Research on establishing the neural stimulation system and apply for evaluating the spatial response of hippocampal place cells

a system for stimulating and recording the electrical activity of neurons based on electronics engineering. - Building mathematical algorithms of neuronal stimulation for 4 practical exercises on mice. 3. Research subjects and scope - Research subjects: Neurons in a defined place of the Hippocampus is electrically stimulated and recorded by using the built system. Supporting devices and integrated programs for stimulating and recording are synchronous to form a complete system. The stimulating and recording processes are automatically performed and analyzed. - Research scope: Developing a system that allows stimulating and recording neurons. The stimulating programs associated with mathematical algorithms are integrated into the system. Simulation and analyzing the results are based on electronic engineering. 4. Research Methodology Data collecting programs, simulations, and practical exercises are used on mice for evaluating the system and exercises. 5. Research contents and thesis structure The main research contents: - The overview of the electrical activity of neurons - Modeling neurons with equivalent circuits and building algorithms of electrical simulations for neurons - Evaluating the algorithms and system by simulation programs and practical exercises on mice. 6. Scientific and practical significance - Proposing proper stimulating parameters for studying subjects. - Developing a system for stimulating and recording the electrical activity of neurons with 4 algorithms and 4 practical exercises on animals. - Performing simulations and practical exercises on mice to evaluate the proposed system and programs - Providing fundamentals of medical issues for studying the central nervous system. 3 CHAPTER 1: THE OVERVIEW OF THE ELECTRICAL ACTIVITY OF NEURONS 1.1. Membrane potential of neurons Neurons are analogous to other cells, which have structural components of cell membranes, nuclei and organelles. The electrical activity of normal cells as well as neurons is highly related to the structure and characteristics of the cell membrane. 1.2. Electrical nerve stimulation and medical significance The development of nerve stimulating and recording system with proper algorithms is based on studying electrical properties of the cell membrane, the influence of electrical stimulating parameters, the response of cell membranes, and electrical stimulations in medical research. Figure 1.1. The change in membrane potential by the influence of stimulating pulses. 1.3. The response of cell membrane to the electrical stimulation The plasma membrane potential changes when neurons are stimulated. The membrane potential will return to its initial resting value after responding to the stimulus. If the electrical stimulation is insufficient to create a transmembrane potential larger than a threshold, the membrane 4 will not be activated. The amplitude and frequency of the electrical stimulation mainly influence the intracranial electrical stimulation, which are used to determine the stimulating threshold and maximum response of cells. In this work, electrical stimulating pulses are positive pulses with their variable amplitudes and frequencies. 1.4. The recording methods of the neuronal action potential The neuronal potential recording technique was developed in the 1940s. During this period, extracellular microelectrodes were used to determine the potential characteristics of a neuron. Recent studies of neurons associated with the neural stimulation and response have proven the relation of neurons in different places within brain. Current technologies and equipment have also been developed for more accurate and convenient analyses of neuronal activities. 1.5. Hippocampus and hippocampal place cells Neuron studies have proven that hippocampal place cells play a vital role in information store, short-term to long-term memory conversion and spatial orientation. 1.6. The basic electronic circuit of neurons To more comprehensively study and understand the action potential of cell membrane on electric stimulation, an equivalent conducting model of neurons has been modeled as an electronic circuit. 1.7. Relevant research - National: currently, there is not any neuronal stimulating system, which allows both stimulating and recording the electrical activity of neurons. - Global: systems of manual neuronal stimulation and recording have been often found, however most of them are not synchronized and complete systems, which leads to the lack of accuracy in data analysis. Some advanced systems can only describe the system function and stimulation results, but not evaluate the system. 1.8. Chapter conclusion 5 CHAPTER 2: THE EQUIVALENT ELECTRICAL CIRCUITS OF NEURONS AND ALGORITHMS FOR ELECTRICAL NEURAL STIMULATION. 2.1. The electronic circuit of neuron membrane and the investigation of electrical stimulating parameters The action potential of cell membrane can be analogously modelled as an electronic circuit. 2.1.2. The simulation of electrical stimulating parameters with the Maeda-Makino model Figure 2.3. The electronic circuit model of neurons by Maeda and Makino. 2.1.3. Simulation results and discussions The amplitude and frequency parameters of stimulation pulses are applied to the electric circuit for simulating the action membrane potential based on the Maeda – Makino model by the NI Multisim 14.0 progam. 2.1.3.1. The relationship between current and the membrane potential of stimulation pulses at a fixed 80Hz frequency The simulation results show that the membrane potential is directly proportional to the stimulating intensity (Figure 2.6). However, the potential only dramatically rises over the current of less than about 10μA (the "bursting" range of potential response) before gradually increasing in the current range from 10 to 110 μA. In addition, while the stimulating current increases 110 μA, the potential increases suddenly and oscillates. R1 100kΩ R2 200Ω R3 100kΩ XSC1 A B Ext Trig + + _ _ + _ C1 0.5µF Q1 2N3904 Q2 2N3906 Q5 2N3904 V1 5V R4 1kΩ R8 100kΩ R6 300Ω XMM1 I1 0.07mA C3 1µF V2 0.4V XFG2 COMSC1 PWM IO1 IO2 IO3 Q3 2N3904 Q4 2N3904 C4 0.2µF R5 10kΩ 6 Theoretically, this explains the risk of breakdown voltage of electronic components and the demolition of cell membrane. Figure 2.6. The change of potential depends on the stimulating intensity at the frequency of 80Hz. 2.1.3.2. The membrane potential depends on the frequency of stimulating current at a fixed current. Figure 2.7 shows the dependence of the membrane potential over the frequencies of current from 0 to 180 Hz when the current intensity is fixed. It can be seen that the membrane potential increases and reaches a maximum value at the frequency of 100 Hz before slightly reducing at higher frequencies. Figure 2.7. The change of potential depends on the stimulating frequency at a fixed intensity of 70μA. 7 2.2. The stimulation and recording system for electrical activity of neurons Figure 2.8. The illustration of the stimulation and recording system for electrical activity of neurons - The behavioral observation system: Consisting of a CCD camera for monitoring movements, behavior and positions of mice - The stimulation system: a pulse generator establishes the form and parameters of pulses (Stimulator), which are sent to the isolator and DAC via USB 6501 before delivering to the nerve cells of mice by stimulation electrodes. -The recording system: The neuronal membrane potential is also recorded by stimulating electrodes. The recorded signal from neurons is doubly amplified and processed by a signal processing unit (Plexon). The action potential is recorded and synchronously counted together with the stimulation pulse as well as the co-ordinates of the mice by control signals (TTL form) from the processing program installed in the computer to the Plexon system via USB device 6501. The developed system with integrated programs forms a complete device, which allows both electrically stimulating neurons and recording the neuronal membrane potential with designed task algorithms. 8 2.3. Algorithms of electrical stimulations for neurons 2.3.1.1. The model of electrical stimulations for neurons with the NPT task Isolator Stimulator USB 6501 Processing Circuit Sensor Central processing system Mice Stim ulating electrode Monitor Figure 2.10. The model of electrical stimulations and nose – poking responses. The input is the nose-poking behavior of the mice, which is transmitted via an optical sensor hiddenly located in a circular hole with the size diameter of 1.5 cm inside the chamber. The sensor operating mode is set at a high logic status. When the nose-poking happens, the sensor will be switched to a lower logic level. The response signal from the sensor is sent to a processing circuit for counting the number of nose- poking if conditions of the task are completed as described in model 2.10. 2.3.1.2. The electrical stimulations for neurons with NPT task a) The significance of the NPT exercise The NPT task algorithm is based on the strict requirements of reward conditions. The intensity and frequency parameters of electrical stimulating pulses are evaluated from practical tasks in order to compare with the simulated parameters. The NPT stimulation algorithm is shown in Figure 2.11 with variable intensity and frequency parameters. The program completely monitors the reward conditions and automatically rewards when the conditions of the task are reached. The number of rewards or nose-poking behaviour is updated and visually displayed on a bar graph. These values are stored in a file and objectively analyzed to evaluate the most appropriate parameters for spatial response tasks. The expectation of the NPT task is to find the optimal parameters of the stimulating electrical pulse, which makes mice interest and poking the most in a period time of the task. 9 b) The NPT task c) The algorithm flowchart t++; delta++; chammui = 0; pt++; ptDelta++ lưu ptDelta; ptDelta = 0; tDelta = 0 tDelta = delta yes yes no t == interval*(countInterval + 1) no no chammui =1 yes yes no Pt == maxPt || t== maxT countInterval++ Reading data Pt = 0; maxPt; chammui = 0; t = 0; maxT; ptDelta = 0; tDelta = 0; delta; countInterval = 0; interval Start Chammui End Figure 2.11. Algorithm flowchart for the NPT task. 2.3.2. The electrical stimulation models and algorithms for neurons with the spatial response task 2.3.2.1. The model of electrical stimulations for neurons with DMT task Figure 2.12. The model for the DMT task. 10 2.3.2.2. The constructions of electrical stimulation algorithms for neurons with the DMT task a) The significance of the DMT exercise The algorithm of the DMT task (Distance Movement Task) is based on strict requirements of reward conditions. The movements of mice will be trained from easy to difficult requirements by the experimental tasks for asserting the optimal intensity and frequency parameters of the stimulating electrical pulses. Those parameters were already determined in the NPT task in the aforementioned part. b) The DMT task c) Algorithm flowchart for the DMT task Pt = 0; s = 0; t = 0; xt-1 = x0; yt-1 = y0; xt = x0; yt = y0; maxpt; maxT; delta t++; xt-1 = xt; yt-1 = yt s+ = sqrt[(yt-yt-1) 2 + (xt-xt-1) 2] s = 0 Pt++ Pt = maxPt || t = maxT yes yes no no s >= delta Reading data xt; yt Start End Figure 2.13. The algorithm flowchart of stimulations for the DMT task. 11 2.3.2.3. The model of electrical stimulations for neurons with the RRPST and PLT tasks. Isolator Stimulator USB 6501 Plexon Central processing system AMP C C D c a m e ra Monitor S tim u la tin g e le c tro d e re c o rd in g e le c tro d e Mice Figure 2.14. system for stimulating and recording the electrical activity of neurons on the mice The algorithm of electrical stimulations for the RRPST task a) The significance of the RRPST exercise The algorithm of the RRPST task (Random Reward Place Search Task) is based on the strict requirements of reward conditions. The movement and reward motivation of mice are evaluated by the algorithm of electrical stimulations for building the program and content of the RRPST task. This experimental exercise will train the mice to move for searching rewards, which appear randomly. The number of rewards or moving distances will be simulated to display the tracking path of mice and to update the reward number. The obtained results are stored in a file and objectively analyzed for assessing the movements of mice in a particular space. b) The RRPST task c) The algorithm flowchart of stimulations for the RRPST task 12 Pt = 0; t= 0; xt = x0; yt = y0; xzt = xz0; yzt = yz0; wz deltaTime = 0; delayTime; maxwidth; maxPt; maxT; taovungpt = false t++; deltaTime++; delta = sqrt[(xt – xzt) 2 + (yt – yzt) 2 ] delta = 0 taovungpt = true Pt++; deltaTime =0 delta <= wz & taovungpt = false yes yes no deltaTime>= delayTime & taovungpt = true xzt = rand(0,maxwidth) yzt = rand(0,maxwidth) Pt == maxPt || t == maxT no no yes Reading data xt; yt End Start Figure 2.15. The algorithm flowchart of stimulations for the RRPST task. The algorithm of stimulations for the PLT task a) The significance of the PLT exercise The algorithms of the PLT task (Place Learning Task) is based on strict requirements of reward conditions. The movements of mice for searching fixed rewards will be trained by experimental exercises. In addition, a program is built which can strictly monitor the reward conditions and automatically reward when the conditions are reached. The number of rewards or moving distances will be simulated to display the tracking path of mice and update the reward number. The obtained 13 results are stored in a file and objectively analyzed for assessing the movements of mice in a particular space. b) The PLT task c) The algorithm flowchart of stimulations for the PLT task Pt = 0; t = 0; xz1; yz1; xz2; yz2; wz vungphanthuong = 1; deltaTime = 0; delayTime; maxPt; maxT; delta t++; deltaTime++; delta = sqrt[(xt - xz1) 2 +(yt - yz1) 2] delta = 0; deltaTime =0 Pt++; vungphanthuong = 2 delta <= wz & deltaTime => delayTime yes yes no Pt == maxPt || t== maxT no no vungphanthuong =1 delta = sqrt[(xt - xz2) 2 +(yt - yz2) 2] delta <= wz & deltaTime => delayTime delta = 0; deltaTime =0 Pt++; vungphanthuong = 1 yes yes no Reading data xt; yt Start End Figure 2.16. The algorithm flowchart of stimulations for the PLT task. 14 CHAPTER 3: EVALUATING THE STIMULATION ALGORITHMS AND THE SYSTEM BY BEHAVIOURAL RESPONSES AND PRACTICAL EXERCISES ON MICE 3.1. Materials and methods Animals: male mice weighed 26 - 29g are obtained from the Central Institute of Hygiene and Epidemiology. Electrode Implantation: monopolar stimulating electrodes, (100µm in diameter, stainless steel) are implanted into the medial forebrain bundle on both sides of the posterior lateral hypothalamic area for intracranial self-stimulation (anteroposterior, -2,3mm; mediolateral, ± 0,7 to 0,75mm; and dorsoventral, -5,3 to 5,4mm). The recording electrodes consist of 8 single electrodes, which can be implanted into the CA1 area of the Hippocampus (region (2,1mm posterior to Bregma, 1,8 mm lateral to Bregma, and 1,4mm below the skull surface). The recording electrodes are checked before implanting. The electrodes are gold plated to ensure a low contact resistance of 100 – 300 kΩ at 1kHz frequency. Three screws (1,2 × 3mm, Matsumoto Industry Co., Ltd., Japan) are also attached to the animal skull for making a reference electrode and reinforcing the implanted electrodes into the head of the mouse. (Figure 3.1). Figure 3.1. The illustration show the implanted stimulating and recording electrodes Research facilities:  The task for recording the nose - poking behaviors response 15 Figure 3.2. The recording chamber for the ICSS response and nose-poking behaviors of mice.  Spatial tasks and memorability Suitable parameters for performing the spatial response exercises are determined by studying the ICSS response and nose-poking behaviors of mice. Figure 3.3. The illustration of the model and arrangement of the spatial tasks. The spatial behavior of mice is investigated in an open round box of 80 cm in diameter and 25 cm in height (figure 3.3). 3.2. Simulation results From the built algorithms of 4 practical exercises (NPT, DMT, RRPST, and PLT) in chapter 2, the stimulation and recording programs for the electrical activity of neurons with 4 respective practical exercises on mice as shown in Figure 3.4. 16 3.2.1. Simulation of the NPT exercise Figure 3.4. The progam for simulating and recording of nose-poking response. Recorded data is stored and analyzed to evaluate the response of mice to the intensity and frequency of the stimulating current. 3.3. Analysis and evaluation of practical results on mice 3.3.1. Practical results of the NPT exercise The ICSS response of mice is recorded after implanting the electrode for one week. During experiment, animals are kept inside a cage with a 1.5 cm hole at the middle and an optical sensor at the bottom (Omron EE- SPX303). Each time the mice poke their nose to the hole, a serial of stimulating pulses for 0,5s is activated (each is a 0,3ms negative square Cathode pulse) It can be seen from Figure 3.8 that, the recorded curve is consistent with the Gompertz model. It is also suitable to the response trend of electrical circuit model of neurons for the stimulation parameters evaluated by simulations in chapter 2. * The intensity of stimulation It can be seen from the Figure 3.8 that the average number of nose- poking behaviors in a minute depends on the intensities of the stimulating current (the blue lozenge dots) as compared to calculated values (the 17 orange square dots) by the Gompertz model. The experiments are performed on 7 mice (each mouse is repeated 2 twice in 2 days). The variation shown in graph is SE (standard error) for the statistical analysis of experiments. Figure 3.8. The dependence of nose-poking response on the stimulating intensity *The Stimulating frequency Figure 3.9. The dependence of nose-poking response on the stimulating frequency. From the Figure 3.9, the average number of nose-poking behaviors in a minute depends on the frequencies of the stimulating current (the blue 18 lozenge dots) as compared to the calculated values (the orange square dots) by the Gompertz model. The experiments are performed on 6 mice (each mouse is repeated in two different days). The variation shown in graph is SE (standard error) for the statistical analysis of experiments. 3.3.2. Experimental results for the spatial response tasks Figure 3.10. Results in the spatial response tasks 3.4. The results of stimulating and recording experiments of the neuronal electronic activity in the Hippocampus on mice The depth of the recording electrodes is increased by 20 µm per day during the measurements. * The general characteristics of the Hippocampal place cells Figure 3.11. The neuron activity are recorded and isolated using an offline- sorter program (Plexon). 19 3.5. The evaluation of the algorithms, stimulation and recording systems for the electrical activity of neurons. 3.5.1. The evaluation of algorithms This research proposes 4 exercises with 4 respective algorithms for studying the electrical activity of neurons on mice. The algorithms were strictly based on the requirements of specific conditions in order to ensure to train the mice from easy to difficult tasks. The processing and controlling programs were then built by using these algorithms. 3.5.2. The evaluation of the stimulating and recording system for the electrical activity of neurons. * The stability and accuracy of the system: The system will offer a reward (100% efficiency) to the mouse when the reward conditions are reached during each measurement for 4 exercises. The recording system regularly monitors the electrical activity of neurons to ensure a complete evaluation in correlation with the stimulation. Moreover, the recording system has sensitivity of an mV range, which can run stably and errorlessly. *The delay of system + The NPT task Isolator Stimulator USB 6501 Central processing system Monitor Labchart v8.1.8 Stim ulating electrode Processing circuit Se ns or t2 t1 Figure 3.13. The evaluation of the stability and delay of the system for the NPT task by the Labchart Pro v8.1.8. The delay of the NPT system: 𝛥𝑡𝑁𝑃𝑇 is a period when the mouse has its nose-poking behaviour until it receives the stimulating signal. The 𝛥𝑡𝑁𝑃𝑇 is determined by 60 ms while a period for receiving a reward is 20 0.74 s. It is clearly noticed that the 𝛥𝑡𝑁𝑃𝑇 is much smaller than the period of one reward acquisition. Figure 3.14. The illustration for pulses of the reward condition, reward delivery, and the delay time of the system. + The DMT, RRPST and PLT tasks: The stability and delay evaluation of the system for The DMT, RRPST and PLT tasks are described in Figure 3.15. The t1 is time when the reward conditions are reached, which is determined by: the movement time of mice in a defined distance (the DMT task); the touch of mice on random reward areas (the RRPST task); the touch of mice on defined reward areas (red and green) with successive conditions (the PLT task). The t2 is the time when mice receive the reward. monitor Isolator Stimulator USB 6501 Central processing system S tim u latin g electrod e CCD cam era t1 t2 Figure 3.15. The evaluation of the stability and delay of the system for the DMT, RRPST and PLT tasks. 21 The simulation program for evaluating the delay time of the DMT, RRPST and PLT tasks. Figure 3.16. The program for evaluating the stability and delay time of the DMT task. Figure 3.16. The program for evaluating the stability and delay time of the RRPST task. 22 Figure 3.20. The program for evaluating the stability and delay time of the PLT task. The delay time between t1 and t2 averagely calculated for the DMT, RRPST, and PLT tasks are 4,88 ± 2,01ms; 4,44 ± 1,91ms and 4,91 ± 2,12ms respectively. Thus the maximum movement of mice during the average delay time for all three tasks is: 7,03 × 0,083  0,58 mm  200 mm (the diameter of the reward area) Moreover, the delay time is statistically estimated from 3 to 7ms in 1500 measurements for all three tasks of the DMT, RRPST, and PLT. There is only one delay time of 24,28ms when the maximum movement of mice is 2,02mm (24,28 × 0,083)  200mm, the radius of the reward area. This means during the delay time for all tasks (DMT, RRPST and PLT), mice almost do not move out of their current place. Therefore, it can be assumed that the delay time of the system is equal to zero (mice receive the reward immediately when the reward conditions are reached). 23 CONCLUSION In the scope of this thesis, a neural stimulation system has been established to evaluate the spatial response of the Hippocampal place cells by practical experiments. The main results obtained in this project are summarized by these following contributions: 1. Results The electrical activity of neurons has been overviewed in the first part of this thesis, which focuses on the electrical stimulation of nerve cells. The equivalent circuit model of neurons is also presented for better understanding the influence of stimulation parameters on the neuronal electrical activity. The Maeda-Mekino model has been used to study the intensity and frequency of the stimulating current by the NI Multisim simulation program, version 14.0. To evaluate the influence of the intensity and frequency of the stimulation on the electrical activity of neurons, both simulation and practical experiments on animals have been performed in this research. The studies of nose poking behavior of mice associated with the stimulation parameters allow selecting the optimal stimulation values: An intensity of 100 µA and a frequency of 100 Hz. Proper stimulation parameters (80% optimal values) were used to study the H