Timing of drug infusions relative to behavioral testing

As shown in Figure 2A, for MUS, we matched the time between infusion and probe test to ensure an equal diffusion of drug between conditions. To accomplish this, we inserted a 30 min delay between infusion and probe test, when the drug was infused after satiation but before the probe. This strategy is similar to that used in the amygdala and orbitofrontal cortex (Wellman et al., 2005; West et al., 2011). The duration of action of MUS when microinjected into the brain easily exceeds 90 min; in a prior study, we found that effects of MUS infusion increased over the duration of a 90 min infusion period and observed behavioral responses that in some cases lasted for several hours (Dybdal et al., 2013). Thus, it was not possible to allow sufficient time for clearance of MUS prior to the probe test.

In order to determine, what, if any, effect the MD plays exclusively during the period of satiation, we required a compound that would be cleared by the time the probe test occurred. For this reason, we turned to KYNA, which is quickly cleared in the brain and has a relatively short duration of action (Forcelli et al., 2014; Vécsei and Beal, 1990; Turski and Schwarcz, 1988). In rodent studies, the half-life of KYNA in brain has been reported to range from 5 to 30 min (Turski and Schwarcz, 1988). Consistent with this timecourse, in a prior study, we observed normalization of cognitive function within 45 min of infusion of KYNA (Forcelli et al., 2014) (and unpublished observations) in the primate brain. To maximize time for clearance of KYNA when infused prior to satiation, we inserted a 15 min delay following the satiation procedure; thus the total time from the end of the infusion to the start of the probe was ~45 min. Finally, we infused KYNA at a lower dose (150 nmol) in two animals (EL/LO). This dose of KYNA was without effect on the proportion shifted when infused before the probe test (0.71 and 0.72 after KYNA, as compared to 0.73 and 0.85 after control, respectively; see Figure 2—figure supplement 1). Given the half-life of KYNA in brain, even if sparse amounts of KYNA remained when animals were infused before satiation, the concentration of KYNA remaining would not be sufficient to produce the significant deficits observed.