Short- and Long-Term Effects of mGluRs on NMDA Responses
Nathanael O'neill, Catherine McLaughlin, Noboru Komiyama, and Sergiy Sylantyev
(see pages 9840–9855)
Neurons contain multiple types of glutamate receptors, including ionotropic NMDA receptors (NMDARs) and G-protein-coupled metabotropic receptors (mGluRs). Often multiple receptor types are expressed at the same synapse and are interlinked by molecular scaffolds. Thus, activating one receptor type can influence the function of others. For example, activation of mGluRs enhances NMDAR currents in some cells, and suppresses them in others.
The effects of mGluRs are usually assumed to result from signaling downstream of G-proteins. But in cerebellar granule cells, type I mGluRs potentiate NMDARs independently of these pathways, instead depending on interactions requiring Homer-family scaffolding proteins. These Homer-dependent influences occur more rapidly (within 1 ms) than is required for G-protein signaling, so mGluRs might exert additional effects on NMDARs via G-protein signaling at later time points.
O'neill et al. examined this possibility in cerebellar and dentate granule cells. Brief (∼1 ms) application of type I mGluR agonists increased the amplitude of NMDA currents in both cell types. These effects were blocked by overexpressing Homer1a, a short Homer isoform that interferes with scaffold formation, but not by pertussis toxin, which blocks G-protein-mediated signaling. Longer application (hundreds of milliseconds) of mGluR agonists had different effects in cerebellar and dentate granule cells. In dentate cells, prolonged mGluR activation increased NMDA responses when either Homer- or G-protein-dependent signaling was blocked, but not when both were blocked. Activating mGluRs also increased NMDA-induced spiking and synaptic plasticity in dentate cells, and both Homer scaffolds and G-protein signaling contributed to these effects. In cerebellar cells, prolonged mGluR activation had no effect on NMDAR responses or NMDA-induced spiking when Homer scaffolds were intact, but the responses and spiking were reduced by mGluR activation when scaffold interactions were disrupted.
These results indicate that Homer scaffolds enable type I mGluRs to enhance NMDA responses rapidly in both cerebellar and dentate granule cells. In dentate cells, Homer- and G-protein-dependent effects are similar, and both enhance NMDAR-evoked spiking and plasticity. In cerebellar cells, however, G-protein-dependent mGluR signaling suppresses NMDAR function and this effect can be masked by Homer-dependent potentiation. Thus, endogenous regulation of Homer1a expression, which occurs during neuronal activity, might serve as a mechanism for modulating the effects of mGluRs on NMDAR function.
Role of Methyl-CpG-Binding Domain Protein 1 in Pai
Kai Mo, Shaogen Wu, Xiyao Gu, Ming Xiong, Weihua Cai, et al.
(see pages 9883–9899)
Peripheral nerve injury sensitizes nociceptors, so they respond to formerly painless stimuli. This sensitization stems from altered expression and/or function of neurotransmitter receptors and ion channels that shape neuronal activity and excitability. Although posttranslational modifications underlie some of these effects, injuries also trigger epigenetic modifications that produce long-term changes in the transcription of receptor and channel genes, resulting in persistent pain. One such modification is methylation of DNA on cytosines adjacent to guanines (CpG sites). This modification is catalyzed by DNA methyltransferases (DNMTs), and it enables recruitment of methyl-CpG-binding domain (MBD) proteins, which repress transcription of nearby genes. People with mutations in one MBD protein, MeCP2, have reduced pain sensitivity. And Mo et al. now report that another MBD protein, MBD1, is involved in both normal and neuropathic pain.
mbd1 mRNA (red) is expressed in a variety of neurons in the DRG, including some peptidergic nociceptors (green). See Mo, Wu, Gu, Xiong et al. for details.
MBD1 mRNA was present in multiple neuron types in mouse dorsal root ganglia (DRG), including some nociceptors. Knocking down MBD1 reduced the frequency and latency of paw withdrawal from mechanical and thermal stimuli, respectively, suggesting it impaired nociception. Conversely, overexpressing MBD1 in DRG increased nocifensive responses and increased time spent in a lidocaine-paired chamber, suggesting it produced spontaneous pain.
MBD1 levels increased in DRG after peripheral nerve injury, whereas—consistent with previous work—levels of mu opioid receptors (MORs) and KV1.2 voltage-gated potassium channels were reduced. Similarly, overexpressing MBD1 reduced MOR and KV1.2 expression in uninjured neurons, whereas MBD1 deficiency increased expression of these genes in both injured and uninjured mice. Consequently, MBD1 deletion increased potassium current density, hyperpolarized the resting membrane potential, and decreased neuronal excitability; it also potentiated MOR-agonist effects and reduced morphine tolerance. Finally, MBD1 was shown to bind to, and promote the binding of DNMT3a to, MOR and KV1.2 gene promoters.
These results suggest that MBD1 works with DNMT3a to repress transcription of MORs, KV1.2, and possibly other proteins, to set the excitability level of nociceptors. Upregulation of MBD1 after peripheral nerve injury likely contributes to heightened pain sensitivity. Reducing MBD1 activity might therefore help to reduce neuropathic pain. Because MBD1 is expressed in many neuron types in DRG, however, development of potential therapies should include assessment of effects on the perception of non-painful somatosensory stimuli.
Footnotes
This Week in The Journal was written by Teresa Esch, Ph.D.
