/10.3389/fncel.2019.00520

    • PubMed
    • Google Scholar
44. 1. Lee CR
    2. Najafizadeh L
    3. Margolis DJ

    (2020) Investigating learning-related neural circuitry with chronic in vivo optical imaging

    Brain Structure & Function 225:467–480.

    https://doi.org/10.1007/s00429-019-02001-9

    • PubMed
    • Google Scholar
45. 1. Levelt CN
    2. Hübener M

    (2012) Critical-Period plasticity in the visual cortex

    Annual Review of Neuroscience 35:309–330.

    https://doi.org/10.1146/annurev-neuro-061010-113813

    • PubMed
    • Google Scholar
46. 1. Lewitus GM
    2. Pribiag H
    3. Duseja R
    4. St-Hilaire M
    5. Stellwagen D

    (2014) An adaptive role of TNFα in the regulation of striatal synapses

    The Journal of Neuroscience 34:6146–6155.

    https://doi.org/10.1523/JNEUROSCI.3481-13.2014

    • PubMed
    • Google Scholar
47. 1. Linden ML
    2. Heynen AJ
    3. Haslinger RH
    4. Bear MF

    (2009) Thalamic activity that drives visual cortical plasticity

    Nature Neuroscience 12:390–392.

    https://doi.org/10.1038/nn.2284

    • PubMed
    • Google Scholar
48. 1. Litwin-Kumar A
    2. Doiron B

    (2014) Formation and maintenance of neuronal assemblies through synaptic plasticity

    Nature Communications 5:5319.

    https://doi.org/10.1038/ncomms6319

    • PubMed
    • Google Scholar
49. 1. Ma Z
    2. Turrigiano GG
    3. Wessel R
    4. Hengen KB

    (2019) Cortical circuit dynamics are homeostatically tuned to criticality in vivo

    Neuron 104:655–664.

    https://doi.org/10.1016/j.neuron.2019.08.031

    • Google Scholar
50. 1. Marder E
    2. Prinz AA

    (2002) Modeling stability in neuron and network function: the role of activity in homeostasis

    BioEssays 24:1145–1154.

    https://doi.org/10.1002/bies.10185

    • PubMed
    • Google Scholar
51. 1. Meyer AF
    2. Poort J
    3. O’Keefe J
    4. Sahani M
    5. Linden JF

    (2018) A head-mounted camera system integrates detailed behavioral monitoring with multichannel electrophysiology in freely moving mice

    Neuron 100:46–60.

    https://doi.org/10.1016/j.neuron.2018.09.020

    • PubMed
    • Google Scholar
52. 1. Miller JK
    2. Ayzenshtat I
    3. Carrillo-Reid L
    4. Yuste R

    (2014) Visual stimuli recruit intrinsically generated cortical ensembles

    PNAS 111:E4053–E4061.

    https://doi.org/10.1073/pnas.1406077111

    • PubMed
    • Google Scholar
53. 1. Mrsic-Flogel TD
    2. Hofer SB
    3. Ohki K
    4. Reid RC
    5. Bonhoeffer T
    6. Hübener M

    (2007) Homeostatic regulation of eye-specific responses in visual cortex during ocular dominance plasticity

    Neuron 54:961–972.

    https://doi.org/10.1016/j.neuron.2007.05.028

    • PubMed
    • Google Scholar
54. 1. Okun M
    2. Steinmetz N
    3. Cossell L
    4. Iacaruso MF
    5. Ko H
    6. Barthó P
    7. Moore T
    8. Hofer SB
    9. Mrsic-Flogel TD
    10. Carandini M
    11. Harris KD

    (2015) Diverse coupling of neurons to populations in sensory cortex

    Nature 521:511–515.

    https://doi.org/10.1038/nature14273

    • PubMed
    • Google Scholar
55. 1. Pakan JM
    2. Lowe SC
    3. Dylda E
    4. Keemink SW
    5. Currie SP
    6. Coutts CA
    7. Rochefort NL

    (2016) Behavioral-state modulation of inhibition is context-dependent and cell type specific in mouse visual cortex

    eLife 5:e14985.

    https://doi.org/10.7554/eLife.14985

    • PubMed
    • Google Scholar
56. 1. Pakan JMP
    2. Currie SP
    3. Fischer L
    4. Rochefort NL

    (2018a) The impact of visual cues, reward, and motor feedback on the representation of behaviorally relevant spatial locations in primary visual cortex

    Cell Reports 24:2521–2528.

    https://doi.org/10.1016/j.celrep.2018.08.010

    • PubMed
    • Google Scholar
57. 1. Pakan JM
    2. Francioni V
    3. Rochefort NL

    (2018b) Action and learning shape the activity of neuronal circuits in the visual cortex

    Current Opinion in Neurobiology 52:88–97.

    https://doi.org/10.1016/j.conb.2018.04.020

    • PubMed
    • Google Scholar
58. 1. Pandey A
    2. Hardingham N
    3. Fox K

    (2022) Differentiation of hebbian and homeostatic plasticity mechanisms within layer 5 visual cortex neurons

    Cell Reports 39:110892.

    https://doi.org/10.1016/j.celrep.2022.110892

    • PubMed
    • Google Scholar
59. 1. Panzeri S
    2. Brunel N
    3. Logothetis NK
    4. Kayser C

    (2010) Sensory neural codes using multiplexed temporal scales

    Trends in Neurosciences 33:111–120.

    https://doi.org/10.1016/j.tins.2009.12.001

    • PubMed
    • Google Scholar
60. 1. Panzeri S
    2. Harvey CD
    3. Piasini E
    4. Latham PE
    5. Fellin T

    (2017) Cracking the neural code for sensory perception by combining statistics, intervention, and behavior

    Neuron 93:491–507.

    https://doi.org/10.1016/j.neuron.2016.12.036

    • PubMed
    • Google Scholar
61. 1. Petrus E
    2. Rodriguez G
    3. Patterson R
    4. Connor B
    5. Kanold PO
    6. Lee HK

    (2015) Vision loss shifts the balance of feedforward and intracortical circuits in opposite directions in mouse primary auditory and visual cortices

    The Journal of Neuroscience 35:8790–8801.

    https://doi.org/10.1523/JNEUROSCI.4975-14.2015

    • PubMed
    • Google Scholar
62. 1. Poort J
    2. Khan AG
    3. Pachitariu M
    4. Nemri A
    5. Orsolic I
    6. Krupic J
    7. Bauza M
    8. Sahani M
    9. Keller GB
    10. Mrsic-Flogel TD
    11. Hofer SB

    (2015) Learning enhances sensory and multiple non-sensory representations in primary visual cortex

    Neuron 86:1478–1490.

    https://doi.org/10.1016/j.neuron.2015.05.037

    • PubMed
    • Google Scholar
63. 1. Rose T
    2. Jaepel J
    3. Hübener M
    4. Bonhoeffer T

    (2016) Cell-specific restoration of stimulus preference after monocular deprivation in the visual cortex

    Science 352:1319–1322.

    https://doi.org/10.1126/science.aad3358

    • PubMed
    • Google Scholar
64. 1. Roth MM
    2. Dahmen JC
    3. Muir DR
    4. Imhof F
    5. Martini FJ
    6. Hofer SB

    (2016) Thalamic nuclei convey diverse contextual information to layer 1 of visual cortex

    Nature Neuroscience 19:299–307.

    https://doi.org/10.1038/nn.4197

    • PubMed
    • Google Scholar
65. 1. Saleem AB
    2. Diamanti EM
    3. Fournier J
    4. Harris KD
    5. Carandini M

    (2018) Coherent encoding of subjective spatial position in visual cortex and hippocampus

    Nature 562:124–127.

    https://doi.org/10.1038/s41586-018-0516-1

    • PubMed
    • Google Scholar
66. 1. Sammons RP
    2. Clopath C
    3. Barnes SJ

    (2018) Size-Dependent axonal bouton dynamics following visual deprivation in vivo

    Cell Reports 22:576–584.

    https://doi.org/10.1016/j.celrep.2017.12.065

    • PubMed
    • Google Scholar
67. 1. Stellwagen D
    2. Malenka RC

    (2006) Synaptic scaling mediated by glial TNF-alpha

    Nature 440:1054–1059.

    https://doi.org/10.1038/nature04671

    • PubMed
    • Google Scholar
68. 1. Stringer C
    2. Pachitariu M
    3. Steinmetz N
    4. Reddy CB
    5. Carandini M
    6. Harris KD

    (2019) Spontaneous behaviors drive multidimensional, brainwide activity

    Science 364:255.

    https://doi.org/10.1126/science.aav7893

    • PubMed
    • Google Scholar
69. 1. Tetzlaff C
    2. Kolodziejski C
    3. Timme M
    4. Wörgötter F

    (2011) Synaptic scaling in combination with many generic plasticity mechanisms stabilizes circuit connectivity

    Frontiers in Computational Neuroscience 5:47.

    https://doi.org/10.3389/fncom.2011.00047

    • PubMed
    • Google Scholar
70. 1. Todd TP
    2. Mehlman ML
    3. Keene CS
    4. DeAngeli NE
    5. Bucci DJ

    (2016) Retrosplenial cortex is required for the retrieval of remote memory for auditory cues

    Learning & Memory 23:278–288.

    https://doi.org/10.1101/lm.041822.116

    • PubMed
    • Google Scholar
71. 1. Tohmi M
    2. Meguro R
    3. Tsukano H
    4. Hishida R
    5. Shibuki K

    (2014) The extrageniculate visual pathway generates distinct response properties in the higher visual areas of mice

    Current Biology 24:587–597.

    https://doi.org/10.1016/j.cub.2014.01.061

    • PubMed
    • Google Scholar
72. 1. Turrigiano GG
    2. Leslie KR
    3. Desai NS
    4. Rutherford LC
    5. Nelson SB

    (1998) Activity-dependent scaling of quantal amplitude in neocortical neurons

    Nature 391:892–896.

    https://doi.org/10.1038/36103

    • PubMed
    • Google Scholar
73. 1. Turrigiano G

    (2011) Too many cooks? intrinsic and synaptic homeostatic mechanisms in cortical circuit refinement

    Annual Review of Neuroscience 34:89–103.

    https://doi.org/10.1146/annurev-neuro-060909-153238

    • PubMed
    • Google Scholar
74. 1. Turrigiano GG

    (2017) The dialectic of hebb and homeostasis

    Philos Trans R Soc Lond, B, Biol Sci 372:258.

    https://doi.org/10.1098/rstb.2016.0258

    • Google Scholar
75. 1. Vann SD
    2. Aggleton JP
    3. Maguire EA

    (2009) What does the retrosplenial cortex do?

    Nature Reviews. Neuroscience 10:792–802.

    https://doi.org/10.1038/nrn2733

    • PubMed
    • Google Scholar
76. 1. van Versendaal D
    2. Rajendran R
    3. Saiepour MH
    4. Klooster J
    5. Smit-Rigter L
    6. Sommeijer JP
    7. De Zeeuw CI
    8. Hofer SB
    9. Heimel JA
    10. Levelt CN

    (2012) Elimination of inhibitory synapses is a major component of adult ocular dominance plasticity

    Neuron 74:374–383.

    https://doi.org/10.1016/j.neuron.2012.03.015

    • Google Scholar
77. 1. Vitureira N
    2. Letellier M
    3. Goda Y

    (2012) Homeostatic synaptic plasticity: from single synapses to neural circuits

    Current Opinion in Neurobiology 22:516–521.

    https://doi.org/10.1016/j.conb.2011.09.006

    • PubMed
    • Google Scholar
78. 1. Wilson DE
    2. Whitney DE
    3. Scholl B
    4. Fitzpatrick D

    (2016) Orientation selectivity and the functional clustering of synaptic inputs in primary visual cortex

    Nature Neuroscience 19:1003–1009.

    https://doi.org/10.1038/nn.4323

    • PubMed
    • Google Scholar
79. 1. Wu YK
    2. Hengen KB
    3. Turrigiano GG
    4. Gjorgjieva J

    (2020) Homeostatic mechanisms regulate distinct aspects of cortical circuit dynamics

    PNAS 117:24514–24525.

    https://doi.org/10.1073/pnas.1918368117

    • Google Scholar
80. 1. Yoshimura Y
    2. Dantzker JLM
    3. Callaway EM

    (2005) Excitatory cortical neurons form fine-scale functional networks

    Nature 433:868–873.

    https://doi.org/10.1038/nature03252

    • PubMed
    • Google Scholar
81. 1. Yoshitake K
    2. Tsukano H
    3. Tohmi M
    4. Komagata S
    5. Hishida R
    6. Yagi T
    7. Shibuki K

    (2013) Visual map shifts based on whisker-guided cues in the young mouse visual cortex

    Cell Reports 5:1365–1374.

    https://doi.org/10.1016/j.celrep.2013.11.006

    • PubMed
    • Google Scholar
82. 1. Zenke F
    2. Hennequin G
    3. Gerstner W

    (2013) Synaptic plasticity in neural networks needs homeostasis with a fast rate detector

    PLOS Comput Biol 9:e1003330.

    https://doi.org/10.1371/journal.pcbi.1003330

    • Google Scholar
83. 1. Zhuang X
    2. Sun W
    3. Xu-Friedman MA

    (2017) Changes in properties of auditory nerve synapses following conductive hearing loss

    The Journal of Neuroscience 37:323–332.

    https://doi.org/10.1523/JNEUROSCI.0523-16.2016

    • PubMed
    • Google Scholar

Decision letter after peer review:

Thank you for submitting your article "Homeostatic regulation through the strengthening of neuronal network-correlated synaptic inputs" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Lu Chen as the Senior Editor. The following individuals involved in the review of your submission have agreed to reveal their identity: Yukiko Goda (Reviewer #1); Keith B Hengen (Reviewer #3).

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

Essential revisions:

All three reviewers have found the work to be important and of considerable general interest. Please carefully address individual points raised by the three reviewers, which can be dealt with by text edits or minor figure modifications. In particular, the fact that the present study used adult animals and that homeostatic plasticity in the adult brain could follow different rules from juvenile animals should be stressed.

Reviewer #1 (Recommendations for the authors):

1) Figure 2C – It would be important to include also a histogram of % visual stimuli eliciting time-locked responses for all deprived spines following deprivation as a measure of how well the system is responding to deprivation. In other words, is there an effect of enucleation on the reliability with which visual stimuli trigger responses?

2) Figure 2 – Related to the point above, only panel B appears to show results pertaining to post-enucleation, which is quite confusing. One should reconsider the panel order, perhaps moving panels B and C with both plots including the respective control and deprived data.

3) Figure 3C – Some network spines do show loss of activity upon sensory deprivation. Given that the proportion of network spines is 65% compared to 24% of visual spines, to begin with, one would expect that increase in the proportion of inactive spines by 10% for network spines (Figure 3F) represents a larger total number of spines than 20% increase for visual spines (Figure 3D). Could the increase in normalized amplitude shown by network spines be a consequence of the reduced activity of some of the network spines and secondary to the direct effects of loss of visual inputs? This point requires consideration/discussion. It would be informative to include a blind assessment of relative proportions of visual, network, and unclassified spines following enucleation and a comparison of the numbers to those of baseline in a graphical representation.

Reviewer #2 (Recommendations for the authors):

1. Methods: "Intrinsic signal imaging before enucleation was used to localize the visual cortex…" Were the recorded spines/dendrites in binocular or monocular V1? Baseline visual stimulation (gratings) was bilateral? It seems surprising that each spine's ocular dominance was not determined at baseline – please explain. I would expect different reactions to enucleation in a neuron that was mainly driven by the missing eye vs the intact eye.

2. In the example trace Figure S1B, please label all spine events that were counted for the frequency analysis (Figure S1E).

3. Figure 1E Are these spine calcium transients' responses to a visual stimulus (aligned to stimulation time) or are they 'spontaneous' events aligned by peak or max. slope? If the latter, please don't call them responses but find a more neutral term (e.g. EPSCaTs).

4. Figure 2B: "When we examined the functional strengthening of the maintained visually responsive inputs, we found an inverse relationship, such that the spines with the lowest visual responsiveness prior to deprivation underwent the strongest functional strengthening after deprivation." Why is there a data point at zero visual responses, wouldn't that be an unclassified or network spine?

5. Figure 3 G-I: It is not well explained why you show this lack of correlation and what can be concluded from it. To me, frequency changes point to presynaptic plasticity (either release at the observed synapse or spike frequency of the presynaptic neuron), and amplitude changes to a postsynaptic mechanism. It seems you have both, but rarely in the same spines. I don't like the binning – why not show a cloud of all spines and the correlation coefficient? Try drawing a conclusion from these plots.

6. Figure 4 B: Please show raw traces under conditions of TNF α block. Was the amplitude of Ca transients at baseline affected by the inhibitor? It is my understanding that you use the blocker mainly as a diagnostic criterion for "true homeostatic plasticity", as you draw no further conclusions (e.g. about the potential involvement of glia cells).

7. Figure 4J: How many branches were analyzed? (N missing in Table S4).

8. L.337: "Further work is required to better understand the key neural circuit and molecular features that allow TNF-α to selectively modulate homeostatic adaption and even more specifically, the network correlated spines." You see the enhanced activity of network-correlated spines, but homeostatic plasticity could well happen elsewhere, e.g. in L4. It cannot be concluded that TNF-α selectively acts on the spines you happen to look at (a.k.a. streetlight effect). It would be good to spell out that the locus of homeostatic plasticity could be upstream of the synapses under scrutiny (unless you have good arguments that plasticity is local).

Reviewer #3 (Recommendations for the authors):

1. Do the authors have the capacity to separate V1m and V1b in their imaging sessions? The responses to monocular deprivation (via enucleation) in V1m should be purely homeostatic, while the effect in V1b is most likely a combination of homeostatic and Hebbian. The TNF α experiments suggest that a significant component of the data is homeostatic (assuming all TNF α plasticity is homeostatic), but it would be powerful to show that the key effects are evident in V1m. This is certainly not necessary for the paper, but if it were possible, it would be quite nice to see and would complement the TNF experiments.

2. Line 64: change "Following deprivation" to "Following sensory deprivation". At this point, there's been no definition of deprivation so readers from outside the field might be confused.

3. Figure 4: the yellow and gray bars (e.g., 4B) are very confusing – my first impression was that they indicated normal vision and 48h dark deprivation. Then, in F through I, I read the figure as if animals were simultaneously deprived of vision and auditory input (note that there is both a gray bar and plugged ears), but the legend says, "timeline for dark exposure (B) OR auditory ear plug (F)", which seems to suggest that these were not performed concurrently.

4. The authors cite Barnes 2017 and Bridi 2018 as the basis of the hypothesis that subsets of spines are subject to compensatory changes. Barnes 2017 clearly shows this, as well as being in adult animals. Bridi 2018 is based on CP animals and, as far as I can tell, doesn't make any claims about synaptic specificity: they measured extracellular spikes and mEPSCs (in whole cell patch), which cannot resolve synaptic subsets. Their key finding (to my eye) is a novel mechanism for compensatory plasticity, not synaptic specificity:

"In sum, our manipulations of visual input combined with manipulations of spontaneous neuronal activity reveal that (i) changes in overall neuronal firing rate alone are not sufficient to mediate the lowered modification threshold for LTP/LTD nor the increased mEPSC amplitude observed after visual deprivation, (ii) changes in mEPSC amplitude induced by visual deprivation reflect Hebbian potentiation of excitatory synapses enabled by the reduced synaptic modification threshold, and (iii) changes in mEPSC amplitude induced by extreme reductions in neuronal activity reflect NMDAR-independent synaptic scaling."

Essential revisions:

All three reviewers have found the work to be important and of considerable general interest. Please carefully address individual points raised by the three reviewers, which can be dealt with by text edits or minor figure modifications. In particular, the fact that the present study used adult animals and that homeostatic plasticity in the adult brain could follow different rules from juvenile animals should be stressed.

We thank the reviewers for their positive and helpful comments. We agree that the age of the animals is a key consideration for our results and we have changed the manuscript to emphasize this point. We have first made sure that the age of the animals is clear throughout the manuscript (abstract, introduction, results, and discussion). We have also highlighted the specifics of adult plasticity, and its relation to juvenile plasticity, in the introduction and discussion to provide context for our findings, as well as how our work is complementary with previous studies in adult animals.

We have addressed the comments of the individual reviewers below.

Reviewer #1 (Recommendations for the authors):

1) Figure 2C – It would be important to include also a histogram of % visual stimuli eliciting time-locked responses for all deprived spines following deprivation as a measure of how well the system is responding to deprivation. In other words, is there an effect of enucleation on the reliability with which visual stimuli trigger responses?

We have double checked and there are no visually-evoked responses following deprivation. It is important to clarify that the visual cortex experiments in our study were done in monocular visual cortex, which we have now specified throughout the manuscript. Because all visual spines also have significant network signals (Figure 2), the spines that were identified as visually responsive during baseline (prior to sensory deprivation), only have network responses following visual deprivation, since there is no longer visual input from the contralateral eye. None of the spines have visually-evoked responses above chance after deprivation. We have specified this in the text now. Lines:

“It is important to note that following sensory deprivation, we no longer observed responses to visual stimuli in the visually responsive spines, since the feedforward input had been permanently removed. The responses of the previously categorized visually responsive spines consisted of network responses following deprivation.”

2) Figure 2 – Related to the point above, only panel B appears to show results pertaining to post-enucleation, which is quite confusing. One should reconsider the panel order, perhaps moving panels B and C with both plots including the respective control and deprived data.

We have made this adjustment and switched the order of the panels 2B and 2C, which we agree makes the manuscript easier to follow.

3) Figure 3C – Some network spines do show loss of activity upon sensory deprivation. Given that the proportion of network spines is 65% compared to 24% of visual spines, to begin with, one would expect that increase in the proportion of inactive spines by 10% for network spines (Figure 3F) represents a larger total number of spines than 20% increase for visual spines (Figure 3D). Could the increase in normalized amplitude shown by network spines be a consequence of the reduced activity of some of the network spines and secondary to the direct effects of loss of visual inputs? This point requires consideration/discussion. It would be informative to include a blind assessment of relative proportions of visual, network, and unclassified spines following enucleation and a comparison of the numbers to those of baseline in a graphical representation.

We thank the reviewer for raising this important point. We have adjusted the text to reflect this point that while the relative fraction of visual spines that become inactive is higher than the relative fraction of inactive network spines, the absolute number of network spines that become inactive is larger. Lines:

“While a larger fraction of visually responsive spines become inactive, the absolute number of network-correlated spines that became inactive was higher, since network-correlated spines reflect a larger proportion of the total spine population. These results suggest that the increase in spine loss previously reported following enucleation (Barnes et al., 2017) may include both spines that were visually responsive and network-correlated prior to deprivation (Figure 3D-F).”

Secondly, the increase in the number of inactive network spines and subsequent strengthening only occurs in deprived animals, so even if the effect is secondary to visual deprivation, our data indicate that it is triggered by deprivation. We consistently see that there are not systematic changes in density/size/functional response strength of spines without the visual deprivation in our previous work (Barnes et al., 2017), as well as in this study.

As for the proportion of spines in each category following enucleation, the spines fall into only two categories, network and unclassified, since we do not observe visually-evoked responses following enucleation, as we are recording in monocular visual cortex. We do not see a difference in the unclassified spines following deprivation (0% of unclassified spines (categorized prior to deprivation) correlate with the network signal after deprivation), so there is no change in categorization, other than the visual spines would now be categorized as network-correlated.

Reviewer #2 (Recommendations for the authors):

1. Methods: "Intrinsic signal imaging before enucleation was used to localize the visual cortex…" Were the recorded spines/dendrites in binocular or monocular V1? Baseline visual stimulation (gratings) was bilateral? It seems surprising that each spine's ocular dominance was not determined at baseline – please explain. I would expect different reactions to enucleation in a neuron that was mainly driven by the missing eye vs the intact eye.

We apologize for the confusion on this point and have now clarified in the manuscript that the visual cortex experiments in the study were done in monocular visual cortex. All visual stimulation was bilateral, which we have now clarified in the methods. Given that we were measuring in monocular visual cortex, we did not measure ocular dominance. Furthermore, given that we see no visually-evoked responses to bilateral visual stimulation following sensory deprivation, we think that it is unlikely that spines are responding to inputs from the intact ipsilateral eye, consistent with expectations in monocular visual cortex.

2. In the example trace Figure S1B, please label all spine events that were counted for the frequency analysis (Figure S1E).

We thank the reviewer for this helpful suggestion and have denoted events that were included in our analysis with an asterisk above the trace.

3. Figure 1E Are these spine calcium transients' responses to a visual stimulus (aligned to stimulation time) or are they 'spontaneous' events aligned by peak or max. slope? If the latter, please don't call them responses but find a more neutral term (e.g. EPSCaTs).

This is a great point and we have changed the figure to refer to them as spine events. These examples include all spine events, which were aligned by the peak. We have clarified this in the figure legend.

4. Figure 2B: "When we examined the functional strengthening of the maintained visually-responsive inputs, we found an inverse relationship, such that the spines with the lowest visual responsiveness prior to deprivation underwent the strongest functional strengthening after deprivation." Why is there a data point at zero visual responses, wouldn't that be an unclassified or network spine?

We thank the reviewer for catching this mistake. This figure includes all spines, not just the visually responsive ones, and we have changed this text accordingly.

5. Figure 3 G-I: It is not well explained why you show this lack of correlation and what can be concluded from it. To me, frequency changes point to presynaptic plasticity (either release at the observed synapse or spike frequency of the presynaptic neuron), and amplitude changes to a postsynaptic mechanism. It seems you have both, but rarely in the same spines. I don't like the binning – why not show a cloud of all spines and the correlation coefficient? Try drawing a conclusion from these plots.

We have included this comparison, as previous work has indicated that frequency and amplitude could be modulated within a single synapse (Lee et al., 2010). In other words, the frequency of responses in a spine goes down (likely presynaptically mediated) and then the response amplitude of that spine increases. The correlation we are measuring here tests that directly and finds that decreases in frequency in an individual spine is unlikely to be driving the increases in amplitude. We have included this conclusion in the manuscript now. Lines:

“Previous work has shown that spine specific decreases in response frequency drive synapse-specific increases in response amplitude (Lee et al., 2010). We therefore examined if frequency changes were indicative of amplitude changes in individual spines. We observed no correlation between changes in frequency and amplitude within individual spines (Figure 1 – —figure supplement 1F), indicating that the changes in frequency are not likely to be driving the changes in amplitude within a single spine.”

6. Figure 4 B: Please show raw traces under conditions of TNF α block. Was the amplitude of Ca transients at baseline affected by the inhibitor? It is my understanding that you use the blocker mainly as a diagnostic criterion for "true homeostatic plasticity", as you draw no further conclusions (e.g. about the potential involvement of glia cells).

We have added these traces to Figure 4B. The baseline amplitude of Ca transients is not altered by the TNF-α inhibitor (TNF-α 79.2 ± 4.2 vs Control 81.8 ± 3.1 % ∆F/F, p>0.5, t-test). As the reviewer suggests, we have used this experiment to determine whether or not our effect is dependent on TNF-α, which is often used as an indicator of synaptic scaling. It is still unclear exactly how TNF-α interacts with synaptic strengthening, but we have included the known role of glia in this interaction in the discussion.

7. Figure 4J: How many branches were analyzed? (N missing in Table S4).

In order to measure the global dendritic signals, we measured the dendrite signal adjacent to the dendritic spine. Thus, the number of spine and dendrite measurements are the same. This particular approach is critical to avoid any ambiguity that occurs in identifying contiguous dendritic branches when a dendrite exits and re-enters the imaging plane. We have clarified this point in the methods. Lines:

“Global dendritic signals were measured for each spine in the dendritic area adjacent to that spine. This approach avoids the ambiguity associated with determining whether or not dendritic branches are contiguous when they enter and exit the imaging depth plane.”

8. L.337: "Further work is required to better understand the key neural circuit and molecular features that allow TNF-α to selectively modulate homeostatic adaption and even more specifically, the network correlated spines." You see the enhanced activity of network-correlated spines, but homeostatic plasticity could well happen elsewhere, e.g. in L4. It cannot be concluded that TNF-α selectively acts on the spines you happen to look at (a.k.a. streetlight effect). It would be good to spell out that the locus of homeostatic plasticity could be upstream of the synapses under scrutiny (unless you have good arguments that plasticity is local).

We agree with the reviewer and have adjusted the language about TNF-α and added text to highlight that the locus of plasticity could be outside of layer 5. Lines:

“Finally, it is important to note that while we made these measurements in layer 5 pyramidal cells, the homeostatic changes mediated by TNF-α could occur outside of layer 5, including changes to upstream inputs or changes to the presynaptic responses, either through changes in presynaptic release (Vitureira et al., 2012) or through a change in activity patterns of the presynaptic cell (e.g., bursts compared to single spikes) (Linden et al., 2009). Given that both sensory responsive spines (which do not strengthen) and network-correlated spines (which do strengthen) exhibit only network-correlated responses after sensory deprivation, changes in presynaptic activity patterns of network signals would require targeted changes to the network-corelated spines, but not the previously sensory responsive spines. Further research is necessary to identify the role of presynaptic plasticity in homeostatic synaptic plasticity in vivo.”

Reviewer #3 (Recommendations for the authors):

1. Do the authors have the capacity to separate V1m and V1b in their imaging sessions? The responses to monocular deprivation (via enucleation) in V1m should be purely homeostatic, while the effect in V1b is most likely a combination of homeostatic and Hebbian. The TNF α experiments suggest that a significant component of the data is homeostatic (assuming all TNF α plasticity is homeostatic), but it would be powerful to show that the key effects are evident in V1m. This is certainly not necessary for the paper, but if it were possible, it would be quite nice to see and would complement the TNF experiments.

We again apologize for this confusion. We have now clarified throughout the manuscript that the visual cortex experiments in our study were conducted in monocular visual cortex and so all of the TNF-α effects in visual cortex are in monocular visual cortex.

2. Line 64: change "Following deprivation" to "Following sensory deprivation". At this point, there's been no definition of deprivation so readers from outside the field might be confused.

We have changed this as suggested.

3. Figure 4: the yellow and gray bars (e.g., 4B) are very confusing – my first impression was that they indicated normal vision and 48h dark deprivation. Then, in F through I, I read the figure as if animals were simultaneously deprived of vision and auditory input (note that there is both a gray bar and plugged ears), but the legend says, "timeline for dark exposure (B) OR auditory ear plug (F)", which seems to suggest that these were not performed concurrently.

We thank the reviewer for pointing out this confusing issue. The dark exposure and auditory ear plug were not performed concurrently, but are separate sets of experiments. We have changed Figure 4 and Figure 4 – —figure supplement 1 so that the background colors are different for the two experiments. We have also clarified in the text that these are separate experiments.

4. The authors cite Barnes 2017 and Bridi 2018 as the basis of the hypothesis that subsets of spines are subject to compensatory changes. Barnes 2017 clearly shows this, as well as being in adult animals. Bridi 2018 is based on CP animals and, as far as I can tell, doesn't make any claims about synaptic specificity: they measured extracellular spikes and mEPSCs (in whole cell patch), which cannot resolve synaptic subsets. Their key finding (to my eye) is a novel mechanism for compensatory plasticity, not synaptic specificity:

"In sum, our manipulations of visual input combined with manipulations of spontaneous neuronal activity reveal that (i) changes in overall neuronal firing rate alone are not sufficient to mediate the lowered modification threshold for LTP/LTD nor the increased mEPSC amplitude observed after visual deprivation, (ii) changes in mEPSC amplitude induced by visual deprivation reflect Hebbian potentiation of excitatory synapses enabled by the reduced synaptic modification threshold, and (iii) changes in mEPSC amplitude induced by extreme reductions in neuronal activity reflect NMDAR-independent synaptic scaling."

We thank the reviewer for this helpful point and we have taken this point out of the manuscript.

Author details

1. Samuel J Barnes

   1. Department of Brain Sciences, Division of Neuroscience, Imperial College London, Hammersmith Hospital Campus, London, United Kingdom
   2. UK Dementia Research Institute at Imperial College, London, United Kingdom

   Contribution

   Conceptualization, Data curation, Formal analysis, Methodology, Writing - original draft, Writing - review and editing

   Competing interests

   No competing interests declared

2. Georg B Keller

   Friedrich Miescher Institute for Biomedical Research, Basel, Switzerland

   Contribution

   Resources, Software, Formal analysis, Funding acquisition, Investigation, Methodology, Writing - review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-1401-0117

3. Tara Keck

   Department of Neuroscience, Physiology and Pharmacology, University College London, London, United Kingdom

   Contribution

   Conceptualization, Data curation, Formal analysis, Supervision, Funding acquisition, Investigation, Methodology, Writing - original draft, Project administration, Writing - review and editing

   For correspondence

   t.keck@ucl.ac.uk

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-6623-1037

• 986

  Page views

• 150

  Downloads

• 0

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.