Depression is a debilitating mental disorder, and selective serotonin reuptake inhibitors (SSRIs) constitute the first-line antidepressant treatment choice for the clinical management of this illness; however, the mechanisms underlying their therapeutic actions and side effects remain poorly understood. Here, we compared the effects of two SSRIs, fluoxetine and citalopram, on synaptic connectivity and the efficacy of cholinergic synaptic transmission between identified presynaptic and postsynaptic neurons from the mollusc Lymnaea. The in vitro paired cells were exposed to clinically relevant concentrations of the two SSRIs under chronic and acute experimental conditions, and the incidence of synapse formation and the efficacy of synaptic transmission were tested electrophysiologically and with fluorescent Ca(2+) imaging. We demonstrate that chronic exposure to fluoxetine, but not to citalopram, inhibits synapse formation and reduces synaptic strength, and that these effects are reversible following prolonged drug washout. At the structural level, we demonstrate that fluoxetine, but not citalopram, prevents the expression and localization of the presynaptic protein synaptophysin. Acute exposure to fluoxetine substantially reduced synaptic transmission and synaptic plasticity (post-tetanic potentiation) in established synapses, whereas citalopram reduced synaptic transmission, but not short-term synaptic plasticity. We further demonstrate that fluoxetine, but not citalopram, directly inhibits voltage-gated Ca(2+) currents in the presynaptic neuron, as well as postsynaptic responsiveness to exogenously applied neurotransmitter. This study provides the first direct evidence that fluoxetine and citalopram exert characteristic, non-specific side effects that are unrelated to their function as SSRIs, and that fluoxetine is more detrimental to synaptic physiology and structure than citalopram.