ROCK1 is a multifunctional factor maintaining the primordial follicle reserve and follicular development in mice.

The follicle is the basic structural and functional unit of the ovary in female mammals. The excessive depletion of follicles will lead to diminished ovarian reserve or even premature ovarian failure, resulting in diminished ovarian oogenesis and endocrine function. Excessive follicular depletion is mainly due to loss of primordial follicles. Our analysis of published human ovarian single-cell sequencing results by others revealed a significant increase in rho-associated protein kinase 1 (ROCK1) expression during primordial follicle development. However, the role of ROCK1 in primordial follicle development and maintenance is not clear. This study revealed a gradual increase in ROCK1 expression during primordial follicle activation. Inhibition of ROCK1 resulted in reduced primordial follicle activation, decreased follicular reserve, and delayed development of growing follicles. This effect may be achieved through the HIPPO pathway. The present study indicates that ROCK1 is a key molecule for primordial follicular reserve and follicular development.NEW & NOTEWORTHY ROCK1, one of the Rho GTPases, plays an important role in primordial follicle reserve and follicular development. ROCK1 was primarily expressed in the cytoplasm of oocytes and granulosa cell in mice. Inhibition of ROCK1 significantly reduced the primordial follicle reserve and delayed growing follicle development. ROCK1 regulates primordial follicular reserve and follicle development through the HIPPO signaling pathway. These findings shed new lights on the physiology of sustaining female reproduction.

[p53 regulates primordial follicle activation through the mTOR signaling pathway].

This paper aimed to investigate the role and potential mechanism of p53 on primordial follicle activation. Firstly, the p53 mRNA expression in the ovary of neonatal mice at 3, 5, 7 and 9 days post-partum (dpp) and the subcellular localization of p53 were detected to confirm the expression pattern of p53. Secondly, 2 dpp and 3 dpp ovaries were cultured with p53 inhibitor Pifithrin-µ (PFT-µ, 5 µmol/L) or equal volume of dimethyl sulfoxide for 3 days. The function of p53 in primordial follicle activation was determined by hematoxylin staining and whole ovary follicle counting. The proliferation of cell was detected by immunohistochemistry. The relative mRNA levels and protein levels of the key molecules involved in the classical pathways associated with the growing follicles were examined by immunofluorescence staining, Western blot and real-time PCR, respectively. Finally, rapamycin (RAP) was used to intervene the mTOR signaling pathway, and ovaries were divided into four groups: Control, RAP (1 µmol/L), PFT-µ (5 µmol/L), PFT-µ (5 µmol/L) + RAP (1 µmol/L) groups. The number of follicles in each group was determined by hematoxylin staining and whole ovary follicle counting. The results showed that the expression of p53 mRNA was decreased with the activation of primordial follicles in physiological condition. p53 was expressed in granulosa cells and oocyte cytoplasm of the primordial follicles and growing follicles, and the expression of p53 in the primordial follicles was higher than that in the growing follicles. Inhibition of p53 promoted follicle activation and reduced the primordial follicle reserve. Inhibition of p53 promoted the proliferation of the granulosa cells and oocytes. The mRNA and protein expression levels of key molecules in the PI3K/AKT signaling pathway including AKT, PTEN, and FOXO3a were not significantly changed after PFT-µ treatment, while the expression of RPS6/p-RPS6, the downstream effectors of the mTOR signaling pathway, was upregulated. Inhibition of both p53 and mTOR blocked p53 inhibition-induced primordial follicle activation. Collectively, these findings suggest that p53 may inhibit primordial follicle activation through the mTOR signaling pathway to maintain the primordial follicle reserve.

ASH1L contributes to oocyte apoptosis by regulating DNA damage.

In female mammals, the size of the initially established primordial follicle pool within the ovaries determines the reproductive life span. Interestingly, the establishment of the primordial follicle pool is accompanied by a remarkable programmed oocyte loss for unclear reasons. Here, we identify a new role of ASH1-like histone lysine methyltransferase (ASH1L) in controlling the apoptosis of oocytes during meiotic prophase I in mice. Our results showed that overexpression of Ash1l led to a dramatic loss of fetal oocytes via apoptosis, which subsequently resulted in a reduced capacity of the primordial follicle pool. Overexpression of Ash1l also led to a deficiency in DNA double-strand break repair associated with premature upregulation of p63 and phosphorylated checkpoint kinase 2 (p-CHK2), the major genome guardian of the female germline, following Ash1l overexpression in fetal ovaries. In summary, ASH1L is one of the indispensable epigenetic molecules that acts as a guardian of the genome. It protects oocyte genome integrity and removes oocytes with serious DNA damage by regulating the expression of p63 and p-CHK2 during meiotic prophase I in mice. Our study provides a perspective on the physiological regulatory role of DNA damage checkpoint signaling in fetal oocyte guardianship and female fertility.

MNL-Network: A Multi-Scale Non-local Network for Epilepsy Detection From EEG Signals.

Epilepsy is a prevalent neurological disorder that threatens human health in the world. The most commonly used method to detect epilepsy is using the electroencephalogram (EEG). However, epilepsy detection from the EEG is time-consuming and error-prone work because of the varying levels of experience we find in physicians. To tackle this challenge, in this paper, we propose a multi-scale non-local (MNL) network to achieve automatic EEG signal detection. Our MNL-Network is based on 1D convolution neural network involving two specific layers to improve the classification performance. One layer is named the signal pooling layer which incorporates three different sizes of 1D max-pooling layers to learn the multi-scale features from the EEG signal. The other one is called a multi-scale non-local layer, which calculates the correlation of different multi-scale extracted features and outputs the correlative encoded features to further enhance the classification performance. To evaluate the effectiveness of our model, we conduct experiments on the Bonn dataset. The experimental results demonstrate that our MNL-Network could achieve competitive results in the EEG classification task.

A One-Dimensional CNN-LSTM Model for Epileptic Seizure Recognition Using EEG Signal Analysis.

Frequent epileptic seizures cause damage to the human brain, resulting in memory impairment, mental decline, and so on. Therefore, it is important to detect epileptic seizures and provide medical treatment in a timely manner. Currently, medical experts recognize epileptic seizure activity through the visual inspection of electroencephalographic (EEG) signal recordings of patients based on their experience, which takes much time and effort. In view of this, this paper proposes a one-dimensional convolutional neural network-long short-term memory (1D CNN-LSTM) model for automatic recognition of epileptic seizures through EEG signal analysis. Firstly, the raw EEG signal data are pre-processed and normalized. Then, a 1D convolutional neural network (CNN) is designed to effectively extract the features of the normalized EEG sequence data. In addition, the extracted features are then processed by the LSTM layers in order to further extract the temporal features. After that, the output features are fed into several fully connected layers for final epileptic seizure recognition. The performance of the proposed 1D CNN-LSTM model is verified on the public UCI epileptic seizure recognition data set. Experiments results show that the proposed method achieves high recognition accuracies of 99.39% and 82.00% on the binary and five-class epileptic seizure recognition tasks, respectively. Comparing results with traditional machine learning methods including k-nearest neighbors, support vector machines, and decision trees, other deep learning methods including standard deep neural network and CNN further verify the superiority of the proposed method.