During a whole-cell voltage clamp recording experiment an alteration in synaptic efficacy can be detected as an increase or a decrease in the postsynaptic current triggered in response to a presynaptic action potential. However, this change in post-synaptic response may arise from multiple sources. There can be an increase in the number of active synapses or conversion of silent synapses to active ones. There may also be an alteration in the number or neurotransmitter sensitivities of post-synaptic receptors. All-or-none changes in synaptic strength as well as postsynap-tic mechanisms typically scale synaptic responses without substantial alterations in their frequency-dependent characteristics. In contrast, an alteration in the probability of neurotransmitter release may result in the same outcome (i.e., an increase or a decrease in synaptic strength) but in turn can fundamentally alter synaptic responses to physiological trains of action potentials. A change in the probability of neuro-transmitter release is the most common way neurotransmitter release kinetics are altered during synaptic plasticity. Such a change may arise from an alteration in the number of vesicles available for release or a modification in their fusion propensity in response to an action potential. These alterations can alter the rate of short-term synaptic plasticity. For instance, an increase in the probability of neurotransmitter release, in the absence of rapid replenishment of fused vesicles, may result in faster depletion of readily releasable vesicles and cause enhanced depression. A decrease in the release probability may make synapses less prone to vesicle depletion and tend to decrease the extent of depression. Therefore, systematic measurement of the extent of synaptic depression induced by a train of action potentials at moderate to high frequency stimulation may yield critical information in regard to neurotransmitter release kinetics (Zucker and Regehr, 2002).
In addition to measuring the kinetics of short-term synaptic depression or facilitation from a population of synapses, one can also obtain a more direct measure of neurotransmitter release probability if vesicle fusion from a single release site can be detected. Such a setting can be achieved during paired recordings between sparsely connected neurons (albeit in a labor-intensive way), or using the minimal stimulation method via application of low-intensity stimulation to activate a single release site. Under these conditions one can typically observe a substantial amount of neurotransmission failures arising from the probabilistic nature of vesicle fusion in single synapses. Over a large number of independent trials, a decrease in the number of failures would suggest an increase in probability of release or vice versa.
Use-dependent block of NMDA receptors by MK-801 provides an additional method to estimate the probability neurotransmitter release. MK-801 is a well-characterized blocker of NMDA receptors that penetrates open NMDA channel pores and impairs the ion flux through NMDA receptors (Huettner and Bean, 1988). This highly selective open channel block renders MK-801-dependent inhibition of NMDA currents extremely use-dependent. Therefore, the time course of MK-801 block of synaptically evoked NMDA currents is proportional to the activation of NMDA receptors by vesicle fusion events, providing a measure of the probability of neurotransmitter release (Hessler et al., 1993; Rosenmund et al., 1993). However, recent studies suggest that NMDA receptors in given synapse may not be saturated by the glutamate released in response to a single action potential (Mainen et al., 1999; Oertner et al., 2002). This observation may complicate direct estimation of the absolute release probability. Nevertheless, this method may still allow a relative measure by comparing different experimental conditions as long as there are no alterations in the number of NMDA receptors and in their sensitivities to glutamate between the conditions in question.
As indicated above, detection of evoked quantal responses (either through minimal stimulation or paired recordings) provides a suitable setting to determine neurotransmitter release probability and alterations in rate of vesicle fusion. However, in synapses with multiple release sites, such as the calyx of Held, isolation of evoked quantal responses is nearly impossible and truly quantal release is hard to detect except in the case of spontaneous neurotransmission. Therefore, under these conditions, the rate of synaptic vesicle fusion can be determined by deconvolution of synaptic currents with the quantal unitary current. This approach is valid only when the synaptic current can be assumed to result from the convolution between a quantal current and quantal release rates. This assumption is not valid in cases where post-synaptic mechanisms, such as receptor saturation and desensitization, alter quantal events and thus shape synaptic responses during repetitive stimulation (Neher and Sakaba, 2001).
Despite its wide use and versatility, electrophysiological approaches also possess several caveats. First, this form of detection is limited to fast neurotransmitters that are released rapidly and activate closely juxtaposed ligand-gated channels.
Using this approach it is not possible to examine release kinetics of neuromodu-latory transmitters such as dopamine or catecholamines, as they typically activate G-protein-coupled receptors and only generate slow electrical responses that are not necessarily linearly proportional to release kinetics. Second, in the case of ionotropic receptors the reliance of this physiological setting on receptor properties such as receptor desensitization and saturation brings in a major caveat that hinders direct estimation of presynaptic release properties. Third, synaptic events, which occur at distal sites as well as on thin dendritic branches, are hard to detect due to dendritic filtering. Fourth, in most cases information is difficult to obtain from single release sites; therefore, experimental readout typically provides an analysis of a population of synapses. Lastly, electrical detection of neurotransmitter release is not inherently sensitive to use history of vesicles (i.e., synaptic vesicle recycling). However in cases when endocytosis is impaired (Delgado et al., 2000) or vesicle re-acidification is inhibited using blockers of vacuolar ATPase, electrophysiological readout of neurotransmitter release can provide information in regard to synaptic vesicle recycling (Ertunc et al., 2007).
Was this article helpful?