Prefrontal inositol triphosphate is molecular correlate of working memory in nonhuman primates

Abstract

Working memory (WM) is a process of actively maintaining information in the mind for a relatively short period of time, and prefrontal cortex (PFC) has been thought to play a central role in its function. However, our understanding of underlying molecular events that translate into WM behavior remains elusive. To shed light on this issue, we have used three distinct nonhuman primate models of WM where each model represents three WM conditions: normal control, WM-deficient, and recuperated to normal from WM deficiency. Based on the hypothesis that there is a common molecular substrate for the coding of WM behavior, we have studied the relationship of these animals' performance on a WM task with their PFC levels of molecular components associated with Gq-phospholipase C and cAMP pathways, with the idea of identifying the footprints of such biomolecules. We observed that in all of the primate models WM deficiency was strongly related to the reduced concentration of IP(3) in PFC, whereas recuperation of WM-deficient animals to normal condition was associated with the normalization in IP(3) level. However, this correlation was absent or weak for cAMP, active protein kinase A, dopamine D(1) receptor, and Gq protein. In addition, WM deficiency related not only to pharmacological conditions but also to aging. Thus, it is suggested that optimal IP(3) activity is essential for normal WM function and the maintenance of intracellular IP(3)-mediated Ca(2+) level in PFC may serve as biochemical substrate for the expression of WM behavior.

Figure 1.

WM status of nonhuman primate models before they were killed and their PFC IP₃ concentration. A, As shown previously (Castner et al., 2000, 2005), both amphetamine (black bars in amphetamine young group) and haloperidol (black bars in haloperidol young and haloperidol aged groups) treatments produced significant loss in WM. Six months after withdrawal in amphetamine-treated monkeys (Castner et al., 2005) (hatched bars in amphetamine young group) and after dopamine D₁ agonist administration to haloperidol-treated monkeys (Castner et al., 2000) (hatched bars in haloperidol young group) the WM deficit observed in animals recovered to a normal level. B, A significant reduction in the amount of IP₃ was observed in animals with WM deficiency. Recuperation in WM was associated with the normalization in IP₃ concentration. * results are significantly different from the control group; ^♦ results are significantly different from the WM-deficient group.

Figure 2.

Levels of cAMP and active PKA in PFC. A, The cAMP level in amphetamine young group did not coincide with the behavioral performance seen in these animals; however, there were coincidence in cases of haloperidol young and haloperidol aged animal groups where WM-deficient animals showed reduced cAMP levels. Bi, ii, Representative blots of active PKA (i) and their optical density plot (ii) are shown. The amphetamine young and haloperidol aged animal groups did not show any change in active PKA level. However, a reduction in WM-deficient animals and an increase in recuperated normal animals of the haloperidol young group were observed. The arrow in the blots in i indicates the immunoreactive band of 39 kDa corresponding to a catalytically active subunit of PKA protein size. * results are significantly different from control group; ^♦ results are significantly different from WM-deficient animals.

Figure 3.

Amount of dopamine D₁ receptor and Gq protein in PFC. A, A correlation of D₁ receptor level with WM status was seen in WM-deficient animals of all the groups but this correlation was lacking in the recuperated normal animals of amphetamine young group. Bi, ii, Representative blots of Gq protein (i) and their optical density plot (ii) are shown. Gq protein level was not changed in any of the groups. The arrow in the blots in i indicates the immunoreactive polypeptide band of 42 kDA corresponding to Gq protein size. * results are significantly different from control group; ^♦ results are significantly different from WM-deficient animals.

Figure 4.

IP₃ concentration in the striatum. IP₃ level in this area did not correlate with the WM status seen in the animals. An increase in IP₃ was observed in recuperated normal animals of the haloperidol young group. However, this may be due to the activation of dopamine D₁ receptor while D₁ agonist is on board in these animals. Values are expressed as the mean ± SEM of four to five animals. * results are significantly different from control; ^♦ results are significantly different from WM-deficient animals.