SEARCH

SEARCH BY CITATION

Keywords:

  • Drosophila;
  • memory consolidation;
  • olfactory memory;
  • rugose;
  • short-term memory

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  7. Acknowledgments

Extensive investigations show several molecular and neuroanatomical mechanisms underlying short-lived and long-lasting memory in Drosophila. At the molecular level, the genetic pathway of memory formation, which was obtained through mutant research, seems to occur sequentially. So far, studies of Drosophila mutants appear to support the idea that mutants defective in short-term memory (STM) are always associated with long-term memory (LTM) impairment. At the neuroanatomical level, distinct memory traces are partially independently distributed. However, whether memory phase dissociation also exists at the molecular level remains unclear. Here, we report on molecular separation of STM and consolidated memory through genetic dissection of rugose mutants. Mutants in the rugose gene, which encodes an evolutionarily conserved A-kinase anchor protein, show immediate memory defects as assayed through aversive olfactory conditioning. Intriguingly, two well-defined consolidated memory components, anesthesia-resistant memory and protein synthesis-dependent LTM, are both normal in spite of the defective immediate memory after 10-session massed and spaced training. Moreover, rugose genetically interacts with cyclic AMP-protein kinase A signaling during STM formation. Considering our previous study that AKAP Yu specifically participates in LTM formation, these results suggest that there exists a molecular level of memory phase dissociation with distinct AKAPs in Drosophila.

Memory formation occurs through several temporal phases. In Drosophila, behavioral, pharmacological and genetic manipulations have dissected olfactory memory formation into four components: short-lived forms including short-term memory (STM) and middle-term memory (MTM), whereas long-lasting forms including anesthesia-resistant memory (ARM) and protein synthesis-dependent long-term memory (LTM) (Margulies et al. 2005; Tully et al. 1994).

Over the past 30 years, extensive investigations have elucidated several mechanisms underlying short-lived and long-lasting memory in Drosophila. At the molecular level, the genetic pathway of memory formation, which was yielded through single-gene mutants that selectively disrupted individual memory phases, seems to occur sequentially. So far, studies of Drosophila mutants show that mutant disruption of STM always affects ‘downstream’ phases. For instance, mutations in rutabaga (adenylyl cyclase), dunce [cyclic AMP (cAMP) phosphodiesterase] and NF1 genes cause STM defect and subsequently LTM impairment (Blum et al. 2009; Dauwalder & Davis 1995; Guo et al. 2000; Ho et al. 2007). To date, no mutant that disrupts STM without affecting LTM has been identified.

At the neuroanatomical level, differential memory traces are at least partially stored in spatially distinct compartments. A prominent site involving memory formation is the mushroom body (MB), which consists of kenyon cells that can be classified into αβ, α′β′ and γ neurons (Crittenden et al. 1998; de Belle & Heisenberg 1994; Heisenberg et al. 1985). The MB γ neurons support STM trace, whereas αβ neurons support LTM trace (Blum et al. 2009; Trannoy et al. 2011). In addition, short-lived and long-lasting memory can occur in other neuroanatomical components, including antennal lobes, the ellipsoid body and DAL neurons (Ashraf et al. 2006; Chen et al. 2012; Thum et al. 2007; Wu et al. 2007), supporting the hypothesis of an independent neuronal pathways of these memory phases. The strategy of distributed memory traces has been well documented with a variety of species and is meaningful for survival (Thompson & Krupa 1994). However, whether the dissociation between the different memory phases at the neuroanatomical level also exists at the molecular level remains unclear.

This study of the rugose (rg) gene sheds light on this issue. The rg gene encodes an evolutionarily conserved A-kinase anchoring protein (DAKAP550), which is important in regulating eye development, synaptic architecture and brain morphology (Shamloula et al. 2002; Volders et al. 2012; Wech & Nagel 2005). Moreover, rg regulates both aversive and appetitive olfactory STM in the MB (Volders et al. 2012). We then made an interesting observation that rg mutants specifically disrupted STM, leaving two consolidated memories intact. Epistasis experiments indicated that rg interacted with protein kinase A (PKA) signaling pathway during STM formation. Considering our previous analysis of AKAP Yu (Lu et al. 2007), we proposed that there exists a molecular level of memory phase dissociation with distinct AKAPs.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  7. Acknowledgments

Fly stocks and maintenance

All flies, except for those that carried tubulin-Gal80ts (Gal80ts), were reared at 25°C with 60% humidity in a 12 h light–dark cycle on standard medium. rg1 was kindly provided by Dr Tadmiri R. Venkatesh. rgKG02343 and PKA-C1H2 were obtained from Bloomington Stock Center (Bloomington, IN, USA). elav; tubulin-Gal80ts was extant stock in our laboratory. To eliminate genetic background differences, all strains used in behavioral experiments were outcrossed with w1118 (isoCJ1) wild-type flies for five generations, except that rg1 mutant was outcrossed with balancers with wild-type genetic background.

Pavlovian aversive olfactory conditioning

Training and test procedures were performed at 25°C with 70% relative humidity, as described previously (Tully et al. 1994). During one-cycle training, around 100 flies were exposed sequentially to two aversive odors, 3-octanol (OCT; Fluka, St. Louis, MO, USA, 1.5 × 10−3 dilution in heavy mineral oil) and 4-methylcyclohexanol (MCH; Fluka, 1 × 10−3 dilution in heavy mineral oil), for 60 seconds with 45-second flush of fresh air after each odor. Flies received electric foot shock (twelve 1.5-second pulse of 60 V) in the presence of the first odor (CS+) but not the second (CS−). The massed training procedure consisted of 10 consecutive training cycles with no rest interval between them, while the spaced training protocol consisted of the same number of sessions separated by 15-min rest periods.

During the behavioral test, trained flies were allowed to choose between CS+ and CS− in a T-maze for 120 seconds. A performance index (PI) was calculated from the distribution of flies in the two T-maze arms. To eliminate odor bias, each experiment (n = 1) consisted of two reciprocal groups, with one trained to associate OCT with shock and the other to associate MCH with shock. The final PI was the average of PIs from the two groups.

Sensorimotor responses

To test olfactory acuity, odor-avoidance responses were quantified by exposing groups of 100 untrained flies to the test odor (either OCT or MCH, 1.5 × 10−3 and 1 × 10−3 dilutions, respectively) vs. fresh air in the T-maze. To test shock reactivity, groups of 100 untrained flies were exposed to two T-maze arms with 60 V electric foot shock delivered to one of the arms but not the other. In both tests, flies were allowed to make a choice between the two arms for 120 seconds and PI was calculated as described in the memory test.

Cold amnesia

The procedures were performed as previously described (Tully et al. 1994). Briefly, flies were transferred to empty vials and cooled in ice water (0°C) for 2 min, which halted locomotor activity immediately, and were then allowed to recover in fresh food vials at room temperature (25°C) for 1 h before they were subjected to the memory test.

Generation of rg RNAi transgenic flies

To generate the UAS-rgRNAi construct, the target cDNA sequences (∼600 bp) were amplified from the genomic DNA of w1118 (isoCJ1) wild-type flies with primers containing unique restriction sites: 5′-aaaatctagagcgctggccgttagagagata-3′ and 5′-aaaatctagagggggctggcgatgttt-3′. The fragment was carefully chosen and no additional genes were targeted based on an established algorithm (Flockhart et al. 2006). This fragment was subcloned into the pWIZ vector (Lee & Carthew 2003) and germ-line transformation was performed as previously described (Rubin & Spradling 1982). The UAS-rgRNAi transgenic flies were backcrossed with w1118 (isoCJ1) flies for five generations to equilibrate the genetic background.

Heat shock regimen

Flies carrying the tubulin-Gal80ts were raised and maintained in 18°C to minimize ‘leaky’ expression in development. Progenies (2–5 days old) were collected and divided into two groups. The induced group was transferred to a 30°C incubator for 3 days, whereas the uninduced control group was kept at 18°C. Both groups were allowed to recover at 25°C for at least 1 h before behavioral experiments.

mRNA quantification

Total RNA was isolated from fly heads using the classic TRIzol method (Invitrogen, CA, USA). cDNA was synthesized directly from the mRNA with the SuperScript first-strand synthesis Kit (Invitrogen Carlsbad, CA, USA). Relative rg transcript abundance was quantified using SYBR Green reagents (Tiangen Biotech, Beijing, China) on icycler iQ5 real-time polymerase chain reaction (PCR) machine (Bio-Rad, Berkeley, CA, USA). The sequences of primers were as follows: rg, 5′-atcgcgcgacagtagcat-3′ and 5′-aactacgataagggcaatgtgg-3′; internal control of the RpL32, 5′-gctaagctgtcgcacaaatgg-3′ and 5′-cggcgacgcactctgtt-3′ (Volders et al. 2012). rg expression levels were first normalized to the loading control RpL32 and then to that of the control group. Data analysis was performed from four independent samples.

Statistics

All data are presented as means ± SEM and analyzed by Student's t-tests or analysis of variance (anova) followed by Tukey's post hoc test in SPSS11.0 (SPSS, Chicago, IL, USA). Asterisks indicate statistically critical value (*P < 0.05 and **P < 0.01); ‘n.s.’ denotes no significant differences (P > 0.05).

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  7. Acknowledgments

We have previously constructed and screened 2021 homozygous P{lacW} transposant lines with defective aversive LTM after spaced training. One identified mutant, AKAP Yu, specifically impaired LTM in the MB (Lu et al. 2007). This finding arouses us to explore the role of the AKAP family during memory formation. Analysis of another AKAP-encoding gene, rg, showed that this AKAP protein contributes to STM, rather than LTM formation.

STM is impaired in rg mutants

Two rg hypomorphic mutants were used in our study: rg1, which is a spontaneous mutant (Lindsley & Zimm 1992; Shamloula et al. 2002), and rgKG02343, which is caused by P {SUP or P} insertion in the 3′-terminal of the rg locus (Fig. 1a). Mutation site of rg1 has not been molecularly characterized; however, multiple new rg alleles are isolated on the basis of their failure to complement the rg1 allele (Shamloula et al. 2002). Quantitative PCR verified that rg transcript in the adult heads was significantly reduced to approximately 30% of that of wild-type flies in both rg1 and rgKG02343 mutants (Fig. 1b) (anova, F2,9 = 58.13, P < 0.001, n = 4).

image

Figure 1. STM is impaired in rg mutants. (a) The gene structure of rg. Black boxes indicate exons. rgKG02343 has a P element insertion in 3′-terminal of the rg gene, and rg1 is spontaneous mutant, whose mutation site has not been molecularly indicated. (b) Quantitative PCR showed that the rg transcript in two mutants was decreased to about 30% of that of wild-type flies (P < 0.001, n = 4, anova). (c) rg mutants displayed significant memory defect at 3 min, 30 min and 3 h after one-cycle training (P < 0.01, n ≥ 6, anova). (d) Genetic complementation analysis showed that the rg1 and rgKG02343 alleles were recessive to wild-type, and rg1/rgKG02343 double heterozygote was non-complement for defective STM, suggesting that rg is essential for STM formation (P < 0.001, n ≥ 6, anova). (e) The diminished memory by cold-amnesia treatment was reduced to the similar level in rg mutants and wild-type flies, indicating that the defective memory in rg mutants is part of labile memory (P = 0.015 for no cold, P = 0.17 for cold shock, n ≥ 11, anova). Average data are presented as means ± SEM. *P < 0.05; **P < 0.01.

Download figure to PowerPoint

Next, we tested memory at multiple time points after the well-established aversive olfactory conditioning (Tully & Quinn 1985). Two mutants exhibited significant memory impairment relative to wild-type flies at 3 min, 30 min and 3 h after one-cycle training (Fig. 1c) (anova, F2,35 = 32.52, F2,17 = 42.86, F2,26 = 7.86 for 3 min, 30 min and 3 h, respectively, P < 0.01, n ≥ 6). This phenotype did not result from abnormal shock or odor avoidance that is essential for the task (Table 1) (P > 0.41 for olfactory acuity and P > 0.84 for shock avoidance, n ≥ 6, anova). Furthermore, genetic complementation analysis using rg1 and rgKG02343, suggesting that two mutants failed to complement each other for STM defect, also confirmed that rg is essential for STM (Fig. 1d) (anova, F2,23 = 41.61, P < 0.001, n ≥ 6).

Table 1. Sensorimotor responses in Pavlovian olfactory learning (related to Figs 1-3)
GenotypeOlfactory acuityShock avoidance (60 V)
OCT (1.5 × 10−3)MCH (1 × 10−3)
  1. All flies showed similar odor acuities and electric shock sensitivity, P > 0.41 for olfactory acuity, P > 0.84 for shock avoidance. Data are shown as means ± SEM, anova, n ≥ 6.

W1118(isoCJ1)44.7 ± 3.145.5 ± 2.965.0 ± 2.1
rg141.1 ± 3.241.2 ± 3.362.0 ± 2.3
rgKG0234338.6 ± 2.539.9 ± 2.865.2 ± 3.7

To determine the nature of the defective memory, we applied a cold shock treatment, which should eliminate anesthesia-sensitive early memory forms (STM and MTM), leaving only the ARM component of 3-h memory (Quinn & Dudai 1976). As seen in Fig. 1e, both rg1 and rgKG02343 flies had 3-h memory deficit in the absence of cold shock, indicating that at least one memory component of 3-h memory was decreased (F2,31 = 4.75, P = 0.015; n ≥ 11, anova). When cold shock was applied at the second hour after one-cycle training, both mutants showed a further reduced 3-h memory. More importantly, the diminished memory in both wild-type flies and in rg mutants was reduced to the similar level (Fig. 1e) (F2,30 = 0.17, P = 0.83, n = 11, anova), indicating that the defective memory in rg mutants is part of labile memory.

Consolidated memory remains intact in rg mutants

We next examined two well-defined consolidated memories, ARM and LTM, which were induced by massed training or spaced training, respectively (Tully et al. 1994). Interestingly, 24-h memory was normal in rg mutants subjected to massed or spaced training (Fig. 2a) (anova, F2,18 = 1.16, P = 0.33 for ARM; F2,19 = 1.39, P = 0.27 for LTM, n ≥ 6). This phenotype was further confirmed using the genetic complementation experiment. As expected, single heterozygotes (rg1/+ and rgKG02343/+) and double heterozygote (rg1/rgKG02343) all showed normal 24-h memory after spaced training (Fig. 2b) (anova, F3,21 = 0.82, P = 0.49, n ≥ 6). Taken together, these results suggested that rg mutants have defective STM but normal consolidated memory.

image

Figure 2. Consolidated memory remains intact in rg mutants. (a) rg mutants displayed normal 24-h memory after massed training (P = 0.33, n ≥ 6, anova) or spaced training (P = 0.27, n ≥ 6, anova). (b) Genetic complementation analysis showed that both double heterozygote (rg1/rgKG02343) and single heterozygotes had normal LTM after spaced training (P = 0.49, n ≥ 6, anova). Average data are presented as means ± SEM. *P < 0.05; **P < 0.01.

Download figure to PowerPoint

Intact consolidated memory is not a result of the differences in training schedules

In aversive olfactory conditioning, STM and consolidated memory are elicited by different training schedules (Tully et al. 1994). To exclude the possibility that intact consolidated memory was a result of the differences between training paradigms, we detected immediate memory after massed training. Similar to memory after one-cycle training, immediate memory in rg1 mutant was lower than in wild-type flies after massed training (Fig. 3a) (t-test, t10 = 6.03, P < 0.001, n = 6). Furthermore, we compared memory retention curves at various time points after spaced training (Fig. 3b). rg1 mutant exhibited significant memory decay in the first 3 h after spaced training (at 3 min and 3 h), but had normal memory at later time points from 6 h up to 24 h (Fig. 3b) (t-test, t10 = 5.18, 4.06, 0.67, 0.44 and P = 0.001, 0.002, 0.52, 0.67 for 3 min, 3 h, 6 h, 24 h, respectively, n = 6.). Because anesthesia-sensitive memory decays rapidly, the remaining predominant memory phases after spaced training are ARM and LTM. The unaltered memory in the later time points suggested that the rugose-dependent STM component is separated from consolidated memory.

image

Figure 3. Intact consolidated memory is not a result of the differences between training schedules. (a) rg1 mutant exhibited immediate memory deficit after massed training (P < 0.001, n = 6, t-test). (b) Retention curves were generated at various time points after spaced training. rg1 mutant exhibited significant memory decay in the first 3 h but showed normal memory at later time points (P = 0.001, 0.002, 0.52 and 0.67 for 3 min, 3 h, 6 h and 24 h, respectively, n = 6, t-test). Average data are presented as means ± SEM. *P < 0.05; **P < 0.01.

Download figure to PowerPoint

Endogenous Rg plays a physiological role during STM formation

A previous study reported that rg regulates the MB morphology (Volders et al. 2012). Therefore, to rule out the possibility that the observed memory phenotype was resulting from abnormal development of the nervous system, we used the Gal4/Gal80ts system to acutely manipulate rg RNAi expression in adult flies (Brand & Perrimon 1993; McGuire et al. 2003). Using this system, the Gal4-induced expression is suppressed by a ubiquitously expressed Gal80ts protein at the permissive temperature (18°C), but not at the restrictive temperature (30°C). UAS-rgRNAi driven by pan-neuronal Gal4 effectively attenuated rg mRNA expression to 34% of that of wild-type flies (elav/+; UAS-rgRNAi/+) (Fig. 4a) (t-test, t6 = 14.01, P < 0.001, n = 4). Then, immediate memory was impaired by knocking down Rg expression during adulthood in nervous system (elav/+; UAS-rgRNAi/+; tubulin-Gal80ts/+, 30°C), whereas LTM remained unaltered, which phenocopied the mutants (Fig. 4b,c) (anova, F2,15 = 21.95, P < 0.001 for STM; F2,15 = 0.38, P = 0.68 for LTM, n = 6). Consistently, no significant difference was detected for uninduced groups kept at the permissive temperature (Fig. 4b,c) (anova, F2,16 = 0.89, P = 0.43 for STM; F2,15 = 0.15, P = 0.86 for LTM, n ≥ 6). Task-related sensorimotor responses were not significantly altered in all lines, no matter whether at the restrictive temperature or at the permissive temperature (Table 2) (P > 0.48 for olfactory acuity, P > 0.88 for shock avoidance, anova, n ≥ 6). The above results suggested that the dissociation of rugose-relevant STM component from consolidated memory is independent of developmental function.

image

Figure 4. Endogenous Rg plays a physiological role during STM formation. (a) Quantitative PCR showed that rg expression in elav/+; UAS-rgRNAi/+ flies was diminished to 34% of that of elav/+ control group (P < 0.001, n = 4, t-test). (b) Adulthood knocking down Rg expression by rg RNAi in nervous system impaired STM (elav/+; UAS-rgRNAi/+; tubulin-Gal80ts/+ compared with control groups, P = 0.43 for uninduced, P < 0.001 for induced, n ≥ 6, anova). (c) Adulthood knocking down Rg expression by rg RNAi in nervous system had normal LTM after spaced training (P = 0.86 for uninduced, P = 0.68 for induced, n = 6, anova). Average data are presented as means ± SEM. *P < 0.05; **P < 0.01.

Download figure to PowerPoint

Table 2. Sensorimotor responses in stains used in Fig 4
GenotypeTemperature (°C)Olfactory acuityShock avoidance (60 V)
OCT (1.5 × 10−3)MCH (1 × 10−3)
  1. All flies showed similar odor acuities and electric shock sensitivity. elav/+; UAS-rgRNAi/+; tubulin-Gal80ts/+ compared with parental controls, P > 0.48 for olfactory acuity, P > 0.88 for shock avoidance. Data are shown as means ± SEM, anova, n ≥ 6.

elav/+; tubulin-Gal80ts/+3039.2 ± 2.843.1 ± 3.266.3 ± 3.8
1837.9 ± 2.739.6 ± 4.469.0 ± 3.9
UAS-rgRNAi/+3037.4 ± 3.145.4 ± 4.264.4 ± 2.1
1840.0 ± 4.142.2 ± 3.866.6 ± 2.8
elav/+; UAS-rgRNAi/+; tubulin-Gal80ts/+3038.3 ± 2.843.1 ± 3.266.3 ± 3.8
1838.7 ± 3.541.2 ± 2.563.9 ± 3.3

rg genetically interacts with the cAMP-PKA pathway during STM formation

Rg belongs to the AKAP family, which determines the subcellular localization of PKA and thereby spatiotemporally restricts cAMP signaling (Wong & Scott 2004). Because the cAMP-PKA signaling cascade plays a crucial role in mediating learning and memory processes (Davis et al. 1995), we hypothesized the existence of an epistatic relationship between rg and cAMP-PKA pathway during STM formation. PKA-C1H2 is a strong hypomorph, carrying a point mutation in the third exon of the catalytic subunit of PKA (Lane & Kalderon 1993). As reported and finding in our data, PKA-C1H2 heterozygote (PKA-C1H2/+), which has a 40% decrease in PKA activity, has no effect on immediate memory and LTM (Horiuchi et al. 2008) (Fig. 5a,b). However, the rg1/+; PKA-C1H2/+ double-heterozygous flies showed STM defect after one-cycle training, but had normal LTM subjected to space training (Fig. 5a,b) (anova, F3,28 = 6.28, P = 0.002 for STM, n ≥ 7; F3,24 = 0.43, P = 0.72 for LTM, n = 7). These observations suggested that rg genetically functions with the cAMP-PKA pathway during STM formation.

image

Figure 5. rg genetically interacts with cAMP-PKA pathway during STM formation. (a) The double heterozygote (rg1/+; PKA-C1H2/+) showed significant STM deficits after one-cycle training, whereas the single heterozygote (rg1/+ or PKA-C1H2/+) showed normal memory (P = 0.002, n ≥ 7, anova), suggesting that rg genetically interacts with PKA pathway. (b) The double heterozygote (rg1/+; PKA-C1H2/+) and the single heterozygote (rg1/+ or PKA-C1H2/+) all showed normal 24-h memory after spaced training (P = 0.72, n = 7, anova). Average data are presented as means ± SEM. *P < 0.05; **P < 0.01.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  7. Acknowledgments

Here, we reported that STM was separated from consolidated memory through genetic dissection of rg mutants at the molecular level, suggesting that divergent memory formations are mediated by functionally distinct AKAPs.

Molecular separation of STM and consolidated memory

Impressive progress in Drosophila memory research at the neuroanatomical level has found multiple memory traces, which form in the olfactory nervous system after conditioning, acting parallel to distinct neuronal cell types, whereas studies at the molecular level have showed that distinct memory phases appear to occur sequentially. To elucidate the logic of how the nervous system organizes different memory phases, it is meaningful to investigate whether the strategy of distributed memory trace also exists at the molecular level.

Our genetic dissection of rg mutants confirmed this hypothesis. First, rg mutant showed immediate memory defects as assayed through aversive olfactory conditioning. Intriguingly, two well-defined consolidated memory components, ARM and LTM, were both normal. Second, two consolidated memory components were both normal, while the defect of STM persisted with 10-cycle massed and spaced training, respectively. These results rule out the possibility that this phenotype was induced by different training schedules. However, without using rg null allele, we could not rule out the possibility that rg is also required for LTM or ARM, but the STM has higher requirement. However, it was hard to get a clear conclusion through null mutant because rg null allele has severe abnormal MB morphology and immediate memory is close to zero in aversive and appetitive olfactory conditioning (Volders et al. 2012). We took advantage of weak hypomorphic mutants (rg1 and rgKG02343), which yielded no detectable structural defects of MB (Volders et al. 2012 and our unpublished data), to show contributions to memory function, but not development. Furthermore, to rule out the developmental role, we used RNAi to knock down rg expression at adult stage, which phenocopied the mutants. Both findings indicated that rg plays a physiological role in olfactory memory independent of the MB development. Certainly, a more effective RNAi tool, which almost knocked out the rg expression, will help solve this issue entirely in the future.

It is widely accepted that LTM is formed through the consolidation process (Dudai 2004; McGaugh 2000), whereas it is still in debate whether LTM is formed sequentially after STM formation (Trannoy et al. 2011). Therefore, two explanations may account for the relationship between rugose-dependent STM and LTM. If STM and LTM are formed and processed parallel, our findings indicated that rg gene is only involved in STM formation, not in LTM formation. If STM and LTM are sequential processes and that LTM formation is built on the short-term trace, our findings suggested that rugose-dependent STM is not further consolidated to form LTM. Taken together, our study indicated that the dissociation between STM and consolidated memory exists both at the molecular and neuroanatomical levels, which will improve the current understanding of the relationship between short-lived and long-lasting memory.

Different AKAPs organize distinct memory

The cAMP-PKA signaling is essential for memory formation across various species (Davis 2005; Kandel 2001). In Drosophila olfactory memory, mutants influencing different steps of the PKA pathway have been identified to affect distinct memory phases, including STM (rutabaga and dunce), MTM (DC0X4), ARM (DC0B3) and LTM (CREB) (Dudai et al. 1976; Horiuchi et al. 2008; Li et al. 1996; Livingstone et al. 1984; Yin et al. 1994). Therefore, the complex temporal requirements of PKA pathway during memory processing raise a question about which molecular mechanism spatiotemporally restricts this pathway.

Our study showed that the rg gene, which encodes AKAP550, genetically interacted with the PKA pathway during STM formation, but not in later temporal phases. Given our previous study that AKAP Yu also genetically functions with cAMP-PKA signaling specifically during LTM formation (Lu et al. 2007), we predicted that two isoforms of the AKAP family play distinct roles in STM or LTM formation. However, the exact downstream mechanism remains unclear. One explanation is organelle- or tissue-specific AKAPs spatiotemporally restrict the PKA subcellular or cellular targeting, thus initiating downstream pathway during different memory phases.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  7. Acknowledgments

Acknowledgments

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  7. Acknowledgments

We are grateful for comments and discussion from Lei Wang, Zhiyong Xie, Shannon Zhao and Yunchuan Zhang. We thank Hui-Hao Lin and Ying-Hsiu Chen for invaluable technical assistance. We also thank Dr Tadmiri R. Venkatesh (City University of New York) and the Bloomington Stock Center for providing stocks. This work was supported by grants from the National Basic Research Project (973 program) of the Ministry of Science and Technology of China (2006CB500806 and 2009CB941301) and the Tsinghua University Initiative Scientific Research Program (20111080956, all to Y.Z.).