WO2018039816A1

WO2018039816A1 - Neural network electrical activity detection system and screening method for neuropsychiatric drugs based on system

Info

Publication number: WO2018039816A1
Application number: PCT/CN2016/000576
Authority: WO
Inventors: 刘长亮; 高明; 江成世
Original assignee: 刘长亮; 高明; 江成世
Priority date: 2016-08-30
Filing date: 2016-10-17
Publication date: 2018-03-08
Also published as: CN106190841B; CN106190841A

Abstract

A neural network electrical activity detection system and a screening method for neuropsychiatric drugs based on the system. The detection system mainly consists of a neural network system constructed in vitro, a photoelectric signal conversion system, and a monitoring and analysis system. On this basis, the detection system can also be used for theoretical research of a nervous system and early screening and testing of potential neuropsychiatric drugs. In the detection system, a neural network is constructed in vitro, which facilitates large-scale preparation; a pathological neural network model can be established, stimulation intervention is not required, the natural physiological working state of the neural network can be kept, network electrical activities can be regulated and controlled in a certain range, drug targets can be introduced by means of viruses, electric signals are converted into optical signals, a collection system is simplified, costs are reduced, high-throughput screening is easy to implement, high-throughput data analysis is performed by means of targeted software, labor costs are reduced, and the present invention has a very good prospect.

Description

Neural network electrical activity detection system and screening method of neuropsychiatric drugs based on the same

Technical field

The invention belongs to the technical field of biomedicine, and particularly relates to a neural network electrical activity detecting system and a screening method of a neuropsychiatric drug based on the system.

Background technique

Neuropsychiatric diseases (including neurodegenerative diseases, epilepsy, depression, anxiety, schizophrenia, bipolar disorder, addiction, etc.) account for about 14% of all human diseases, affecting 2.5-3.0% of the global population And its incidence is increasing year by year (The Lancet. 2007, 370, 859; Plos Med. 2006, 3, e442). Due to the complexity of the nervous system itself, the imperfection of the basic theory, and the lack of an effective and reliable screening platform, the cycle and risk of neuropsychiatric drug development are significantly higher than those of other drugs (Neuron. 2014, 84, 546). . The statistical results show that only 8% of pre-clinical test drugs can finally pass clinical trials. As a result, many pharmaceutical companies have reduced their R&D investment in neuropsychiatric drugs in recent years to avoid risks (Net Rev Drug Discov. 2015, 14,815). However, human needs for mental health and quality of life are inseparable from the support of more and better drugs. Therefore, the market needs a more accurate and efficient pre-clinical neuropsychiatric evaluation platform to speed up the development of such drugs.

At present, there are four main methods for pre-clinical development and screening of neuropsychiatric drugs (Neuron. 2014, 84, 546; Net Rev Drug Discov. 2015, 14, 815; Neuropharmacology research techniques and methods. People's Medical Publishing House 2015, ISBN:9787117066051):

1. Screening for biochemical levels for specific targets.

This method is based primarily on our knowledge of specific receptors in the nervous system to screen for compounds that interact with the receptor. For example, if a study shows that a receptor has the effect of treating a disease, then the compounds that interact with the receptor theoretically have the potential to develop into a drug. Disadvantages of this method: 1) The theoretical level lacks effective support. The role of specific receptors in the nervous system is not well defined due to differences in neuronal and receptor subtypes and differences in expression levels. Secondly, for the vast majority of neuropsychiatric diseases, we currently lack effective and reliable drug targets, so even if the compound has a clear interaction with the receptor, it can not predict its function in the nervous system; 2) the rate of missed detection is very high. Since the screening system only performs binding rate tests for specific targets, new targets cannot be found. In addition, since compounds often bind to multiple receptors, simply detecting the ability of a compound to bind to a receptor does not exclude whether the compound interacts with other receptors.

2. Cell level screening

The method evaluates the efficacy by culturing nerve cells in vitro by detecting the effects of the drug on the morphology or molecular parameters of the nerve cells (eg, survival rate, specific protein levels, etc.). This technology has the advantage of high throughput and high repeatability. But because Most neurological diseases do not have reliable markers at the protein or gene level, so the detection indicators are vague and poorly guided, not functional screening.

3. Animal behavior screening.

The method mainly observes the human disease by establishing an animal pathological model in vivo, and then tests the effect of the drug on the specific behavioral paradigm of the animal to observe the efficacy. The disadvantages of this technique are: 1) The animal's nervous system is different from humans. For advanced cognitive function diseases (such as schizophrenia, autism, etc.) can not be effectively simulated; 2) long cycle, poor stability; 3) there is no clear guiding significance of the mechanism of action of the drug to be selected.

4. Screening for neuronal electrical activity.

The nervous system is a functional network formed by the interconnection of a large number of neurons. Neurons communicate and function through electrical activity and chemical transmitters. This method mainly uses this property to test the effect of drugs on nerve electrical activity by electrophysiological means (extracellular recording, patch clamp, etc.). This technology belongs to the functional screening of nervous system, with high sensitivity, stability and good repeatability. However, the prior art system is complicated, the cost is extremely high, and the efficiency is low, and it is necessary to continue to improve and optimize.

Summary of the invention

In view of the problems existing in the prior art, the present invention provides a detection system for neural network electrical activity and a screening method for neuropsychiatric drugs based on the system. The detection system simulates the physiological and pathological state of the nervous system by establishing a neural network in vitro and generating self-generated activity, and overcomes the way of relying on external stimulation to induce electrical activity of the neurons. The conversion of specific neuronal electrical activity signals into optical signals simplifies the complex techniques and specialized equipment required for current neuronal electrical activity monitoring, and enables the monitoring and analysis of large-scale neuronal electrical activity at the single cell level. At the same time, the invention can be used for theoretical research of the nervous system and early screening and testing of potential neuropsychiatric drugs.

The invention adopts the following technical solutions:

A neural network electrical activity detecting system, which is composed of a neural network system, an optical signal conversion system and a monitoring and analysis system constructed in vitro;

The in vitro constructed neural network system is constructed by the following method:

A. Obtaining primary neurons from the brain of newborn mice

1) The whole operation is performed under aseptic conditions. According to the experimental needs, the neurons can be taken from the cortex, hippocampus, striatum, midbrain or multi-brain;

2) Newborn mice, 2-3 groups, were disinfected with 70% alcohol, and placed on ice for 5 minutes, then the head was broken, the brain was taken, and the corresponding brain tissue was taken under dissection and placed at 0-4 °C. 10 ml/2 mice in HBS buffer;

3) Wash the brain tissue twice with HBS buffer, add tissue lysate and place in a cell incubator at 37 ° C, 5% CO ₂ for 20 minutes;

4) vacuum aspirate the tissue lysate and rinse the tissue with MEM buffer twice;

5) adding plating buffer to the tissue in an amount of 2 ml/2 mice and pipetting 40 times with a pipette;

6) filtering the tissue suspension with a cell filter to obtain suspended primary neurons;

B. Primary neuron plating

1) The coverslip was coated on the surface with an extracellular matrix, and then placed in an incubator at 37 ° C, 5% CO ₂ for 1 hour;

2) After the surface treatment is finished, the suspended primary neurons obtained in step A are uniformly plated onto the coverslip, the cell density is 200-1000/mm ² , and then placed in an incubator for 1 hour;

3) adding the plating buffer and placing it in the incubator;

C. Follow-up

1) 12 hours after plating, replace all plating buffer as growth buffer;

2) 48-72 hours after plating, check the density of glial cells every 2 hours, add 1-4 μM cytarabine when the growth density of glial cells is 60%-80%;

3) On the 3-5th day after plating, virus infection is carried out;

4) 6-10 days, mature synapse formation, neural network growth;

5) After 14 days, the network is mature and the in vitro neural network system is completed.

The photoelectric signal conversion system is carried out by the following method:

1) Viral design: The lentiviral or adenoviral vector was transformed by molecular cloning. The Synapsin promoter was used to express the target protein only in neurons but not in glial cells. The 2D expression system was used to simultaneously express tdTomato red fluorescent protein and calcium ion. Agent protein, the expression level of both is strictly in accordance with 1:1, red fluorescent protein is used to indicate the position of a single neuron, and calcium ion indicator protein is used to detect electrical activity;

2) Virus preparation: The vector plasmid and packaging plasmid, 600 ug DNA/75 ml culture dish were co-transduced in the HEK293 cell line by calcium phosphate transfection. The density of HEK293 cells was 60-70% when transfected, and the growth buffer was replaced the next day; For lentivirus, the supernatant was collected for 72 hours after transfection to obtain virus replacement; for adenovirus, HEK293 cell bodies were collected 72 hours after transfection, and lysed by freeze-thaw method, incubated at 37 ° C / -40 ° C for 3 times each time. After 10 min, the DNA/RNase was added after centrifugation, and the cells were incubated at 37 ° C for 30 minutes, and the supernatant was taken to obtain virus for use; the obtained virus was filtered through a 0.2 μm filter, and then stored at -80 ° C;

3) virus infection: 3-5 days after the primary neuron plating, add 50-100μl / ml step 2) prepared virus, so that the final titer of the virus is 10 ⁸ -10 ¹² VP / ml;

4) 3-5 days after the virus infection, the target protein begins to be expressed, and a single fluorescent neuron labeled single neuron can be seen under a fluorescence microscope;

The monitoring and analysis system is composed of two parts: video acquisition and data processing;

The pH of the HBS buffer is 7.3;

The tissue lysate is configured in an amount of 10 ml and is configured from the following components:

10ml HBS buffer

10ul 0.5M EDTA pH 8.0

10ul 1M CaCl ₂

100ul papain

100ul deoxyribonuclease;

The plating buffer, prepared in an amount of 1 L, is prepared from the following components:

900ml MEM buffer

25ml 20% glucose

2.5ml 8% NaHCO ₃

100mg transferrin

100ml FBS

10ml 200mM L-Glutamine

25mg insulin;

The growth buffer, prepared in an amount of 1 L, is prepared from the following components:

900ml MEM buffer

25ml 20% glucose

2.5ml 8% NaHCO ₃

100mg transferrin

50ml FBS

20ml cell culture additive B 27

2.5 ml of 200 mM L-glutamine.

The specific method of the video collection is as follows:

1) 15-18 days after plating, remove the coverslip loaded with the constructed neural network and place in the standard extracellular fluid for 10 minutes, using a thermostat to control the temperature at 25-30 ° C;

2) Place the neural network under a fluorescence microscope, using a 4x objective lens, with a field of view of 2 mm x 2 mm, including 200-400 neurons, imaging with an EMCCD or CMOS system, and activating the calcium indicator protein with a 470 nm LED or a 488 nm laser;

3) Using the transistor logic circuit to synchronously control the light source and image acquisition system, the exposure time is 100-500ms, the acquisition frequency is 2.5s/frame, and the acquisition system finally generates the TIFF format file, the single image pixel 2048x2048;

4) Record the basic network electrical activity for 1 hour using the above method;

The standard formula for extracellular: 140mM NaCl, 5mM KCl, 2mM MgCl 2, 2mM CaCl 2, 10mM HEPES, 10mM Glucose, pH 7.4.

The specific method of data processing is:

1) Stabilizing the acquired image using the Lucas-Canad algorithm to eliminate the displacement caused by long-term recording;

2) subtract the background of the image;

3) Using two-dimensional Dajin exhibition algorithm to identify the individual neurons and generate a coordinate set to be analyzed;

4) analyzing the fluorescence intensity of each neuron to generate a phase diagram;

5) Statistical analysis of the phase diagram, the main analysis contents include: discharge intensity, discharge time history, oscillation frequency, overall electrical activity, correlation between single neurons, peak time of single neuron discharge, peak velocity , oscillating activity propagation path, propagation speed, individual neurons non-oscillation spontaneous activity 10 indicators.

A screening method for neuropsychiatric drugs by using the above-mentioned neural network electrical activity detecting system, which is to continue to adopt the same after adding the drug to be tested after recording the network basic electrical activity in the video collecting part of the monitoring and analysis system. Methods The electrical activity of the neural network system was recorded and the data was processed according to the system method. The phase analysis data before and after the drug were compared to obtain the drug screening results. The basic oscillating electrical activity of the neural network system was adjusted according to the purpose of the selected drugs. The specific control methods are as follows:

1) Adjust the standard extracellular fluid to a calcium ion concentration of 1-8 mM to control signal intensity and network conduction efficiency;

2) Adjusting the concentration of magnesium ions in the standard extracellular fluid at 0-4 mM to control the open rate of the NMDA receptor, thereby controlling the network excitability;

3) adjusting the concentration of potassium ions in the standard extracellular fluid at 0-40 mM to control the membrane potential of the neurons;

4) Add specific receptor agonists or blockers to control network activity, including: acetylcholine receptor, dopamine receptor, glutamate receptor, GABA receptor, calcium channel, potassium channel, Sodium ion channel, kinase.

The theoretical basis of the present invention is that the human nervous system is a complex network composed of a large number of neurons (~10 ¹¹ ) and glial cells (~10 ¹⁴ ) interconnected. The basic mode of activity of a neuron is to generate an action potential (a short-term discharge behavior; typically lasting 1-5 ms). A single neuron receives the information transmitted by the superior neuron through the dendrites and computes and integrates it, and then sends the signal to the next neuron by generating an action potential. The normal work of the nervous system relies on mutual coordination and smooth communication between neurons. The communication between neurons is controlled and regulated by various factors: for example, the degree of ion channel opening, the amount of receptor expression and activation, and the level of transmitter and tempering will have a significant impact on the electrical activity of neurons.

Neuropsychiatric diseases are a group of neurological diseases characterized by disorders in behavior and mental activity. Their causes are often diverse: either due to a decline in the nervous system or death of specific neurons (eg, Alzheimer's and Parkinson's disease); others due to abnormal development or connection of the nervous system (eg epilepsy and autism) Symptoms; there are functional disorders due to transmitter or receptor levels (eg, schizophrenia and addiction). But regardless of the causes of these diseases, Their common feature is that neurons cannot communicate properly. Drugs that fight these diseases, without exception, need to play a therapeutic role by affecting the electrical activity of neurons at the network level. At present, research on neural network electrical activity mainly relies on electrophysiological methods (such as patch clamp, extracellular recording, EEG recording, etc.). These methods are difficult to achieve high-throughput screening of potential drugs due to their technical complexity. Therefore, a rapid and accurate method for evaluating the effects of compounds or biological products on the electrical activity of neural networks is of great significance for promoting the development of neuropsychiatric drugs.

The present invention constructs a neural network by using a method of culturing neurons in vitro. Neurons grow under specific conditions in vitro to form synapses with each other to form a network. When the network connection reaches a certain scale and there are pacing neurons, the in vitro neural network can generate spontaneous oscillating activity (a regular cluster discharge behavior). Once the network oscillating activity is formed, the in vitro neural network can integrate and calculate information without external intervention (stimulus). Using this property, we have large-scale construction of neural networks with functional electrical activities in vitro. The basic unit of activity of a neuron is the action potential (produced by sodium, calcium ion transmembrane motion). Along with the action potential is the inflow of a large amount of calcium ions. Since calcium ion levels are positively correlated with neuronal electrical activity levels, the concentration of calcium ions in neurons can be quantified by calcium ion indicators. A calcium ion indicator is a fluorescent dye that binds to calcium ions, some are chemical agents, and some are proteins. These indicators change in spectral properties (eg, fluorescence enhancement) upon binding to calcium ions. Based on this principle, we convert the electrical activity of neurons into light signals by means of calcium ion indicators. The present invention uses a virus-infected neuron to introduce a calcium ion indicator protein (GCaMP6, which was introduced in 2013 and is currently the most sensitive indicator protein in the world, and this protein has enhanced fluorescence intensity after binding with calcium ions) to convert electrical signals. After the electrical activity of the neurons is converted into a light signal, the electrical activity of the neurons can be acquired by a fluorescence microscope imaging system to generate video data. The invention optimizes the optical acquisition system, can monitor neuron activity on a large scale, and generate video data. Based on this, the present invention develops related software to perform targeted analysis on video data. The system is used for early screening and testing of potential neuropsychiatric drugs.

In addition, the function and application range of the present invention can be expanded in various aspects: First, the in vitro neural network construction can also be directly prepared using GCaMP6 transgenic mice, so that the virus infection step can be omitted. Calcium ion imaging can also be replaced with chemical dyes. To achieve targeted drug screening, the present invention can express a candidate drug target in a particular neuron by a virus, and these targets can be receptors, ion channels or enzymes.

In order to achieve drug screening for specific diseases, it is necessary to simulate neural network electrical activity under pathological conditions. The present invention mainly adopts the following measures to construct a pathological neural network.

1) Primary neurons are obtained directly from pathological model animals for culture to form a pathological network.

2) Infect healthy neurons with viral systems (mainly lentiviruses and adenoviruses) and introduce specific pathogenic genes.

3) Add a specific drug or change the ion environment, temperature, pH, etc. in a normal neural network.

The beneficial effects of the invention are: 1) constructing a neural network in vitro to facilitate large-scale preparation. 2) A pathological neural network model can be established. 3) Maintaining the natural physiological working state of the neural network without stimulating intervention. 4) can carry out network electrical activities A range of adjustments and controls. 5) Drug targets can be introduced by viruses. 6) Converting electrical signals into optical signals simplifies the acquisition system and saves costs. 7) Easy to achieve high throughput screening. 8) Targeted software for high-throughput data analysis to reduce labor costs.

DRAWINGS

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a schematic illustration of a neural network system constructed in vitro in the basic composition of the detection system of the present invention. Among them, A-D is the morphology of neurons at various time points after plating. A is the first day after the boarding, B is the third day after the boarding, and C is the fifth day after the boarding, and the virus is added on this day. The 14th day is a red fluorescent protein fluorescence imaging (D) showing individual neurons and well-developed axons and dendrites.

2 is a schematic diagram of a photoelectric signal conversion system in the basic composition of the detection system of the present invention. Among them, virus packaging (A1), neuronal infection (A2) and signal acquisition optical path (B), wherein B1 is a neural network, B2 is a light source, B3 is a dichroic mirror, B4 is a synchronous control and signal acquisition, and B5 is an imaging system.

3 is a schematic diagram of a monitoring and analysis system in the basic composition of the detection system of the present invention. The figure is an example of calcium ion indicator protein (GCaMP6) imaging, basic electrical activity level (top) and post-administration (bottom) electrical activity (A), and data processing schematic (B).

4 is a laser confocal imaging image of the indicated protein using the immunofluorescence cytochemistry method of the mature neural network constructed in Example 1 of the present invention.

Fig. 5 is a view showing an example of imaging of a self-generated active calcium ion indicator protein of a mature neural network constructed in Example 1 of the present invention. The original image (top) and the enlarged image (bottom) have an exposure time of 400ms.

6 is a flow chart of data processing of the monitoring and analysis system in Embodiment 2 of the present invention, wherein 1 is image stabilization processing and subtracting background, 2 is single neuron recognition, 3 is defining a quantization region, and 4 is for each Time-phase analysis is performed on the quantized areas, and 5 is the parameter measurement and a statistical report is generated.

Fig. 7 is a diagram showing the results of an example of the regulation of oscillating electrical activity in a neurological activity detecting system constructed by the present invention. In the figure, A is the level of basic activity, and B is the level of electrical activity after the addition of picrotoxin.

detailed description

The content of the present invention will be further described in detail below with reference to specific embodiments.

Example 1

A neural network electrical activity detecting system, which is composed of a neural network system, a photoelectric signal conversion system and a monitoring and analysis system constructed in vitro, and the specific composition and workflow diagram are shown in FIG. 1;

A. Obtaining primary neurons from the brain of newborn mice

1) The whole operation is carried out under aseptic conditions. According to the experimental needs, the neurons can be taken from the cortex, hippocampus, striatum, Mixed in the midbrain or multi-brain;

B. Primary neuron plating

3) adding the plating buffer and placing it in the incubator;

C. Follow-up

1) 12 hours after plating, replace all plating buffer as growth buffer;

3) On the 3-5th day after plating, virus infection is carried out;

4) 6-10 days, mature synapse formation, neural network growth;

The pH of the HBS buffer is 7.3;

10ml HBS buffer

10ul 0.5M EDTA pH 8.0

10ul 1M CaCl ₂

100ul papain

100ul deoxyribonuclease;

900ml MEM buffer

25ml 20% glucose

2.5ml 8% NaHCO ₃

100mg transferrin

100ml FBS

10ml 200mM L-Glutamine

25mg insulin;

900ml MEM buffer

25ml 20% glucose

2.5ml 8% NaHCO ₃

100mg transferrin

50ml FBS

20ml cell culture additive B 27

2.5 ml of 200 mM L-glutamine.

The specific method of the video collection is as follows:

The specific method of data processing is:

2) subtract the background of the image;

After constructing a mature neural network system, the present invention uses the immunofluorescence cytochemistry method to perform laser confocal imaging of the indicated protein, as shown in FIG. 2, in which the PQ-type calcium channel is a calcium ion channel unique to the synapse. The gamma-aminobutyric acid vesicle transporter is an inhibitory neuronal synaptic marker and the glutamate vesicle transporter is an excitatory synaptic marker. The presence of these proteins indicates that the in vitro neural network already has a complete synaptic structure. Since synapses are the basic unit of communication between neurons, the formation of a large number of synapses suggests that neural networks in vitro have become quite complex.

In addition, the present invention also studies the self-generating activity of the in vitro neural network in the system, and the specific results are shown in FIG. It can be seen from the figure that the neural network constructed in vitro has a small number of neurons in the resting state and has self-generating activity (left; bright point-like objects). At intervals of time (3-10 min), the neural network will explode a cluster discharge (middle; a large number of neurons become brighter). These activities can be completely blocked by tetrodotoxin (sodium channel blocker, which blocks the release of neuronal action potentials) (right), indicating calcium signal-dependent action potentials. This also indicates that the recorded calcium signal represents network electrical activity.

Example 2

A screening method for a neuropsychiatric drug using the neural network electrical activity detecting system described in Embodiment 1, which is to add a drug to be tested after recording the basic electrical activity of the neural network in the video capturing part of the monitoring and analysis system. Then continue to use the same method to record the electrical activity of the neural network system and process the data according to the system method, compare the phase-phase map analysis data before and after the drug to obtain the drug screening results, and adjust the neural network system according to the purpose of the selected drugs. The basic oscillating electrical activities, the specific control methods are as follows:

In the drug screening process of the present invention, the specific data processing workflow is shown in FIG. 4 . In the figure, 1 is an example of the original image, and each bright spot in the figure is a neuron. The captured image is stabilized using the Lucas-Kanade algorithm to eliminate displacement caused by long-term recording. 2. After subtracting the background of the image, the 2D-OTsu's method is used to identify the individual neurons to generate a binary image. The black point object is the identified individual neuron. 3. Generate a set of position coordinates to be analyzed based on the identified neuron coordinates. 4. Analyze the fluorescence intensity of each quantized region to generate a phase diagram. For example, using 0.4Hz to capture images for 5 hours, this will generate 7200 high-resolution photos, quantify the corresponding areas of the 7200 photos, then take the quantized value as the ordinate, and collect the time point as the abscissa for the phase. Figure. 5. An example of a statistical report generated by a computer after statistical analysis. Data before (open circles) before dosing (closed circles) were compared.

The invention can regulate the basic oscillating electrical activity according to different drug screening targets. For example, a bitter taste (a drug that blocks inhibitory transmission) is added to the neural network, and as a result, as shown in Fig. 5, the drug can significantly enhance the intensity and frequency of discharge activity. In practice, the present invention utilizes the drug to enhance the signal and to create a pathological model (this drug is a tool for the manufacture of animal models of epilepsy in vivo).

Due to the different purposes of screening, the indicators observed for specific drugs to be selected will also be emphasized. For example, when screening anti-epileptic drugs, it is hoped that the basic oscillating activity is stronger, so that the effect of the drug is more easily reflected, thereby increasing the sensitivity of the system. This requires us to be able to regulate the oscillators. The invention can realize the regulation of the basic oscillating activity to improve the sensitivity of the system, and thus can well screen the neuropsychiatric drugs.

Claims

A neural network electrical activity detecting system, characterized in that it is composed of a neural network system, an optical signal conversion system and a monitoring and analysis system constructed in vitro;

The in vitro constructed neural network system is constructed by the following method:

A. Obtaining primary neurons from the brain of newborn mice

1) The whole operation is performed under aseptic conditions. According to the experimental needs, the neurons can be taken from the cortex, hippocampus, striatum, midbrain or multi-brain;

2) Newborn mice, 2-3 groups, were disinfected with 70% alcohol, and placed on ice for 5 minutes, then the head was broken, the brain was taken, and the corresponding brain tissue was taken under dissection and placed at 0-4 °C. 10 ml/2 mice in HBS buffer;

3) Wash the brain tissue twice with HBS buffer, add tissue lysate and place in a cell incubator at 37 ° C, 5% CO 2 for 20 minutes;

4) vacuum aspirate the tissue lysate and rinse the tissue with MEM buffer twice;

5) adding plating buffer to the tissue in an amount of 2 ml/2 mice and pipetting 40 times with a pipette;

6) filtering the tissue suspension with a cell filter to obtain suspended primary neurons;

B. Primary neuron plating

1) The coverslip was coated on the surface with an extracellular matrix, and then placed in an incubator at 37 ° C, 5% CO 2 for 1 hour;

2) After the surface treatment is finished, the suspended primary neurons obtained in step A are uniformly plated onto the coverslip, the cell density is 200-1000/mm 2 , and then placed in an incubator for 1 hour;

3) adding the plating buffer and placing it in the incubator;

C. Follow-up

1) 12 hours after plating, replace all plating buffer as growth buffer;

2) 48-72 hours after plating, check the density of glial cells every 2 hours, add 1-4 μM cytarabine when the growth density of glial cells is 60%-80%;

3) On the 3-5th day after plating, virus infection is carried out;

4) 6-10 days, mature synapse formation, neural network growth;

5) After 14 days, the network is mature and the in vitro neural network system is completed.

The photoelectric signal conversion system is carried out by the following method:

1) Viral design: The lentiviral or adenoviral vector was transformed by molecular cloning. The Synapsin promoter was used to express the target protein only in neurons but not in glial cells. The 2D expression system was used to simultaneously express tdTomato red fluorescent protein and calcium ion. Agent protein, the expression level of both is strictly in accordance with 1:1, red fluorescent protein is used to indicate the position of a single neuron, and calcium ion indicator protein is used to detect electrical activity;

2) Virus preparation: co-transfection of vector plasmid and packaging plasmid in HEK293 cell line by calcium phosphate transfection, 600ug DNA/75ml culture dish, the density of HEK293 cells was 60-70% when transfected, and changed to growth buffer the next day; for lentivirus, the supernatant was collected after 72 hours of transfection to obtain virus replacement; for adenovirus, transfection 72 After hours, the cells of HEK293 cells were collected and lysed by freeze-thaw method, incubated at 37 ° C / -40 ° C for 3 times, 10 min each time. After centrifugation, DNA/RNase was added, incubated at 37 ° C for 30 minutes, and the supernatant was taken to obtain virus. Spare; the virus obtained was filtered through a 0.2 μm filter, and stored at -80 ° C;

3) virus infection: 3-5 days after the primary neuron plating, add 50-100μl / ml step 2) prepared virus, so that the final titer of the virus is 10 8 -10 12 VP / ml;

4) 3-5 days after the virus infection, the target protein begins to be expressed, and a single fluorescent neuron labeled single neuron can be seen under a fluorescence microscope;

The monitoring and analysis system is composed of two parts: video acquisition and data processing;

The pH of the HBS buffer is 7.3;

The tissue lysate is configured in an amount of 10 ml and is configured from the following components:

10ml HBS buffer

10ul 0.5M EDTA pH 8.0

10ul 1M CaCl 2

100ul papain

100ul deoxyribonuclease;

The plating buffer, prepared in an amount of 1 L, is prepared from the following components:

900ml MEM buffer

25ml 20% glucose

2.5ml 8% NaHCO 3

100mg transferrin

100ml FBS

10ml 200mM L-Glutamine

25mg insulin;

The growth buffer, prepared in an amount of 1 L, is prepared from the following components:

900ml MEM buffer

25ml 20% glucose

2.5ml 8% NaHCO 3

100mg transferrin

50ml FBS

20ml cell culture additive B 27

2.5 ml of 200 mM L-glutamine.
The neural network electrical activity detecting system according to claim 1, wherein the specific method of video capturing is as follows:

1) 15-18 days after plating, remove the coverslip loaded with the constructed neural network and place in the standard extracellular fluid for 10 minutes, using a thermostat to control the temperature at 25-30 ° C;

2) Place the neural network under a fluorescence microscope, using a 4x objective lens, with a field of view of 2 mm x 2 mm, including 200-400 neurons, imaging with an EMCCD or CMOS system, and activating the calcium indicator protein with a 470 nm LED or a 488 nm laser;

3) Using the transistor logic circuit to synchronously control the light source and image acquisition system, the exposure time is 100-500ms, the acquisition frequency is 2.5s/frame, and the acquisition system finally generates the TIFF format file, the single image pixel 2048x2048;

4) Record the basic network electrical activity for 1 hour using the above method;

The standard formula for extracellular: 140mM NaCl, 5mM KCl, 2mM MgCl 2, 2mM CaCl 2, 10mM HEPES, 10mM Glucose, pH 7.4.
The neural network electrical activity monitoring system according to claim 1, wherein the data processing specific method is:

1) Stabilizing the acquired image using the Lucas-Canad algorithm to eliminate the displacement caused by long-term recording;

2) subtract the background of the image;

3) Using two-dimensional Dajin exhibition algorithm to identify the individual neurons and generate a coordinate set to be analyzed;

4) analyzing the fluorescence intensity of each neuron to generate a phase diagram;

5) Statistical analysis of the phase diagram, the main analysis contents include: discharge intensity, discharge time history, oscillation frequency, overall electrical activity, correlation between single neurons, peak time of single neuron discharge, peak velocity , oscillating activity propagation path, propagation speed, individual neurons non-oscillation spontaneous activity 10 indicators.
A method for screening a neuropsychiatric drug using the neural network electrical activity detecting system according to claims 1-3, characterized in that, after recording the network basic electrical activity in the video capturing portion of the monitoring and analysis system, After adding the drug to be tested, continue to record the electrical activity of the neural network system by the same method and process the data according to the system method, and compare the data of the phase phase analysis before and after the drug to obtain the drug screening result, according to the purpose of the selected drug. To regulate the basic oscillating electrical activity of the neural network system, the specific control methods are as follows:

1) Adjust the standard extracellular fluid to a calcium ion concentration of 1-8 mM to control signal intensity and network conduction efficiency;

2) Adjusting the concentration of magnesium ions in the standard extracellular fluid at 0-4 mM to control the open rate of the NMDA receptor, thereby controlling the network excitability;

3) adjusting the concentration of potassium ions in the standard extracellular fluid at 0-40 mM to control the membrane potential of the neurons;

4) Add specific receptor agonists or blockers to control network activity, including: acetylcholine receptor, dopamine receptor, glutamate receptor, GABA receptor, calcium channel, potassium channel, Sodium ion channel, kinase.