Theta frequency is a significant oscillatory rhythm in rodents, because it is observed during exploratory behavior and is highly effective in the induction of LTP (Frick et al., 2004, Hoffman et al., 2002, Kelso and Brown, 1986 and Watanabe et al., 2002). The facilitatory action of presynaptic NMDARs on neurotransmission offers a mechanistic rationale as to why theta frequency is effective for LTP induction. These data also resolve the paradox of how it is that synapses with low pr are able to contribute to the induction of LTP. A synapse with a pr of 0.1 might be expected

to release transmitter just twice during a train of 20 APs and might therefore be expected to fail to achieve adequate activation of the postsynaptic neuron. However, the feedback loop generated from check details Ca2+ influx via activation of NMDA autoreceptors will ensure that a low pr synapse achieves augmented release during the course of the stimulus train (Figure 10B). The relationship

between transmitter release and presynaptic NMDAR activation also has utility, because manipulations of pr also change the probability of observing presynaptic NMDAR-mediated large Ca2+ events. Manipulations that reduce pr boutons, such as adenosine, decrease the number of large Ca2+ events, whereas manipulations that increase pr increase the PD173074 number of large events. Importantly, induction of LTP, which is reported to increase pr at active synapses (Antonova et al., 2001, Bolshakov and Siegelbaum, 1995, Emptage et al., 2003, Enoki et al., 2009, Malgaroli et al., 1995 and Ward et al., 2006), increases the incidence of large Ca2+ transients. Therefore, the measurement of the number of large Ca2+ transients in the bouton provides a novel technique with which to measure pr. Whether this approach first has utility at other axon terminals will be dependent on the presence of NMDAR autoreceptors. There is an interesting correlation in the

literature that would seem to suggest that Ca2+ transient variability at the presynaptic boutons and presynaptic NMDARs is a general motif. For example, (1) modulating the frequency of mini EPSPs in the entorhinal cortex (Berretta and Jones, 1996 and Woodhall et al., 2001), layer V of the visual cortex (Sjöström et al., 2003), or CA1 pyramidal neurons of the hippocampus (Madara and Levine, 2008) or (2) enhancing long-term depression (LTD) in the visual cortex (Sjöström et al., 2003), the barrel cortex (Rodríguez-Moreno and Paulsen, 2008), and the cerebellum (Duguid and Smart, 2004) all require presynaptic NMDAR activation and each are regions known to show highly variable presynaptic Ca2+ transients (Frenguelli and Malinow, 1996, Kirischuk and Grantyn, 2002, Llano et al., 1997 and Wu and Saggau, 1994b).