Diversity amongst human cortical pyramidal neurons revealed via their sag currents and frequency preferences

Abstract

In the human neocortex, coherent theta oscillations between superficial and deep cortical layers are driven by deep layer neurons, suggesting distinct intrinsic electrophysiological properties of neurons across cortical layers. Here, we used in vitro whole-cell recordings to characterize pyramidal cells in layer 2/3 (L2/3) and layer 5 (L5) of the human neocortex. We found that human L5 pyramidal cells were more excitable and were endowed with a more prominent sag voltage and larger Ih currents relative to L2/3 neurons, that were abolished through direct pharmacological blockade. Although no peak in subthreshold resonance was observed for either L2/3 or L5 cells, we found that L5 neurons demonstrated greater spiking gain at low frequencies. Integrating patient-level demographic features revealed larger sag amplitudes in pyramidal cells recorded from older patients. These data suggest that sag is prominently expressed in L5 pyramidal cells and is a dynamic feature of human cortical microcircuits.

Modeling reveals human-rodent differences in h-current kinetics influencing resonance in cortical layer 5 neurons

“…This process yields a full spiking, biophysically detailed, multi-compartment model of a distinctly human L5 cortical pyramidal cell. The resulting model matches the electrophysiological data from hyperpolarizing current clamp experiments in the primary cell remarkably well, while also demonstrating repetitive PIR spiking properties characteristic of human L5 pyramidal cells from Chameh et al (2019). Further, our model makes possible a comparison of these neurons in the human and rodent setting, revealing important differences in both the conductance and kinetics of the h-current across species.…”

“…Recently it has been shown that increased expression of h-channels contribute to the observed subthreshold resonance in supragranular layer human pyramidal cells not seen in their rodent counterparts (Kalmbach et al, 2018). Such differential expression of h-channels also appears to be present between superficial and deep layer neurons of human cortex, with layer 5 (L5) pyramidal cells demonstrating a larger sag mediated by h-channels when compared to those in layer 2/3 (L2/3) (Chameh et al, 2019). However, despite the presence of large sag currents in human L5 pyramidal cells, they very rarely exhibit subthreshold resonance (Chameh et al, 2019).…”

“…Such differential expression of h-channels also appears to be present between superficial and deep layer neurons of human cortex, with layer 5 (L5) pyramidal cells demonstrating a larger sag mediated by h-channels when compared to those in layer 2/3 (L2/3) (Chameh et al, 2019). However, despite the presence of large sag currents in human L5 pyramidal cells, they very rarely exhibit subthreshold resonance (Chameh et al, 2019). This result is surprising considering recent human work in L2/3 (Kalmbach et al, 2018), as well as findings that rodent L5 pyramidal cells commonly exhibit subthreshold resonance (Silva et al, 1991; Ulrich, 2002; Dembrow et al, 2010; Schmidt et al, 2016).…”

“…Concerted efforts to understand these relationships are the subject of ongoing theoretical research (Rotstein and Nadim, 2014; Rotstein, 2014). Toward this end, untangling the relationship between the h-current and subthreshold resonance is crucial, especially considering the unexpected results reported by Chameh et al (2019). Unfortunately, while the experimental literature (Kalmbach et al, 2018; Chameh et al, 2019) has revealed that differences in the capacity for subthreshold resonance exist between human and rodent neurons, the reason(s) why these differences arise remains unclear.…”

“…As these recordings were obtained in the presence of a voltage-gated sodium channel blocker, spiking was not present in this data set. Correspondingly, we analyzed post-inhibitory rebound (PIR) spiking and repetitive firing characteristics from another set of human L5 neurons (Chameh et al, 2019) to impart constraints on the model’s spiking characteristics. A particular emphasis is placed on ensuring our model neuron exhibits reasonable PIR spiking behavior, a dynamic which is shown by Chameh et al (2019) to be reliant upon h-channels in human L5 neurons.…”