Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/springer-journals/age-related-attenuation-of-cortical-synaptic-tagging-in-the-acc-is-VRli5ZwLbR
- Publisher
- Springer Journals
- Copyright
- Copyright © The Author(s) 2023
- eISSN
- 1756-6606
- DOI
- 10.1186/s13041-022-00992-x
- Publisher site
- See Article on Publisher Site
Abstract
Long‑term potentiation (LTP) is a key cellular mechanism for learning and memory, and recent studies in the hip ‑ pocampus found that LTP was impaired in aged animals. Previous studies of cortical LTP have focused primarily on the homosynaptic plasticity in adult mice, while fewer studies have looked at heterosynaptic plasticity—such as synaptic tagging in aged mice. In the present study, we investigated synaptic tagging in adult and middle‑aged mice’s anterior cingulate cortex (ACC) using the 64‑ channel multielectrode dish (MED64) recording system. We found that synaptic tagging was impaired in the ACC of middle‑aged male mice as compared to adult mice. Both the network late ‑phase LTP (L‑LTP) and the recruitment of inactive responses were reduced in the ACC of middle ‑aged male mice. Similar results were found in female middle‑aged mice, indicating that there is no gender difference. Furthermore, bath application of brain‑ derived neurotrophic factor (BDNF) or systemic treatment with newly developed TrkB receptor agonists R13, was shown to rescue both synaptic tagging, and L‑LTP, in middle ‑aged mice. To determine the distribu‑ tion of synaptic LTP within the ACC, a new visualization method was developed to map the Spatio‑temporal variation of LTP in the ACC. Our results provide strong evidence that cortical potentiation and synaptic tagging show an age‑ dependent reduction, and point to the TrkB receptor as a potential drug target for the treatment of memory decline. Keywords Synaptic tagging, LTP, BDNF, R13, ACC , Middle‑aged mice *Correspondence: Pathology and Laboratory Medicine, Emory University School Xu‑Hui Li of Medicine, Atlanta, GA, USA lixuhui19@xjtu.edu.cn Faculty of Life and Health Sciences, Brain Cognition and Brain Disease Min Zhuo Institute, Shenzhen Institute of Advanced Technology, Chinese Academy min.zhuo@utoronto.ca of Sciences, Shenzhen, China 1 7 Center for Neuron and Disease, Frontier Institutes of Science Oujiang Laboratory, Zhejiang Lab for Regenerative Medicine, Vision and Technology, Xi’an Jiaotong University, Xi’an, China and Brain Health, Wenzhou 325000, Zhejiang, China 2 8 Institute of Brain Research, Qingdao International Academician Park, Department of Physiology, Faculty of Medicine, University of Toronto, Qingdao, Shandong, China Medical Science Building, 1 King’s College Circle, Toronto, ON M5S 1A8, Institute of Artificial Intelligence and Robotics, Xi’an Jiaotong University, Canada Xi’an, China CAS Key Laboratory of Brain Connectome and Manipulation, Interdisciplinary Center for Brain Information, The Brain Cognition and Brain Disease Institute, Shenzhen‑Hong Kong Institute of Brain Science‑Shenzhen Fundamental Research Institutions, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences Shenzhen Institute of Advanced Technology, Shenzhen, China © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Zhou et al. Molecular Brain (2023) 16:4 Page 2 of 16 In the present study, we employed the MED64 record- Introduction ing system to investigate synaptic tagging in the ACC of Long-term potentiation (LTP), an activity-dependent both adult and middle-aged mice. Both male and female long-lasting increase of synaptic efficacy caused by high- mice were used. We found that synaptic tagging in the frequency stimulation or theta burst stimulation (TBS), ACC was significantly reduced in middle-aged animals has been established as a cellular model of memory in as compared to adults. Cumulative data has demon- different regions of the brain, including the hippocam - strated that brain-derived neurotrophic factor (BDNF) pus, prefrontal cortex, and the anterior cingulate cor- is crucially involved in synaptic plasticity in the adult tex (ACC) [1–3]. LTP has at least two distinct temporal brain [29]. Furthermore, by using BDNF—or selective phases: protein synthesis-independent early-phase LTP trkB receptor agonist R13—we were able to reverse the (E-LTP), and protein synthesis-dependent late-phase LTP loss of synaptic tagging in middle-aged animals. Finally, (L-LTP) [4–6]. Furthermore, it has been reported that we developed a novel method to better visualize the Spa- E-LTP and L-LTP can interact with each other in a ’syn- tio-temporal signals of fEPSP signals and multiple LTP aptic tagging-like manner. Weak tetanus-inducing E-LTP responses within the ACC circuit from low-resolution sets a “tag”, which can capture the plasticity-related MED64 inputs. proteins (PRPs) synthesized following the strong teta- nus-inducing L-LTP [7–9]. A weak stimulus can induce Materials and methods L-LTP if it is preceded or followed by strong tetanus Animals given to a separate, independent pathway that converges For the animal groups, we divided them into two major into the same neuronal population. This finding has been groups by age: adult mice (6–8 weeks), and middle-aged subsequently repeated and extensively investigated [10]. mice (50–60 weeks). All mice were done on male and Synaptic tagging is not just limited to the hippocampus. female C57BL/6 mice purchased from the Experimental In the ACC, a cortical region that is important for pain Animal Center of Xi’an Jiaotong University (6–8 weeks) perception and emotional memory process [11–17], syn- and Charles River Laboratories in Beijing (50–60 weeks). aptic tagging has also been reported. Similar to our find - All mice were randomly housed by three to four per cage ings in the hippocampus, our previous studies reported under standard laboratory conditions (12 h light/12 h that weak TBS can also induce heterosynaptic synap- dark, temperature 22–26 °C, air humidity 55–60%). Food tic tagging in the ACC of adult mice, which depends on and water were available ad-lib. All research protocols a certain time window and the synthesis of new pro- performed in this experiment were approved by the Eth- teins[18]. Functionally, several lines of evidence suggest ics Committee of Xi’an Jiaotong University. that synaptic tagging may contribute to memory alloca- tion and storage [19–21]. It may also provide a synaptic Preparation of the multi‑electrode array mechanism for emotional tagging [22]. For example, Liu There is an array of 64 square planar microelectrodes et al. previously reported that tail amputation-induced (50 × 50 µm/each) arranged in an 8 × 8 pattern in the peripheral injury caused a loss of hetero-synaptic L-LTP MED64 probe (P515A, chamber depth 10 mm, Alpha in the ACC [18]. Vecsey et al. found that sleep depriva- MED Scientific, Japan), with an interpolar distance of tion impaired synaptic tagging in the mouse hippocam- 150 μm. Since the surface of the MED64 probe is rela- pus [23]. Most of the previous synaptic tagging studies tively hydrophobic, in order to attach the slice to the have focused primarily on adult mice—it is unclear to MED64 probe well, the new MED64 probe received this point whether synaptic tagging is affected by aging. hydrophilic treatment. Before experiments, we treated Age-related synaptic LTP and memory impairment has the surface of the MED64 probe with 0.1% polyethyl- been reported in the hippocampus and hippocampus- eneimine (Sigma, St. Louis, MO; P-3143) in 25 mmol/L dependent behavioral tests [24–26]. Interestingly, age- borate buffer (pH 8.4) overnight at room temperature. related impairment of behavioral tagging and synaptic Then we used sterile distilled water to flush the probe tagging has also been reported in the hippocampus [27, surface three times to remove any harmful substances 28]. Wong et al. (2021) reported that synaptic tagging that may affect the activity of brain slices [30, 31]. was attenuated in the hippocampal region of middle- aged mice [27]. There is no report of age-related changes Brain slice preparation in synaptic tagging in the ACC. Our recent studies using The general procedures for making the ACC slices were animal models of amputation found that synaptic tagging similar to that in our previous study [30, 31]. Acute in the ACC was either reduced or abolished in ACC slices coronal brain slices (300 μm) containing ACC were pre- after tail amputation [18], suggesting that cortical syn- pared from C57BL/6 mice. In brief, we anesthetized aptic tagging is plastic, and may possibly be affected by C57BL/6 mice with 1–2% isoflurane and sacrificed them either peripheral injury or aging. Zhou et al. Molecular Brain (2023) 16:4 Page 3 of 16 by decapitation. The entire brain was quickly removed For test stimulation, constant current pulses (0.2 ms from the skull and submerged in an ice-cold oxygenated in duration) generated by a data acquisition software (equilibrated with 95% O and 5% C O ) cutting solution (Mobius, Panasonic Alpha-Med Sciences) were applied 2 2 containing (in mM) 252 sucrose, 2.5 KCl, 6 MgSO , 0.5 to either deep layers (layer V-VI, for tagging TBS input) CaCl , 25 NaHCO , 1.2 NaH PO , and 10 glucose, pH and/or superficial layers (layer II-III, for strong TBS 2 3 2 4 7.3 to 7.4 for a short time. After a brief cooling, the brain input) of the ACC slice. Bipolar constant current pulse was trimmed, and the remaining brain block was glued stimulation (6–10 µA, 0.2 ms) was applied to the stimula- onto the ice-cold stage of a vibrating tissue slicer (Leica, tion site, and the intensity was adjusted so that we could VT1200S). In this way, the brain coronal brain slices evoke a half-maximum field excitatory postsynaptic (300 μm) containing ACC were obtained, and the slices potential (fEPSP) in the channel nearest to the stimula- were then transferred to a submerged recovery cham- tion site. The channel with fEPSP was regarded as an acti - ber with oxygenated (95% O and 5% C O ) artificial cer - vated channel, and its fEPSP response was sampled every 2 2 ebrospinal fluid (ACSF) containing (in mM) NaCl 124, 2 min and averaged every 4 min. The ‘slope’ parameter KCl 2.5, CaCl 2, MgSO 2, NaHCO 25, NaH PO 1 and represented the average slope of each fEPSP recorded by 2 4 3 2 4 glucose 10, pH 7.3–7.4 at room temperature for at least the activated channel. Stable baseline responses (varia- 1.5 h. This ACSF was used throughout the experiment, tion in the baseline response of a single channel is < 5% including the recording stage. and the number of channels with unstable baseline responses was ≤ 10% of the total number of active chan- Field potential recording nels) were first recorded for 60 min. Then, a strong TBS A commercial 64-channel recording system (MED64, (five trains of bursts with four pulses at 100 Hz at 200 ms Panasonic Alpha-Med Sciences, Japan) was used to intervals; repeated five times at intervals of 10 s, 4 × 5×5) record extracellular field potential in ACC in male and with the same intensity as the baseline stimulation was female C57BL/6 mice. After incubation, one slice con- applied to the same stimulation channel to induce LTP. taining ACC was transferred to the prepared recording After a half hour of strong TBS, a tagging TBS (five trains probe. The ACC part was placed on the MED64 probe’s of bursts with four pulses at 100 Hz at 200 ms intervals, electrodes, and the 64 electrodes were covered by differ - 4 × 5×1) with the same intensity as the baseline stimula- ent layers of the ACC. We can easily choose the super- tion in the deep layer was applied to the same stimulation ficial layers (layers II-III) or deep layers (layers V-VI) as channel to induce synaptic tagging. After the induction the stimuli site. Once the slice was settled, a mesh and of LTP or/and tagging LTP, the fEPSP responses were an anchor (Warner Instruments, Harvard) were care- continued recording for another 4–4.5 h. The fEPSPs of fully positioned to ensure the stability of the slice dur- sites S1 and S2 will be recorded in turn, with an interval ing recording. The slice was perfused continuously with of one minute. oxygenated (95% O and 5% CO ) ACSF at 26–28 ℃ We also applied BDNF to the brain slices of middle- 2 2 and maintained at a 2–3 ml/min flow rate with the aid aged mice. We added 1 µg BDNF to 20 ml oxygenated of a peristaltic pump (Minipuls 3, Gilson) throughout the ACSF to perfuse the brain slice. The ACSF with BDNF experiments. Figure 1a shows a micrograph of an ACC was applied for 1 h, from 15 min after strong TBS to slice placed on a MED64 probe. Before the experiment, 45 min after tagging TBS. Then the BDNF was washed the slices were kept in the probe for at least 1 h. out. Synaptic tagging in R13 oral gavage of middle-aged (See figure on next page.) Fig. 1 Synaptic tagging was induced in the ACC by tagging TBS in middle‑aged male mice. Tagging was recorded from a slice of a male mouse by extracellular field potential recording. a Microscopy photograph showing the relative location of ACC slice and MED64 probe and the arrangement of the microelectrodes (electrode size 50 × 50 μm, the interpolar distance of electrodes 150 μm). Schematic diagram of stimulation site (S1: blue circle for the strong TBS; S2: red circle for the weak TBS) of microelectrodes in the ACC. Spatial distribution of extracellular field potential induced by strong TBS on channel 26 (marked as a blue circle) in layers II and weak TBS on channel 46 (marked as a red circle) in layers V of the ACC. b Schematic diagram of the recording procedure for sites S1 and S2. c The Markov Random Field model was introduced to smooth and visualize the low‑resolution 8 × 8 MED64 data. The graph model consists of the observation at site i (O ), the state at site i (S ), and the state at a neighboring site i i j (S ). d As a generative process, the value of O is generated from S , while the value of S also depends on its 4‑ connection neighborhood denoted j i i i as S . The original 8 × 8 MED64 record for a single time frame. Then, a 256 × 256 projection of the original MED64 data that is both spatially and temporally smoother with a color bar. Finally, the temporal evolving of the 3D surface of the projected 256 × 256 MED64 recording. e The temporal evolving of the 3D surface on the site S1 in the middle‑aged mice. f Summarized plot of the time ‑ varying fEPSP slope in all activated channels for site S1 from 1 slice of a middle‑aged male mouse. g The temporal evolving of the 3D surface on site S2 of the middle ‑aged mice. h Summarized plot of time‑ varying fEPSP slopes in all activated channels for tagging TBS protocol from the 1 slice Zhou et al. Molecular Brain (2023) 16:4 Page 4 of 16 Fig. 1 (See legend on previous page.) Zhou et al. Molecular Brain (2023) 16:4 Page 5 of 16 mice was also studied. After R13 oral administration where the unary term considers the consistency from (concentration: 43.6 mg/kg/d) for 15 days, we performed observation and is defined by the same field potential recording protocol as above 8×8 mentioned. R13 was dissolved in pure DMSO, then sus- � (X) = (o − x ) (2) unary i i=0 pended in 0.5% methylcellulose at a final concentration of 5% DMSO/0.5% methylcellulose. and the pairwise term considers the interdependency between sites and is defined by Western blot N×N After the fEPSPs recording, the ACC was collected from � (X) = a (x − x ) pairwise i,j j (3) i=0 the brain slices on ice in cold PBS and homogenized in j∈N (i) lysis buffer (10 mM Tris–HCl (pH 7.4), 2 mM EDTA, Specifically, the smoothness weight a between site i and i,j 1% SDS including a protease inhibitor cocktail). Samples j is defined using the Lanczos [ 34] filter by were then centrifuged (12,000g, 20 min, 4 °C) for superna- tant. Western blot was performed as previously described αsin π(i − j) sin π(i − j)/α [32]. Sample protein concentrations were quantified a = i,j (4) π (i − j) using Bradford assay (Beyotime), and electrophoresis of equal amounts of protein (30 μg) was performed on 7.5% and N (i) denotes a neighborhood of site i and is con- SDS–polyacrylamide gel. Separated proteins were trans- trolled by α . In our implementation, we set α = 3 and ferred to polyvinylidene fluoride (PVDF) membranes, hence N (i) includes sites from within a 5 × 5 grid region followed by blocking with 5% skim milk in TBS-T (Tris- centering at i. buffered saline with Triton X-100) at room temperature The following figures (Fig. 1d) illustrate the original for 1 h, and were then probed with primary antibody: 8 × 8 Med64 data, the obtained 256 × 256 estimation, anti-Trk B receptor (1:1000, rabbit polyclonal, Santa Cruz and visualization of the 256 × 256 surface with some Biotechnology), anti-CREB (1:1000, rabbit polyclonal, additional information. Abcam) and anti-GAPDH (1:10,000, rabbit polyclonal, Since we have a continuous recording of the MED64 Abcam) at 4 °C overnight. The membranes were incu - signal, we could actually further consider a more com- bated with horseradish peroxidase-coupled anti-rabbit/ plete spatio-temporal relationship between sites to get a mouse lgG secondary antibody diluted at 1:5000 (Milli- more stable, smooth and possibly more accurate estima- pore) for 1 h, followed by enhanced chemiluminescence tion of the site value. Mathematically this is achievable by detection of the proteins with Enhanced chemilumines- defining a 3D neighborhood system in addition to Eq. ( 3), cence, ECL (GE Healthcare). ImageJ software (National we have Institute of Health) was used to assess the intensity of N×N immunoblots. 2 � (X) = a (x − x ) pairwise i,j,t j,t i,t i=0 T S t∈N (i) j∈N (i) (5) A novel fEPSP signal modeling and visualization system Specifically, the smoothness weight a between site i We consider the recorded 8 × 8 Med64 data as a partially i,j,t and j spanning over t frame is defined as observed Markov Random Field [33]. Specifically, we assume that the original data is from a larger field with a = a a i,j,t i,j t (6) N × N sites. Typically, N ≥ 8 but technically N can be any integer value greater than zero. We define the state where a is calculated by Eq. 4, while t considers the tem- i,j space to be continuous in [m ,m ] , and for visualiza- poral distance between two sites as min max tion purpose, we stretch the range to be discrete within β − t [ 0, 255] . Among the N × N sites, 8 × 8 state has observa- a = t (7) tionsO = {o , . . . ,o } . Therefore, the MRF has the fol - 0 63 lowing structure as shown in Fig. 1c. where β controls the size of the temporal neighborhood, We aim to estimate the state value of all the sites and in our implementation, we set β = 3 . Now, com- X = {x , . . . , x } with the Maximum a Posteriori 0 N×N pared to the spatial case, in spatio-temporal case our (MAP) criterion based on the partial observation and the N (i) includes sites from within a 5 × 5 × 5 grid region interdependency between sites. The MAP estimation of centering at i. the MRF strives to minimize the following loss, �(X)= � (X) + � (X) unary pairwise (1) Zhou et al. Molecular Brain (2023) 16:4 Page 6 of 16 Statistical analysis in all 5 adult mice, but only 3 mice showed L-LTP in 5 The data, either in 8 × 8 or more generally as N × N recorded middle-aged mice (Fig. 2a). The fEPSP slopes channels, is presented by channel-wise means ± SEM. of the last 30 min were 148.01 ± 5.79% in adult mice and Statistical comparisons between the two groups were 128.04 ± 6.88% in middle-aged mice (Fig. 2c). At site S2, performed using Student’s t-test to identify significant most channels showed tagging-like response after tag- differences. In all cases, *p < 0.05 was considered sta- ging TBS in all 5 adult mice. However, in middle-aged tistically significant. All statistical analysis was done mice, only 3 mice out of 5 mice showed 1 or 2 tagging using SPSS Statistics. L-LTP, and most activated channels showed the E-LTP (Fig. 2b). The fEPSP slopes were 175.20 ± 8.92% in adult mice and 114.35 ± 7.75% in middle-aged mice (t = 42.54, Results **p < 0.01, n = 5 mice for each group). The conclusion Impaired synaptic tagging LTP observed in the ACC could also be supported by analyzing the percentage of of middle‑aged mice channels with tagging-like responses over all activated We employed a 64-channel field potential record - channels. The ration of the induction of tagging LTP was ing system and analyzed the induction probability and significantly higher in the adult group (73.33 ± 4.67%) properties of synaptic tagging in vitro ACC slices from than in the middle-aged group (33.33 ± 4.24%, t = 6.446, middle-aged male mice (50–60 weeks). Two different *p = 0.023 < 0.05, Fig. 2d). channels were stimulated in different layers of the ACC (S1: superficial layer; S2: deep layer). Simultaneously, the Less recruitment of inactive responses after synaptic evoked multi-channel fEPSP around the stimulation sites tagging in middle‑aged male mice were recorded (Fig. 1a, b). Specifically, we first applied a Previous studies have shown that the strong TBS- strong TBS (4 × 5×5) to site S1, and 30 min later, a weak induced L-LTP could recruit newly-activated channels in TBS (tagging TBS, 4 × 5×1) was delivered to site S2. We the ACC [31, 35]. Consistently, we also found that there found that normal LTP can be induced by strong TBS in was clear enlargement of the response areas surrounding middle-aged mice (Fig. 1f ). There are 2 channels show - the strong TBS at site S1 both in adult mice and middle- ing E-LTP and 6 channels showing L-LTP in 8 activity aged mice (Fig. 3a). After tagging TBS on site S2, the channels around site S1. However, the synaptic tagging enlargement of response areas can be observed in adult LTP induced by weak TBS was impaired in middle-aged mice, but not in middle-aged mice (Fig. 3b). In adult mice, although there were 2 channels showing tagging- mice, the number of recruited channels both on sites S1 like response, while the other 4 activated channels only and S2 gradually increased after TBS induction (Fig. 3c, showed the E-LTP on site S2 (Fig. 1h). d, n= 5 mice). However, in middle-aged mice, the tagging A novel fEPSP signal modeling and visualization sys- TBS on site S2 failed to increase many recruited chan- tem was developed to monitor the Spatio-temporal nels. The recruited channels around site S2 are less than properties tagging LTP. We assumed that the 8 × 8 obser- those in adult mice. (t = 2.473, t = 4.750, t = 10.633, 1h 2h 3h vations from the MED64 system, corresponding to the 63 th t = 18.174, t = 22.136, *p < 0.05, **p < 0.01, n = 5 mice). 4h 5h recording channels plus the 64 channel as the stimula- The time courses of the changed fEPSP slope in the tion input, are only a sparse observation of the complete recruited channels were further shown in Fig. 3g, h. In fEPSP signal field residing in latent state space. Such these recruited channels, strong TBS-induced fEPSPs latent state space could then project into another obser- on site S1 were gradually potentiated, and the amplitude vation space with N × N resolution where N is usually finally became as large as 24.79 ± 0.81 μV in adult mice greater than 8 and was set to 256 in our experiments. In and 23.85 ± 0.79 μV in middle-aged mice at 4.5 h after this way, the originally recorded MED64 values of fEPSP TBS induction (Fig. 3g). It is worth noting that the ampli- slopes were reconstructed into high-density 3D signal tudes of fEPSPs in tagging-TBS recruited channels were sequences (Fig. 1d) As shown in Fig. 1e and g, the peak no significant difference in adult mice (23.90 ± 0.87 μV ) intensity and spatial distribution of the fEPSP slopes and in middle-aged mice (24.09 ± 0.81 μV) when the were significantly increased at 120 min and 240 min channels occurred. (Fig. 3h). after strong TBS (Fig. 1e). However, the peak intensity and spatial distribution ware slight increased at 120 min Synaptic tagging LTP in middle‑aged female mice after tagging TBS in middle-aged mice (Fig. 1g). These We further tested the synaptic tagging LTP in middle- 3D maps clearly showed that the synaptic tagging was aged female mice. Similar results were obtained in impaired in the ACC of middle-aged mice. middle-aged female mice and male mice (Fig. 4). A sum- Next, we compared synaptic tagging response in adult marized plot of the fEPSP slope showed that both strong mice (6–8 weeks) and middle-aged (50–60 weeks) mice. TBS at the site S1 and tagging TBS at the site S2 can At site S1, strong TBS induced late-phase LTP (L-LTP) Zhou et al. Molecular Brain (2023) 16:4 Page 7 of 16 Fig. 2 Synaptic tagging was observed in the ACC at different ages in male mice. Summarized plot of the fEPSP slope demonstrates that both strong TBS at site S1 and tagging TBS at site S2. Either adult male mice (n = 5 slices/5 mice) or middle‑aged male mice (n = 3 slices/3 mice) could induce L‑LTP on site S1 (a). The tagging‑like response could be induced in adult male mice (n = 5 slices/5 mice) but not in middle‑aged male mice on site S2 (n = 3 slices/3 mice) (b). The fEPSP slope in the last 30 min on site S1 and site S2 in each group (c). Percentage of activated channels that has tagging induced by tagging TBS (d) (**p < 0.01, Student’s t‑test) In adult male mice, there were 55 channels exhibited induce L-LTP in adult female mice (n = 5 mice). How- clear synaptic responses from 5 slices (5 mice) at the ever, in the middle-aged female mice, there was no tag- baseline, and 26 new channels were recruited at 4.5 h ging LTP, and the L-LTP occurred only in site S1 in 4 after strong TBS in site S1 (Fig. 4c). Meanwhile, in mid- slices from 5 mice (Fig. 4a, b). These results suggest that dle-aged male mice, only 28 channels exhibited clear syn- there is no gender difference for synaptic tagging LTP in aptic responses from 4 slices (4 mice) at the baseline, and middle-aged mice. 8 new channels were recruited at 4.5 h after strong TBS Next, we showed the spatial distribution of the active in site S1 (Fig. 4e). For tagging LTP, in adult male mice, 53 responses in the ACC before and after strong TBS and channels exhibited clear synaptic responses from 5 slices tagging TBS application across both male and female at the baseline, and 26 new channels were recruited at 4 h mice. The distribution of all observed activated channels after tagging TBS in site S2 (Fig. 4d). In middle-aged male was displayed by a polygonal diagram on a grid repre- mice, 26 channels exhibited clear synaptic responses senting the channels (Fig. 4c–j). Zhou et al. Molecular Brain (2023) 16:4 Page 8 of 16 Fig. 3 New responses were recruited after synaptic tagging in the male mice ACC. Sample slices showed the distribution of the basal activated channels (blue) and the TBS‑recruited (a) or tagging TBS (b) recruited channels (red) in adult and middle ‑aged male mice. The temporal changes of the number (c, d) and amplitude (g, h) of the recruited fEPSPs on sites S1 and S2. (*p < 0.05, **p < 0.01, n = 5 mice for each group, Student’s t‑test.). Samples showed the recruited responses induced by strong TBS and tagging TBS in adult and middle‑aged male mice (e, f) from 4 slices at the baseline, and only 3 new channels the site S2, in adult female mice, 93 channels exhibited were recruited at 4 h after tagging TBS site S2 (Fig. 4f ). clear synaptic responses from 5 slices in 5 mice at the In adult female mice, 60 channels exhibited clear syn- baseline, and 20 new channels were recruited at 4 h after aptic responses from 5 slices in 5 mice at the baseline, tagging TBS (Fig. 4h). In the middle-aged female mice, and 22 new channels were recruited at 4.5 h after strong 72 channels exhibited clear synaptic responses from 4 TBS in site S1 (Fig. 4g). In middle-aged female mice, only slices in 4 mice at the baseline, and only 1 new channel 47 channels exhibited clear synaptic responses in 4 slices was recruited at 4 h after tagging TBS (Fig. 4j). Taken from 4 mice at the baseline, and 4 new channels were together, these results suggest tagging TBS can signifi - recruited at 4.5 h after strong TBS in site S1 (Fig. 4i). For cantly enhance the spatial distribution of active responses (See figure on next page.) Fig. 4 Synaptic tagging occurred in female mice. Summarized plot of the fEPSPs slope demonstrates that both strong TBS at site S1 and tagging TBS at site S2. Either adult female mice (n = 5 slices/5 mice) or middle‑aged female mice (n = 4 slices/4 mice) could induce L‑LTP on site S1 (a). Tagging‑like response could be induced in adult female mice (n = 5 slices/5 mice) but not in middle‑aged female mice on site S2 (n = 3 slices/3 mice) (b). Polygonal diagrams of channel data in each group (4 groups, n = 5) pooled from slices with the tagging‑like response. c, e, g, i The channels were activated in the baseline (blue, pre) and at 4.5 h after strong TBS of the site S1 (red, post) in the slice. Black dots represent the 64 channels in the MED64 system. Vertical dashed lines indicate the layers in the ACC slice. d, f, h, j Polygonal diagrams of the channels of pool data that were activated in the baseline state (blue, pre) and at 4 h after tagging TBS (red, post) at site S2. Delivery of strong TBS resulted in an enlargement of the activation area in all groups, while tagging TBS recruit less or no new synaptic responses in the ACC in adult and aged groups Zhou et al. Molecular Brain (2023) 16:4 Page 9 of 16 Fig. 4 (See legend on previous page.) Zhou et al. Molecular Brain (2023) 16:4 Page 10 of 16 Fig. 5 L‑LTP was more easily induced by weak TBS in adult mice than in middle ‑aged mice. The induction of E‑LTP (a) and L ‑LTP (b) in adult male mice. Summarized plot of the fEPSP slope demonstrates that weak TBS could induce L‑LTP in adult male mice (n = 5 slices/5 mice) (c). The induction of E‑LTP by weak TBS lasts for at least 2.5 h. Individual data of E‑LTP (d) were induced by weak TBS in middle ‑aged male mice. Summarized plot of the fEPSP slope demonstrates that weak TBS could induce E‑LTP in middle ‑aged male mice (n = 5 slices/5 mice) (e). The induction of E‑LTP (f) and L‑LTP (g) by weak TBS in adult female mice. Summarized plot of the fEPSP slope demonstrates that weak TBS could induce L ‑LTP in adult female mice (n = 5 slices/5 mice) (h). Individual data of E‑LTP (i) were induced by weak TBS in middle ‑aged female mice. Summarized plot of the fEPSP slope demonstrates that weak TBS could induce E‑LTP in middle ‑aged female mice (n = 5 slices/5 mice) (j). The bar plots are a quantitative representation of the number of activated channels induced by weak TBS in different ages and genders (k) in the adult mice, but failed to change spatial distribution whether L-LTP could be induced by weak TBS alone, and of active responses in middle-aged mice. whether it is age dependent or not. We attempted only delivering a single weak TBS, the same as tagging TBS (five trains of a burst with four 100 Hz pulses at 200 ms intervals), to the deep layer of Age difference of weak TBS‑induced L‑LTP within ACC ACC after obtaining a stable baseline for an hour. Inter- In the above results, it was observed that synaptic tag- estingly, it was found that L-LTP could be induced by ging induced by tagging TBS (weak TBS) was more weak TBS in some channels in adult mice. However, easily found in younger mice. Therefore, we wonder Zhou et al. Molecular Brain (2023) 16:4 Page 11 of 16 there was no L-LTP induced by weak TBS in middle- L-LTP peaks (vertices), representing the spatial strength aged mice (Fig. 5). In adult male mice, there were 99 and spatial frequency of the synaptic tagging changes in channels activated in total after weak TBS, among brain slices after the application of BDNF. This is consist - which there were 21 channels showing E-LTP, 54 chan- ent with our previous work [31] that the network LTP is nels showing L-LTP and 24 channels showing non-LTP often formed as a ring distribution surrounding the stim- (n = 5 mice, Fig. 5a–c). The ratio of L-LTP channels ulation site. was 54.55% in all activated channels (Fig. 5k). In mid- dle-aged male mice, there were 34 channels activated, TrkB agonist R13 rescued synaptic tagging LTP among which there were only 6 channels showing in middle‑aged mice E-LTP and 28 channels showing non-LTP (n = 5 mice, R13 is a prodrug for 7,8-dihydroxyflavone(7,8-DHF), Fig. 5d, e). No channel showed L-LTP in the middle- which is a flavone found in plants and has a similar func - aged mice. tion to BDNF [39, 40]. We also tested the roles of R13 in Similarly, in adult female mice, there were 91 chan- the impaired tagging LTP in middle-aged mice. R13 was nels activated, among which 22 channels were showing orally administered for 15 days in middle-aged mice, and E-LTP, 59 channels showing L-LTP, and 10 channels then the brain slices were used for fEPSPs recording as showing non-LTP (n = 5 mice, Fig. 5f–h), and the ratio above mentioned. As shown in Fig. 7, in R13-treated mid- of L-LTP channels was 65.56% (Fig. 5k). In the middle- dle-aged mice, all the brain slices from 5 mice with R13 aged female mice, there were 27 channels activated, treatment had synaptic tagging response, while no brain among which 15 channels were showing E-LTP and slices from control mice showed tagging-like response 12 channels showing non-LTP (Fig. 5i, j), and again (Fig. 7a). On site S1, we found that R13 enhanced the no channel showed L-LTP. In summary, L-LTP can be L-LTP in middle-aged mice (Fig. 7b). All the R13 treated induced by weak TBS in the ACC and shows an age- mice shown L-LTP responses after strong TBS (n = 5 dependent manner in both male and female mice. slices/ 5 mice), while only 4 mice showed E-LTP in con- trol mice. BDNF rescued synaptic tagging LTP in middle‑aged mice We also analyzed the recruited channels in R13-treated Our previous study reported that BDNF contributes to middle-aged mice. As shown in Fig. 7c, d, 4.5 h after synaptic potentiation in the ACC of adult mice (Miao strong TBS on site S1, 2.4 ± 1.4 channels were recruited in et al. [36]). In this study, we tested whether BDNF can each slice with R13 treated mice, and 1.6 ± 0.9 channels in rescue the impaired tagging LTP in middle-aged mice. untreated mice(t = 1.395, p = 0.235 > 0.05; t = 0.196, 1h 1h 2h As shown in Fig. 6a, all the BDNF–incubated brain slices p = 0.854 > 0.05; t = 0.93, p = 0.405 > 0.05; t = 1, 2h 3h 3h 4h had synaptic tagging responses in middle-aged mice, and p = 0.374 > 0.05; t = 1.089, p = 0.338 > 0.05, n = 5 4h 5h 5h 35 channels (34.3%) exhibited clear synaptic tagging LTP. mice for each group, Student’s t-test). After tagging TBS, In control mice, only 2 brain slices from 5 middle-aged 6.2 ± 2.6 channels were recruited in brain slices from mice showed tagging-like response, and only 11 chan- R13 treated mice, and it was 0.4 ± 0.4 channels for the nels (9.73%) exhibited synaptic tagging LTP. In site S1 of untreated mice (t = 1.360, p = 0.245 > 0.05; t = 0.222, 1h 1h 2h strong TBS, it was also found that BDNF improved the p = 0.835 > 0.05; t = 1.037, p = 0.359 > 0.05; t = 1.6 2h 3h 3h 4h L-LTP in the middle-aged mice (Fig. 6b). All the BDNF 01, p = 0.185 > 0.05; t = 2.558, *p = 0.043 < 0.05, n = 5 4h 5h incubated brain slices had L-LTP after strong TBS, while mice for each group, Student’s t-test.). These results sug - only 3 brain slices from 5 mice showed L-LTP in no gest that R13 enhances the network propagation of syn- BDNF application slices. aptic tagging responses in the ACC of middle-aged mice. In addition, the change of tyrosine kinase receptor B In addition, as revealed in Fig. 7e, f, where the Spatio- (Trk B) and cAMP-response element binding protein temporal distribution of synaptic plasticity response after (CREB) were also tested in the middle-aged mice slices R13 oral administration was visualized as a 3D surface. after the application of BDNF. The result from the west - Taken together, these results show that the TrkB agonist ern blot showed that both the Trk B and the CREB in the rescued synaptic tagging and LTP in middle-aged mice. ACC increased after BDNF incubation in the middle- aged mice (Fig. 6c, d). These results are consistent with Discussion the previous reports [37, 38]. The ACC is a critical forebrain structure involved in a Next, we compared the Spatio-temporal distribution variety of high-level brain functions, such as pain per- of the synaptic tagging LTP after BDNF in middle-aged ception, memory storage, and emotional processing. mice by using the developed modeling and visualiza- Synaptic plasticity in the ACC has been proven to be the tion system (Fig. 6e, f ). After incubating BDNF, it is then key cellular and synaptic substrate for chronic pain, fear, observed that the adopted tagging-TBS exhibits multiple and anxiety [12, 41, 42]. In the present study, we report Zhou et al. Molecular Brain (2023) 16:4 Page 12 of 16 Fig. 6 BDNF rescued the synaptic tagging in middle‑aged mice. BDNF was applied in the brain slices of middle ‑aged male mice. Tagging LTP recorded around site S2 from the brain slices of the BDNF applied /unapplied in middle‑aged male mice (a). LTP was recorded around site S1 from the brain slices of the same mice (b). Trk B, CREB levels were detected by western blot in three groups, adult mice, middle‑aged mice, and BDNF‑applied middle ‑aged mice (c). The intensity level of Trk B and CREB in each group, n = 5 for each group (d). The Markov Random Field model was introduced to smooth and visualize the low‑resolution 8 × 8 MED64 data. The temporal evolving of the 3D surface on site S2 on the brain slice of the control middle‑aged male mouse (e). The temporal evolving of the 3D surface on site S2 on the brain slice of the BDNF‑incubated middle‑aged male mouse (f) Zhou et al. Molecular Brain (2023) 16:4 Page 13 of 16 Fig. 7 R13 rescued the synaptic tagging in middle‑aged mice. Tagging was recorded around site S2 from the brain slices of the R13 treated/ untreated middle‑aged male mice (a). LTP around S1 was recorded around site S1 from the same brain slices (b). The temporal changes of the average number on sites S1 and S2 of the recruited channels in R13 treated and untreated mice. n = 5 slices from 5 R13 treated mice and n = 4 slices from 5 untreated mice, respectively (c, d). (*p < 0.05, n = 5 mice for each group, Student’s t‑test.). The temporal evolving of the 3D surface on site S2 of the R13 untreated mouse brain slice (e). The temporal evolving of the 3D surface on site S2 of the R13 treatment mouse brain slice (f) that synaptic tagging, hetero-synaptic plasticity, exists found that TrkB agonists, both BDNF and R13, rescued in adult excitatory synapses of the ACC. We found that the impaired synaptic tagging in middle-aged mice. Our synaptic tagging LTP in the ACC was impaired in both study provides strong evidence that impaired cortical male and female middle-aged mice. The loss of synaptic synaptic tagging may contribute to memory dysfunction tagging in mice indicates its relevance to the physiology in aged mice. changes in the brain of middle-aged mice. In addition, we Zhou et al. Molecular Brain (2023) 16:4 Page 14 of 16 Loss of synaptic tagging in middle‑aged mice One possible explanation is that, younger mice have bet- Many studies have reported on the basic mecha- ter synaptic plasticity that favors a lower fire threshold, nisms and behavioral relevance of synaptic tagging while the induction protocols we used might be over in the hippocampus [43–46]. Synaptic tagging seeks strong and therefore biased over L-LTP. This result was to explain how neural signaling at a particular syn- also reported in previous work [51]. This result gives us a apse creates a target for subsequent plasticity-related new way to define the strong effect of the TBS protocol. product (PRP) trafficking, which is essential for sus - tained LTP. Recently, our group found that synaptic TrkB agonist rescued synaptic tagging in middle‑aged tagging also occurred in the ACC [18]. Loss of synap- mice tic tagging in ACC due to amputation may contribute BDNF plays a diverse, and broad, role in regulating neu- to injury-related cognitive changes and phantom limb ronal structure and function in the central nervous sys- sensation and pain. It was also discovered that the tem [52]. Our previous study reported that BDNF can spatial extent of tagging LTP propagation was greatly contribute to synaptic potentiation in the ACC of adult dependent on the interval time window. However, it mice (Miao et al. [36]). In present studies, we adopt has been unclear to this point if synaptic tagging cor- BDNF to explore whether it can contribute to synap- relates with aging. tic tagging in the ACC or not. We found that BDNF can In this study, we first demonstrated that synaptic tag - rescue the impaired synaptic tagging and network LTP ging channels can perform in the ACC synapses in mid- in middle-aged mice. Previous studies reported that 7,8- dle-aged mice. The synaptic tagging LTP was impaired in DHF can imitate the function of BDNF [39, 40]. R13, the ACC in middle-aged mice. Both the network L-LTP as an optimal prodrug of 7,8-DHF, increases the half- and the recruitment of inactive responses were reduced. life, oral bioavailability, and brain exposure of 7,8-DHF As age increases, the ACC synapses may have some phys- [53]. We found that similar to BDNF, R13 also revised iological change in response to stimulation, which leads the impaired synaptic tagging and network LTP in mid- to a change in synaptic tagging. Considering the impor- dle-aged mice. These results showed that R13 may be a tance of ACC synapses in pain perception, fear memory, potential therapeutic agent for the treatment of memory and anxiety [2, 12, 47, 48], these findings suggest that loss, which is consistent with the previous reports [53]. synaptic tagging may have a more general physiological function in the brain. It is of great interest to further elu- BDNF‑TrkB signaling in middle‑aged mice cidate the molecular mechanisms of this cortical synaptic In the present study, we found that BDNF and a TrkB tagging and to dissect its behavioral relevance. receptor agonist R13 both rescued aged-related reduc- tion of synaptic tagging. These results indicate that No differences between gender in synaptic tagging BDNF-related signaling pathway maybe altered dur- Gender differences in responses to brain diseases have ing aging. In consistent with this observation, reduced been reported in both animal and human studies. In the BDNF levels and altered BDNF-TrkB signaling have been present work, we find that there is no significant differ - reported in aged mice. For example, total BDNF mRNA ence in synaptic tagging in the ACC between male and and the expression level of TrkB receptor were reduced female mice. This is similar to findings in our previous in 12-month-old rat hippocampus [54]. It is known that work [49], which indicated that there is no gender-related BDNF is expressed in the hippocampus and cortex in the difference in LTP in the ACC. However, in AD patients, brain, and can be synthesized by both neurons and glial some reports have shown that synaptic plasticity and cells. While a previous report showed that BDNF gene associative memory impairments are more prominent in expression in microglia is at very low level in the adult females due to the difference in hippocampus LTP [50]. brain [55], Zhou et al. reported that microglial BDNF These inconsistent results of gender differences in syn - deletion prevents high-frequency stimulation-induced aptic plasticity may be due to the different regions of the LTP [56]. Future studies are clearly needed to determine brain or various induction paradigms employed. the exact cellular mechanism for BDNF and its release in the ACC of both adult and aged animals. L‑LTP induced by weak TBS In the present study, it is quite likely that BDNF-TrkB Our previous studies found that weak TBS only induced signaling contributes to the rescuing of synaptic tag- declining E-LTP in the ACC for 2–3 h in adult mice [18]. ging by regulating the AMPAR expression and function, We wonder whether weak TBS was able to induce L-LTP including synaptic AMPAR subunit trafficking and phos - and whether it was age-dependent. However, it was found phorylation. Our previous study has found that selective that L-LTP could be induced by weak TBS only in adult phosphorylation of AMPAR at the PKA phosphoryla- mice, but not (or much more rarely) in middle-aged mice. tion site serine 845 contributes to the network of LTP Zhou et al. Molecular Brain (2023) 16:4 Page 15 of 16 expression in the ACC [31]. Furthermore, Miao et al. recruitment of cortical circuitry in the ACC. Application found that BDNF-induced enhancement in the ACC is of TrkB agonist can rescue the synaptic tagging in mid- dependent on L-VGCC, calcium-stimulated adenylyl dle-aged mice. Since L-LTP is important for the process cyclase subtype 1(AC1) and postsynaptic incorporation of anxiety and fear memory, further studies should inves- of calcium-permeable AMPA receptors (CP-AMPARs) tigate the detection of the relationship between strong [36]. In the hippocampus, it has also been reported that TBS site and tagging TBS site in synaptic tagging, as well BDNF-TrkB signaling enhances synaptic transmission by as an expanded application such as the possible changed upregulating the AMPAR subunit trafficking on the post - spatial–temporal properties of network L-LTP in anxiety synaptic membrane [57, 58]. and fear memory. Acknowledgements The novel fEPSP signal modeling and visualization system The authors would like to thank Yong‑Min Liu for his construction of the In our research, by using the MED64 system, the LTP experiment. could be recorded from up to 64 channels for one ACC Author contributions slice, as compared to the traditional methods that MZ, SBZ, and XHL designed the experiments. SBZ, MX, WL, and QYC per‑ are based on two-electrode (one for stimulation and formed behavioral experiments and analyzed data. JSL and YXC perform the western blot. JW built the fEPSP signal modeling and visualization system. KY one for recording). Unfortunately, to obtain a more provided R13 reagents. SBZ, XHL, and MZ drafted the manuscript and finished comprehensive understanding of the Spatio-tempo- the final version of the manuscript. All authors read and approved the final ral dynamics of signals and responses in a real net- manuscript. work, the number of available activations is only 8 × 8 Funding which is still insufficient. To overcome the constraint, M.Z. is in part supported by grants from the Canadian Institute for Health we consider the original MED64 recording only as a Research (CIHR) project grants (PJT‑148648 and 419286). X.H.L. is supported by grants from the National Science Foundation of China (32100810) and the sparse observation from a latent state space that could China Postdoctoral Science Foundation (2022M710111). be either discrete or continuous, and by estimating the values of the state variables, we can easily project Availability of data and materials All data generated or analyzed during this study are included in this published them back into a smoother observation space such that article. visualizing and analyzing the Spatio-temporal dynam- ics of activation and response becomes easier. In our Declarations previous work, we attempted modeling such a pro- cess for a single time frame with a Gaussian Mixture Ethics approval and consent to participate Animal experiments followed protocols approved by the Animal Care and Use Model [31]. In this paper, we generalize the problem Committee of Xi’an Jiaotong University. with a state space model (“A novel fEPSP signal mod- eling and visualization system” section), and by adopt- Consent for publication Not applicable. ing the graph theory, we can build a Markov Random Field based on the MED64 observation to estimate the Competing interests latent state variable, and then project the original 8 × 8 The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. input into a higher resolution output (256 in our exper- iments). Such an observation-state-projection pipe- line is generic enough to take observations from both Received: 23 November 2022 Accepted: 27 December 2022 single time frame (the spatial case) as well as multiple time frames (the Spatio-temporal case), and can also project into a single higher resolution estimation (this References paper) as well as multiple higher resolution estima- 1. Bliss and Collingridge. Expression of NMDA receptor‑ dependent LTP in tions (on-going work). In fact, we believe that our pro- the hippocampus: bridging;the divide. J Molecular Brain. 2013;6(1):5–5. posed model aligns well with the unknown mechanism 2. Bliss TV, et al. Synaptic plasticity in the anterior cingulate cortex in acute and chronic pain. Nat Rev Neurosci. 2016;17(8):485–96. of ACC where the recorded MED64 signal (the obser- 3. Asok A, et al. Molecular mechanisms of the memory trace. Trends Neuro‑ vation) depends on the behavior of the ACC neurons sci. 2019;42(1):14–22. (the state), and once we obtain an accurate estimation 4. Chen T, et al. Adenylyl cyclase subtype 1 is essential for late‑phase long term potentiation and spatial propagation of synaptic responses in the of the state value, we can further estimate responses anterior cingulate cortex of adult mice. Mol Pain. 2014;10(1):65. at a finer granularity (the high-resolution projection). 5. Park P, et al. Calcium‑permeable AMPA receptors mediate the induction Such a generic model could apply to many other similar of the protein kinase a‑ dependent component of long‑term potentiation in the hippocampus. J Neurosci. 2016;36(2):622. scenarios. 6. Li XY, et al. Alleviating neuropathic pain hypersensitivity by inhibiting In summary, our results demonstrate that middle-aged PKMzeta in the anterior cingulate cortex. Science. 2010;330(6009):1400–4. mice have less ability to induce the synaptic tagging and Zhou et al. Molecular Brain (2023) 16:4 Page 16 of 16 7. Frey U, Morris RG. Synaptic tagging and long‑term potentiation. Nature. 36. Miao HH, et al. Brain‑ derived neurotrophic factor produced long‑term 1997;385(6616):533–6. synaptic enhancement in the anterior cingulate cortex of adult mice. Mol 8. Frey U, Morris RGM. Synaptic tagging: implications for late maintenance Brain. 2021;14(1):140. of hippocampal long‑term potentiation. Trends Neurosci. 1998;21(5):181. 37. Caldeira MV, et al. BDNF regulates the expression and traffic of NMDA 9. Reymann KG, Frey JU. The late maintenance of hippocampal LTP: require‑ receptors in cultured hippocampal neurons. Mol Cell Neurosci. ments, phases, “synaptic tagging”, “late‑associativity” and implications. 2007;35(2):208–19. Neuropharmacology. 2007;52(1):24–40. 38. Leal G, Comprido D, Duarte CB. BDNF‑induced local protein synthesis and 10. Park P, et al. On the Role of Calcium‑Permeable AMPARs in Long‑ Term synaptic plasticity. Neuropharmacology. 2014;76:639–56. Potentiation and Synaptic Tagging in the Rodent Hippocampus. Front 39. Wurzelmann M, Romeika J, Sun D. Therapeutic potential of brain‑ derived Synaptic Neurosci. 2019;11:4. neurotrophic factor (BDNF) and a small molecular mimics of BDNF for 11. Zhuo M. Cortical excitation and chronic pain. Trends Neurosci. traumatic brain injury. Neural Regen Res. 2017;12(1):7–12. 2008;31(4):199–207. 40. Liu C, Chan CB, Ye K. 7,8‑ dihydroxyflavone, a small molecular TrkB agonist, 12. Zhuo M. Neural Mechanisms Underlying Anxiety‑ Chronic Pain Interac‑ is useful for treating various BDNF‑implicated human disorders. Transl tions. Trends Neurosci. 2016;39(3):136–45. Neurodegener. 2016;5:2. 13. Vogt BA. Pain and emotion interactions in subregions of the cingulate 41. Zhao MG, et al. Roles of NMDA NR2B subtype receptor in prefron‑ gyrus. Nat Rev Neurosci. 2005;6(7):533–44. tal long‑term potentiation and contextual fear memory. Neuron. 14. Tang J, et al. Pavlovian fear memory induced by activation in the anterior 2005;47(6):859–72. cingulate cortex. Mol Pain. 2005;1:6. 42. Koga K, et al. Coexistence of two forms of LTP in ACC provides a synaptic 15. Vetere G, et al. Spine growth in the anterior cingulate cortex is necessary mechanism for the interactions between anxiety and chronic pain. for the consolidation of contextual fear memory. Proc Natl Acad Sci U S A. Neuron. 2015;85(2):377–89. 2011;108(20):8456–60. 43. Young JZ, et al. Metaplasticity of the late‑phase of long‑term potentia‑ 16. Frankland PW, Teixeira CM. A pain in the ACC. Mol Pain. 2005;1:14. tion: a critical role for protein kinase A in synaptic tagging. Eur J Neurosci. 17. Tanimizu T, et al. Functional connectivity of multiple brain regions 2006;23(7):1784–94. required for the consolidation of social recognition memory. J Neurosci. 44. Ramachandran B, Frey JU. Interfering with the actin network and its effect 2017;37(15):4103–16. on long‑term potentiation and synaptic tagging in hippocampal CA1 18. Liu MG, Song Q, Zhuo M. Loss of synaptic tagging in the anterior neurons in slices in vitro. J Neurosci. 2009;29(39):12167–73. cingulate cortex after tail amputation in adult mice. J Neurosci. 45. Sajikumar S, Korte M. Metaplasticity governs compartmentalization of 2018;38(37):8060–70. synaptic tagging and capture through brain‑ derived neurotrophic factor 19. Rogerson T, et al. Synaptic tagging during memory allocation. Nat Rev (BDNF) and protein kinase Mzeta (PKMzeta). Proc Natl Acad Sci U S A. Neurosci. 2014;15(3):157–69. 2011;108(6):2551–6. 20. Kastellakis G, et al. Synaptic clustering within dendrites: an emerging 46. Dasgupta A, et al. Substance P induces plasticity and synaptic tag‑ theory of memory formation. Prog Neurobiol. 2015;126:19–35. ging/capture in rat hippocampal area CA2. Proc Natl Acad Sci U S A. 21. Luboeinski J, Tetzlaff C. Memory consolidation and improvement by 2017;114(41):E8741–9. https:// doi. org/ 10. 1073/ pnas. 17112 67114. synaptic tagging and capture in recurrent neural networks. Commun 47. Keum S, et al. A Missense variant at the Nrxn3 locus enhances empathy Biol. 2021;4(1):275. fear in the mouse. Neuron. 2018;98(3):588‑601.e5. 22. Bergado JA, Lucas M, Richter‑Levin G. Emotional tagging–a simple 48. Sellmeijer J, et al. Hyperactivity of anterior cingulate cortex areas 24a/24b hypothesis in a complex reality. Prog Neurobiol. 2011;94(1):64–76. drives chronic pain‑induced anxiodepressive ‑like consequences. J Neuro ‑ 23. Vecsey CG, Huang T, Abel T. Sleep deprivation impairs synaptic tagging in sci. 2018;38(12):3102–15. mouse hippocampal slices. Neurobiol Learn Mem. 2018;154:136–40. 49. Liu RH, et al. Sex difference in synaptic plasticity in the anterior cingulate 24. Bach ME, et al. Age‑related defects in spatial memory are correlated with cortex of adult mice. Mol Brain. 2020;13(1):41. defects in the late phase of hippocampal long‑term potentiation in vitro 50. Navakkode S, et al. Sex‑specific accelerated decay in time/activity‑ and are attenuated by drugs that enhance the cAMP signaling pathway. dependent plasticity and associative memory in an animal model of Proc Natl Acad Sci U S A. 1999;96(9):5280–5. Alzheimer’s disease. Aging Cell. 2021;20(12): e13502. 25. Wimmer ME, et al. Aging impairs hippocampus‑ dependent long‑term 51. Radulescu CI, et al. The aging mouse brain: cognition, connectivity and memory for object location in mice. Neurobiol Aging. 2012;33(9):2220–4. calcium. Cell Calcium. 2021;94: 102358. 26. Shoji H, et al. Age‑related changes in behavior in C57BL/6J mice from 52. Panja D, Bramham CR. BDNF mechanisms in late LTP formation: a synthe‑ young adulthood to middle age. Mol Brain. 2016;9:11. sis and breakdown. Neuropharmacology. 2014;76:664–76. 27. Wong LW, et al. Age‑related changes in hippocampal‑ dependent synap‑ 53. Chen C, et al. The prodrug of 7,8‑ dihydroxyflavone development and tic plasticity and memory mediated by p75 neurotrophin receptor. Aging therapeutic efficacy for treating Alzheimer’s disease. Proc Natl Acad Sci U Cell. 2021;20(2): e13305. S A. 2018;115(3):578–83. 28. Gros A, Wang SH. Behavioral tagging and capture: long‑term memory 54. Perovic M, et al. BDNF transcripts, proBDNF and proNGF, in the cortex decline in middle‑aged rats. Neurobiol Aging. 2018;67:31–41. and hippocampus throughout the life span of the rat. Age (Dordr). 29. Bramham CR, Messaoudi E. BDNF function in adult synaptic plasticity: the 2013;35(6):2057–70. synaptic consolidation hypothesis. Prog Neurobiol. 2005;76(2):99–125. 55. Bennett ML, et al. New tools for studying microglia in the mouse and 30. Xue M, et al. NMDA receptor‑ dependent synaptic depression in potenti‑ human CNS. Proc Natl Acad Sci U S A. 2016;113(12):E1738–46. ated synapses of the anterior cingulate cortex of adult mice. Mol Pain. 56. Zhou LJ, et al. Microglia are indispensable for synaptic plasticity in the 2021;17:17448069211018044. spinal dorsal horn and chronic pain. Cell Rep. 2019;27(13):3844‑3859.e6. 31. Song Q, et al. Selective phosphorylation of AMPA receptor contributes to 57. Zhang H, et al. Modulation of AMPA receptor surface diffusion restores the network of long‑term potentiation in the anterior cingulate cortex. J hippocampal plasticity and memory in Huntington’s disease models. Nat Neurosci. 2017;37(35):8534–48. Commun. 2018;9(1):4272. 32. Li XH, et al. Oxytocin in the anterior cingulate cortex attenuates neuro‑ 58. Keifer J. Regulation of AMPAR trafficking in synaptic plasticity by pathic pain and emotional anxiety by inhibiting presynaptic long‑term BDNF and the impact of neurodegenerative disease. J Neurosci Res. potentiation. Cell Rep. 2021;36(3): 109411. 2022;100(4):979–91. 33. Kindermann R, Snell JL. Markov random fields and their applications. American Mathematical Society; 1980. Publisher’s Note 34. Turkowski K. Filters for common resampling tasks. Graphics Gems; 1990. Springer Nature remains neutral with regard to jurisdictional claims in pub‑ p. 147–65. lished maps and institutional affiliations. 35. Chen T, et al. Pharmacological rescue of cortical synaptic and network potentiation in a mouse model for fragile X syndrome. Neuropsychop‑ harmacology. 2014. https:// doi. org/ 10. 1038/ npp. 2014. 44.
Journal
Molecular Brain – Springer Journals
Published: Jan 6, 2023
Keywords: Synaptic tagging; LTP; BDNF; R13; ACC; Middle-aged mice