DHHC2 regulates fear memory formation, LTP, and AKAP150 signaling in the hippocampus

Summary

Palmitoyl acyltransferases (PATs) have been suggested to be involved in learning and memory. However, the underlying mechanisms have not yet been fully elucidated. Here, we found that the activity of DHHC2 was upregulated in the hippocampus after fear conditioning, and DHHC2 knockdown impaired fear induced memory and long-term potentiation (LTP). Additionally, the activity of DHHC2 and its synaptic expression were increased after high frequency stimulation (HFS) or glycine treatment. Importantly, fear learning selectively augmented the palmitoylation level of AKAP150, not PSD-95, and this effect was abolished by DHHC2 knockdown. Furthermore, 2-bromopalmitic acid (2-BP), a palmitoylation inhibitor, attenuated the increased palmitoylation level of AKAP150 and the interaction between AKAP150 and PSD-95 induced by HFS. Lastly, DHHC2 knockdown reduced the phosphorylation level of GluA1 at Ser845, and also induced an impairment of LTP in the hippocampus. Our results suggest that DHHC2 plays a critical role in regulating fear memory via AKAP150 signaling.

Graphical abstract

Highlights

• •Augmentation of DHHC2 activity in the hippocampus mediated fear memory formation
• •DHHC2 preferentially palmitoylated AKAP150, not PSD-95, in response to fear learning
• •Treating with 2-BP abolished the AKAP150 signaling in the hippocampus
• •DHHC2 is required for AMPARs delivery during fear learning and LTP

Biochemistry; Molecular biology; Neuroscience

Introduction

Palmitoylation, a reversible lipid modification, is the post-translational addition of saturated 16-carbon palmitic acid to cysteine residue through formation of a labile thioester linkage.^1,2 Palmitoylation allows for rapid structural modulation, localization, trafficking, membrane association, and cellular signaling transduction of synaptic proteins.^3,4 A large amount of evidence shows that palmitoylation might be involved in multiple physiology and pathological process,⁵ such as X-linked intellectual disability and 22q11.2 deletion syndrome,⁶ and our previous results revealed that palmitoylation is required for fear memory formation and the anxiolytic action of benzodiazepine.^7,8

There are 23 mammalian palmitoyl acyltransferases (PATs) which have a common aspartate-histidine-histidine-cysteine domain (DHHC proteins),⁹ such as DHHC2, -3, and -5, are widely distributed in the brain, and each of DHHCs shows substrate specificity.^10,11 Additionally, some studies noted that knockout of PATs leads to a deficiency in prepulse inhibition, the knockout mice display a decrease in exploratory activity in a new environment,¹² and PATs activity is necessary to modify neural plasticity, a mechanism required for memory acquisition and consolidation.¹³ However, direct evidence about whether PATs mediates palmitoylation of synaptic proteins in fear memory is lacking.

A-kinase anchoring protein 79/150 (AKAP79/150, human 79/rodent 150), one member of AKAPs family, is widely distributed in the central nervous system.¹⁴ In recent years, accumulating studies have shown that AKAP150 plays an important role in learning, memory, and synaptic plasticity.^{15,16,17,18,19} Particularly, as an anchoring protein, AKAP150 targets protein kinase A (PKA), PKC, and calcineurin (CaN) to the postsynaptic membrane to regulate GluA1 phosphorylation, trafficking, and activity associated with long-term potentiation (LTP) and depression (LTD).^20,21 In addition, infusion to the lateral amygdala of a peptide St-Ht31, which blocks PKA anchoring onto AKAPs, impairs memory consolidation in cue fear conditioning,¹⁵ and our previous result also revealed that this peptide reversed the depression-like behaviors induced by chronic stress.²² However, the role of AKAP150 in synaptic plasticity related to memory, especially in LTP, is still far from fully known.

Previous reports revealed that palmitoylation of AKAP150 targets it to recycling endosomes in dendrites, and promotes the delivery of AMPARs to the plasma membrane and the maintenance of synaptic transmission in hippocampal neurons.^23,24,25,26 Importantly, intrahippocampal injection of peptides disrupting PKA-anchoring impairs consolidation and facilitates extinction of contextual fear memories.²⁷ Furthermore, CaMKII-mediated de-palmitoylation of AKAP150 is required for both AKAP79 removal and structural LTD.²⁸ Thus, we speculated that AKAP150 palmitoylation plays a critical role in LTP/LTD plasticity. However, most of researches on the mechanism of this is mainly focused on cultured hippocampal neurons, in vivo study is rare.

In the present study, using a multifaceted approach involving acyl-biotin exchange assay (ABE), western blotting, synaptic fractionation, co-immunoprecipitation, electrophysiology and chemogenetic manipulation, in combination with glycine-induced chemical LTP (cLTP) and high frequency stimulation (HFS)-induced LTP assays, we explored mechanisms underlying DHHC2 in regulating fear memory. Firstly, the phenotype of DHHC2 in fear memory was explored, and the role of DHHC2 in LTP induced by in vitro treatment (HFS or glycine) also was assessed. Next, given that AKAP150 and PSD-95 were the substrates of DHHC2, we examined the effect of fear learning and HFS on AKAP150 and PSD-95 palmitoylation in the hippocampus. Lastly, we verified the function of DHHC2 mediated AKAP150 signaling in LTP and fear memory by chemogenetic knockdown and palmitoylation inhibitor. In summary, these results expand our understanding of DHHC2 underlying fear memory and LTP in the hippocampus, and provide new targets for research focused on learning and memory.

Results

DHHC2 is required for fear memory formation

DHHC2, a member of PATs family, is highly expressed in the hippocampus and controls palmitoylation of many neuronal proteins.²⁹ However, the role of DHHC2 in fear memory and related hippocampal synaptic plasticity is still unknown. To explore this, we performed contextual fear conditioning, and collected the hippocampus tissue to detect its expression (Figure 1A), while western blotting results showed that there is no significant change in DHHC2 expression (Figure 1B). Additionally, according to previous study, DHHC2 activity, which detected by its auto-palmitoylation,³⁰ thus we performed ABE assay and found there was a significant increase in the hippocampus after fear conditioning (Figure 1B).

Figure 1.

DHHC2 mediates fear memory formation in the hippocampus

(A) The contextual fear conditioning and ABE test procedure.

(B) Representative western blots and statistical analysis of DHHC2 expression and its palmitoylation in the hippocampus after fear conditioning (n = 4 per group).

(C) The schematic timeline of the experimental procedures.

(D) Schematic representation for the construct of LV-shDHHC2.

(E) Stereotaxic injection and immunofluorescence of coronal brain sections from hippocampus.

(F) Representative western blotting and quantification of DHHC2 expression from LV-eGFP and LV-shDHHC2 groups (n = 7 per group).

(G) Time-course of LTP induced by HFS in hippocampal slices from the FC + LV-eGFP/FC + LV-shDHHC2 groups after fear conditioning, and statistical histogram showing LTP magnitude averaged from the last 15 min of recordings from the two groups (n = 4 per group).

(H) The freezing behavior of mice tested on day 3 after DHHC2 knockdown (n = 5 per group).

All statistical comparisons were performed via Student’s t tests. Data represented are means ± SEM. ∗p < 0.05 and ∗∗∗p < 0.001.

To further confirm the role of DHHC2 in fear memory formation, we constructed lentiviral vector-mediated shRNA to knockdown DHHC2 in the hippocampus and designed a series of experiments (Figures 1C and 1D). We then performed bilateral hippocampus injections of LV-shDHHC2 or LV-eGFP and identified its local transfection by immunofluorescence (Figure 1E). Additionally, as the results showed, DHHC2 was appreciably diminished in LV-shDHHC2 group compared with LV-eGFP group (Figure 1F). Next, we trained the mice after four weeks, as we expected, the results of LTP recording showed that there was obvious impairment in the slope of DHHC2 knockdown mice (Figure 1G). Importantly, fear memory of DHHC2 knockdown mice was also impaired in the testing on day 3 (Figure 1H). Together, these results indicate that DHHC2 mediated fear memory formation by regulating related LTP in the hippocampus.

DHHC2 activity is increased during LTP in vitro

Our results suggest that DHHC2 may play an important role in hippocampal LTP. To further confirm the findings described above, HFS and chemical drugs such as glycine, have been widely used to induce LTP to explore the underlying mechanisms of synaptic activity.³¹ We treated the hippocampal slices of mice with glycine for 5 min to induce the cLTP, and detected the global palmitoylation at 30, 60 and 90 min after glycine stimulation. As the results showed, significant enhancements of global palmitoylation occurred not only at 30 min after glycine treatment, but also at 60 and 90 min, indicating that synaptic protein was stably palmitoylated when neurons were activated (Figure 2A). To confirm whether the LTP induced by HFS leads to similar changes, a standard protocol of HFS was applied for reliable LTP, and the global palmitoylation was examined by ABE assay at 30, 60 and 90 min after synaptic activation. Similar to cLTP, the palmitoylation of global proteins was increased significantly, peaked at 30 min, and maintained up to 90 min after LTP induction (Figure 2B). In summary, these results indicate that acute increase in neuronal activity enhances global palmitoylation.

Figure 2.

DHHC2 activity is increased after LTP induction in the hippocampus

(A and B) Western blots of global palmitoylation at different time points after glycine (A) or HFS (B) treatment.

(C and D) Representative western blots and quantification of DHHC2 expression, its palmitoylation (C) and its synaptic expression (D) in the hippocampus at 30 min after glycine treatment (n = 6 per group).

(E and F) Representative western blots and quantification of DHHC2 expression, its palmitoylation (E) and its synaptic expression (F) in the hippocampus at 30 min after HFS (n = 4 per group).

All statistical comparisons were performed via Student’s t tests. Data represented are means ± SEM. ∗p < 0.05 and ∗∗p < 0.01.

Next, to further evaluate the possible role of DHHC2 in a more intact system, we detected the total protein level and palmitoylation of DHHC2 after glycine treated slices. We found that the expression of DHHC2 after glycine treatment was unchanged compared with the Veh group, suggesting that cLTP did not affect DHHC2 protein expression. However, via ABE assay, we observed a remarkable increase in the palmitoylation level of DHHC2 after glycine treatment (Figure 2C). Furthermore, combining synaptic fractionation and western blotting assay, synaptic expression level of DHHC2 was detected, and we found that cLTP induced a significant increase of DHHC2 synaptic expression (Figure 2D). Similar to cLTP, analogous results were observed in LTP induced by HFS (Figures 2E and 2F). Together, these results suggest that LTP did not affect total DHHC2 expression, but enhanced DHHC2 activity and synaptic expression by promoting its auto-palmitoylation.

DHHC2 selectively palmitoylated AKAP150 in response to fear learning or HFS

Given that AKAP150 and PSD-95 are the main substrate of DHHC2,²⁵ we wondered which one mediated the alteration of synaptic plasticity in response to fear learning or HFS. We carried out ABE assay and found that there was a significant increase in AKAP150 palmitoylation level, without effects on its expression after fear conditioning (Figure 3A), while an elevating tendency on PSD-95 expression, not its palmitoylation, was detected in the hippocampus after fear conditioning (Figure 3B). Further, to examine whether acute increase of neuronal activity regulates AKAP150 and PSD-95 palmitoylation, we treated hippocampal slices with HFS, and evaluated their palmitoylation levels by ABE assay. Interestingly, similar to fear learning, the palmitoylation level of AKAP150, not PSD-95, was increased significantly after LTP induction, and HFS did not affect their total expression (Figures 3C and 3D). These results revealed that AKAP150 is the target of DHHC2 in response to fear conditioning or HFS.

Figure 3.

DHHC2 preferentially palmitoylated AKAP150 in response to fear learning or HFS

(A and B) Representative western blots and statistical results of AKAP150 (A) and PSD-95 (B) expression, and its palmitoylation level in the hippocampus after fear conditioning (n = 6 per group).

(C and D) Representative western blots and statistical results of AKAP150 (C) and PSD-95 (D) expression, and its palmitoylation level in the hippocampus after HFS (n = 6 per group).

(E) Representative western blots and statistical results of AKAP150 expression and its palmitoylation level in the hippocampus after DHHC2 knockdown and/or fear conditioning treatment (n = 4 per group).

Student’s t test for (A–D), and two-way ANOVA for (E). Data represented are means ± SEM. ∗∗p < 0.01 and ∗∗∗p < 0.001, and ^###p < 0.001.

To further confirm the findings described above, we microinjected LV-shDHHC2 into the hippocampus, and trained the mice after four weeks. As we expected, the increase in AKAP150 palmitoylation induced by fear conditioning was reversed by silencing DHHC2 in the hippocampus, and without little effects on AKAP150 expression (Figure 3E). Taken together, these results demonstrated that DHHC2-mediated AKAP150 palmitoylation in response to fear learning or HFS.

Palmitoylation promotes AKAP150 signaling during LTP

To further evaluate whether palmitoylation promotes AKAP150 signaling during LTP, we used 2-BP, a protein palmitoylation inhibitor,³² to incubate the hippocampal slices for 30 min, and examined AKAP150 palmitoylation and synaptic expression after HFS and 2-BP treatment. As the data showed, a significant increase in AKAP150 palmitoylation level was observed in HFS group compared to Veh group, and this was reversed by 2-BP (Figure 4A). Additionally, we assessed the synaptic expression of AKAP150 in the hippocampus after LTP induction by synaptic fractionation assay, and observed there was a significant elevation in synaptic expression of AKAP150 after HFS (Figure 4B), indicating that HFS promotes the translocation of AKAP150 by palmitoylation.

Figure 4.

HFS facilitates AKAP150 signaling by palmitoylation

(A and B) Representative western blots and quantification of AKAP150 palmitoylation (A) and its synaptic expression (B) in the hippocampus after HFS and/or 2-BP treatment (n = 4 per group).

(C) Representative western blots and statistical results of co-immunoprecipitation between PSD-95 and AKAP150 in the hippocampus after HFS and/or 2-BP treatment (n = 4 per group).

(D and E) Representative western blots and quantification of CaN (D) and PKA (E) synaptic expression in the hippocampus after HFS and/or 2-BP treatment (n = 4 per group).

(F) Representative western blots and statistical results of co-immunoprecipitation between AKAP150 and PKA in the hippocampus after HFS and/or 2-BP treatment (n = 4 per group).

All statistical comparisons were performed via One-way ANOVA. Data represented are means ± SEM. ∗p < 0.05 and ∗∗p < 0.01, ^#p < 0.05 and ^##p < 0.01.

PSD-95 is a key scaffold protein that contains PDZ domains and locates close to the postsynaptic membrane, and AKAP150 can interact with PSD-95 to regulate synaptic plasticity.^33,34 Interestingly, our results showed that the synaptic expression of AKAP150 was increased significantly after HFS. We hypothesized that the interaction of AKAP150 and PSD-95 might be upregulated in LTP, and used co-immunoprecipitation assay to detect the interaction. The results identified that the association between AKAP150 and PSD-95 was increased after HFS (Figure 4C). The above findings indicate that synaptic expression of AKAP150 is palmitoylation-dependent. Thus, we next assessed whether 2-BP treatment can reduce the interaction between AKAP150 and PSD-95. Strikingly, 2-BP treatment decreased the HFS-induced association between AKAP150 and PSD-95 remarkably (Figure 4C). Together, these results demonstrated that neuronal activation enhances the synaptic expression of AKAP150, and the activity-dependent changes can be reserved by de-palmitoylation.

Considering that the interactions of AKAP150 with CaN and PKA mediate the processes of LTD and LTP,^17,20,21,26 respectively, the enhancement of neuronal activity may regulate the expression of CaN or PKA.³⁵ To confirm this, HFS and 2-BP were used to examine how LTP impacts the total protein level and synaptic expression of CaN and PKA in the hippocampal slices. As our results showed, there was little difference in the total protein level or synaptic expression of CaN before and after HFS, meanwhile, 2-BP had no effect on these processes (Figure 4D). These results suggested that the total level and the synaptic expression of CaN were not regulated by palmitoylation. Additionally, the total expression of PKA was not affected by HFS, but the synaptic expression of PKA increased remarkably after HFS, which was reversed by 2-BP treatment (Figure 4E). These results suggest that the synaptic PKA, not CaN, is regulated by AKAP150 palmitoylation.

Previous studies indicate that the interaction of PKA and AKAP150 may control the AMPARs during plasticity,^17,20,23 and we observed that the expression of PKA and AKAP150 at synapses were increased significantly after HFS. However, there is a great need to know what happens about the interaction of PKA and AKAP150 after HFS. Strikingly, a significant elevation in the interaction between AKAP150 and PKA after HFS was observed, but it was abolished by 2-BP treatment (Figure 4F). Together, these results indicated that DHHC2 mediated AKAP150 palmitoylation promotes AKAP150 signaling during LTP in the hippocampus.

DHHC2 is required for surface delivery of GluA1 during LTP

To further examine the effect of DHHC2 on AKAP150 signaling in the hippocampus during LTP, the LV-shDHHC2 or LV-eGFP was microinjected into the hippocampus, and the interaction of AKAP150 and PKA was examined by co-immunoprecipitation assay 2 weeks later. Compared with LV-eGFP group, the binding of AKAP150 to PKA was decreased significantly in LV-shDHHC2 group, suggesting that knockdown of DHHC2 disrupts the interaction between AKAP150 and PKA (Figures 5A and 5B).

Figure 5.

DHHC2 is required for GluA1 surface expression in response to LTP

(A and B) Representative western blots (A) and statistical results (B) of co-immunoprecipitation between AKAP150 and PKA in the hippocampus after DHHC2 knockdown (n = 4 per group).

(C) Representative western blots of GluA1, phosphorylated GluA1 at Ser831 and Ser845, and GluA1 surface expression after treating with HFS and LV-shDHHC2.

(D–G) Statistical results of GluA1 expression (D), phosphorylation level of GluA1 at Ser831 (E) and Ser845 (F), and GluA1 surface expression (G) (n = 5 per group).

(H) Time-course of LTP induced by HFS in hippocampal slices from the LV-eGFP/LV-shDHHC2-treated groups.

(I) Statistical histogram showing LTP magnitude averaged from the last 15 min of recordings from the two groups (n = 4 per group).

Two-way ANOVA for (A–G), and Student’s t test for (I). Data represented are means ± SEM. ∗p < 0.05, ∗∗p < 0.01 and ∗∗∗p < 0.001; ^#p < 0.05 and ^##p < 0.01.

Evidence noted that Ser831 and Ser845, the key serine residues on GluA1, are critical for the surface expression of GluA1 and involved in synaptic plasticity.^36,37 Previous finding show that DHHC2 promoted AKAP150 to cluster into postsynaptic membrane,²⁵ thus, we hypothesized that DHHC2 might regulate the phosphorylation of GluA1 at Ser831 or Ser845 and its surface expression by AKAP150/PKA signaling during LTP. After knockdown DHHC2 for 2 weeks, western blotting results showed that there was little difference after LV-shDHHC2 and HFS treatment (Figures 5C and 5D). Under the same conditions, we found that the phosphorylation level of GluA1 at Ser831 and Ser845 sites was significantly increased during LTP, meanwhile, LV-shDHHC2 treatment reversed the enhancement of GluA1-Ser845 phosphorylation but did not affect the phosphorylation level of GluA1-Ser831 (Figures 5E and 5F). Furthermore, the surface expression of GluA1 was increased remarkably as expected during LTP. Interestingly, knockdown DHHC2 caused a significant basal increase in the surface expression of GluA1, whereas the HFS-induced increase in surface expression of GluA1 was abolished by LV-shDHHC2 (Figure 5G). In addition, our fEPSP recordings from mice hippocampal slices showed that DHHC2 knockdown significantly decreased fEPSP slopes (Figures 5H and 5I). These findings demonstrated that DHHC2 mediated AKAP150 signaling is functionally involved in hippocampal synaptic transmission during LTP. Together, our finding demonstrated that DHHC2-mediated AKAP150 palmitoylation plays an important role in regulating hippocampal LTP and fear memory.

Discussion

In the present study, we demonstrated that DHHC2 activity is required for fear memory formation, knockdown DHHC2 impaired fear memory and related LTP in the hippocampus. In vitro, the activity of DHHC2 was increased in glycine-induced cLTP and HFS-induced LTP. Interestingly, AKAP150, not PSD-95, was selectively palmitoylated by DHHC2 in response to fear learning and HFS. Downregulation of the activity of DHHC2 with shRNA silencing significantly decreased the palmitoylation level and synaptic expression level of AKAP150, along with lower fEPSP slopes during LTP in vitro, suggesting that DHHC2-mediated AKAP150 palmitoylation plays important roles in regulating glutamatergic synaptic plasticity and fear memory formation.

Histologically, synaptic protein palmitoylation is regulated by neuronal activity, and palmitoylated proteins move into synapses.^38,39,40,41 For example, blocking synaptic activity rapidly induces PSD-95 palmitoylation, and mediates synaptic clustering of PSD-95 and AMPA receptors,³⁸ while AKAP150 palmitoylation was obviously increased after kainate seizures and mediated postsynaptic lipid raft localization.²⁶ LTP is associated with increased synaptic activity and strength, meanwhile, synaptic activity also increases the turnover of proteins necessary for LTP, thereby modulating LTP stability.⁴² Here, we used HFS and glycine to induce two kinds of LTP, respectively, and found that the level of global palmitoylation rapidly increased following acute stimulation. Importantly, our previous results showed that palmitoylation is required for synaptic strengthening and fear memory formation, and incubation with 2-BP abolished the enhancement of LTP induced by fear conditioning in the hippocampus.⁷ These results suggest that palmitoylation plays an important role in LTP-induced exocytosis of recycling endosomes. AKAP150, mainly locates at recycling endosome, is targeted to plasma membrane, along with recycling endosomes in dendrites during LTP.^26,43 However, the possible underlying mechanisms involved in palmitoylation are still unknown.

DHHC2, distributed prominently in recycling endosomes, is responsive to a lot of synaptic molecules, including PSD-95, SNAP23/25 and AKAP150.⁴⁴ From prior studies we know that DHHC2 is activity-dependently translocated from shaft to spine upon activity blockade,³⁸ and mediates PSD-95 palmitoylation and postsynaptic clustering to promote the synaptic incorporation of AMPARs.⁴⁵ Conversely, the activity of DHHC2, detected by its auto-palmitoylation,³⁰ displayed pronounced upregulation after HFS. Considering that HFS treatment recapitulated LTP-stimulated recycling endosome exocytosis in the hippocampus,⁴⁶ we reasoned that activity-dependent auto-palmitoylation of DHHC2 was involved in its targeting to AKAP150 in recycling endosome and formed a complex, and then moved into synaptic compartment. To address this issue, we knockdown DHHC2 with lentivirus vector in the hippocampus. Strikingly, a lower level of DHHC2 reversed the increase in the interaction between AKAP150 and PKA induced by HFS. To further examine the effects of DHHC2 on AMPARs delivery during LTP, we used biotinylation assay and found that the phosphorylation levels of GluA1 at Ser831 and Ser845 were significantly increased after HFS in the hippocampus. However, it should be noted that knockdown DHHC2 abolished the increase in phosphorylation level of GluA1 at Ser845, but not Ser831, induced by HFS in the hippocampus. Interestingly, an obvious elevation of basal GluA1 surface expression was observed after DHHC2 knockdown in the hippocampus. Similarly, another study also showed that there were increased basal mEPSC amplitudes in DHHC2 knockdown neurons.²⁵ Our findings suggest that the prominent increases in basal mEPSC activity and AMPARs surface expression may be mediated by a combination of new synapse formation and postsynaptic GluA1 surface delivery.

DHHC2 regulates the strength and plasticity of excitatory synaptic transmission by palmitoylating AKAP150 during LTP in the hippocampus.²³ However, the mechanism underlying the effect of HFS on the activity of DHHC2 has not been illustrated. A possible explanation for this might be that, similar to Fyn-mediated phosphorylation of DHHC5, some other kinases rapidly phosphorylate DHHC2 and induce its auto-palmitoylation, which subsequently results in AKAP150 palmitoylation during LTP. Furthermore, DHHC2 was identified as a regulatory factor of AKAP150 palmitoylation,²⁵ and palmitoylation of synaptic proteins is regulated by neuronal activity dynamically.⁴⁷ In addition, the level of global palmitoylation was increased at 30 min after HFS, and then a lower increase was observed at 60 and 90 min. These results suggest that de-palmitoylation was also activated during LTP. However, enzymes and mechanisms responsible to synaptic protein de-palmitoylation need to be illustrated in future study.

In summary, our study identifies a critical role for DHHC2-mediated AKAP150 palmitoylation which dynamically regulates LTP and fear memory. Glycine and HFS induce AKAP150 palmitoylation by activating DHHC2, and then the palmitoylated AKAP150 anchors PKA, not CaN, into synapse to regulate GluA1 phosphorylation and trafficking, and eventually promotes LTP expression and fear memory formation. In contrast, knockdown DHHC2 disrupts the effects of fear learning or HFS on the phosphorylation and surface expression of GulA1, and induces an impaired LTP. Taken together, these findings strongly suggest that the DHHC2 mediated AKAP150 signaling pathway regulates membrane surface stability of AMPARs in response to fear learning and HFS, and may serve as a potential target for improvement of learning and memory.

Limitations of the study

The present study demonstrated the important role for DHHC2 mediated AKAP150 palmitoylation cycling in fear memory formation and LTP, and identified a novel mechanism underlying the crosstalk between AKAP150 palmitoylation and LTP via DHHC2 in a synaptic activity dependent manner. Further studies are needed to examine the de-palmitoylating enzymes of AKAP150 de-palmitoylation, and its role in fear memory and synaptic plasticity.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Rabbit monoclonal anti-CaN	Abcam	Cat#ab52761
Rabbit monoclonal anti-GluA1	Cell Signaling Technology	Cat#13185
Rabbit polyclonal anti-PKA	Cell Signaling Technology	Cat #4782
Rabbit monoclonal anti-GluA1-Ser831	Cell Signaling Technology	Cat#75574
Rabbit monoclonal anti-GluA1-Ser845	Cell Signaling Technology	Cat#8084
Rabbit monoclonal anti-PSD-95	Cell Signaling Technology	Cat#3409
Mouse monoclonal anti-β-actin	Santa Cruz Biotechnology	Cat#sc-47778
Mouse monoclonal anti-AKAP150	Santa Cruz Biotechnology	Cat#sc-377055
Mouse monoclonal anti-DHHC2	Santa Cruz Biotechnology	Cat#sc-515204
anti-HRP-Streptavidin	BOSTER	Cat#BA1088
Bacterial and virus strains
LV-eGFP	GeneChem	This paper
LV-shDHHC2	GeneChem	This paper
Chemicals, peptides, and recombinant proteins
2-BP	Sigma	Cat#238422
Glycine	Sigma	Cat#G8898
Hydroxylamine	Sigma	Cat#431362
MMTS	Sigma	Cat#208795
HPDP-biotin	Thermo Fisher Scientific	Cat#21341
Experimental models: Organisms/strains
Mouse: C57BL6/J	Experimental Animal Center of Fujian Medical University	N/A
Software and algorithms
Any-maze	Stoeling	RRID:SCR_014289
Microsoft Excel	Microsoft	RRID:SCR_016137
GraphPad Prism version 7.0	GraphPad Software	RRID:SCR_002798
ImageJ	NIH	https://imagej.net/
SPSS	IBM	RRID:SCR_002865

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Zu-Cheng Shen (shenzc2019fjmu@163.com).

Materials availability

This study did not generate new unique reagents.

Experimental model and subject details

Animals

The experimental animals were adult male C57BL/6J mice (aged 8–10 weeks, weight 22–25 g) from Experimental Animal Center of Fujian Medical University in China. The mice were group housed in individual ventilated cages (IVC) at 22 ± 2°C and 50%–60% relative humidity under a 12 h light/dark cycle switched off at 18:00, and could access to enough food and water freely. All experimental procedures of using animals were performed according to the Guide for Care and Use of Laboratory Animals (Ministry of Health, China). All efforts were made for an aim which was reducing the number of experimental animals and their suffering.

Method details

Agents

2-bromopalmitic acid (2-BP), Methyl methane thiosulfonate (MMTS), hydroxylamine (HA) and streptavidin-agarose were purchased from Sigma-Aldrich (St. Louis, MO, USA). N-(6-(Biotinamido) hexyl)-3’-(2′-pyridyldithio)-propionamide (HPDP)-biotin was supplied by Thermo Fisher Scientific (Waltham, MA, USA). Syn-PER Synaptic Protein Extract Reagent was purchased from Thermo Scientific (Rockford, USA). Other agents were obtained from commercial suppliers.

Contextual fear conditioning

Contextual fear conditioning experiment was conducted over 3 days, day 1 (habituation), day 2 (fear conditioning) and day 3 (testing). Briefly, on day 1, mice were exposed to the chamber for 5 min. On day 2, mice were given five 1-s foot shocks at 0.75 mA with the 90 s-intertrial intervals. Mice were removed from the chambers immediately after the session. On day 3, mice were placed into the same chambers and observed for 5 min. The percentage of freezing time was recorded by ANY-maze software (Stoelting Co., IL, USA).

Slice preparation

Mice were anesthetized with isoflurane and transcardially perfused with cutting solution (4°C) containing (in mM): sucrose 209, ascorbic acid 11.6, sodium pyruvate 3.1, MgSO₄ 4.9, NaHCO₃ 26.2, NaH₂PO₄ 1.0, glucose 20, pH 7.4, osmolarity 290–310 mOsm, and saturated with 95% O₂ + 5% CO₂. After quickly decapitation, the brains were moved into the ice-cold oxygenated cutting solution and cut into 300-μm coronal sections on a vibratome (VT 1200S, Leica, Wetzlar, Germany). The slices containing the hippocampus were incubated in artificial cerebrospinal fluid (aCSF) at 34°C for 30 min. After that, slices were incubated at 28°C in aCSF which was continuously perfused with oxygen. Compositions of aCSF were as follows (in mM): 119.0 NaCl, 1.3 MgSO₄, 3.5 KCl, 1.0 NaH₂PO₄, 26.2 NaHCO₃, 2.5 CaCl₂ and 11.0 glucose, pH 7.4 osmolarity 290–310 mOsm.

Chemical LTP induction

After incubation at 37°C in aCSF, the slices were treated with aCSF containing TTX (1 μM) at 37°C for 5 min, and then the solution was removed and changed with 100 μM glycine in Mg²⁺-free aCSF for 5 min, followed by recovery in aCSF with Mg²⁺ and TTX.

HFS-LTP recording

After recovery, one slice was placed in the recording chamber where it was continuously perfused with oxygenated aCSF at the rate of 2 mL/min. Field excitatory postsynaptic potentials (fEPSPs) were evoked by a constant stimulation in the Schaffer collaterals with a bipolar stimulating electrode and recorded in the stratum radiatum layer of CA1 region with a glass electrode (3–5 MΩ) filled with 3 M NaCl. To evoke LTP induction, HFS was introduced after recording baseline fEPSPs for at least 20 min. The HFS consisted of three trains of 100 pulses at 100 Hz separated by 30 s and delivered at test intensity. Testing of fEPSPs post-HFS with single pulse was then lasted for 90 min to determine the level of stable LTP. The fEPSPs were monitored for 60 min and the fEPSPs slopes were normalized to the average of slopes of fEPSPs obtained during the baseline period. All data were calculated as the average of the last 15 min of the 60 min recordings.

Acyl-biotin exchange assay

Palmitoylated proteins were extracted and detected through ABE method. The specific steps were modified appropriately on the basis of previous literature. Ice-cold RIPA buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.5% sodium deoxycholate, 1% NP-40, 2% sodium dodecyl sulfate, 1 mM ethylene diamine tetraacetate) containing protease inhibitor and 20 mM MMTS was added to the hippocampal tissue to block free thiols. The tissues were homogenized and centrifuged, and the supernatant fractions were obtained. MMTS (2 μL) was added to all samples and homogeneously mixed, and shaken gently at 50°C and kept out of light for 30 min to block free thiols. Four times volume of pre-colded acetone was added and incubated at −20°C for 30 min to remove the residual MMTS, and centrifuged at 3000 g for 10 min to further remove MMTS. The precipitates were collected and resuspended in RIPA buffer again. This process was repeated twice. The RIPA buffer resuspending the final precipitates was divided into two portions. One portion was incubated with HPDP-biotin in RIPA buffer to serve as control. The other portion was incubated in four times volume of RIPA buffer with 0.7 mM HA to cleave thioester bonds and 1 mM HPDP-biotin to link biotin to newly exposed cysteine thiols, and used to detect protein palmitoylation level. All samples were placed in the dark with frequent vortexing for 2 h at room temperature. Acetone precipitation was used to remove the remaining HPDP-biotin and HA. The precipitates were resuspended in RIPA buffer. Coomassie blue protein-binding assay was used to measure the concentration of protein, and global palmitoylation was detected using HRP-streptavidin by Western blotting steps. Protein samples after ABE were divided into two portions. One portion was directly added to SDS-PAGE sample-loading buffer to serve as “input”, and the other was purified with streptavidin-agarose. After purified for 10 times, streptavidin-agarose was washed three times with RIPA buffer to affirm reaction completed. After being eluted and denatured by SDS-PAGE sample-loading buffer at 95°C for 30 min, the biotinylated samples were stored at −80°C. The palmitoylated proteins were normalized to the input proteins (total proteins), with β-actin as internal reference.

Surface biotinylation assay

Surface biotinylation assay was performed as described below: Sulfo-NHS-LC-Biotin (1 mg/mL, dissolved in aCSF) was added into the hippocampal tissue and reacted for 30 min on ice. Then, the tissue was washed with ice-cold aCSF for three times. Finally, it was homogenized with RIPA buffer including a cocktail of protease and phosphatase inhibitors, and then centrifuged at 12 000 g for 15 min at 4°C. The supernatant was measured via Coomassie blue protein-binding assay. Biotinylated protein was incubated with streptavidin for 4 h at 4°C. Streptavidin-protein complexes were washed three times with RIPA buffer and centrifuged at 4000 g for 3 min at 4°C. Bound proteins were separated from the beads, and denatured by SDS-loading buffer at 95°C for 5 min.

Co-immunoprecipitation

Hippocampal slices were homogenized with extraction buffer [in mM: 150 NaCl, 0.2% NP-40 (v/v), 1 Na₃VO₄, 50 NaF, 3 Na-pyrophosphate, 6 Na-deoxycholate, 1% cocktail of protease inhibitor (v/v), 1% phosphatase inhibitor (v/v)] for 40 min on ice, and then centrifuged (12 000 g) for 20 min at 4°C and the supernatant was obtained. Then the supernatant was served as the primary material for co-immunoprecipitation. The supernatant was incubated with IgG, PSD-95 or AKAP150 antibody overnight at 4°C with gentle rotation, and then the antibody-antigen complexes were precipitated by incubation with Protein A/G Agarose bead at 4°C for 3 h. Collection of samples from the beads was performed by the means of centrifuging (4000 g, 3 min) and washing with extraction buffer for three times. Finally, SDS-loading buffer added, the eluents were boiled at 95°C for 5 min before being analyzed by Western blotting method.

Synaptic fractionation

The hippocampal tissue was homogenized with Syn-PER Reagent (10 mL/mg) for 10 min. Then, the homogenate was centrifuged at 1 200 g for 10 min at 4°C, and the supernatant (supernatant 1) was collected into a new tube. Synaptosome pellet and cytosolic fraction were separated from supernatant 1 by centrifugation at 15 000 g for 20 min at 4°C. Synaptosome pellet was resuspended in RIPA buffer with protease inhibitors cocktail, and then added the SDS-loading buffer and boiled at 95°C.

Western blotting

Total proteins were extracted from the hippocampus of mice using RIPA lysis buffer and quantified by Coomassie bright blue assay. Equivalent protein denatured by 3× loading buffer (95°C, 5 min) was separated by SDS-PAGE, and the bands were transferred to NC membranes (250 mA, 90 min). Cut the NC membranes to obtain the target protein and put it in 5% BSA-TBST solution for 1 h. The membranes washed by TBST for three times were incubated respectively with primary antibodies overnight at 4°C. The membranes were incubated with horseradish peroxidase-conjugated secondary antibodies at room temperature for 2 h. After washing, add ECL color solution, incubate for 1–2 min, and select appropriate exposure time for detection by Micro Chemi gel imaging system (DNR Bio-imaging systems, Jerusalem, Israel). The gel images were analyzed by ImageJ, and protein expression levels in different samples were compared using β-actin as internal reference.

Stereotaxic injections

After anesthesia with isoflurane (Shenzhen RWD Life Science), the mice were mounted in an automated stereotactic apparatus (RWD68001, Shenzhen RWD). For local knockdown of DHHC2, green fluorescent protein (GFP)-tagged lentiviral vector encoding specific short hairpin RNA (shRNA) against DHHC2 (LV-shDHHC2) was injected into the hippocampal region of the adult mice (AP: −2.3 mm; ML: ±2.0 mm; DV: −2.0 mm). Then, the mice were individually housed and allowed to recover for at least seven days after the surgical operation.

Quantification and statistical analysis

All results were expressed as the “mean ± SEM”. We used Student’s t test to compare differences between unpaired two groups, and one- or two-way analysis of variance (ANOVA) to compare the means of multiple groups of samples. Statistical significance level was set at p value less than 0.05.

Published: August 9, 2023

Data and code availability

• •All data reported in this paper will be shared by the lead contact upon request.
• •This paper does not report original code.
• •Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.