All flies were tested at 30 days of age for learning during training or short-term memory. Mean CIs (+/− SEM) are plotted, Ns are indicated above each bar for all groups. The levels of significance are indicated (one asterisk * indicates p < 0.05, two asterisks ** indicate p < 0.01, three asterisks *** indicate p < 0.001). A triangle in indicates a 40% or greater reduction in courtship activity during the training session with a previously mated female (A–C). A) Heterozygousity for the IPPase loss of function mutant IPPase 1 impairs learning during training, while heterozygousity for IPPase 3 does not. B) Heterozygousity of the IPPase loss of function mutants rescues learning during training in psn-het B3. C) Heterozygousity for the IPPase 3 loss of function mutant rescues learning during training in psn-het I2 flies, while heterozygousity for IPPase 1 does not. D) Heterozygousity for the IPPase loss of function mutants does not impair short-term memory. Heterozygousity of the IPPase loss of function mutants rescues short-term memory E) in psn-het B3 and F) in psn-het I2 flies.

All flies were tested at 30 days of age for learning during training or short-term memory. Mean CIs (+/− SEM) are plotted, Ns are indicated above each bar for all groups. The levels of significance are indicated (one asterisk * indicates p < 0.05, two asterisks ** indicate p < 0.01, three asterisks *** indicate p < 0.001). A triangle in indicates a 40% or greater reduction in courtship activity during the training session with a previously mated female (A–C). DmGluRA¹¹² is a null allele and DmGluRA^2b is a precise excision wild type control allele of DmGluRA. A) Heterozygousity for the DmGluRA null impairs learning during training. Heterozygousity of DmGluRA rescues learning during training in B) psn-het B3 and C) psn-het I2 flies. D) Heterozygousity for the DmGluRA null impairs short-term memory. Heterozygousity of DmGluRA rescues short-term memory in E) psn-het B3 and F) psn-het I2 flies.

All flies were tested at either 30 or 40 days of age for learning during training or short-term memory. Mean CIs (+/− SEM) are plotted, Ns are indicated above each bar for all groups. The levels of significance are indicated (one asterisk * indicates p < 0.05, two asterisks ** indicate p < 0.01, three asterisks *** indicate p < 0.001). A triangle in indicates a 40% or greater reduction in courtship activity during the training session with a previously mated female (A–D). A) Heterozygousity for the InsP₃R loss of function mutants 361 and 90B0 impairs learning during training, while heterozygousity for the InsP₃R loss of function mutant 703 does not completely abrogate learning during training. B) Heterozygousity of the InsP₃R loss of function mutants 703 and 361 rescues learning during training in psn-het B3 flies. C) Heterozygousity of the InsP₃R loss of function mutants 703 and 361 fails to rescue learning during training in psn-het I2 flies. D) Heterozygousity for the InsP₃R loss of function mutant 90B0 does not rescue learning during training in psn-het B3 or psn-het I2 flies. E) Heterozygousity for the InsP₃R loss of function mutants does not impair short-term memory. Heterozygousity of the InsP₃R loss of function mutants 703 and 361 rescues short-term memory F) in psn-het B3 and G) in psn-het I2 flies. H) Heterozygousity of the InsP₃R loss of function mutant 90B0 rescues short-term memory in psn-het B3 and psn-het I2 flies at 30 days and I) at 40 days of age.

Psn-het males for the following psn alleles (psn B3, psn C4, psn I2, psn K2, psn S3, psn C1, psn C2.1, psn EMS46, psn L(3)15, psn L(3)14) were aged for 30 days post eclosion. Mean CIs (+/− SEM) are plotted. Ns are indicated above each bar for all groups. The levels of significance are indicated (one asterisk * indicates p < 0.05, two asterisks ** indicate p < 0.01, three asterisks *** indicate p < 0.001). A) Naïve courtship levels remained robust and only slightly reduced from 5 day old psn-het flies (compare to Figure 1F) in all psn-het males. B–C) Psn-het males were tested for learning during training, and none of the psn-het males demonstrated at least a 40% reduction in courtship activity during the training session or a statistically significant reduction in courtship behavior. Therefore, psn-het flies have an age-dependent impairment in learning during training. D–E) Immediate recall remained intact in all psn-het males. F–G) Psn-hets for the psn B3 and psn I2 alleles were also tested for short-term memory at 60 minutes post training. Both lines failed to exhibit short-term memory, in contrast to their behavior at 5 days of age (Figure 1K–L).

A–E) Naïve courtship, learning during training and short-term memory were examined in WT flies at 5 days of age and at 45 days of age. Mean CIs (+/− SEM) are plotted, Ns are indicated above each bar for all groups. The levels of significance are indicated (one asterisk * indicates p < 0.05, two asterisks ** indicate p < 0.01, three asterisks *** indicate p < 0.001). WT flies are Ore R. A) Naïve courtship in 5 day-old WT (black bar) and 45 day-old WT (open bar) flies. B) Learning during training in 5 day-old WT flies. C) Learning during training in 45 day-old WT flies. D) Short-term memory in 5 day-old WT flies. E) Short-term memory in 45 day-old WT flies. Thus at 45 days of age there is no significant change in naïve courtship behavior or cognition in WT flies. The behavior of young adult (5 day-old) psn-het flies.
F–L) Naïve courtship, learning during training, immediate recall and short-term memory were examined in psn-het flies at 5 days of age. Psn-het males for the following psn alleles (psn B3, psn C4, psn I2, psn K2, psn S3, psn C1, psn C2.1, psn EMS46, psn L(3)15, psn L(3)14) were aged for 5 days post eclosion. F) Psn-het males were tested for naïve courtship, G–H) learning during training, and I–J) immediate recall (0–2 minutes post training). K–L) Psn-hets for the psn B3 and psn I2 alleles were also tested for short-term memory at 60 minutes post training. No deficits in behavior or cognitive capabilities were detected in any of the tests performed on the young adult psn-het flies.

A–C) Analysis of cell death in aged psn-het, Ore R and Adar mutant brains. TUNEL staining was performed on 50-day old (negative control) OreR A), 50-day old psn-het B), and 5-day old (positive control) Adar mutant brains C). The brains were examined by confocal microscopy to identify cells undergoing cell death. Foci containing at least one cell undergoing cell death are denoted by white arrows. Maximum projections, as shown in A–C, made from confocal stacks of entire brains were examined for cell death foci. The number of foci in the central brain region for each brain was scored. The distribution of the number of cell death foci versus the percentage of brains with such number of foci is shown in D). The majority of psn-het brains lacked any detectable cell death (n=41). Those that did had less than 5 foci and there was no consistent location in the brain where the foci were detected. Similar results were obtained with the Ore R brains (n=33). Cell death foci were detected in all Adar mutant brains (n=5), which have previously been described to undergo massive cell loss via cell death (Palladino et al., 2000). 50 day old Ore R E) and psn-het F) brains were stained with anti-FasII and examined by confocal microscopy to examine the integrity of the mushroom bodies. This antibody strongly labels the α-lobes (white arrow with *, shown in E), β-lobes (arrowhead shown in E) and weakly stains the γ-lobes (white arrow with #, shown in E). There was no detectable difference in the morphology of the mushroom bodies between Ore R (n=21) and psn-het (n=18) brains. G–P) Treatment with LY341495, MPEP, or lithium reverses the age-dependent impairments in learning during training and short-term memory observed in psn-het flies. Psn-hets for the psn B3 and psn I2 alleles were raised on control food, then at day 30 post eclosion were placed on food containing either 8.6 µM MPEP (G and L), 400 nM LY341495 (H and M), 5.0 mM lithium (I and N), 5.0 mM NaCl (J and O) or 0.004% DMSO (K and P) until day 44 and were tested on day 45. Mean CIs (+/− SEM) are plotted, Ns are indicated above each bar for all groups. The levels of significance are indicated (one asterisk * indicates p < 0.05, two asterisks ** indicate p < 0.01, three asterisks *** indicate p < 0.001). A triangle in indicates a 40% or greater reduction in courtship activity during the training session with a previously mated female (G–P). Psn-het flies that were treated with LY341495, MPEP or lithium demonstrated intact learning during training (G–I) and intact short-term memory (L–N) at 45 days of age. In contrast, psn-het flies that were on food containing NaCl or DMSO did not display learning during training (J and K) or short-term memory (O and P) at 45 days of age.

A diagram of the putative signaling in our AD model as well as that from the literature is shown. Arrows indicate activation or increases, and blunt ends indicate inhibition or decreases in activity or protein levels. The double line between wild type presenilin and InsP₃R indicates an interaction that is neutral or inhibitory. Previous studies have established that group I mGluRs in mammals activate the Gq pathway, while the group II and group III mGluRs activate the Gi signaling pathway. However, there is accumulating evidence that in mammals group II mGluRs may activate the Gq signaling pathway and induce LTD in a manner dependent on PLC and InsP₃R activity (Huang et al., 1997; Huang et al., 1999a; Huang et al., 1999b; Otani et al., 1999; Otani et al., 2002), and group I mGluRs are capable of activating Gi (Kreibich et al., 2004). We are outlining in this model the actions of DmGluRA in Drosophila, which is known to activate Gi activity and recent studies indicate may also activate Gq activity, just as is the case of Group II mGluRs in mammals (Pan et al., 2008). Upon activation, Gq activates PLC, which produces DAG and InsP₃. DAG and InsP₃ can activate PKC and the InsP₃R to release calcium from the endoplasmic reticulum. Gi alpha activation inhibits adenylate cyclase (AC), thereby preventing an increase in cAMP in response to synaptic stimulation, which prevents the activation of CREB by PKA. Phosphodiesterase 4 (PDE-4) breaks down cAMP levels. MPEP is a group I mGluR5 non-competitive antagonist, which we use at high concentrations to block DmGluRA. LY341495 is a very specific competitive antagonist of DmGluRA(Bogdanik et al., 2004). LiCl has two main actions at therapeutic levels in cells, one of which is the lowering of InsP₃ levels by inhibiting inositol monophosphatase and inositol polyphosphatase, thereby decreasing the synthesis of InsP₃, as well decreasing the rate at which it is recycled (Berridge, 1993; Takei et al., 1998; Williams et al., 2002). The second action of lithium is the inhibition of GSK-3B activity (Klein and Melton, 1996). It is through the aforementioned activities that we propose that these drugs may alter signaling in the Drosophila brain. B–G) Treatment WT control flies (OreR) with the DmGluRA antagonist LY341495, the putative DmGluRA antagonist MPEP, or lithium during aging results in impairment of learning during training and short-term memory. WT control flies were raised on control food, then after day 6 post eclosion were placed on food containing either 8.6 uM MPEP(MP) or vehicle (Veh) (B and E), 400 nM LY341495 (LY) or vehicle (Veh) (C and F), 5.0 mM lithium (Li) or 5.0 mM NaCl (Na) (D and G) until day 29 then were tested on day 30. H–Q) Treatment of psn-hets with the DmGluRA antagonist LY341495, the putative DmGluRA antagonist MPEP, or lithium prevents the age-dependent impairments in learning during training and short-term memory observed in psn-het flies. Psn-hets for the psn B3 and psn I2 alleles were raised on control food, then at day 6 post eclosion were placed on food containing either 8.6 µM MPEP (H and M), 400 nM LY341495 (I and N), 5.0 mM lithium (J and O), 5.0 mM NaCl (K and P) or 0.004% DMSO (L and Q) until day 29. Mean CIs (+/− SEM) are plotted. Ns are indicated above each bar for all groups. The levels of significance are indicated (one asterisk * indicates p < 0.05, two asterisks ** indicate p < 0.01, three asterisks *** indicate p < 0.001). A triangle in indicates a 40% or greater reduction in courtship activity during the training session with a previously mated female (H–L). OreR flies that were treated with MPEP, LY341495 and lithium lacked detectable learning during training (B–D) or short-term memory (E–G). OreR flies treated with DMSO or NaCl displayed normal learning during training (B–D) and short term memory (E–G). Psn-het flies that were treated with LY341495, MPEP, or lithium demonstrated intact learning during training (H–J) and intact short-term memory (M–O) at 30 days of age. In contrast, psn-het flies that were on food containing NaCl or DMSO did not display learning during training (K and L) or short-term memory (P and Q) at 30 days of age.