Surprisingly, our data show no evidence for a learning-induced increase in synaptic strength. However, when more glutamatergic release events were induced by adding sucrose, we observed a 1.55-fold increase in the frequency of mEPSCs from AMPH CPP animals compared to controls. These new findings, an increase in the frequency of high osmolarity-induced release of glutamate after AMPH CPP, in the absence of a change in release probability, are consistent with an increase in the number of functional excitatory synapses contacting BLA pyramidal neurons.
Other studies have shown an increase in afferent drive to the lateral amygdala after cue specific reward learning (Tye et al. 2008
) as well as to the BLA after drug-induced learning (Tye et al. 2010
; Rademacher et al. 2010
). Any differences are unlikely to be caused by an overshadowing impact of handling or injection-induced stress, as the frequency and kinetics of the EPSCs measured in control groups were similar to control groups of other studies (Läck et al. 2007
; Tye et al. 2008
), and were similar to untreated animals in the same age range (data not shown). A few studies also examined the effect of other stimulants on the modification of BLA synapses, mostly focusing on the lateral amygdala (LA). Tye et al. (2010)
have shown that acute administration of methylphenidate facilitated the strengthening of cortico-amygdala synapses to the LA after drug-induced learning through a postsynaptic increase in AMPAR-mediated currents. In addition, Numachi et al. (2007)
found a significant increase in EphA5 mRNAs in the amygdala 9 and 24 h after acute methamphetamine treatment. They therefore suggested that methamphetamine could affect patterns of synaptic connectivity. Goussakov and colleges (2006)
showed that withdrawal (21 h) after repeated cocaine injections was accompanied by enhancements of glutamatergic synaptic transmission within the LA.
The absence of a change in spontaneous release is inconclusive by itself. To further analyze possible changes in synaptic strength to BLA pyramidal neurons after AMPH CPP, we analyzed the PPR as well as the CV of evoked EPSC amplitudes. These measures reflect potential differences in release probability. The PPRs were comparable to previous work, recorded in diverse brain regions (Schlüter et al. 2006;
Tye et al. 2010
). There was no difference in the CV or the PPR among the groups. Combined with the absence of an effect on EPSCs, those data suggest no pre- or postsynaptic changes in synaptic strength to BLA pyramidal neurons after AMPH CPP. This complements the results of our previous finding of increased excitatory drive of BLA neurons in vivo
(Rademacher et al. 2010
). However, that study was not designed to dissociate differences arising from the number of synapses versus the strength of synapses. The current results indicate that the changes observed in vivo
likely arise by a change in synapse number and not an increase of synaptic strength. Other factors may also contribute to the changes observed in vivo
that contrast with ex vivo
conditions, including the presence of extracellular synaptically released GABA that can modulate release of glutamate, and the in vivo
study focused on afferents arising from the hippocampal formation, while in this current study we used local stimulation within the BLA, activating diverse excitatory inputs, independent of their source. It is therefore possible that selective changes, which might occur only at afferents from the hippocampal formation, are masked in the presence of other inputs that might be unchanged. In addition, methodological differences exist between these studies, including the loss of a significant amount of synaptic connectivity in slices compared to in vivo
, the differences in temperature, intracellular versus whole-cell recording techniques, and the presence of an abundance of neuromodulators in vivo
that are absent in the slice.
While we did not find evidence for increased synaptic strength following the CPP testing, previous studies support a role for an increase of synaptic strength immediately following a single conditioning in the amygdala (Tye et al. 2008
), and over the course of several days in other brain regions (Stuber et al. 2008
). Additionally, further studies describe a potentiation of glutamatergic synapses to neurons of other brain areas caused by an exposure to drugs of abuse (Ungless et al. 2001
, Saal et al. 2003
, Faleiro et al. 2004
, Mameli and Lüscher 2011
). It is possible that there is a similar transient increase in synaptic strength that occurs during early components of CPP that is then translated into long-term structural changes. Furthermore, an increase of synaptic strength may occur in a subset of inputs, but these changes might be overshadowed by the relative absence of a change in the remaining inputs, masking the change of synaptic strength in this subset of inputs.
Taken together, our data indicate that it is unlikely that AMPH CPP leads to changes of synaptic release probability. A change in the number of excitatory synapses may be more readily observed in conditions when more synaptic terminals release their contents. Action potential-independent exocytosis can be stimulated in a manner that is independent from internal or external calcium concentration by increasing the extracellular osmolarity with sucrose (Rosenmund and Stevens 1996
). There was no change in the amplitude of the sucrose-evoked mEPSCs, similar to other studies, and consistent with a presynaptic site of effect (Stevens and Tsujimoto 1995
; Rosenmund and Stevens 1996
). Elevating the osmolarity of the extracellular saline increased the mEPSC frequency in each group, but neurons from AMPH CPP treated animals showed a mEPSC frequency in the presence of sucrose that was significantly higher than that of the two control groups. These results suggest a presynaptic change that might be caused by either a modification in release probability (Rosenmund and Stevens 1996
) or by an increase in the number of excitatory synapses per neuron. The difference in the sucrose-evoked frequency of mEPSCs alone does not allow one to discriminate between these two possibilities. But, considering the increased frequency of sucrose-evoked mEPSCs in the absence of a change in the PPRs and the CV of EPSC amplitude, as well as no changes in the frequency of spontaneous or mEPSCs, a modification in release probability seems to be unlikely. Consequently, together, those data strongly support the hypothesis that AMPH CPP leads to an increase in the number of excitatory synapses rather than a change in the release probability. Thus, these data support our previous results. The 1.55-fold greater increase in the frequency of sucrose-evoked excitatory events after AMPH CPP fits very nicely with the previously described learning-induced 1.6-fold increase in the total number of asymmetric synapses contacting BLA neurons (Rademacher et al. 2010
). We suggest that the lack of changes in the frequency of excitatory events after AMPH CPP without sucrose might be due to the low release probability of these new synapses.
During the preference test the animals were exposed to the drug-paired environmental context, but in the absence of the drug. A CPP memory can be extinguished, as repeated drug-free exposure to a drug-associated context reduces CPP behavior (Bardo et al. 1986
; Schroeder and Packard 2003
). Extinction is an active process that involves new memory formation (Rescorla 2001
; Schroeder and Packard 2003
). The BLA plays an important role in the extinction of CPP. For instance, excitotoxic lesions of the BLA prior to extinction sessions attenuated the extinction of CPP (Fuchs et al. 2002
). Due to the importance of the BLA in extinction of CPP, and the possibility that this form of learning induces synaptic changes, we cannot rule out a potential contribution of extinction to our findings.
This study demonstrates that changes induced by AMPH CPP occurred at the level of mEPSC, not spontaneous or evoked EPSCs ex vivo. Therefore, these changes are unlikely to involve global activity of neuronal networks that provide input to the BLA. Instead, our ex vivo data reveal a change that is consistent with an increased number of synapses after AMPH CPP. Combined with previous studies, this indicates that several changes contribute to the potency of drug-associated contexts in the BLA-dependent relapse to drug-seeking. Factors include specific enhancement of glutamatergic inputs from the hippocampal formation, in the absence of a generalized increase of synaptic strength. This is coupled to, and perhaps caused by, an increase in the number of excitatory synapses in the BLA. Perhaps, in the context associated with drug availability, the BLA is under a steady stream of excitatory drive, and the potent hippocampal inputs overwhelmingly drive the BLA. The expected result would be a BLA that is readily controlled by hippocampal inputs at the expense of other afferents. This imbalance may underlie the strong context-driven drug-seeking behavior seen in those individuals addicted to drugs.