Getting hot under the calyx
Year: 2007
Type: article
Abstract: Most mammals are homeothermic, and finely tune their core body temperature within a narrow range (36–39°C). Unsurprisingly, brain function is temperature dependent and it is thought that brain cooling can ameliorate damage following trauma, such as stroke. A variety of structural and physiological properties of neurons are influenced by temperature, including spine morphology and motility, vesicular release of neurotransmitter, action potential generation, and the kinetics of neurotransmitter receptors and transporters (Klyachko & Stevens, 2006). However, a full description of the effects of temperature at central synapses is still missing. What is more, the majority of in vitro studies are conducted at room temperature to improve tissue viability and facilitate electrophysiological recordings. A drawback of this is that the interpretation of the results may not hold in the warmer surroundings of the functioning brain. In a recent issue of The Journal of Physiology, Postlethwaite et al. (2007) investigate how changes in temperature affect fast neurotransmission at the calyx of Held, a giant glutamatergic terminal in the mammalian auditory pathway that conveys information about sound-source localization. Their initial experimental observation is in agreement with what is already known from membrane biophysics and from pioneering work on nicotinic acetylcholine receptors at the neuromuscular junction. Excitatory postsynaptic currents (EPSCs) become larger and have faster kinetics when the temperature in the recording chamber is increased from 25 to 35°C. By comparing the experimental results with data from computer-simulated neurotransmission in a 3D reconstruction of the calyx, the authors are able to identify the most temperature-sensitive mechanism that shapes EPSCs. This turns out to be the kinetics of postsynaptic AMPA-subtype glutamate receptors. In fact, taking a kinetic model for the receptor and simply scaling all the rate constants by 2.4 is sufficient to explain the physiological data. In agreement with this, puffing glutamate directly onto the postsynaptic cell, which bypasses presynaptic release mechanisms, results in AMPA receptor currents with a similar temperature sensitivity to that of miniature EPSCs (mEPSCs). That this mechanism describes the data well has interesting implications for the conformational changes that would be responsible for such strong increases in affinity. Furthermore, it raises the question as to why this behaviour is so different from recombinant receptors, which appear to have independent binding sites (i.e. K0↔1=K1↔2=K2↔3=K3↔4; Robert & Howe, 2003). Accessory proteins like stargazin can increase the apparent affinity of AMPA receptors for glutamate, and they could also alter the cooperativity of binding in native channels. A further caveat of the proposed mechanism is that the cyclic reactions do not obey microscopic reversibility. While some ion channels have been shown to violate this principle, there is no such evidence for AMPA receptors and this is unlikely to be necessary to describe the data. The observed Q10 of 2.4 is well within the range seen in previous studies on muscle nicotinic and glutamate receptors. Significant changes in protein conformation, which would be expected to have a Q10 of 2 or more, are likely to accompany channel opening and desensitization. In contrast, binding rates might be expected to show weak temperature dependence, because this process is related to diffusion, which has a Q10 value of ∼1.3. However, structural studies tell us that the binding domain in glutamate receptors is shaped like a clam shell, and that binding induces the clam shell to close its jaws firmly on the agonist. The bound lifetime of the receptor depends as much on newly formed interdomain bonds that stabilise this conformation, as it does on ligand–receptor interactions. So it is reasonable that association and dissociation rates show a similar temperature sensitivity to other rates in the mechanism. Indeed, the experimentally observed changes in kinetics and amplitude of mEPSCs cannot be reproduced when scaling the rate constants for binding and unbinding by 1.3 rather than 2.4 (our unpublished simulations of synaptic currents using the authors' model). Although postsynaptic mechanisms are sufficient to explain the observed changes with temperature, can all the other components of neurotransmission be disregarded? The authors test this by modelling the influence of increasing the rate of glutamate diffusion, or the rate of fusion pore opening. In contrast to the temperature-dependent increase in EPSC amplitude observed experimentally, these manipulations predict a decrease in the peak current. Deconvolution analysis of mEPSCs also suggests that glutamate release rate does not change much with temperature. However, if the properties of neurotransmitter release differ during evoked and miniature events, other scenarios could wield a presynaptic influence. For example, improved synchronization of neurotransmitter release across different active zones or an increase in the probability of multiple fusion events could lead to faster and larger synaptic currents. In addition, recent experiments have directly demonstrated that both ‘kiss-and-run’ and full-fusion events can occur at this synapse (He et al. 2006). If temperature affects the nature of the release process, or alters the relative proportions of the two types of fusion events, there could be a change in the shape of the synaptic glutamate transient. Ultrastructural studies reveal glutamate transporters in glial cells that ensheath the calyx of Held (Renden et al. 2005). While these transporters are not close enough to release sites to directly influence the glutamatergic transmission in the time course of a mEPSC, their activity can affect resting levels of glutamate. In young animals, like those used in this study, tonic activation of presynaptic metabotropic glutamate receptors can inhibit Ca2+ influx and exocytosis from the calyx (Renden et al. 2005). At 35°C transporters cycle glutamate 3 times faster than at 25°C, and this could reduce the activation of presynaptic metabotropic receptors by ambient glutamate, thereby leading to an increase in EPSC amplitude (and mEPSC frequency). For this reason, it would be interesting to check whether the temperature sensitivity of the EPSCs recorded by Postlethwaite et al. (2007) holds in the presence of glutamate transporter blockers. How can we use the present study to extrapolate physiological information from experimental data obtained at room temperature? The use of a single scaling factor is tempting in its simplicity. This is likely to account for the temperature sensitivity of AMPA receptors with different subunit composition and so may be generally applicable throughout the brain. Strikingly, the proposed mechanism suggests that AMPA receptor EPSCs measured at room temperature have a larger contribution from low-conductance open states. This implies less desensitization at room temperature and could lead to an underestimate in the number of channels contributing to an EPSC. Another extrapolation of this study is to ask how temperature might affect the response to trains of stimuli rather than single events. The calyx of Held can fire action potentials at frequencies of up to 1 kHz and it is likely that, under such conditions, the temperature sensitivity of presynaptic factors, such as vesicle recycling and calcium channel inactivation, could have more influence on neurotransmission. Given the unusual geometry and properties of the calyx, it is possible that other synapses have divergent temperature dependence. Most neurons in the brain receive multiple synaptic inputs distributed along their dendrites, and express neurotransmitter transporters within the synaptic cleft. Although AMPA receptors will be likely to show a similar temperature dependence to that reported in this paper, they may no longer be the sole determinants for the temperature sensitivity of synaptic transmission. However, the study by Postlethwaite and colleagues emphasizes that quantitative analysis of synaptic behaviour, taking into account the synaptic environment in its entirety, can lead to valuable insights about neurotransmission as it really is in our brains.
Source: The Journal of Physiology
Language: English
Subfield: Cellular and Molecular Neuroscience
Field: Neuroscience
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W2010416360