Conditioning Makes Some Neurons More Excitable
Joseph Ziminski, Sabine Hessler, Gabriella Margetts-Smith, Meike C. Sieburg, Hans S. Crombag, et al.
(see pages 3160–3170)
Repeated co-occurrence of an otherwise neutral stimulus with a reward or punishment promotes the creation of neural ensembles that drive an animal to seek the reward or avoid the punishment when the conditioned stimulus appears. The neural ensembles include neurons that encode the stimulus and reward and neurons that motivate specific behaviors. For rewards, the ensembles include neurons in the orbitofrontal cortex (OFC) and nucleus accumbens (NAc).
The importance of synaptic plasticity in driving ensemble formation during learning is well studied and widely appreciated. Less well studied is how changes in the intrinsic properties of a neuron, such as ion channel expression and conductance, contribute to associative conditioning, particularly reward conditioning. Ziminski et al. addressed this question using mice that express GFP under control of the fos promoter, which allowed them to visualize and record neurons activated during exposure to an auditory cue and training context.
More neurons in the NAc and PFC were activated by the presence of an auditory cue and training context in mice that had learned that the cue predicted food availability (trained mice) than in mice that had not undergone conditioning (untrained mice). After extinction training, however, there was no difference in the number of activated (GFP+) neurons in NAc or PFC between previously trained and untrained mice. In the NAc of trained mice, GFP+ neurons were more excitable than surrounding inactive (GFP–) neurons. Specifically, GFP+ neurons fired more action potentials in response to a given current injection, likely because they had a higher input resistance (suggesting fewer channels were open at rest) than GFP– neurons. No such difference in excitability and input resistance was seen between GFP+ and GFP– neurons in the NAc of untrained mice, the NAc of mice that had undergone extinction training, or in the OFC of any mice.
These results suggest that an increase in the excitability of NAc neurons contributes to reward conditioning. This heightened excitability might increase the likelihood that a conditioned stimulus will motivate reward-seeking behaviors. Future studies should investigate the molecular basis for this change in excitability, and determine whether blocking those mechanisms disrupts learning.
Reduced K+ Current Underlies Cold Allodynia
Alejandro González, Gonzalo Ugarte, Carlos Restrepo, Gaspar Herrera, Ricardo Piña, et al.
(see pages 3109–3126)
Cold-sensitive sensory neurons allow us to discern a broad range of temperatures, from pleasantly cool to painfully cold. Cold sensitivity is conferred primarily by the expression of TRPM8 channels, but the activation threshold is determined largely by the expression level of voltage-gated KV1 channels. Because the slowly inactivating KV1 current (IKD) dampens neuronal excitability, neurons that express high levels of these channels with low levels to TRPM8 have higher thresholds (that is, are activated at lower temperatures) than neurons with high levels of TRPM8 and low levels of KV1.
Lowering bath temperature (black traces) causes cold-sensitive neurons (red trace) to spike. Treatment with a KV1 antagonist (black bar above traces) raises the temperature threshold (arrowheads) and increases firing rate of cold-sensitive neurons. It also imparts cold sensitivity to some formerly insensitive neurons (blue trace). See González, Ugarte, et al. for details.
People become hypersensitive to cold during several pathological conditions, including peripheral nerve injury. The molecular basis of this hypersensitivity is poorly understood. Addressing this question in mice, González, Ugarte, et al. found that chronic constriction of the sciatic nerve produced behavioral and physiological effects similar to those produced in uninjured mice by injecting a KV1 antagonist. Both conditions produced behavioral manifestations of cold allodynia, both shifted the activation threshold of cold-sensitive neurons to warmer temperatures, and both increased the maximum firing rate of these neurons. The number of neurons that were sensitive to temperature reduction also increased. Some of these newly cold-sensitive neurons were activated by a TRPV1 activator and some exhibited spike waveforms indicative of nociceptors.
Importantly, nerve constriction occluded the effects of the KV1 antagonist. Moreover, IKD current density was lower in cold-sensitive neurons from injured mice than in controls, whereas TRPM8 current density was similar in the two groups. Finally, although a TRPM8 antagonist greatly reduced cold sensitivity, the proportion of cold-sensitive neurons that were activated by a TRPM8 agonist and the amplitude of these neurons' responses were unaffected by nerve constriction.
These results suggest that peripheral nerve injury produces cold hypersensitivity by reducing IKD in cold-sensitive neurons, as well as in some normally cold-insensitive nociceptive neurons that express TRPV1. The plausibility of this hypothesis was confirmed with a computational model. Future studies will be needed to determine how nerve injury reduces IKD and whether the same mechanism underlies cold hypersensitivity in other neuropathies, such as those resulting from cancer drugs or diabetes.
Footnotes
This Week in The Journal was written by Teresa Esch, Ph.D.
