Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, NEUROSCIENCE ( (c) Oxford University Press USA, 2019. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

date: 19 October 2019

Homeostatic Plasticity in the CNS

Summary and Keywords

Homeostatic plasticity refers to a collection of mechanisms that function to homeostatically maintain some feature of neural function. The field began with the view that homeostatic plasticity exists predominantly for the maintenance of spike rate. However, it has become clear that multiple features undergo some form of homeostatic control, including network activity, burst rate, or synaptic strength. There are several different forms of homeostatic plasticity, which are typically triggered following perturbations in activity levels. Homeostatic intrinsic plasticity (HIP) appears to compensate for the perturbation with changes in membrane excitability (voltage-gated conductances); synaptic scaling is thought to be a multiplicative increase or decrease of synaptic strengths throughout the cell following an activity perturbation; presynaptic homeostatic plasticity is a change in probability of release following a perturbation to postsynaptic receptor activity. Each form of homeostatic plasticity can be different in terms of the mechanisms that are engaged, the feature that is homeostatically regulated, the trigger that initiates the compensation, and the signaling cascades that mediate these processes. Homeostatic plasticity is often described in development, but can extend into maturity and has been described in vitro and in vivo.

Keywords: synaptic scaling, homeostatic intrinsic plasticity, homeostatic presynaptic plasticity, compensation, stability

Homeostatic Intrinsic Plasticity

The intrinsic excitability of a neuron is defined by its constellation of ion channels expressed at the membrane, which ultimately determines a cell’s propensity to fire an action potential. Seminal studies identified that the intrinsic membrane excitability of a cell can be regulated to homeostatically maintain a cell’s spiking activity level (Desai, Rutherford, & Turrigiano, 1999; Turrigiano, Abbott, & Marder, 1994; Turrigiano, LeMasson, & Marder, 1995). This type of plasticity is called homeostatic intrinsic plasticity (HIP), which represents a set of mechanisms that maintain spiking activity in a cell in response to an activity perturbation through compensatory changes in the function and/or expression of certain ion channels (Figure 1). The first study to demonstrate HIP was carried out in the stomatogastric ganglion (STG) of the lobster (Turrigiano et al., 1994). In the intact system, STG neurons display a bursting activity pattern that is dependent on synaptic signaling, modulators, and intrinsic membrane excitability. When a single cell from this network was isolated and plated in a petri dish, synaptic input and neuromodulators were lost. Initially the cell did not fire any action potentials. Remarkably, after one day in culture, the cell began to tonically spike in response to current depolarizations, and following three days of isolation cells returned to a burst activity pattern that was largely similar to cells from the intact network. Because this isolated cell was devoid of any synaptic input, it was clear that the observed recovery of firing activity was mediated through changes in intrinsic properties. Consistent with this expectation, the authors found that various voltage-gated ion channels had changed conductance in a compensatory direction, which mediated the recovery of bursting activity (Turrigiano et al., 1995). Importantly, this homeostatic recovery of neuronal activity through HIP was bidirectional.

Homeostatic Plasticity in the CNSClick to view larger

Figure 1. Schematic showing an activity perturbation that leads to compensatory changes in ion channel constellation and therefore excitability through HIP mechanisms.

Do synaptically connected networks also recover spiking activity patterns through homeostatic changes in intrinsic properties? Subsequent studies have addressed this question in cortical and hippocampal neuronal cultures or organotypic slices. These studies demonstrated that chronic pharmacological blockade of spiking activity (using the voltage-gated sodium channel blocker tetrodotoxin [TTX]) led to an increase in firing frequency and reduced the action potential threshold in response to depolarizing current injections (Desai et al., 1999; Karmarkar & Buonomano, 2006; Lee & Chung, 2014). Importantly, this increase in excitability was observed even in the presence of synaptic blockers, consistent with the idea that changes in intrinsic membrane properties mediate the compensatory increases in excitability. Studies have shown that these compensatory changes, like that in the STG, are bidirectional, as chronic increases in excitability (e.g., by blockade of the alpha subunit of the gamma-Aminobutyric acid receptor [GABAARs]) led to reduced cellular excitability (Desai et al., 1999; Lee & Chung, 2014). The studies referred to thus far describe HIP using in vitro systems, but studies have now demonstrated that HIP is also expressed in response to chronic activity perturbations in vivo (Lambo & Turrigiano, 2013; Wilhelm, Rich, & Wenner, 2009). For example, if rats are chronically deprived of visual inputs from one eye (e.g., through lid suture), compensatory changes in the intrinsic membrane excitability of visual cortical neurons is observed in acute slices (Lambo & Turrigiano, 2013; Maffei & Turrigiano, 2008). However, these homeostatic changes are cell-type and layer specific, underscoring the complexity of HIP’s expression mechanisms.

Expression Mechanisms

In one of the original studies, it was identified that both voltage-gated sodium and various potassium currents were altered in response to pharmacological isolation of the STG neuron in a manner consistent with the time course of recovery of bursting activity patterns. Specifically, it was observed that while the peak voltage-gated Na+ conductance increased, both fast and delayed rectifier type K+ channels exhibited decreased conductance (Turrigiano et al., 1995). Similarly, in synaptically coupled networks there were changes in voltage-gated ion channel currents following activity perturbation (Aizenman, Akerman, Jensen, & Cline, 2003; Desai et al., 1999; Lee & Chung, 2014; Wilhelm et al., 2009). Interestingly, although some channels appear to be uniformly regulated across tissue/cell types (e.g. voltage-gated Na+ channels), the majority of channels are not (see Table 1). For example, while activity blockade reduces the delayed rectifier K+ current in primary visual cortical neuronal cultures, it remains unchanged in chick spinal motoneurons. The opposite relationship is true for the fast-activating K+ current IA. Whether these differences represent variation in the developmental stage, tissue/preparation, or type of activity perturbation remains to be answered. But what is clear is that voltage-gated ion channels are regulated across very different regions of the central nervous system in response to activity perturbation, suggesting that HIP is a general mechanism utilized by the brain and spinal cord to maintain targeted activity levels.

Table 1. Activity Perturbations Trigger Diverse Changes in Ion Channel Expression and Function









STG cell

Pharm isol

Turrigiano, 1995


Prim vis cort neurons

TTX, 24 hr

Desai, 1999



Visual stim

Aizenman, 2003

Correlated with number of spikes and rise time slope


Chick spinal cord

Gabazine (act block)

Wilhelm, 2009

Mediated by changes in the activity of a translation repressor, Pumilio. Is specific to fast Na channels


Drosophila motoneuron

Genetic manipulation

Mee et al., 2004


Rat hippocampal slice cultures

TTX, 48 hr

Aptowicz et al., 2004

Increase seen specifically in dendrites



STG cell

Pharm iso

Turrigiano, 1995



Prim vis cort


Desai, 1999


No change

neurons Tadpole

Visual stim

Aizenman, 2003


No change

Chick spinal cord

Gabazine (act block)

Wilhelm, 2009


Prim hipp mouse neurons


Lee et al., 2015


Chick auditory cortex

Auditory deprivation

Kuba, 2015



STG cell

Pharm iso

Turrigiano, 1995

No change

Prim vis cort neurons


Desai, 1999


No change


Visual stim

Aizenman, 2003


Chick spinal cord

Gabazine (act block)

Wilhelm, 2009



STG cell

Pharm iso

Turrigiano, 1995

No change

Prim vis cort neurons


Desai, 1999

No change

Chick spinal cord

Gabazine (act block)

Wilhelm, 2009

No change

Drosophila motoneuron

Increase/decrease in Pumilio levels

Muraro et al., 2008


Slight increase

STG cell

Pharm iso

Turrigiano, 1995



Prim vis cort neurons


Desai, 1999



Chick spinal cord

Gabazine (act block)

Wilhelm, 2009

I-K-slow (M-type)


Prim hipp mouse neurons


Lee et al., 2015



Auditory chick cortex

Auditory deprivation

Kuba, 2015

Shal (Kv4) (I-K-fast)



Nicotine antagonist

Ping & Tsunoda, 2012

Nic block à increase nAchR expression à increase Ca influx à CAMKII à upreg Shal à stabilized synaptic potentials



Acute hipp mouse slices

3 Hz stimulation (LTD)

Brager & Johnston, 2007

Mediated by protein kinase C pathway

Leak K channels (voltage independent)


Acute cerebellar mouse slices

Deletion of GABAR delta subunit (tonic GABA current)

Brickley et al., 2001

Demonstrating that neurons can engage HIP to compensate for deletions of certain receptor subunits to maintain normal excitability and presumably behavioral function (e.g., walking)

The expression of HIP is thought to act relatively quickly, as homeostatic changes in excitability and ion channel function have been reported as early as one hour (Turrigiano et al., 1995), four hours (Aizenman et al., 2003), or 12 hours (Wilhelm et al., 2009) following chronic activity perturbation. HIP appears to occur prior to synaptic scaling (Karmarkar & Buonomano, 2006; Wilhelm et al., 2009), suggesting that it may be particularly important for an initial fast recovery of neuronal activity.

Modeling HIP using Calcium as a Sensor

The initial experiments demonstrating HIP showed that changes in spike rate or pattern triggered HIP (Turrigiano et al., 1994, 1995). This suggested the presence of an intracellular sensor, whose concentration levels correlated with spiking activity. The clearest candidate for such a role was intracellular calcium levels, as suggested in the initial study using the isolated STG neuron (Turrigiano et al., 1994). Calcium levels are clearly influenced by spiking activity patterns (Ross, 1989), and can control ion channel expression by influencing signaling cascades that regulate transcription and translation (Schulz & Lane, 2017). Critically, seminal work modeling neurons of rhythmically active networks demonstrated the capacity of calcium as a key mediator of HIP (LeMasson, Marder, & Abbott, 1993). The model used the calcium level as a sensor of activity and various ion channel conductances were regulated in response to changes in calcium to restore target calcium levels associated with normal activity. For example, when calcium levels were below this target level (lower activity) the calcium error signal would drive increases in inward and decreases in outward ion channel conductances. The strength of the model was based on the observation that following a perturbation the system recovered its normal activity pattern over time, suggesting that calcium could act as a reliable sensor of spiking activity patterns. Several later models, with more sophisticated approaches (e.g., less constrained channel variables) were able to provide even more accurate results (Liu, Golowasch, Marder, & Abbott, 1998; O’Leary, Williams, Caplan, & Marder, 2013). These observations are multifold and are now discussed.

The models have several strengths. First, they have the flexibility to recover spiking activity patterns using many different combinations of ion channel densities from a range of starting ion channel conductances. This matches the observations that cells with essentially identical activity patterns have dramatically different combinations of ion channel conductances (Marder, Goeritz, & Otopalik, 2015; Prinz, Bucher, & Marder, 2004). For instance, one cell might have a large delayed rectifier K+ channel conductance but only a weak Ia conductance, while another cell may have the opposite arrangement but express the same pattern of activity. Second, these models coordinate to regulate channel conductances due to the calcium error signal, which drives correlations of different ion channel conductances (O’Leary et al., 2013). These channel correlations, described in rhythmically active neurons, are thought to be an important part of the homeostatic process (Temporal, Lett, & Schulz, 2014; Tobin, Cruz-Bermudez, Marder, & Schulz, 2009). Third, certain models incorporate the calcium-dependent control of ion channel translation via regulation of messenger ribonucleic acid (mRNA) (O’Leary, Williams, Franci, & Marder, 2014). The strengths of the models are reflected in their similarity to real biological systems. Importantly, the models have also led to important predictions that are informing and shaping future scientific experiments. The models explain how altering certain ion channels leads to different responses in terms of recovering or maintaining activity levels. If ion channels serve redundant roles then their loss can be easily compensated (Bergquist, Dickman, & Davis, 2010; Lane, Samarth, Ransdell, Nair, & Schulz, 2016; Ransdell, Nair, & Schulz, 2012), but certain channels were found to be irreplaceable and their absence led to pathological consequences (O’Leary et al., 2014). This is consistent with the observation that certain channel mutations leading to loss of function can drive neuronal networks into aberrant states of function, for example seizures (Meisler & Kearney, 2005). In addition, when the polarity of channel regulation was inverted for one channel in an antihomeostatic direction (e.g., Ia increased following reduced activity), the models were still able to homeostatically recover activity through the regulation of other channels (O’Leary et al., 2013, 2014). These complex and sometimes antihomeostatic changes in channel conductances have been described in homeostatic studies on biological systems and support the robust nature of HIP (Table 1).

Calcium-Dependent Pathways that Mediate HIP in Biological Systems

The modeling studies inspired a great deal of work assessing calcium’s role in mediating HIP in cultured neural networks. Calcium can enter the intracellular compartments of a cell through various routes: (a) voltage-gated calcium channels, (b) N-methyl-D-aspartate receptors (NMDARs), and (c) calcium-permeable α‎-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors (AMPARs). Blocking L-type voltage-gated Ca2+ channels does not trigger HIP (increased cell excitability) in hippocampal cultures; however, these channels may play a role in recovering excitability levels after chronic GABAergic blockade, which initially elevates activity levels (Lee et al., 2015). This could suggest that other calcium channels may be involved or, alternatively, that calcium influx through calcium-permeable receptors (NMDAR or Ca2+-permeable AMPAR) is important for triggering HIP. In fact, blocking just NMDAR activity with the antagonist (2R)-amino-5-phosphonovaleric acid (APV) resulted in compensatory changes in intrinsic excitability that were indistinguishable from treatment with TTX, a combination of TTX and APV, or even a combination of APV and GABAergic blockade (Lee & Chung, 2014; Lee et al., 2015). This result suggested that triggers for HIP in these cultures were highly influenced by calcium entry through NMDARs. Although many studies have demonstrated the importance NMDAR transmission in triggering homeostatic signaling cascades (Lee et al., 2015; Reese & Kavalali, 2015; Sutton et al., 2006), one study has given an example of how changes in calcium influx through calcium-permeable AMPARs can also trigger HIP (Aizenman et al., 2003).

Intracellular calcium concentration is known to correlate with cellular activity levels, and calcium can induce gene expression and protein phosphorylation. While studies have looked at the dependence of transcription or translation on homeostatic synaptic plasticity, less is known with regard to HIP. One study reported that activity blockade (TTX) led to an increase in intrinsic excitability (HIP), which was dependent on calcium/calmodulin-dependent protein kinase type IV (CAMKIV) signaling and presumably transcription (Joseph & Turrigiano, 2017). However, transcription-dependent processes work over a slower timescale, which is inconsistent with some of the fast changes in intrinsic properties observed in HIP. Thus, an alternative possibility is that changes in intrinsic properties are triggered through non-transcription dependent pathways, such as fast phosphorylation, trafficking, or local translation of ion channel proteins. One study conducted in the drosophila, found that increasing or decreasing motoneuron activity led to bidirectional changes in the expression/activity of the translational repressor Pumilio (Mee et al., 2004). This change mediated the compensatory changes measured in the function of the fast voltage-gated Na+ channel. A later study demonstrated that Pumilio could also lead to a reduction in a K+ channel (Shal) through an indirect pathway, potentially as a compensation to the reduced Na+ channel conductance (Muraro et al., 2008). Separate studies have shown that rapid changes in intracellular calcium levels can modify the phosphorylation status and function of voltage-gated Na+ channels, with consequences for spike threshold and frequency (Evans, Dumitrescu, Kruijssen, Taylor, & Grubb, 2015).

Thus, it is clear that HIP can be triggered through diverse signaling routes and can lead to the regulation of many different ion channels. However, what seems to be an overarching theme is the importance of activity-dependent changes in calcium influx. Importantly, in studies on HIP, calcium conductance is rarely changed in response to activity perturbation (see Table 1), which may suggest that calcium is at least one of the features being monitored and maintained by homeostatic mechanisms.

Perturbation-Specific Signaling Cascades Underlying HIP

The synaptic scaling literature has clearly demonstrated that different types of activity perturbation (spike vs. transmission) can trigger different underlying signaling cascades (e.g., transcription dependence vs. transcription independence) (Ibata, Sun, & Turrigiano, 2008; Sutton et al., 2006). It is possible that similar perturbation–specific-signaling cascades exist in HIP. This possibility is supported by one study which found that expression of certain mRNA transcripts and proteins were affected either by TTX or APV treatment, but not both (Lee et al., 2015). However, either TTX or APV could reduce the function of different K+ channels but still increase firing frequency, suggesting that different signaling cascades could mediate increased excitability through similar proteins. It is interesting to note that in response to at least activity deprivation, all studies find some reduction in K+ channel function (see Table 1). However, what differs across the studies is the specific K+ channels that are being altered (Table 1). One possibility is that different perturbations (spike blockade vs. transmission blockade) trigger different versions of HIP that may rely on different signaling cascades and functions to restore distinct aspects of neuronal activity. Going forward, it will be important to parse out the triggers, goals, and expression mechanisms elicited by different types of activity perturbations.

The Compartmentalization of HIP

It is clear that activity perturbations can lead to changes in ion channel function and/or expression, that are thought to mediate the compensatory recovery of excitability. However, for a long time it remained unknown if these changes in ion channel function occurred throughout the cell or in local compartments. Recent studies have started answering this question. A set of seminal studies demonstrated that the axon initial segment (AIS) responds locally to chronic changes in neuronal activity levels (Grubb & Burrone, 2010; Kuba, Oichi, & Ohmori, 2010). The AIS contains the highest density of voltage-gated Na+ channels on the cell, and is the initiation site of action potential generation. The length and position of the AIS can significantly influence the intrinsic excitability of a cell, and in response to activity perturbation (either chronic increases or chronic decreases), these features can be changed in a compensatory direction to homeostatically maintain at least some aspects of cellular excitability. Not only are morphological aspects of the AIS altered by activity perturbations, but there is also a redistribution of certain voltage-gated K+ channels that are also specifically expressed in the AIS (Kuba, Yamada, Ishiguro, & Adachi, 2015).

Compartmentalized compensatory changes in ion channel function have also been identified in the dendrites. One study found that in response to activity deprivation, cultured hippocampal slices upregulated voltage-gated Na+ channel expression preferentially in the dendrites and this correlated with increased excitability of the neurons (Aptowicz, Kunkler, & Kraig, 2004). Another study made the observation that hyperpolarization-activated cyclic nucleotide-gated (HCN1) channels were upregulated in what appears to be a homeostatic manner (Brager & Johnston, 2007). HCN1 channels are predominantly found on the dendrites, consistent with the idea of the compartmentalized expression of HIP.

Interestingly, when the conductance of a single class of ion channel within the AIS of the drosophila motoneuron was altered, it led to compensatory changes in a channel expressed on the dendrites of the same cell (Bergquist et al., 2010). This is an important observation, because it tells us that even though we might observe the homeostatic changes in one compartment of the cell (dendrites vs. AIS), it may not reflect the trigger initiation site for HIP. Rather, a cell relies on local and global integration of changes in neuronal activity, calcium influx, and ion channel conductances, and the summation of these changes will determine when, where, and how HIP is expressed.

The Complexities of HIP

As addressed (see Table 1), several studies suggest that certain conductances can be regulated similarly in response to different types of activity perturbations across different cell types and preparations (isolated STG cells, cortical primary cultures, spinal cord). However, the homeostatic trajectory of any given cell will depend on its starting constellation of conductances, which, as has been discussed, can be dramatically different for cells sharing a similar pattern of activity. Therefore, the homeostatic adjustments in channel conductances for a cell will be complex, depending on the initial ion constellation as well as activity-dependent and activity-independent (e.g., genetic) influences on channel regulation. This may be particularly relevant for cells or neural networks during development. Neurons have genetic programs that change with development, and can start out with a wide variety of channel conductances, but ultimately the maturation of their excitability will be defined by the homeostatic trajectory of the cell as it matures (O’Leary & Wyllie, 2011; O’Leary et al., 2014).

While we have focused on activity-dependent calcium entry regulating ion channel expression or function via HIP, activity-independent mechanisms can also play a role in ion channel regulation. One elegant study demonstrated that activity-independent mechanisms exist for the regulation of ion channel correlations (MacLean, Zhang, Johnson, & Harris-Warrick, 2003). Here the authors showed that Ia and Ih were coregulated such that an alteration in one led to a similar alteration in the other which stabilized activity levels; however, they also demonstrated that this coregulation was carried out at the level of the mRNA and was not dependent on spiking activity.

Neuromodulatory input can regulate ion channel conductances, and when it was abolished in the STG, rhythmic activity was lost but then homeostatically recovered in the following days (Thoby-Brisson & Simmers, 1998). This neuromodulatory input was important in maintaining certain ion channel correlations and both the recovery of activity and the maintenance of these correlations appeared to be under the control of both activity-dependent and activity-independent mechanisms (Golowasch, 2014; Khorkova & Golowasch, 2007; Schulz & Lane, 2017; Temporal et al., 2014; Thoby-Brisson & Simmers, 1998; Zhang, Khorkova, Rodriguez, & Golowasch, 2008). These findings have significant functional implications as the loss of neuromodulatory input rendered the network less capable of sustaining appropriate activity patterns when faced with further perturbations such as changes in temperature (Haddad & Marder, 2018). It is clear that several activity-dependent and activity-independent processes shape and constrain the expression profile of ion channels.

Synaptic Scaling

There had been a concern for many years that Hebbian types of synaptic plasticity should lead to a runaway strengthening of synapses: inputs that drive a cell to fire would get stronger (fire together wire together), and with no further brake on this plasticity these inputs would simply continue to grow in strength, driving cells to fire excessively. It was therefore proposed that there must be some mechanism that limited the spike rate of these cells by preventing synapses from becoming too strong. These concerns predicted the existence of a form of synaptic plasticity that was distinct from Hebbian forms (Turrigiano, 2017). Homeostatic synaptic scaling fulfilled such a prediction; it was first defined and reported by Turrigiano, Leslie, Desai, Rutherford, and Nelson in a 1998 Nature paper. In it they demonstrated that one- to two-day spike blockade in visual cortical cultures caused an increase in AMPAergic miniature excitatory postsynaptic current (mEPSC) amplitudes (quantal amplitudes) in excitatory pyramidal-like cells. In this initial study the authors showed that mEPSC amplitudes from cultured cortical neurons treated with a Na+ channel blocker (TTX) or AMPA receptor blocker cyanquixaline (6-cyano-7 nitroquinoxaline-2,3-dione) (CNQX) became twice as strong as those under control conditions. One of the ways the multiplicative nature of the “scaling” effect was demonstrated was through comparisons of the distribution of mEPSC amplitudes. When the distribution of control amplitudes was multiplied by a scaling factor (plus an additive value) one could transform the control cumulative distribution of amplitudes into the TTX-treated cumulative distribution and this led to the term synaptic scaling (Figure 2).

Homeostatic Plasticity in the CNSClick to view larger

Figure 2. Synaptic scaling.

Note. A schematic representation (top) of synaptic scaling where all inputs are scaled by some multiplicative factor. The cumulative histogram shows how mPSC amplitudes are increased across their entire distribution.

In addition, the study demonstrated that the opposite scaling relationship (downscaling) of mEPSCs occurred when activity was increased in cultured neurons by blocking inhibition. Unlike TTX treatment where activity cannot recover because action potentials are blocked, spiking activity was able to homeostatically return to original levels following the blockade of GABA receptors (GABARs), demonstrating a truly homeostatic response. The idea was then established that all of the synaptic inputs were adjusted by a multiplicative factor following an alteration in spike rate and that this synaptic compensation was engaged in order to compensate for the pharmacological activity perturbation. Because scaling was multiplicative, the idea was that it acted to maintain the relative strengths of different inputs by adjusting synaptic gain to ensure that the runaway strengthening did not happen. In the same year, two other studies reported similar findings in two different culture systems. One study showed that upward or downward scaling occurred in spinal cultures in response to chronic activity blockade (glutamate antagonists) or enhancement (inhibitory antagonists), respectively (O’Brien et al., 1998). The other study demonstrated that following days of treating cultures with an inhibitory receptor antagonist, hippocampal neurons displayed reduced mEPSC amplitudes (Lissin et al., 1998). These other studies suggested there were changes in postsynaptic AMPA receptors that mediated the altered synaptic strength observed. In addition to AMPAergic scaling, it was shown that NMDA mEPSCs undergo strengthening following activity block (TTX, CNQX, APV/CNQX) and a weakening following chronic increases in activity through GABAR block in cortical, hippocampal, or spinal cultures (Ehlers, 2003; Rosen, Moghekar, & O’Brien, 2007; Watt, van Rossum, MacLeod, Nelson, & Turrigiano, 2000); however, see (Lissin et al., 1998).

While the majority of work on synaptic scaling has focused on glutamatergic inputs onto excitatory projection neurons, scaling of GABAergic inputs to excitatory neurons has also been described (Wenner, 2011). Compensatory decreases in the amplitude of GABAergic miniature postsynaptic currents (mPSCs) have been demonstrated following increased network activity in cultured excitatory neurons, while compensatory decreases in miniature inhibitory postsynaptic currents (mIPSC) amplitude are observed following activity blockade (Hartman, Pal, Burrone, & Murthy, 2006; Kilman, van Rossum, & Turrigiano, 2002; Peng et al., 2010).

Synaptic Scaling In Vivo

These in vitro studies have been critical in characterizing scaling and identifying underlying mechanisms; however, these chronic perturbations are quite extreme and it was important to demonstrate that scaling occurs in the living system in response to less extreme perturbations. Several studies have now shown synaptic scaling following perturbations in vivo. This was initially demonstrated again in the Turrigiano lab, but this time in the visual cortex following a two-day monocular deprivation (MD) via TTX intraocular injection (Desai, Cudmore, Nelson, & Turrigiano, 2002). Here, scaling was observed in vitro in slices of visual cortex two days after in vivo TTX injections to silence retinal ganglion cells, the output of the retina. The authors report that scaling was exhibited differently in different layers. Scaling was observed in L4 monocular primary visual cortical pyramidal cells (PCs) following MD from P14-16, but not P21-23. On the other hand, scaling was observed in layer 2/3 PCs following MD from P21-23, but not P14-16. This suggested a development-specific period for scaling in different layers. Others have gone on to show that the scaling capacity is maintained into adulthood in L2/3 PCs in V1 (Goel & Lee, 2007; Goel et al., 2006). While several different forms of visual deprivation have been shown to trigger scaling in the visual cortex, the timing of this plasticity is different depending on the particular form of deprivation. MD by lid suture has been shown to produce weakening of closed eye inputs in the first two to three days of lid suture followed by a strengthening of open eye inputs (Mrsic-Flogel et al., 2007). It is believed that the weakening is dependent on Hebbian mechanisms, while the slower strengthening is synaptic scaling. This was supported by the observation that the slow strengthening of synaptic inputs did not occur in mouse models where scaling does not exist (Kaneko, Stellwagen, Malenka, & Stryker, 2008). Very similar synaptic depression followed by synaptic strengthening has been observed in the somatosensory system with whisker deprivations (Gainey & Feldman, 2017; Glazewski, Greenhill, & Fox, 2017). Again the slow strengthening is absent in mouse models not expressing synaptic scaling (Greenhill, Ranson, & Fox, 2015).

GABAergic scaling has also been demonstrated following in vivo perturbations in the embryonic chick spinal cord (Gonzalez-Islas & Wenner, 2006; Lindsly, Gonzalez-Islas, & Wenner, 2014). When GABAergic currents, which are depolarizing at this early stage of development, are blocked, both AMPAergic and GABAergic scaling are observed in the isolated spinal preparation. While the mechanisms are varied, changes in receptor number contribute to the plasticity mechanisms. In fact, the first indication of GABAergic homeostatic plasticity was suggested through immunocytochemical studies, which showed reduced visual input in vivo led to decreased cortical expression of GABAA receptors, GABA and glutamate decarboxylase (GAD) (Hendry & Jones, 1986; Hendry et al., 1994).

Expression Mechanisms

The mechanisms underlying synaptic scaling (strengthening and weakening) have been examined extensively, and it is clear that there are several different mechanisms identified in different contexts. However, the most commonly identified mechanism involves trafficking transmitter receptors into or out of the synaptic membrane. AMPAergic upscaling has been best studied and the prominent mechanism involves trafficking AMPA receptors (AMPAR) into the postsynaptic membrane. Most of this work was carried out in optically accessible culture systems. There appear to be two different types of AMPAR insertion, where scaling is mediated by increases in either GluA2-containing (Anggono, Clem, & Huganir, 2011; O’Brien et al., 1998; Sun & Wolf, 2009; Sutton et al., 2006; Wierenga, Ibata, & Turrigiano, 2005) or calcium-permeable GluA2-lacking AMPARs (Garcia-Bereguiain et al., 2013; Shepherd et al., 2006; Sutton et al., 2006; Thiagarajan, Lindskog, & Tsien, 2005). Similarly, AMPAergic downscaling has been shown to be mediated by removal of AMPARs from the synaptic membrane following chronic increases in activity through GABAergic inhibition (Anggono et al., 2011; Ibata et al., 2008; Shepherd et al., 2006; Sun & Wolf, 2009).

Some reports have also shown that another contributor to altering quantal amplitude, which mediates up and down scaling, is through changes in the amount of glutamate packaged into vesicles. Following chronic increases or decreases of activity in cultured neurons it has been shown that the amount of glutamate that is released per vesicle is decreased or increased, respectively (De Gois et al., 2005; Erickson, De Gois, Varoqui, Schafer, & Weihe, 2006; Wilson et al., 2005).

GABAergic up- and downscaling, like AMPAergic scaling, have been shown to be mediated by both changes in receptor trafficking as well as GABA vesicle concentration. Cultures where activity is chronically altered show the compensatory changes in several features including GABARs in the synaptic membrane, levels of the enzyme for GAD, levels of vesicular transporter for GABA, and quantal GABA release. These mechanisms have been shown for GABAergic up- and downscaling in both cortical cultures (De Gois et al., 2005; Kilman et al., 2002) and hippocampal cultures (Hartman et al., 2006; Peng et al., 2010).

A separate mechanism for GABAergic up and down scaling has been demonstrated in chick embryo spinal motoneurons and interneurons (Garcia-Bereguiain, Gonzalez-Islas, Lindsly, & Wenner, 2016; Gonzalez-Islas, Chub, Garcia-Bereguiain, & Wenner, 2010; Lindsly et al., 2014). Here it has been demonstrated that in vivo perturbations to spontaneous network activity or GABAergic synaptic signaling can trigger changes in intracellular chloride levels, thus altering synaptic strength not by receptor trafficking, but through changes in the driving force for these receptor channels.

The Triggers, Signaling Pathways, and Potential Goals of Synaptic Scaling

We have discussed the mechanisms that mediate the changes in quantal amplitude, but what are the triggers for initiating these synaptic changes? It is becoming clear that different perturbations can trigger different forms of synaptic scaling, a point that will be important to recognize for a more complete understanding of the goals of these different forms of plasticity. As discussed, AMPAergic scaling can be mediated by either GluA2-containing or GluA2-lacking AMPARs. While there are exceptions (Hou, Zhang, Jarzylo, Huganir, & Man, 2008; O’Brien et al., 1998), which may be dependent on culturing protocol, timing, tissue, or species, there do appear to be some generalities. There seem to be at least two distinct classes of scaling—triggered by spike blockade with chronic TTX (TTX-dependent scaling) and another that is triggered by neurotransmitter receptor blockade (transmission-dependent scaling).

TTX-dependent scaling typically triggers scaling through GluA2-containing AMPARs (Anggono et al., 2011; Sun & Wolf, 2009; Sutton et al., 2006; Wierenga et al., 2005). TTX-mediated scaling is consistent with the initial idea that alterations in spike rate, possibly through large somatic calcium signals, trigger biochemical cascades that lead to changes in GluA2-containing AMPARs (Ibata et al., 2008; Turrigiano et al., 1998). TTX-mediated scaling has been reported to be transcription dependent, involve CamKIV, and be dependent on large calcium transients experienced by the cell body (Ibata et al., 2008). This form of synaptic scaling could accomplish the original idea that scaling is triggered by changes in spike rate in order to homeostatically restore spike rate, although further and less severe perturbations will be necessary to confirm this.

Transmission-dependent scaling tends to be mediated by GluA2-lacking AMPARs. This form of scaling is triggered by reduced glutamatergic synaptic signaling—either by blocking NMDAR-mediated mEPSCs (Aoto, Nam, Poon, Ting, & Chen, 2008; Ju et al., 2004; Sutton et al., 2006) or blocking AMPARs (Jakawich et al., 2010; Thiagarajan et al., 2005). This form of scaling is not transcription-dependent, but rather requires local mRNA translation (Ju et al., 2004; Maghsoodi, Poon, Nam, Aoto, Ting, & Chen, 2008; Sutton et al., 2006). Two pathways have been implicated in TTX/APV-triggered scaling in hippocampal cultures. The first shows that reductions in calcium entry through NMDARs disinhibits the synthesis of retinoic acid (RA), which then binds to its receptor RAR, preventing the inhibition of translation of GluA1 mRNA (Aoto et al., 2008; Maghsoodi et al., 2008). Another study has shown that NMDA mPSCs lead to very small calcium transients, which cause an intracellular calcium-induced calcium release, and when these signals are reduced scaling is triggered; this is dependent on protein translation and eukaryotic elongation factor-2 kinase activity (Reese & Kavalali, 2015). In addition to these pathways several different proteins have been implicated in scaling including tumor necrosis factor alpha (TNFα‎), methyl CpG binding protein 2 (MECP2), guanylate kinase-associated protein (GKAP), fragile X mental retardation protein (FMRP), Arc, but it is unclear how or if these molecules interact (Qiu et al., 2012; Shepherd et al., 2006; Shin et al., 2012; Soden & Chen, 2010; Stellwagen & Malenka, 2006). Interestingly, TNFα‎ has been implicated in HIP as well (Shim et al., 2018). Further, proteomic approaches are now being employed to better identify the scaling proteome (Schanzenbacher, Langer, & Schuman, 2018). The signaling pathway for transmission-mediated scaling is dependent on RA synthesis and FMRP, as demonstrated by the observation that this form of scaling is prevented when RA synthesis is blocked and in the fragile x mental retardation 1 (Fmr1) KO model of fragile X syndrome (Soden & Chen, 2010; Wang, Zhang, Hintze, & Chen, 2011). Interestingly, TTX-triggered scaling remains intact in the Fmr1 KO and during RA synthesis blockade, consistent with the idea that different signaling cascades mediate these two forms of scaling. Transmission-dependent scaling appears to be quite different in its function compared to TTX-dependent scaling, as it seems that neurotransmitter receptor activation may be the critical feature that is homeostatically maintained. Altering spike rate by itself does not seem to impact transmission-dependent scaling, as blocking AMPARs in cortical cultures triggers identical levels of scaling whether spiking activity is reduced or maintained at normal levels (Fong, Newman, Potter, & Wenner, 2015). In contrast to TTX-dependent scaling, transmission-dependent scaling requires the local dendritic translation of GluA1 receptors, and can be influenced by alterations in mEPSCs (Maghsoodi et al., 2008; Reese & Kavalali, 2015; Sutton et al., 2006). These studies suggest that this form of synaptic compensation is actually occurring at each individual synapse or in a specific dendritic location, but is expressed in a global fashion as scaling because the compensations are triggered at all synaptic locations when antagonists are applied. Similar to these local synaptic compensations, a local form of synaptic compensation through GluA2-lacking AMPARs has also been shown when evoked release of individual inputs is reduced via hyperpolarization (Beique, Na, Kuhl, Worley, & Huganir, 2011; Hou, Zhang et al., 2008).

Transmission-dependent scaling has also been shown to exist in vivo in the embryonic chick spinal cord. Here, both GABAergic and AMPAergic scaling have been shown to be triggered by alterations in GABAergic transmission, rather than spiking activity that occurs during bouts of network activity (Wilhelm & Wenner, 2008). Further, spinal neurons express upward and downward scaling when the frequency of spontaneous action-potential independent vesicle release was decreased and increased, respectively (Garcia-Bereguiain et al., 2016). Because this transmission-dependent form of scaling is triggered by altered transmission and can be at the local site of the alteration it is likely that this form of scaling acts to homeostatically control synaptic strength at individual synapses (Beique et al., 2011; Hou, Huang et al., 2008; Sutton et al., 2006); this takes the form of global scaling when all inputs are altered as would occur during a pharmacological blockade.

Developmental Profile of Scaling

Synaptic scaling has been reported routinely in developing circuits. While studies have also suggested that scaling occurs in maturity, others suggest there are restrictions in the expression of scaling during development. Scaling may be most robustly expressed in the embryo when glutamate and GABA are both excitatory, where spinal motor and interneurons demonstrate scaling of both AMPA and GABA mPSCs following increases and decreases in synaptic activity (Wenner, 2014). The majority of studies carried out in culture have been from 7-14 days in vitro (DIV). It has been shown that earlier time points show a stronger increase in quantal amplitude following similar perturbations (Wierenga, Walsh, & Turrigiano, 2006). Further, in hippocampal slices synaptic scaling is observed at earlier stages across all inputs to pyramidal cells (PCs), but the synaptic compensations become more restricted to one set of inputs at later stages (Lee et al., 2013). In vivo AMPAergic scaling following sustained release of TTX in the hippocampus has been reported in juvenile CA1 (Cornu Ammonis region 1) PCs, but is absent in the adult. Scaling has also been shown in the neonatal visual cortex following visual deprivation, which occurs in layer 2/3 principal neurons (Desai et al., 2002), and continues into adulthood (Goel & Lee, 2007). On the other hand, layer 4 PCs do not express scaling following the third postnatal week (Desai et al., 2002). A similar early expression of a scaling-dependent plasticity is observed in neonatal visual cortex following MD, that does not appear to be operating in the adult (Ranson, Cheetham, Fox, & Sengpiel, 2012). Finally, one study has shown a homeostatic capacity of retinal network activity following GABA blockade in the first week of postnatal development that is lost in the second week, although they did not look at scaling in this study (Hennig, Grady, van Coppenhagen, & Sernagor, 2011). Scaling has been shown throughout circuit maturation, but it appears to be most commonly expressed at early stages. This could be due to a loss of the capacity of homeostatic synaptic scaling or it is also possible that the triggers and potentially goals of scaling change during maturation.

Presynaptic Homeostatic Plasticity (Probability of Release)

Another form of homeostatic synaptic plasticity has been referred to as presynaptic homeostatic plasticity, where fast changes (minutes) in probability of release are altered as a compensatory response. An excellent and very complete review of this form of plasticity is available, and so here we will only briefly describe this important form of homeostatic plasticity (Davis & Muller, 2015). An early study showed that when motoneuron innervation of a muscle fiber in the fly was genetically enhanced, the strength of this input was homeostatically maintained through a reduction in the probability of release (Davis & Goodman, 1998). In another study, expression of a potassium channel, which hyperpolarized the postsynaptic muscle fiber, also led to a homeostatic control of the evoked release from the innervating motoneuron, and again this was mediated by a compensation in probability of release (Paradis, Sweeney, & Davis, 2001). These forms of presynaptic homeostatic plasticity occurred over a chronic period of development of the neuromuscular junction. Later, it was discovered that a presynaptic homeostatic plasticity involving changes in probability of release could be expressed within minutes of a pharmacological perturbation (Frank, Kennedy, Goold, Marek, & Davis, 2006). When miniature excitatory postsynaptic potential (mEPSP) amplitudes were reduced with a weak neurotransmitter antagonist, evoked responses were initially reduced as well, but within minutes the motoneuron-evoked response homeostatically recovered. This occurred without a concurrent increase in mEPSP amplitude, and was mediated by an increase in the number of quanta released per action potential (quantal content). Strikingly, this homeostatic phenomenon occurred even in the absence of action potentials; This suggested the muscle determined the degree of mEPSP reduction and then sent a retrograde signal to the innervating motoneuron fiber which led to an increase in the number of vesicles released, and returned the evoked response to its original value. These fast changes in quantal content have been well studied, but it is less clear how this plasticity is expressed over longer periods of time as studied for HIP and scaling (hours to days). Since these original studies, several molecular mediators have been implicated in this process including voltage activated calcium channel subunit 2.1 (Cav2.1), dysbindin, semaphorin-plexin signaling (Dickman & Davis, 2009; Frank et al., 2006; Orr, Fetter, & Davis, 2017). While the fly neuromuscular junction (NMJ) has been crucial in characterizing presynaptic homeostatic plasticity, this plasticity has also been demonstrated at the mammalian NMJ. Presynaptic homeostatic plasticity is not only expressed in the peripheral nervous system, as it has been described at central synapses (Bartley, Huang, Huber, & Gibson, 2008; Burrone, O’Byrne, & Murthy, 2002; Mitra, Mitra, & Tsien, 2011; Murthy, Schikorski, Stevens, & Zhu, 2001; Oleskevich & Walmsley, 2002; Tyler, Petzold, Pal, & Murthy, 2007; Wang et al., 2004; Zhao, Dreosti, & Lagnado, 2011).


Aizenman, C. D., Akerman, C. J., Jensen, K. R., & Cline, H. T. (2003). Visually driven regulation of intrinsic neuronal excitability improves stimulus detection in vivo. Neuron, 39, 831–842.Find this resource:

Anggono, V., Clem, R. L., & Huganir, R. L. (2011). PICK1 loss of function occludes homeostatic synaptic scaling. Journal of Neuroscience, 31, 2188–2196.Find this resource:

Aoto, J., Nam, C. I., Poon, M. M., Ting, P., & Chen, L. (2008). Synaptic signaling by all-trans retinoic acid in homeostatic synaptic plasticity. Neuron, 60, 308–320.Find this resource:

Aptowicz, C. O., Kunkler, P. E., & Kraig, R. P. (2004). Homeostatic plasticity in hippocampal slice cultures involves changes in voltage-gated Na+ channel expression. Brain Research, 998, 155–163.Find this resource:

Bartley, A. F., Huang, Z. J., Huber, K. M., & Gibson, J. R. (2008). Differential activity-dependent, homeostatic plasticity of two neocortical inhibitory circuits. Journal of Neurophysiology, 100, 1983–1994.Find this resource:

Beique, J. C., Na, Y., Kuhl, D., Worley, P. F., & Huganir, R. L. (2011). Arc-dependent synapse-specific homeostatic plasticity. Proceedings of the National Academy of Sciences of the United States of America, 108, 816–821.Find this resource:

Bergquist, S., Dickman, D. K., & Davis, G. W. (2010). A hierarchy of cell intrinsic and target-derived homeostatic signaling. Neuron, 66, 220–234.Find this resource:

Brager, D. H., & Johnston, D. (2007). Plasticity of intrinsic excitability during long-term depression is mediated through mGluR-dependent changes in I(h) in hippocampal CA1 pyramidal neurons. Journal of Neuroscience, 27, 13926–13937.Find this resource:

Brickley, S. G., Revilla, V., Cull-Candy, S., Wisden, W., & Farrant, M. (2001). Adaptive regulation of neuronal excitability by voltage-independent potassium conductance. Nature, 409, 88–92.Find this resource:

Burrone, J., O’Byrne, M., & Murthy, V. N. (2002). Multiple forms of synaptic plasticity triggered by selective suppression of activity in individual neurons. Nature, 420, 414–418.Find this resource:

Davis, G. W., & Goodman, C. S. (1998). Synapse-specific control of synaptic efficacy at the terminals of a single neuron. Nature, 392, 82–86.Find this resource:

Davis, G. W., & Muller, M. (2015). Homeostatic control of presynaptic neurotransmitter release. Annual Review of Physiology, 77, 251–270.Find this resource:

De Gois, S., Schafer, M. K., Defamie, N., Chen, C., Ricci, A., Weihe, E., . . . Erickson, J.D. (2005). Homeostatic scaling of vesicular glutamate and GABA transporter expression in rat neocortical circuits. Journal of Neuroscience, 25, 7121–7133.Find this resource:

Desai, N. S., Cudmore, R. H., Nelson, S. B., & Turrigiano, G. G. (2002). Critical periods for experience-dependent synaptic scaling in visual cortex. Nature Neuroscience, 5, 783–789.Find this resource:

Desai, N. S., Rutherford, L. C., & Turrigiano, G. G. (1999). Plasticity in the intrinsic excitability of cortical pyramidal neurons. Nature Neuroscience, 2, 515–520.Find this resource:

Dickman, D. K., & Davis, G. W. (2009). The schizophrenia susceptibility gene dysbindin controls synaptic homeostasis. Science, 326, 1127–1130.Find this resource:

Ehlers, M. D. (2003). Activity level controls postsynaptic composition and signaling via the ubiquitin-proteasome system. Nature Neuroscience, 6, 231–242.Find this resource:

Erickson, J. D., De Gois, S., Varoqui, H., Schafer, M. K., & Weihe, E. (2006). Activity-dependent regulation of vesicular glutamate and GABA transporters: A means to scale quantal size. Neurochemistry International, 48, 643–649.Find this resource:

Evans, M. D., Dumitrescu, A. S., Kruijssen, D. L. H., Taylor, S. E., & Grubb, M. S. (2015). Rapid modulation of axon initial segment length influences repetitive spike firing. Cell Reports, 13, 1233–1245.Find this resource:

Fong, M.-F., Newman, J. P., Potter, S. M., & Wenner, P. (2015). Upward synaptic scaling is dependent on neurotransmission rather than spiking. Nature Communications, 6, 6339.Find this resource:

Frank, C. A., Kennedy, M. J., Goold, C. P., Marek, K. W., & Davis, G. W. (2006). Mechanisms underlying the rapid induction and sustained expression of synaptic homeostasis. Neuron, 52, 663–677.Find this resource:

Gainey, M. A., & Feldman, D. E. (2017). Multiple shared mechanisms for homeostatic plasticity in rodent somatosensory and visual cortex. Philosophical Transactions of the Royal Society B: Biological Sciences, 372, 1–7.Find this resource:

Garcia-Bereguiain, M. A., Gonzalez-Islas, C., Lindsly, C., Butler, E., Hill, A. W., & Wenner, P. (2013). In vivo synaptic scaling is mediated by GluA2-lacking AMPA receptors in the embryonic spinal cord. Journal of Neuroscience, 33, 6791–6799.Find this resource:

Garcia-Bereguiain, M. A., Gonzalez-Islas, C., Lindsly, C., & Wenner, P. (2016). Spontaneous release regulates synaptic scaling in the embryonic spinal network in vivo. Journal of Neuroscience, 36, 7268–7282.Find this resource:

Glazewski, S., Greenhill, S., & Fox, K. (2017). Time-course and mechanisms of homeostatic plasticity in layers 2/3 and 5 of the barrel cortex. Philosophical Transactions of the Royal Society B: Biological Sciences, 372, 1–11.Find this resource:

Goel, A., Jiang, B., Xu, L. W., Song, L., Kirkwood, A., & Lee, H. K. (2006). Cross-modal regulation of synaptic AMPA receptors in primary sensory cortices by visual experience. Nature Neuroscience, 9, 1001–1003.Find this resource:

Goel, A., & Lee, H. K. (2007). Persistence of experience-induced homeostatic synaptic plasticity through adulthood in superficial layers of mouse visual cortex. Journal of Neuroscience, 27, 6692–6700.Find this resource:

Golowasch, J. (2014). Ionic current variability and functional stability in the nervous system. Bioscience, 64, 570–580.Find this resource:

Gonzalez-Islas, C., Chub, N., Garcia-Bereguiain, M. A., & Wenner, P. (2010). GABAergic synaptic scaling in embryonic motoneurons is mediated by a shift in the chloride reversal potential. Journal of Neuroscience, 30, 13016–13020.Find this resource:

Gonzalez-Islas, C., & Wenner, P. (2006). Spontaneous network activity in the embryonic spinal cord regulates AMPAergic and GABAergic synaptic strength. Neuron, 49, 563–575.Find this resource:

Greenhill, S. D., Ranson, A., & Fox, K. (2015). Hebbian and homeostatic plasticity mechanisms in regular spiking and intrinsic bursting cells of cortical layer 5. Neuron, 88, 539–552.Find this resource:

Grubb, M. S., & Burrone, J. (2010). Activity-dependent relocation of the axon initial segment fine-tunes neuronal excitability. Nature, 465, 1070–1074.Find this resource:

Haddad, S. A., & Marder, E. (2018). Circuit robustness to temperature perturbation is altered by neuromodulators. Neuron, 100, 609–623.Find this resource:

Hartman, K. N., Pal, S. K., Burrone, J., & Murthy, V. N. (2006). Activity-dependent regulation of inhibitory synaptic transmission in hippocampal neurons. Nature Neuroscience, 9, 642–649.Find this resource:

Hendry, S. H., Huntsman, M. M., Vinuela, A., Mohler, H., de Blas, A. L., & Jones, E. G. (1994). GABAA receptor subunit immunoreactivity in primate visual cortex: Distribution in macaques and humans and regulation by visual input in adulthood. Journal of Neuroscience, 14, 2383–2401.Find this resource:

Hendry, S. H., & Jones, E. G. (1986). Reduction in number of immunostained GABAergic neurones in deprived-eye dominance columns of monkey area 17. Nature, 320, 750–753.Find this resource:

Hennig, M. H., Grady, J., van Coppenhagen, J., & Sernagor, E. (2011). Age-dependent homeostatic plasticity of GABAergic signaling in developing retinal networks. Journal of Neuroscience, 31, 12159–12164.Find this resource:

Hou, Q., Huang, Y., Amato, S., Snyder, S. H., Huganir, R. L., & Man, H. Y. (2008). Regulation of AMPA receptor localization in lipid rafts. Molecular and Cellular Neuroscience, 38, 213–223.Find this resource:

Hou, Q., Zhang, D., Jarzylo, L., Huganir, R. L., & Man, H. Y. (2008). Homeostatic regulation of AMPA receptor expression at single hippocampal synapses. Proceedings of the National Academy of Sciences of the United States of America, 105, 775–780.Find this resource:

Ibata, K., Sun, Q., & Turrigiano, G. G. (2008). Rapid synaptic scaling induced by changes in postsynaptic firing. Neuron, 57, 819–826.Find this resource:

Jakawich, S. K., Nasser, H. B., Strong, M. J., McCartney, A. J., Perez, A. S., Rakesh, N., . . . Sutton, M.A. (2010). Local presynaptic activity gates homeostatic changes in presynaptic function driven by dendritic BDNF synthesis. Neuron, 68, 1143–1158.Find this resource:

Joseph, A., & Turrigiano, G. G. (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. Journal of Neuroscience, 37, 6778–6785.Find this resource:

Ju, W., Morishita, W., Tsui, J., Gaietta, G., Deerinck, T. J., Adams, S. R., . . . Malenka, R.C. (2004). Activity-dependent regulation of dendritic synthesis and trafficking of AMPA receptors. Nature Neuroscience, 7, 244–253.Find this resource:

Kaneko, M., Stellwagen, D., Malenka, R. C., & Stryker, M. P. (2008). Tumor necrosis factor-alpha mediates one component of competitive, experience-dependent plasticity in developing visual cortex. Neuron, 58, 673–680.Find this resource:

Karmarkar, U. R., & Buonomano, D. V. (2006). Different forms of homeostatic plasticity are engaged with distinct temporal profiles. European Journal of Neuroscience, 23, 1575–1584.Find this resource:

Khorkova, O., & Golowasch, J. (2007). Neuromodulators, not activity, control coordinated expression of ionic currents. Journal of Neuroscience, 27, 8709–8718.Find this resource:

Kilman, V., van Rossum, M. C., & Turrigiano, G. G. (2002). Activity deprivation reduces miniature IPSC amplitude by decreasing the number of postsynaptic GABA(A) receptors clustered at neocortical synapses. Journal of Neuroscience, 22, 1328–1337.Find this resource:

Kuba, H., Oichi, Y., & Ohmori, H. (2010). Presynaptic activity regulates Na(+) channel distribution at the axon initial segment. Nature, 465, 1075–1078.Find this resource:

Kuba, H., Yamada, R., Ishiguro, G., & Adachi, R. (2015). Redistribution of Kv1 and Kv7 enhances neuronal excitability during structural axon initial segment plasticity. Nature Communications, 6, 8815.Find this resource:

Lambo, M. E., & Turrigiano, G. G. (2013). Synaptic and intrinsic homeostatic mechanisms cooperate to increase L2/3 pyramidal neuron excitability during a late phase of critical period plasticity. Journal of Neuroscience, 33, 8810–8819.Find this resource:

Lane, B. J., Samarth, P., Ransdell, J. L., Nair, S. S., & Schulz, D. J. (2016). Synergistic plasticity of intrinsic conductance and electrical coupling restores synchrony in an intact motor network. eLife, 5, 1–23.Find this resource:

Lee, K. J., Queenan, B. N., Rozeboom, A. M., Bellmore, R., Lim, S. T., Vicini, S., & Pak, D. T. (2013). Mossy fiber-CA3 synapses mediate homeostatic plasticity in mature hippocampal neurons. Neuron, 77, 99–114.Find this resource:

Lee, K. Y., & Chung, H. J. (2014). NMDA receptors and L-type voltage-gated Ca(2)(+) channels mediate the expression of bidirectional homeostatic intrinsic plasticity in cultured hippocampal neurons. Neuroscience, 277, 610–623.Find this resource:

Lee, K. Y., Royston, S. E., Vest, M. O., Ley, D. J., Lee, S., Bolton, E. C., & Chung, H. J. (2015). N-methyl-D-aspartate receptors mediate activity-dependent down-regulation of potassium channel genes during the expression of homeostatic intrinsic plasticity. Molecular Brain, 8, 4.Find this resource:

LeMasson, G., Marder, E., & Abbott, L. F. (1993). Activity-dependent regulation of conductances in model neurons. Science, 259, 1915–1917.Find this resource:

Lindsly, C., Gonzalez-Islas, C., & Wenner, P. (2014). Activity blockade and GABAA receptor blockade produce synaptic scaling through chloride accumulation in embryonic spinal motoneurons and interneurons. PLoS One, 9, e94559.Find this resource:

Lissin, D. V., Gomperts, S. N., Carroll, R. C., Christine, C. W., Kalman, D., Kitamura, M., . . . von Zastrow, M. (1998). Activity differentially regulates the surface expression of synaptic AMPA and NMDA glutamate receptors. Proceedings of the National Academy of Sciences of the United States of America, 95, 7097–7102.Find this resource:

Liu, Z., Golowasch, J., Marder, E., & Abbott, L. F. (1998). A model neuron with activity-dependent conductances regulated by multiple calcium sensors. Journal of Neuroscience, 18, 2309–2320.Find this resource:

MacLean, J. N., Zhang, Y., Johnson, B. R., & Harris-Warrick, R. M. (2003). Activity-independent homeostasis in rhythmically active neurons. Neuron, 37, 109–120.Find this resource:

Maffei, A., & Turrigiano, G. (2008). The age of plasticity: Developmental regulation of synaptic plasticity in neocortical microcircuits. Progress in Brain Research, 169, 211–223.Find this resource:

Maghsoodi, B., Poon, M. M., Nam, C. I., Aoto, J., Ting, P., & Chen, L. (2008). Retinoic acid regulates RARalpha-mediated control of translation in dendritic RNA granules during homeostatic synaptic plasticity. Proceedings of the National Academy of Sciences of the United States of America, 105, 16015–16020.Find this resource:

Marder, E., Goeritz, M. L., & Otopalik, A. G. (2015). Robust circuit rhythms in small circuits arise from variable circuit components and mechanisms. Current Opinion in Neurobiology, 31, 156–163.Find this resource:

Mee, C. J., Pym, E. C., Moffat, K. G., & Baines, R. A. (2004). Regulation of neuronal excitability through Pumilio-dependent control of a sodium channel gene. Journal of Neuroscience, 24, 8695–8703.Find this resource:

Meisler, M. H., & Kearney, J. A. (2005). Sodium channel mutations in epilepsy and other neurological disorders. Journal of Clinical Investigation, 115, 2010–2017.Find this resource:

Mitra, A., Mitra, S. S., & Tsien, R. W. (2011). Heterogeneous reallocation of presynaptic efficacy in recurrent excitatory circuits adapting to inactivity. Nature Neuroscience, 15, 250–257.Find this resource:

Mrsic-Flogel, T. D., Hofer, S. B., Ohki, K., Reid, R. C., Bonhoeffer, T., & Hubener, M. (2007). Homeostatic regulation of eye-specific responses in visual cortex during ocular dominance plasticity. Neuron, 54, 961–972.Find this resource:

Muraro, N. I., Weston, A. J., Gerber, A. P., Luschnig, S., Moffat, K. G., & Baines, R. A. (2008). Pumilio binds para mRNA and requires Nanos and Brat to regulate sodium current in drosophila motoneurons. Journal of Neuroscience, 28, 2099–2109.Find this resource:

Murthy, V. N., Schikorski, T., Stevens, C. F., & Zhu, Y. (2001). Inactivity produces increases in neurotransmitter release and synapse size. Neuron, 32, 673–682.Find this resource:

O’Brien, R. J., Kamboj, S., Ehlers, M. D., Rosen, K. R., Fischbach, G. D., & Huganir, R. L. (1998). Activity-dependent modulation of synaptic AMPA receptor accumulation. Neuron, 21, 1067–1078.Find this resource:

O’Leary, T., Williams, A. H., Caplan, J. S., & Marder, E. (2013). Correlations in ion channel expression emerge from homeostatic tuning rules. Proceedings of the National Academy of Sciences of the United States of America, 110, E2645–2654.Find this resource:

O’Leary, T., Williams, A. H., Franci, A., & Marder, E. (2014). Cell types, network homeostasis, and pathological compensation from a biologically plausible ion channel expression model. Neuron, 82, 809–821.Find this resource:

O’Leary, T., & Wyllie, D. J. (2011). Neuronal homeostasis: Time for a change? Journal of Physiology, 589, 4811–4826.Find this resource:

Oleskevich, S., & Walmsley, B. (2002). Synaptic transmission in the auditory brainstem of normal and congenitally deaf mice. Journal of Physiology, 540, 447–455.Find this resource:

Orr, B. O., Fetter, R. D., & Davis, G. W. (2017). Retrograde semaphorin-plexin signalling drives homeostatic synaptic plasticity. Nature, 550, 109–113.Find this resource:

Paradis, S., Sweeney, S. T., & Davis, G. W. (2001). Homeostatic control of presynaptic release is triggered by postsynaptic membrane depolarization. Neuron, 30, 737–749.Find this resource:

Peng, Y. R., Zeng, S. Y., Song, H. L., Li, M. Y., Yamada, M. K., & Yu, X. (2010). Postsynaptic spiking homeostatically induces cell-autonomous regulation of inhibitory inputs via retrograde signaling. Journal of Neuroscience, 30, 16220–16231.Find this resource:

Ping, Y., & Tsunoda, S. (2012). Inactivity induced increase in nAChRs upregulates Shal K+ channels to stabilize synaptic potentials. Nature Neuroscience, 15, 90–98.Find this resource:

Prinz, A. A., Bucher, D., & Marder, E. (2004). Similar network activity from disparate circuit parameters. Nature Neuroscience, 7, 1345–1352.Find this resource:

Qiu, Z., Sylwestrak, E. L., Lieberman, D. N., Zhang, Y., Liu, X. Y., & Ghosh, A. (2012). The Rett syndrome protein MeCP2 regulates synaptic scaling. Journal of Neuroscience, 32, 989–994.Find this resource:

Ransdell, J. L., Nair, S. S., & Schulz, D. J. (2012). Rapid homeostatic plasticity of intrinsic excitability in a central pattern generator network stabilizes functional neural network output. Journal of Neuroscience, 32, 9649–9658.Find this resource:

Ranson, A., Cheetham, C. E., Fox, K., & Sengpiel, F. (2012). Homeostatic plasticity mechanisms are required for juvenile, but not adult, ocular dominance plasticity. Proceedings of the National Academy of Sciences of the United States of America, 109, 1311–1316.Find this resource:

Reese, A. L., & Kavalali, E. T. (2015). Spontaneous neurotransmission signals through store-driven Ca transients to maintain synaptic homeostasis. eLife, 4, 1–9.Find this resource:

Rosen, K. M., Moghekar, A., & O’Brien, R. J. (2007). Activity dependent localization of synaptic NMDA receptors in spinal neurons. Molecular and Cellular Neuroscience, 34, 578–591.Find this resource:

Ross, W. N. (1989). Changes in intracellular calcium during neuron activity. Annual Review of Physiology, 51, 491–506.Find this resource:

Schanzenbacher, C. T., Langer, J. D., & Schuman, E. M. (2018). Time- and polarity-dependent proteomic changes associated with homeostatic scaling at central synapses. eLife, 7, 1–20.Find this resource:

Schulz, D. J., & Lane, B. J. (2017). Homeostatic plasticity of excitability in crustacean central pattern generator networks. Current Opinion in Neurobiology, 43, 7–14.Find this resource:

Shepherd, J. D., Rumbaugh, G., Wu, J., Chowdhury, S., Plath, N., Kuhl, D., . . . Worley, P. F. (2006). Arc/Arg3.1 mediates homeostatic synaptic scaling of AMPA receptors. Neuron, 52, 475–484.Find this resource:

Shim, H. G., Jang, S. S., Kim, S. H., Hwang, E. M., Min, J. O., Kim, H. Y., . . . Kim, S. J. (2018). TNF-alpha increases the intrinsic excitability of cerebellar Purkinje cells through elevating glutamate release in Bergmann Glia. Scientific Reports, 8, 11589.Find this resource:

Shin, S. M., Zhang, N., Hansen, J., Gerges, N. Z., Pak, D. T., Sheng, M., & Lee, S. H. (2012). GKAP orchestrates activity-dependent postsynaptic protein remodeling and homeostatic scaling. Nature Neuroscience, 15, 1655–1666.Find this resource:

Soden, M. E., & Chen, L. (2010). Fragile X protein FMRP is required for homeostatic plasticity and regulation of synaptic strength by retinoic acid. Journal of Neuroscience, 30, 16910–16921.Find this resource:

Stellwagen, D., & Malenka, R. C. (2006). Synaptic scaling mediated by glial TNF-alpha. Nature, 440, 1054–1059.Find this resource:

Sun, X., & Wolf, M. E. (2009). Nucleus accumbens neurons exhibit synaptic scaling that is occluded by repeated dopamine pre-exposure. European Journal of Neuroscience, 30, 539–550.Find this resource:

Sutton, M. A., Ito, H. T., Cressy, P., Kempf, C., Woo, J. C., & Schuman, E. M. (2006). Miniature neurotransmission stabilizes synaptic function via tonic suppression of local dendritic protein synthesis. Cell, 125, 785–799.Find this resource:

Temporal, S., Lett, K. M., & Schulz, D. J. (2014). Activity-dependent feedback regulates correlated ion channel mRNA levels in single identified motor neurons. Current Biology, 24, 1899–1904.Find this resource:

Thiagarajan, T. C., Lindskog, M., & Tsien, R. W. (2005). Adaptation to synaptic inactivity in hippocampal neurons. Neuron, 47, 725–737.Find this resource:

Thoby-Brisson, M., & Simmers, J. (1998). Neuromodulatory inputs maintain expression of a lobster motor pattern-generating network in a modulation-dependent state: Evidence from long-term decentralization in vitro. Journal of Neuroscience, 18, 2212–2225.Find this resource:

Tobin, A. E., Cruz-Bermudez, N. D., Marder, E., & Schulz, D. J. (2009). Correlations in ion channel mRNA in rhythmically active neurons. PLoS One, 4, e6742.Find this resource:

Turrigiano, G. G. (2017). The dialectic of Hebb and homeostasis. Philosophical Transactions of the Royal Society B: Biological Sciences, 372, 1–7.Find this resource:

Turrigiano, G. G., Abbott, L. F., & Marder, E. (1994). Activity-dependent changes in the intrinsic properties of cultured neurons. Science, 264, 974–977.Find this resource:

Turrigiano, G. G., LeMasson, G., & Marder, E. (1995). Selective regulation of current densities underlies spontaneous changes in the activity of cultured neurons. Journal of Neuroscience, 15, 3640–3652.Find this resource:

Turrigiano, G. G., Leslie, K. R., Desai, N. S., Rutherford, L. C., & Nelson, S. B. (1998). Activity-dependent scaling of quantal amplitude in neocortical neurons. Nature, 391, 892–896.Find this resource:

Tyler, W. J., Petzold, G. C., Pal, S. K., & Murthy, V. N. (2007). Experience-dependent modification of primary sensory synapses in the mammalian olfactory bulb. Journal of Neuroscience, 27, 9427–9438.Find this resource:

Wang, H. L., Zhang, Z., Hintze, M., & Chen, L. (2011). Decrease in calcium concentration triggers neuronal retinoic acid synthesis during homeostatic synaptic plasticity. Journal of Neuroscience, 31, 17764–17771.Find this resource:

Wang, X., Engisch, K. L., Li, Y., Pinter, M. J., Cope, T.C., & Rich, M. M. (2004). Decreased synaptic activity shifts the calcium dependence of release at the mammalian neuromuscular junction in vivo. Journal of Neuroscience, 24, 10687–10692.Find this resource:

Watt, A. J., van Rossum, M. C., MacLeod, K. M., Nelson, S. B., & Turrigiano, G. G. (2000). Activity coregulates quantal AMPA and NMDA currents at neocortical synapses. Neuron, 26, 659–670.Find this resource:

Wenner, P. (2011). Mechanisms of GABAergic homeostatic plasticity. Neural Plasticity, 1–6.Find this resource:

Wenner, P. (2014). Homeostatic synaptic plasticity in developing spinal networks driven by excitatory GABAergic currents. Neuropharmacology, 78, 55–62.Find this resource:

Wierenga, C. J., Ibata, K., & Turrigiano, G. G. (2005). Postsynaptic expression of homeostatic plasticity at neocortical synapses. Journal of Neuroscience, 25, 2895–2905.Find this resource:

Wierenga, C. J., Walsh, M. F., & Turrigiano, G. G. (2006). Temporal regulation of the expression locus of homeostatic plasticity. Journal of Neurophysiology, 96, 2127–2133.Find this resource:

Wilhelm, J. C., Rich, M. M., & Wenner, P. (2009). Compensatory changes in cellular excitability, not synaptic scaling, contribute to homeostatic recovery of embryonic network activity. Proceedings of the National Academy of Sciences of the United States of America, 106, 6760–6765.Find this resource:

Wilhelm, J. C., & Wenner, P. (2008). GABAA transmission is a critical step in the process of triggering homeostatic increases in quantal amplitude. Proceedings of the National Academy of Sciences of the United States of America, 105, 11412–11417.Find this resource:

Wilson, N. R., Kang, J., Hueske, E. V., Leung, T., Varoqui, H., Murnick, J. G., . . . Liu, G. (2005). Presynaptic regulation of quantal size by the vesicular glutamate transporter VGLUT1. Journal of Neuroscience, 25, 6221–6234.Find this resource:

Zhang, Y., Khorkova, O., Rodriguez, R., & Golowasch, J. (2008). Activity and neuromodulatory input contribute to the recovery of rhythmic output after decentralization in a central pattern generator. Journal of Neurophysiology, 101(1), 372–386.Find this resource:

Zhao, C., Dreosti, E., & Lagnado, L. (2011). Homeostatic synaptic plasticity through changes in presynaptic calcium influx. Journal of Neuroscience, 31, 7492–7496.Find this resource: