Transcriptome Profiling Analysis Of The Mechanisms Underlying The BDNF Val66Met Polymorphism Induced Dysfunctions Of The Central Nervous System

Background:The human brain, which can be called the "central processor" of the nervous system, plays an central role in controlling our feeling, emotion, memory, motor movement and logical reasoning. With the development of human society, the brain is suffering from many serious diseases, such as epilepsy, Parkinson’s, Alzheimer’s disease, schizophrenia and depression, all of which are serious threat to human health and life. The neurotrophin family is a kind of widely expressed proteins in the brain playing important roles in promoting cell differentiation, neuronal growth and survival and maintaining normal brain function. The metabolic abnormalities of neurotrophins are usually closely related to development of many neurological diseases. Among these neurotrophins, the brain-derived neurotrophic factor (BDNF) is currently one of the mostly studied proteins.BDNF is one of the most widely distributed neurotrophins in the central nervous system and is widely expressed during the brain development and in adulthood. As a secreted protein, BDNF regulate the structure and function of neurons mainly through binding two different receptors:a high affinity receptor TrkB and a low affinity receptor p75. The binding of BDNF and TrkB can activate intracellular MAPK/ERK, PLCγ and PI3K signaling pathways, thus regulating the axonal growth, neuronal survival and differentiation, synaptic plasticity and learning and memory. The binding of BDNF and p75receptor usually play negative roles in regulating neuronal function, such as the induction of neuronal apoptosis and the neural plasticity induced by depression. Studies have shown that BDNF is involved in a variety of learning and memory processes,such as the Morris water maze, contextual fear memory and so on. In addition, BDNF is also associated with the development of many neurological diseases. Expression changes of BDNF in specific neurons will lead to different diseases, such as depression, epilepsy, Alzheimer’s disease, schizophrenia, Huntington and Parkinson’s disease. These data suggest that BDNF plays an important role in maintenance of the normal central nervous system function.Recently, a common single-nucleotide polymorphism (SNP) has been identified in the human BDNF gene leading to a valine (Val) to methionine (Met) substitution at position66in the prodomain (Val66Met). This variant leads to decreased trafficking of BDNF into secretory granules and thus results in reduced activity dependent BDNF secretion. Individuals with this gene polymorphism usually have significantly impaired structure of the nervous system, such as decreased hippocampal volume and synaptic plasticity. Furthermore, BDNFMet variant is also associated with increased vulnerability to develop anxiety, depression and memory disorders. However, the relationship between this gene polymorphism on BDNFMet with other features (such as spatial memory, working memory) and the potential molecular mechanisms underlying the BDNFMet variant induced structural and functional changes in the central nervous system are still unclear.The construction of BDNF Val66Met transgenic mice provides us with a good experimental model and will help us further understanding the mechanisms involved in the BDNFMet associated behavior and neuroanatomy alternations. In this paper, we aimed to investigate the impact of BDNFMet variant on individual behaviors. Then, by using a massive microarray approach, we systematically examined the gene expression changes across the dorsal hippocampus (HPC), ventromedial prefrontal cortex (PFC) and amygdala (AMY) regions of BDNFMet variant mice brain compared with wild type (WT) and found the potential molecular mechanisms underlying the BDNFMet variant induced structural and functional alterations. Through analysis of the microarray data, we found the chemokine signaling CX3CL1/CX3CR1and the cell adhesion signaling PCDH1were significantly changed in the HPC of BDNFMet variant mice, and then we performed further studies on the potential functions and mechanisms involved in these alterations. This will not only help us to further understand the intracellular and extracellular BDNF signaling, but also to explore new strategies of treatment on BDNFMet variant induced dysfunctions.Objective:1. Examining the impact of BDNFMet variant on spatial and working memory.2. To find the potential molecular mechanisms underlying the BDNFMet variant induced dysfunctions.3. To find the potential effect of CX3CL1on the treatment of memory dysfunctions induced by BDNFMet variant.4. To investigate the expression pattern of PCDH1in the central nervous system and its potential biological functions on synaptic plasticity.Methods:1. Morris Water MazeMorris water maze was employed to evaluate the effect of BDNFMet variant on spatial memory performance. In the hidden platform stage, mice were trained six days to find a hidden platform under the water. The time spent in finding the platform was used to evaluate the acquire ability of mice on spatial memory during training. In the probe test, the platform was removed, time spent in the target quadrant and the numbers of target platform crossings were acquired. 2. T Maze TestT maze test was employed to evaluate the effect of BDNFMet variant on working memory performance. In the training phase, mice were trained in a session of six successive trials to foster the alteration behavioral pattern. Mice were trained two sessions daily and each session included six consecutive alternate behavior training. During test, all the mice were exposed to6successive trials. The alteration rate expressed in percentage was determined and was used as an index of working memory performance.3. Open Field TestSpontaneous exploratory activity was assessed in the open field test. For the test, mice were placed in the center and the total distance traveled in10min was recorded as a measure of locomotor activity.4. Contextual Fear ConditioningMice were fear conditioned in a Panlab mouse fear conditioning chamber. On conditioning day, mice were conditioned with three trials with a2s,0.7mA foot shock. Each trial was separated by a60s inter-trial interval. After the final foot shock, mice remained in the conditioning chamber for1min and were returned to their home cages. Freezing behavior was defined as the complete lack of movement except for respiration.Memory for the contextual dependent fear was tested24h later. On the testing day, mice were placed back to the training chamber described above and were tested without foot shock for5min. Memory was assessed by calculating the percentage of freezing during5min in the training context.5. Quantitative RT-PCRTissues of each brain regions were dissected and the total RNA was extracted. RNA purity and integrity were controlled by OD260/OD280≥1.8and agarose gel electrophoresis with the ratio of28S/18SS1.5. Quantitative real time RT-PCR was performed in a Cycler using SYBR-Green. Each sample was assayed in duplicate and the mRNA levels were normalized for each well to the (3-actin mRNA levels using the2-△△CT method. 6. Western BlotSamples of mice brain were isolated and homogenized. Samples were subjected to SDS-PAGE and transferred to a polyvinylidene difluoride membrane. Membranes were blocked and incubated with primary antibodies and second antibodies and were detected using the ECL chemiluminescence system. Immunoreactive bands were scanned and density quantification was performed using NIH Image J software.7. Surgical Procedures and MicroinjectionMice were anesthetized with5%chloral hydrate, restrained in a stereotaxic apparatus and implanted bilaterally with guide cannula aimed at dentate gyrus (DG) of the dorsal HPC. Mice were allowed two weeks to recuperate and then CX3CL1was microinject to examine the following results:(1) whether CX3CL1can improve the memory performance of BDNFMet/Met mice;(2) whether CX3CL1can promote the survival of newborn neurons in DG of BDNFMet/Met mice.8. Bromodeoxyuridine Administration and ImmunohistochemistryAfter recovered from surgery for two weeks, WT and BDNFMet/Met mice received four intraperitoneal injections of BrdU (50mg/kg) during a4-day time period and were allowed to survive for28days before sacrifice. To estimate the effect of CX3CL1on survival of newborn neurons in BDNFMet/Met mice, CX3CL1(100ng/μl,1μl/side/day) was injected into the HPC beginning at day21or27for single or consecutive7days, after the last BrdU injection. Immunohistochemistry was conducted on floating sections with two primary antibodies including sheep anti-BrdU and mouse anti-NeuN. All BrdU and NeuN double stained cells within the granule cell layer of the DG were counted.9. Microarray Hybridization and Data AnalysisSamples were hybridized to the Mouse WG-6v2.0Expression Bead Chip. The chips were processed according to the manufacturer’s instructions recommended by Illumina. The microarray hybridization work was accomplished by the Shanhai Biotechnology Co., Ltd. Next we performed the gene screening and signaling pathway analysis with the Gene Ontology and KEGG database.Results:1.Transcriptome profiling analysis of the BDNF Va166Met polymorphism induced gene expression and functional alterations in the central nervous system1.1BDNFMetMet mice exhibited impaired spatial and working memoryMorris water maze test was employed to evaluate the effect of BDNFMet variant on spatial memory performance. During the training session, there were no significant differences in escape latencies between BDNF+/Metand WT mice. However, the BDNFMet/Met mice exhibited significantly increased escape latency compared with WT controls, suggestive of a poorer acquisition of spatial memory. In the probe trial, we observed that the BDNF+/Met mice had a trend and BDNFMet/Met mice showed significantly decreased time spent in the target quadrant and number of target platform crossings. In the T maze test, BDNFMet/Met b(jt not BDNF+/Met mjce were sjgnjfjcant|y poOrer in the alteration performances compared with WT mice. These data suggest that the BDNFMet variant could lead to the spatial and working memory impairments in a gene-dose dependent manner.1.2Region-distinctive and gene-dose dependent gene expression and pathway alterations in BDNF Val66Met MiceWe used the Illumina bead microarray across the HPC, PFC and AMY regions in BDNF+/Met, BDNFMet/Met and WT mice. By comparing with the WT controls, we identified148,116, and150genes to be regulated in the HPC, PFC and AMY respectively in BDNF+/Met mice and767,449,476genes in each region of BDNF Met/Met mice.We then analyzed genes regulated across HPC, PFC and AMY of BDNFMet/Met mice using two forms of annotational clustering, GO biological process and KEGG pathway analysis. We observed both common and specific pathways were regulated across these brain regions, suggesting that the BDNFMet variant induced structural and functional changes are highly orchestrated biological processes needing the interaction and integration of multiple neural circuits.1.3The CX3CL1/CX3CR1*signaling was down regulated in HPC of BDNFNlet/Met mjceWe examined the expression levels of CX3CL1and CX3CR1in the HPC, PFC and AMY of BDNFMet/Met mice. Quantitative RT-PCR and western blot analysis revealed a significantly decreased mRNA and protein expression of CX3CL1and CX3CR1in the HPC of BDNFMet/Met mice compared with WT mice. Interestingly, the decreased mRNA levels of CX3CL1and CX3CR1were only found in the HPC but not in the PFC and AMY. These results suggest that the CX3CL1/CX3CR1signaling is selectively impaired in the HPC of BDNFMet/Met mice.1.4Administration of CX3CL1rescued the hippocampal dependent memory deficits in BDNFMet/Met miceChronic CX3CL1injection was effective to rescue the impaired contextual fear memory in BDNFMet/Met mice. CX3CL1treated BDNFMet/Met mice showed significantly increased freezing time compared with the vehicle treated BDNFMet/Met mjc6j suggestjng tnat CX3CL1could rescue the impaired contextual fear memory induced by the BDNFMet variant. Moreover, chronic CX3CL1microinfusion showed a tendency to rescue the deficit in spatial memory acquisition in BDNFMet/Met mice during training, however, did not reach significant difference. In the probe trial, BDNFMet/Met mice treated with CX3CL1showed significantly increased time spent in the target zone and number of target platform crossings compared with the vehicle treated BDNFMet/Metmice, suggesting that CX3CL1was able to rescue the memory retention deficit observed in BDNFMet/Metmice. To exclude the influence of CX3CL1infusion on motor ability, we examined the spontaneous locomotor activity in vehicle and CX3CL1treated WT and BDNFMetyMet mice in the open field test. There were no significant differences in total distance traveled among these groups. Overall, our results suggest that microinjection of CX3CL1into the HPC could rescue the hippocampal dependent memory deficits induced by the BDNFMet variant.1.5Chronic CX3CL1administration reversed the decreased hippocampal neurogenesis in BDNFMet/Met miceWe want to know whether administration of CX3CL1into the HPC could rescue the decreased hippocampal neurogenesis in the BDNFMet/Met mice. Quantification of the number of BrdU and NeuN double labeled cells in the granule cell layer of DG revealed a significantly decreased number of BrdU positive neurons in BDNFMet/Met mice, indicating reduced survival of newborn neurons induced by the BDNFMet variant. However, the decreased survival of newborn neurons could be rescued by chronic CX3CL1microinfusion into the HPC, suggesting an important role of CX3CL1in increasing the maturation and survival of newborn neurons in BDNFMet/Met mice, which may contribute to the improved memory performance.2. Involvement of PCDH1in the BDNFMet variant induced impairments in synaptic plasticity2.1Expression of PCDH1was decreased in the HPC of BDNFMet/MetmiceWe examined the expression level of PCDH1in the HPC of BDNFMet/Met mice. Quantitative RT-PCR and western blot analysis revealed a significantly decreased mRNAand protein expression of PCDH1in the HPC of BDNFMet/Met mice compared with WT mice, but the expression of PCDH7, PCDH8, PCDH9, PCDH10, PCDH17, PCDH19and PCDH20, which are also highly expressed in the HPC, were not changed. Furthermore, we also examined the mRNA expression of SynCAMI, SALM3, Nectin3, Cdh2, Nlgni and NCAM1between the HPC of BDNFMet/Met and WT mice. These molecules are synaptic cell adhesion molecules involved in the regulation of synaptogenesis and synaptic plasticity, and their expression were also unchanged in the HPC of BDNFMet/Met mice. 2.2Expression of PCDH1in the mouse brain during postnatal development and in adultThe expression of PCDH1steadily increased during the postnatal brain development, a pattern similar to that of postsynaptic density protein95kDa (PSD95). The level of PCDH1was significantly increased on postnatal day7(P7) and reached a peak on postnatal day21(P21), and consistently: maintained high levers of expression. In the adult mouse brain, PCDH1was widely distributed and was detected in various subregions of mouse brain, including HPC, PFC, AMY,, striatum (STR), entorhinal cortex (ENT) and lateral septum (LS), whereas in the subregions of hypothalamus (HT) and cerebellum (CBI), the expression of PCDH1was relatively low.Conclusion:Our data provide several new insights into phenotypes and mechanisms associated with the BDNFMet variant.1. BDNFMet/Met but not BDNF+/Met variant will lead to the impairments of spatial and working memory in mice.2. This is the first study to investigate the potential mechanisms underlying the BDNFMet variant induced dysfunctions of the central nervous system using a genome-wide microarray approach.3. The down-regulated CX3CL1/CX3CR1signaling in the HPC might be an important factor contributing to the hippocampal dependent memory deficits in the BDNFMet/Met mice.4. PCDH1is widely expressed in subregions of the mouse brain, and is significantly decreased in the HPC of BDNFMet variant mice.Significance:The altered gene expression profiles will help us further understanding the potential mechanisms involved in the BDNFMet associated behavior and neuroanatomy alternations and will help us to find new treatments for BDNFMet variant induced dysfunctions.