Authors :
Presenting Author: Anna Maslarova, MD, PhD – New York University, Grossman School of Medicine
Jiyun Shin, PhD – Neurology, New York University Langone Health; Andrea Navas-Olive, PhD – Institute of Science and Technology (IST), Austria; Arnd Dörfler, MD – Chair, Neuroradiology, Uniklinikum Erlangen; Hajo Hamer, MD – Spokesman, Epilepsy Centre, Neurology, Uniklinkum Erlangen, Germany; Simon Henin, PhD – Neurology, New York University Langone Health;; György Buzsáki, MD, PhD – Professor, Department of Neuroscience and Physiology, New York University Grossman School of Medicine; Anli Liu, MD – Associate Professor, Neurology, New York University Langone Health
Rationale:
Sharp-wave ripples (SWRs) are key hippocampal oscillations supporting memory and decision making and highly conserved in mammals. While ripples have recently been linked to memory behavior in humans, their equivalence to rodent SWRs is debated due to differences in recording and analytic methods. Unlike rodent SWRs recorded from CA1 with silicon probes, human “ripples” have been detected beyond the hippocampus (HPC) with macroelectrodes measuring large population level activity. Moreover, because human electrodes are often placed in epileptogenic tissue, pathological high-frequency oscillations (pHFOs) and interictal epileptiform discharges (IEDs) can be erroneously labeled as SWRs. To overcome these challenges, we (1) used an Alzheimer’s Disease (AD) mouse model and high-density silicon probes to identify the neurophysiological features distinguishing hippocampal SWRs and IEDs, and (2) applied these features to identify probable SWRs and IEDs in human microwire recordings.
Methods:
We detected SWRs and IEDs from APP/PS1 transgenic mice (n=3) on 1024-channel silicon probes (Neuronexus) spanning the entire dorsal HPC during quiet wakefulness in rodents. Next we examined 312 hours of human microwire data (Behnke-Fried electrodes) from the HPC of epilepsy surgical patients (n=15). The anatomical locations of the microwires and macroelectrodes were determined via CT-MRI co-registration and HPC subfield segmentation (ASHS software). Putative pyramidal cells and interneurons were distinguished based on their waveforms and firing rates.
Results:
In the mouse model, SWRs were primarily localized to CA1 and were characteristically coupled with a positive sharp wave in the pyramidal layer and a negative sharp wave in the dendritic layers. Firing rates of both excitatory and inhibitory neurons displayed a strong excitation followed by inhibition during SWRs. In contrast, IEDs exhibited a higher voltage amplitude (~15 folds) and were recorded across all HPC subfields simultaneously, with interneurons showing a prolonged (~300 ms) decrease in firing rate. We applied these features to distinguish SWRs and IEDs in human HPC microwire data. After automated detection of the sharp-wave events (10-60 Hz) verified by visual inspection, we detected putative SWRs in four out of 15 patients. The small fraction of SWR-positive patients was likely due to the small number of microwires positioned in CA1, and poor sleep quality in epilepsy patients. All four patients also displayed hippocampal IEDs. In these four patients with probable SWRs, we performed single-unit analysis, revealing distinct firing modulation patterns in pyramidal and inhibitory neurons during putative SWRs and IEDs, consistent with our AD mouse model.
Conclusions:
By identifying the features distinguishing SWRs and IEDs from hippocampal silicon probe recordings in an AD mouse model, we can differentiate true SWRs from pathological IEDs in human microwire recordings. This approach enhances the accuracy and reliability of SWR identification and separation from other pHFOs, facilitating translational research in memory-related disorders.
Funding:
DFG MA 10301/1- 1(AM),NYU FACES (AM),R01 NS127954 (AL),K23-NS 104252 (AL),Department of Neurology