Bence Pelyvas, Ervin Wolf, Zita Kepes, Ervin Berenyi, Tamas Papp, Attila Somogyi
Our computer models predict that diagnostic level US treatment does not induce changes in specific membrane resistance and capacitance of neuronal membrane. Correlative analyses of simulated dendritic impulse propagation in US treated (UT) and non-treated control neurons (NT) revealed minor differences in attenuations and delays of somatopetally propagating PSPs in the passive model, and these alterations became even less pronounced when hyperpolarization-activated cyclic nucleotide-gated (HCN) cation channels and A-type potassium (KA) channels were inserted into the model. Synaptic input pattern recognition between UT and NT neurons showed no significant alterations.
INTRODUCTION: Neural development is regulated by several spatiotemporally changing factors, which are essential for enabling neurons to develop functional networks, during early developmental stages. Additionally, external physical stimuli, like scanning prenatal ultrasound (US) examination, may influence neural development. Aim of our study is to examine potential consequences of diagnostic level ultrasound exposure during the early phase of hippocampal CA1 neural network development by using computational models.
METHODS: Using in silico modelling, this study explores and compares multiple features of intraneuronal dendritic signal propagation in US-treated and control CA1 pyramidal neurons to look for possible alterations caused by US. We used morphological data of CA1 neurons based on previous morphometric datasets and built high-fidelity subthreshold passive and active segmental cable models of these neurons in the NEURON simulator. To simulate dendritic signalling either a current was injected or a synapse was activated at hundreds of dendritic points of model neurons, eliciting local postsynaptic potentials (PSPs), and multiple descriptors of dendritic impulse propagation between dendritic points and soma were computed.
RESULTS: Our computer models predict that diagnostic level US treatment does not induce changes in specific membrane resistance and capacitance of neuronal membrane. Correlative analyses of simulated dendritic impulse propagation in US treated (UT) and non-treated control neurons (NT) revealed minor differences in attenuations and delays of somatopetally propagating PSPs in the passive model, and these alterations became even less pronounced when hyperpolarization-activated cyclic nucleotide-gated (HCN) cation channels and A-type potassium (KA) channels were inserted into the model. Synaptic input pattern recognition between UT and NT neurons showed no significant alterations.
DISCUSSION: We conclude that subthreshold somatopetal signalling properties of US-treated CA1 neuronal membranes remain predominantly at the level of non-treated cells. This conservation is due primarily to HCN channels, contributing to net membrane conductance at rest in an inhomogeneous manner over the somato-dendritic surface. The HCN channels balance the effects of US-induced morphological alterations on dendritic signalling. Conservation of signalling properties align with our independent prediction on the conservation of synaptic integration and input pattern recognition in UT neurons.