Changes in theta and gamma oscillations of the medial prefrontal cortex following hippocampal-focused ultrasound stimulation
Article information
Abstract
Objective
Neuromodulation is a rapidly growing field of treatment, encompassing implantable and noninvasive technology-based approaches. Focused ultrasound (FUS), a noninvasive method, has the potential to modulate neural activity in deep brain regions with spatial precision based on the penetrance of sound waves into the bone and soft tissue. However, neuromodulation is limited by the inability to confirm modulation in real-time. Electrophysiological technology is required to overcome this limitation. In this study, we aimed to confirm neural activity using Neuropixel electrodes during FUS sonication.
Methods
Six male C57BL/6 mice were subjected to ultrasound sonication in the hippocampus. Simultaneously, local field potential (LFP) measurements were performed using Neuropixel electrodes. The experimental groups were divided into three categories: before, during, and after FUS sonication and the measured LFP data were compared between groups.
Results
Theta and gamma waves were extracted from the recorded LFP data and compared between the groups. The theta and gamma oscillations changed after FUS sonication compared to before. The theta graph showed significant changes between the pre- and during-treatment groups and between the during- and post-treatment groups (theta: 4–8 Hz, F=23.91, p<0.0001). However, the gamma graph did not show significant changes in any groups (gamma: 30–50 Hz). Additionally, the expression of c-fos, a marker of neuronal activity, increased after FUS sonication.
Conclusion
Although further research is needed—including longer follow-up LFP measurements after FUS sonication and the collection of more LFP data—neuromodulation with FUS can be considered usable in other neurological disease models.
INTRODUCTION
Neuromodulation is a new class of therapy that refers to intervening with the nervous system through electrical, chemical, or optogenetic methodologies for the regulation of neuronal activity and control of symptoms such as tremors seen in movement disorders. Neuromodulation is a rapidly growing field of treatment, which encompasses implantable and noninvasive technology-based approaches such as deep brain stimulation and transcranial magnetic stimulation. Recently, treatment using focused ultrasound (FUS), one of the non-invasive methods, has been reported [1,2]. FUS has the potential to modulate neural activity in deep brain regions with spatial precision based on the penetrance of sound waves into the bone and soft tissue [3-7].
However, neuromodulation is limited because whether modulation has occurred in real-time cannot be confirmed. To overcome this limitation, the development of functional imaging or electrophysiological technology is required [8]. In previous neuromodulation studies using chemical and optogenetic methods, neural activity was more accurately measured using an electroencephalogram than using an extracellular microelectrode [1,9]. Also, electrophysiology in vivo during FUS sonication needs to be challenged due to the vibrations of recording electrodes and mechano-electrical coupling that can occur in the tissue [3-5,8].
In this experiment, local field potential (LFP) was measured using Neuropixels, a multi-channel electrode, to measure brain waves in specific tissues. Neuropixels have a shank length of 1 cm and are electrodes capable of measuring brain waves in 986 channels. In this experiment, we will try to record neuronal scale and high temporal resolution simultaneously in multiple brain regions through the recording length of Neuropixels that are similar in size to the size of rodent brains. This will be an essential step in understanding the overall coordination of activity that is fundamental to brain function.
LFP has an important role in coordinating the activity of different areas of the brain and synchronizing the activity of individual neurons with that of neural networks. For example, theta frequency (4–8 Hz) oscillations and phase-locking discharges of neurons to theta waves are found in the hippocampus and some cortical regions, providing potential mechanisms related to complex functions in the brain, such as memory formation and neuroplasticity.
We wanted to sonicate FUS along with concurrent measurements of LFP. However, we faced a limitation where the hippocampal region was obscured by the transducer. To overcome this, we needed to find a circuit connected to the hippocampus [10]. Ultimately, we decided to conduct measurements in the medial prefrontal cortex (mPFC). Several studies showed that when stimulation was applied to the hippocampus, hippocampal neuronal activity was enhanced, especially theta and gamma oscillations, which led to theta-gamma phase-locking [11-13]. This phase-locking continues through the hippocampus-medial prefrontal cortex (HPC-mPFC) circuit to the mPFC. The hippocampal formation and mPFC have well-established roles in memory encoding and retrieval. The HPC-mPFC circuit has a fundamental role in cognitive functions such as short-term and long-term memory, attention, and decision-making, which are affected in several neurological diseases [10,14-16]. During spontaneous behavior, functional connectivity between the HPC and mPFC can be inferred from electrophysiological recordings of spike activity and LFPs [9,16].
Therefore, in this study, we tested whether the measurement of neural activity could be performed with electrodes during ultrasound sonication without the vibrations of mechano-electrical coupling that can occur in the tissue. We recorded and analyzed the LFP of the mPFC to confirm neuronal activity after FUS-mediated neuromodulation. We also confirmed a neuronal activity marker, c-fos, to determine neuronal activity by a histological method. Under the assumption that theta and gamma oscillations will increase through neuromodulation using ultrasound, it will have a positive effect on improving cognitive function.
MATERIALS AND METHODS
Animals
All animal experiments were performed according to the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health, and approved by the Institutional Animal Care and Use Committee (IACUC) of Yonsei University (South Korea) (IACUC number: 2023-0102). All mice were housed in groups of 2–5 per cage in a temperature/humidity-controlled room with a 12-hour/12-hour light/dark cycle and with access to food and water ad libitum. Every effort was made to minimize the number of mice used and overall animal suffering. Eight-week-old male C57BL/6 mice were categorized into the control group (without FUS modulation), and the pre-, during, and post-treatment groups, which received FUS modulation. Three mice were used in each group; the control group and the ultrasound group. When measuring LFP, each group set up a recording site of 100 Neuropixels and averaged the measured data for analysis, so many mice were not needed. The groups that underwent FUS neuromodulation corresponded to 10 minutes before modulation, 10 minutes during modulation, and 10 minutes after modulation.
Surgery
Ketamine (75 mg/kg), acepromazine (0.75 mg/kg), and xylazine (4 mg/kg) were used to anesthetize the animals. Mice were then fixed to a stereotaxic frame using ear and nose bars. After the skin was incised, a hole was drilled for electrode insertion, and the meningeal membrane was removed. Then, the electrode was inserted into the right mPFC (2.5 mm anterior and 0.2 mm lateral to the bregma, 2.3 mm depth from the skull surface) at an angle of 45 degrees. For FUS modulation, the transducer was placed on the skull at the hippocampus (2.3 mm posterior and 2.5 mm lateral to the bregma) position so as not to damage the electrode inserted into the mPFC.
Electrophysiology
The Neuropixel is the first fully integrated multi-channel digital neuronal extracellular microelectrode developed for non-human primates [17]. The Neuropixel 3B probe consists of a 10-mm long silicon shank with 384 user-selectable channels consisting of 384 recording sites cross-sectioned at 70×20 mm. This electrode can be used to record multiple brain areas at the same time and might be a possible chronic implant for long-term recording [18-21].
The LFP was recorded using a Neuropixel inserted at a 45-degree angle into the right mPFC. The electrode was connected to the ground by a wire soldered to it to hold the ground reference. The ground reference was screwed into the bone above the cerebellum so as not to collide with the transducer for modulation. Before electrode insertion, the shank of the electrode was coated with Dil dye (1,19-dioctadecyl-3,3,39,39-tetramethylindocarbocyanine perchlorate, 100 mg/mL in acetone: DiI), a red lipophilic dye, to confirm the position of the electrode by histological staining. The recording was conducted with Open Ephys (https://open-ephys.org/). The LFP was sampled at 2,500 kHz with a gain of 500. The LFP was measured 10 minutes before FUS modulation, 10 minutes during modulation, and 10 minutes after modulation in each group. The LFP data were band-pass filtered (0.5–200 kHz) and analyzed for each frequency band using a code that applied ‘load_open_ephys’ in Matlab.
Focused ultrasound-induced hippocampus modulation
The transducer was fixed to the skull of the animal into which the electrode was inserted. The transducer was targeted at the right dorsal hippocampus. The FUS device consisted of a 515 kHz single-element spherically focused H-107MR transducer (Sonic Concept Inc.), two waveform generators (33220A and 33500B; Agilent), and a radiofrequency power amplifier (240 L; ENI Inc.). The bursts of pulsed waves, each sinusoidal, were generated using two serially connected arbitrary function generators. The corresponding duty cycle (DC), which is the percentage of sonication that is active compared to continuous sonication, can be described as the product of tone-burst duration (TBD; in msec) and pulse-repetition frequency (PRF; in kHz) as a percentage value. FUS parameters were as follows: 515 kHz fundamental frequency, 50% DC, 1 kHz PRF, 0.2 millisecond TBD, 300 milliseconds sonication duration, and 2 seconds inter-stimulus interval [2,22].
Data analysis
LFP was recorded in three phases: pre-, during, and post-FUS modulation. All LFP data were analyzed by Matlab. The Time-Frequency analysis code was generated based on Matlab code for Neuropixel analysis: ‘load_open_ephys_binary’ (https://github.com/open-ephys/analysis-tools/blob/master/load_open_ephys_binary.m), and ‘npy_matlab’ (https://github.com/kwikteam/npy-matlab) [23]. The LFP data was averaged from 100 channels to approximate the mPFC channel location. First, the data was filtered by a bandpass filter in Open-Ephys, recording software. The power spectrum of the measured data was investigated by convolving the LPF data with the short-time Fourier transform (STFT). The STFT, having a moving window with a window size of 2,500 samples and a sliding step of 500 samples, was applied to all data in every channel. The frequency bands of interest were theta (4–8 Hz) and gamma (30–50 Hz), and the power spectrum in each band was averaged, respectively [24].
All data were analyzed by one-way analysis of variance (ANOVA) with Tukey’s post hoc comparisons using GraphPad Prism 7 (GraphPad Software Inc.). The mean±standard error of the mean was used to present the data. Statistical significance was set at *p<0.05, **p<0.01, ***p<0.001, and ****p<0.0001.
Immunohistochemistry
Ninety minutes after FUS modulation, each animal was sacrificed and perfused with 0.9% saline and 4% paraformaldehyde. The brains were acquired and sectioned into 30 µm slices using a microtome (Leica Biosystems). The slices were placed in a cryoprotectant solution of 0.1 M phosphate buffer (pH 7.2) (30% sucrose, 1% polyvinylpyrrolidone [Sigma-Aldrich], and 30% ethylene glycol [Thermo Fisher Scientific]) and stored at −20 °C.
Brain tissues were subjected to antigen retrieval in 2 N HCl for 1 hour and neutralized twice with 0.1 M borate buffer for 10 minutes to identify c-fos. The tissues were blocked with 10% normal goat serum for 1 hour after washing with phosphate-buffered saline (PBS). Primary antibodies against c-fos (1:300, ab190289; Abcam), diluted in PBS containing 0.3% Triton X-100 (Sigma-Aldrich), were applied to the tissues and incubated overnight at 4 °C, followed by incubation with secondary antibodies conjugated with Alexa Fluor 488 (A21070, 1:600; Thermo Fisher Scientific) [25].
Analyses of c-fos were performed in the dentate gyrus, hilus, cornu ammonis (CA1 and CA3), and mPFC. Staining intensity was visualized using an M2 microscope (Carl Zeiss).
RESULTS
Changes in theta and gamma oscillations of the medial prefrontal cortex by focused ultrasound modulation
All LFP data was measured by Matlab. As a result of FUS modulation in the hippocampus with the set parameters, changes in the theta and gamma oscillations were revealed in the two groups: the during and post-groups after sonication and the pre-group. We found that hippocampal neuronal activity, especially theta and gamma oscillations, was modulated by FUS. Also, theta and gamma oscillations induced in the hippocampus by FUS migrated to the mPFC (Fig. 1) [11]. Theta graph is shown significant changes between the pre- and during group and during and post-group (theta: 4–8 Hz, F=23.91, p<0.0001). However, gamma has not shown significant changes in all groups (gamma: 30–50 Hz). The results of this experiment were analyzed by amplitude analysis by converting the measured LFP to STFS. Therefore, the significance of the change after neuromodulation was seen in theta oscillations, and it could not be confirmed in gamma oscillations. If the analysis had been performed with phase analysis rather than amplitude phase analysis, the results would have been significantly increased in the theta-gamma phase as the change was confirmed in the spectrogram.
Spectrograms of the pre-, during, and post-treatment groups. (A, D) These represent the pre-group. (B, E) These represent the during group. (C, F) These represent the post-group. (A–C) These show the frequency of theta bands in each group. (D–F) These show the frequency of gamma bands (theta band frequency=4–8 Hz, gamma band frequency=30–50 Hz). (G, H) These are comparisons of each frequency for the three groups (G shows theta oscillations and H shows gamma oscillations). Significant differences in theta oscillations were found between the pre- and during-treatment groups. ****p<0.0001.
Histological confirmation after focused ultrasound modulation
Sampling was performed 90 minutes after modulation of the mouse hippocampus with FUS. To measure FUS modulation-related neuronal changes, the neuronal activity marker, c-fos in the mPFC and HPC areas was stained by immunohistochemistry. The expression of c-fos was confirmed in the hippocampus after modulation by FUS and was increased in the mPFC (Fig. 2).
To examine neuronal activity, c-fos staining was performed in the medial prefrontal cortex (mPFC) and hippocampus. (A) It shows the stained mPFC tissues, with the yellow square indicating the left region and the white square indicating the right region in the center of the tissue. (B) It demonstrates the stained hippocampus tissues, observed under fluorescent microscopy at ×5 and ×20 magnifications from left to right, respectively (scale bars for each image=50 μm, 200 μm). FUS: focused ultrasound.
DISCUSSION
Neuromodulation is a new class of therapy that involves intervening with the nervous system to regulate nerve activity, and rapidly growing research is being conducted toward the treatment of neurological diseases [1]. FUS, a non-invasive method, has been studied recently [2,3]. However, it is limited because it is impossible to visually confirm that modulation has occurred in real time [8]. To overcome this, the development of verification electrophysiological methods is needed. Therefore, we recorded and analyzed the LFP of the mPFC to confirm neuronal activity after FUS-mediated neuromodulation [9]. Using this method, we can test whether the measurement of neural activity can be performed with electrodes during ultrasound neuromodulation without the vibrations of mechano-electrical coupling that can occur in the tissues.
We wanted to sonicate FUS along with concurrent measurements of LFP. However, we faced a limitation where the hippocampal region was obscured by the transducer. To overcome this, we needed to find a circuit connected to the hippocampus [10]. Ultimately, we decided to conduct measurements in the mPFC. The hippocampus is the operational hub of the episodic memory system, a spatially distributed brain network that enables us to remember past experiences in rich detail, together with the space and time in which they occurred [14]. Also, the mPFC is related to decision-making and reward-guided learning. Previous research observed that mPFC plays a key role in the retrieval of remote memories [16]. A neurophysiological study reported the relationship and dynamics between the hippocampus and mPFC interactions. The HPC-PFC circuit has a fundamental role in emotional function and decision-making [15].
During hippocampus and mPFC functional interactions, electrophysiological rhythms in these two structures are coupled, particularly in the theta and gamma ranges [12]. Both theta and gamma waves play important roles in brain functions and are associated with different mental states and cognitive processes. Theta waves are typically observed during states of deep relation and mediation. They are also thought to be associated with enhanced learning and problem-solving abilities. The frequency of the theta wave is between 4 and 8 Hz. Gamma waves are thought to be involved in various cognitive functions, including perception, sensory processing, memory formation, and problem-solving. The frequency of gamma waves is between 30 and 50 Hz. Previous studies reported that stimulating the hippocampus increased nerve activity and improved memory storage function. In particular, elevated theta and gamma oscillations in the hippocampus are phase-locked, which is an important neural activity pattern that occurs in the brain. It has an important role in cognitive function and cognitive processes such as learning and memory. Therefore, the increase in theta and gamma oscillations and c-fos with FUS neuromodulation has the positive potential to improve cognitive function, especially in complex brain functions related to memory. In this study, we observed that theta and gamma oscillations were increased; however, we did not analyze phase-locking in our data because it was determined that the duration of LFP measurement was too short to confirm phase-locking. Therefore, we plan to conduct additional recordings over a longer period in a future study.
CONCLUSION
Decades of research in both humans and animals have revealed that two brain areas, HPC, and mPFC, are essential for the encoding and retrieval of episodic memories. Recently, the HPC-mPFC interaction was demonstrated in electrophysiological results: Theta oscillations and Gamma oscillations. During the spatial working memory task, Theta and Gamma oscillations were increased with the learning of spatial working memory. This experiment confirmed that neural activity can be identified by using Neuropixel during FUS neuromodulation in the HPC.
In addition, we were able to confirm changes in hippocampal neuronal activity transmitted through the HPC-mPFC circuit in mPFC. Via the result, it proved that brain waves can be measured using Neuropixel and can be utilized simultaneously with FUS neuromodulation. Amplitude analysis showed a significant increase in theta oscillations, but no significant increases in gamma oscillations. There is a limitation that we tried to measure the phase-locking value, but we could not measure it because we could not measure LFP in the hippocampus. To compensate for this, neuronal activity was confirmed using c-fos, and it was confirmed that there was an increase in both HPC and mPFC domains after neuromodulation using ultrasound. In a further study, Neuropixel will be used as an electrophysiological tool to record various neurological diseases’ neural activity. Neuropixels are developed to be reusable, it is possible to follow up the long term of experimental animals. Therefore, further research is needed to confirm whether cognitive function is improved through long-term follow-up after neuromodulation and animal behavior experiments. In this study, normal mice were used, but neuromodulation using ultrasound can be applied to models of diseases related to cognitive functions such as dementia in further study.
Notes
CONFLICTS OF INTEREST
No potential conflict of interest relevant to this article was reported.
Acknowledgements
This work was supported by the Korea Medical Device Development Fund (project number: RS-2020-KD000103). This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (No. RS-2023-00266075).