Homeostatic signaling: the positive side of negative feedback

https://doi.org/10.1016/j.conb.2007.04.004Get rights and content

Synaptic homeostasis provides a means for neurons and circuits to maintain stable function in the face of perturbations such as developmental or activity-dependent changes in synapse number or strength. These forms of plasticity are thought to utilize negative feedback signaling to sense some aspect of activity, compare this with an internal set point, and then adjust synaptic properties to keep activity close to this set point. However, the molecular identity of these signaling components has not been firmly established. Recent work suggests that there are likely to be multiple forms of synaptic homeostasis, mediated by distinct signaling pathways and with distinct expression mechanisms. These include presynaptic forms that depend on retrograde signaling to presynaptic Ca2+ channels, and postsynaptic forms influenced by BDNF, TNFα and Arc signaling. Current challenges include matching signaling elements to their functions (i.e. as detectors of activity, as part of the set-point mechanism and/or as effectors of synaptic change), and fitting these molecular candidates into a unified view of the signaling pathways that underlie synaptic homeostasis.

Introduction

Homeostatic synaptic plasticity is emerging as an important complement to Hebbian forms of plasticity in the activity-dependent refinement of synaptic connectivity [1, 2, 3]. Loosely defined, a ‘homeostatic’ form of plasticity is one that acts to stabilize the activity of a neuron or neuronal circuit in the face of perturbations, such as changes in cell size or in synapse number or strength, that alter excitability. In the past decade, a growing number of plasticity phenomena have been identified in a wide range of systems that conform to this definition of homeostatic plasticity [4, 5, 6]. Generating and maintaining stability in neuronal circuit function is likely to be so fundamentally important that circuits employ multiple overlapping (and perhaps partially redundant) mechanisms that cooperate to constrain excitability. These mechanisms include: activity-dependent regulation of intrinsic neuronal firing properties [4, 5]; presynaptic and postsynaptic forms of excitatory synaptic plasticity, such as synaptic scaling, that adjust the strength of all excitatory synapses of a neuron up or down to stabilize firing [1, 2]; balancing of excitation and inhibition within complex recurrently connected neuronal networks [7, 8]; and compensatory changes in synapse number [9, 10•]. Importantly, several studies have shown that these compensatory changes in synaptic and neuronal properties act to restore neuronal firing rates to control levels following perturbations [11, 12], indicating that these mechanisms are truly ‘homeostatic’ in nature, and act to conserve some aspect of neuronal firing.

The existence of homeostatic plasticity in a variety of systems is well established, but little is known about the underlying signaling mechanisms. To implement homeostatic plasticity, neurons must be able to sense some aspect of activity, possibly integrate this measure over time, compare this (integrated) signal with a set-point value, and adjust synaptic properties to minimize the difference between actual activity and this set point. Such negative feedback is a fundamental feature of many physiological systems [1, 13], but how it is implemented in synaptic homeostasis is currently unknown. Understanding the molecular underpinnings of these homeostatic signaling loops is likely to be of major importance in understanding the processes that underlie activity-dependent refinement of neuronal circuitry, and also disease states and developmental disorders in which the balance between excitation and inhibition is disrupted. At the moment we are far from understanding the complete signaling pathways that underlie any form of synaptic homeostasis. Important open questions include: the nature of the ‘activity’ signal; the molecular identity of the ‘integrator’; whether synaptic homeostasis is cell-autonomous or requires altered function of entire networks; and whether synaptic homeostasis operates locally (in a synapse-specific manner) or globally on all synapses of a neuron (Figure 1). In this review, I focus on several forms of synaptic homeostasis, and discuss findings from the past two years that are beginning to shed light on these outstanding questions.

Section snippets

Rapid homeostatic signaling at the NMJ

One of the best-described examples of synaptic homeostasis occurs at the neuromuscular junction (NMJ), where perturbations in presynaptic function lead to compensatory changes in postsynaptic excitability, and vice versa [1, 14]. These changes compensate precisely for altered function, so that neuromuscular transmission is maintained. This has led to the idea that during development nerve excitation keeps up with muscle growth through homeostatic compensatory changes in synaptic transmission.

Synaptic homeostasis in the vertebrate CNS

Homeostatic synaptic compensation could occur through a variety of presynaptic and postsynaptic changes, acting either in isolation or synergistically. How synaptic homeostasis is implemented has important consequences for circuit function: for example, presynaptic changes in release probability will strongly affect short-term plasticity and thus information transfer across the synapses, whereas postsynaptic changes in receptor number will tend to scale up or scale down postsynaptic

Presynaptic homeostatic plasticity at central synapses

As I have already discussed, activity-deprivation in older neuronal cultures increases mEPSC frequency, presynaptic vesicle recycling and vesicle release probability during evoked synaptic transmission (inferred from FM-dye destaining rates) [10•, 12, 21, 22]. These studies suggest that under some circumstances, chronic inactivity at central synapses, as at the NMJ, can increase release probability and the number of functional release sites. Most studies that have looked have found that these

Expression mechanisms of synaptic scaling

Most studies of homeostatic plasticity in central neurons have observed that pharmacological manipulations of activity induce bidirectional compensatory changes in mEPSC amplitude at glutamatergic synapses [2]. These changes in quantal amplitude seem to operate uniformly on the entire distribution of synaptic weights, in effect scaling synaptic strength up or down; hence the origin of the term ‘synaptic scaling’ [11]. Synaptic scaling has been observed at cortical synapses following in vivo

Conclusions

The field of homeostatic plasticity is still young, and there are currently more questions about the induction mechanisms and signaling loops involved than there are answers. The field is in a period of almost exponential growth, and it is likely that additional phenomenology, as well as numerous molecular players, will rapidly accumulate over the next few years. A thoughtful approach to a few central challenges in the field will, in my view, greatly aid progress. First, given the wide range of

References and recommended reading

Papers of particular interest, published within the period of review, have been highlighted as:

  • • of special interest

  • •• of outstanding interest

Acknowledgements

My work is supported by grants NS 36853 and EY 014439 and the National Science Foundation.

References (50)

  • D. Moechars et al.

    Vesicular glutamate transporter VGLUT2 expression levels control quantal size and neuropathic pain

    J Neurosci

    (2006)
  • J.F. Guzowski et al.

    Environment-specific expression of the immediate-early gene Arc in hippocampal neuronal ensembles

    Nat Neurosci

    (1999)
  • K.R. Leslie et al.

    Postsynaptic depolarization scales quantal amplitude in cortical pyramidal neurons

    J Neurosci

    (2001)
  • K. Wibrand et al.

    Identification of genes co-upregulated with Arc during BDNF-induced long-term potentiation in adult rat dentate gyrus in vivo

    Eur J Neurosci

    (2006)
  • L.C. Rutherford et al.

    Brain-derived neurotrophic factor mediates the activity-dependent regulation of inhibition in neocortical cultures

    J Neurosci

    (1997)
  • G.W. Davis

    Homeostatic control of neural activity: from phenomenology to molecular design

    Annu Rev Neurosci

    (2006)
  • G.G. Turrigiano et al.

    Homeostatic plasticity in the developing nervous system

    Nat Rev Neurosci

    (2004)
  • L.F. Abbott et al.

    Synaptic plasticity: taming the beast

    Nat Neurosci

    (2000)
  • W. Zhang et al.

    The other side of the engram: experience-driven changes in neuronal intrinsic excitability

    Nat Rev Neurosci

    (2003)
  • A. Maffei et al.

    Selective reconfiguration of layer 4 visual cortical circuitry by visual deprivation

    Nat Neurosci

    (2004)
  • S.A. Kirov et al.

    Slices have more synapses than perfusion-fixed hippocampus from both young and mature rats

    J Neurosci

    (1999)
  • C.J. Wierenga et al.

    Temporal regulation of the expression locus of homeostatic plasticity

    J Neurophysiol

    (2006)
  • G.G. Turrigiano et al.

    Activity-dependent scaling of quantal amplitude in neocortical neurons

    Nature

    (1998)
  • J. Burrone et al.

    Multiple forms of synaptic plasticity triggered by selective suppression of activity in individual neurons

    Nature

    (2002)
  • W.B. Cannon

    The Wisdom of the Body

    (1932)
  • Cited by (249)

    • Intrinsic plasticity and birdsong learning

      2021, Neurobiology of Learning and Memory
    View all citing articles on Scopus
    View full text