Long-term BZ (benzodiazepine) anxiolytic therapy increases the risk of physical dependence manifested as withdrawal anxiety. BZ-induced potentiation of GABA(A)R (gamma-aminobutyric acid type-A receptor) function by 1-week oral administration of FZP (flurazepam) bi-directionally modulates excitatory glutamatergic synaptic transmission in hippocampal CA1 neurons during drug withdrawal. Previous electrophysiological studies on acutely isolated and intact CA1 neurons, as well as immunofluorescence and post-embedding immunogold electron microscopy studies, suggest increased synaptic insertion of GluR (glutamate receptor) 2-lacking AMPARs (alpha-amino-3-hydroxy-5-methylisoxazole-4-propionic acid receptors) in 2-day FZP-withdrawn rats. Preliminary studies indicated a similar increase in GluR1, then phospho-Ser(831)-GluR1, as well as CaMKIIalpha (Ca(2+)/calmodulin-dependent protein kinase IIalpha), but not phospho-Thr(286)-CaMKII levels at the same time point. In our studies, whole-cell recordings in hippocampal slices revealed that AMPAR mEPSC [miniature EPSC (excitatory postsynaptic current)] amplitude was increased in 1-day FZP-withdrawn rats followed by an increase in estimated single-channel conductance in 2-day-FZP-withdrawn rats. Enhanced conductance was no longer observed in slices pre-incubated for 2 h in the CaMKII inhibitor KN-93, but not the inactive analogue KN-92. To evaluate whether CaMKII-mediated AMPA potentiation could occlude LTP (long-term potentiation), LTP was induced by TBS (theta burst stimulation) and recorded using whole-cell and extracellular techniques. LTP was induced in both groups, but only maintained for <15 min in 2-day FZP-withdrawn rats. LTP was fully restored after 7-day withdrawal. Despite the lack of LTP maintenance, impairment of object recognition, place and context was not observed in 2-day-FZP-withdrawn rats. Since L-VGCC (L-type voltage-gated calcium channel) current density was doubled on drug withdrawal and up to 2 days, Ca(2+) entry through L-VGCCs and perhaps subsequently through Ca(2+)-permeable AMPARs are proposed to be responsible for enhanced CaMKIIalpha levels and AMPAR potentiation. Mechanisms associated with several different models of activity-dependent plasticity may underlie BZ physical dependence.