Abstract
Protracted Postnatal Development of Sparse, Specific Dentate Granule Cell Activation in the Mouse Hippocampus
Yu E, Dengler G, Frausto S, Putt M, Yue C, Takano H, Coulter D, J Neurosci 2013;33:2947–2960.
The dentate gyrus (DG) is a critical entry point regulating function of the hippocampus. Integral to this role are the sparse, selective activation characteristics of the principal cells of the DG, dentate granule cells (DGCs). This sparse activation is important both in cognitive processing and in regulation of pathological activity in disease states. Using a novel, combined dynamic imaging approach capable of resolving sequentially both synaptic potentials and action potential firing in large populations of DGCs, we characterized the postnatal development of firing properties of DG neurons in response to afferent activation in mouse hippocampal-entorhinal cortical slices. During postnatal development, there was a protracted, progressive sparsification of responses, accompanied by increased temporal precision of activation. Both of these phenomena were primarily mediated by changes in local circuit inhibition, and not by alterations in afferent innervation of DGCs because GABA(A) antagonists normalized developmental differences. There was significant and frequency-dependent synaptic recruitment of DGC activation in adult, but not developing, animals. Finally, we found that the decision to fire or not fire by individual DGCs was robust and repeatable at all stages of development. The protracted postnatal development of sparse, selective firing properties, increased temporal precision and frequency dependence of activation, and the fidelity with which the decision to fire is made are all fundamental circuit determinants of DGC excitation, critical in both normal and pathological function of the DG.
Commentary
What is the dentate gyrus (DG) good for? The answer may depend on whether or not you are interested in learning and memory, or epilepsy. While the actual circuitry is more complicated, the DG is the initial relay in the hippocampal circuit from the entorhinal cortex (EC). From there the DG projects to hippocampal sub-region Cornu Ammonis 3 (CA3), CA3 projects to hippocampal sub-region Cornu Ammonis 1 (CA1) and CA1 outputs back to the EC. Principal neurons in the EC project in the perforant path (PP) throughout the DG (convergence), while the number of principle neurons in the DG (dentate granule cells, DGCs) greatly exceeds the number of EC principle neurons (divergence) and their CA3 projections (1). While PP stimulation might be expected to result in exuberant DG (and CA3) activation, prior work has detailed that PP stimulation actually results in very spatially-selective and limited activation of the DG due to exuberant feed-forward and feedback inhibition activated by PP stimulation. To the neuroscientist interested in learning and memory, the DG is thus crucial for memory resolution or pattern separation (1, 2). To the epileptologist interested in epileptogenesis and seizure control, the DG is the crucial gate, limiting hippocampal activation in temporal lobe epilepsy (3). Given the impact of epilepsy on cognition, the role and function of the DG should be interesting to both. Surprisingly, how this circuit develops has not been fully explored. Further, comparative determination of single DGC behavior within the population of DGCs has not been studied; thus, it seems very appropriate and relevant to study.
The authors addressed these gaps using two distinct functional imaging approaches applied to developing rodent (mice) hippocampus to address spatial activation of the hippocampal circuit as a whole and to investigate activation of individual DGCs within the circuit upon PP stimulation. While PP axons from EC and inhibitory interneurons are present in the DG from birth, DGCs begin to populate the DG around 2 to 3 weeks of age. The authors therefore investigated ventral/temporal hippocampal slices obtained from approximately 2-, 3-, and 8-week-old rodents. The first imaging approach used a fluorescent dye, chemically introduced into nearly all cell membranes in the slice, which is sensitive to changes in membrane potential (voltage-sensitive dye imaging [VSDI]). This technique allowed the authors to investigate the network activation of the entire circuit from the DG, through the hilus of the DG, to CA3. Importantly, VSDI visualizes the entire circuit and is averaged over several PP stimuli; thus, it represents a temporal and spatial average (4). This allowed the authors to perform an important control experiment. Given the significant developmental changes in DGC numbers, it was important to determine that the stimulus paradigm activating the PP was equivalent across development. The authors compared this visualized circuit activation with that measured with a standard field recording electrode, which measured local current density (near the electrode) in the area where PP axons terminate. The authors found, across development, that they were able to equivalently activate the PP in both the infrapyramidal and suprapyramidal blades of the DG.
Having established through this control experiment equivalent activation across development, the authors used VSDI to compare how equivalent activation of the DG propagates into the hippocampal circuit at different developmental time points. The authors found that the DG progressively gates or limits propagation into CA3 with increasing age. The authors concluded that normal DG gate function, whether in learning and memory processes or in epilepsy, is developmentally regulated and strongly attenuated in immature animals. The authors further observed the temporal activation of DG, as visualized with VSDI, was prolonged in immature animals, compared with adults. They found that this was normalized in the presence of bumetanide, suggesting that the prolongation was due to depolarizing GABAergic neurotransmission, a feature of early development and epileptogenesis in adult animals. The effect of bumetanide on hippocampal network propagation was not reported.
In addition to VSDI, the authors simultaneously applied a second functional imaging technique to investigate activation of individual DGCs within the network at different developmental time points, again under the assumption of equivalent PP activation. In this technique, another fluorescent dye was chemically introduced into most cells in the slice, which is sensitive to changes in intracellular calcium concentration (multicellular calcium imaging [MCI]). Based on their morphology, the authors focused on labeled DGCs. The authors then established that this was a reliable reporter of DGC action-potential firing or activation. This technique allowed the authors to investigate single-trial DGC activation upon PP stimulation, thus allowing the determination of accurate spatial and temporal DGC activation, not possible with VSDI. This formed the basis for their next key observation. Upon PP activation, a greater proportion of DGCs were individually activated in immature DG compared with adult; this was normalized in the presence of picrotoxin (PTX), which blocked GABAergic inhibition. This was largely a feature of the infrapyramidal blade of the DG, where there are fewer inhibitory interneurons than the suprapyramidal blade. While the role of GABAergic inhibition as a key regulator of DG function was not surprising, the dramatic impact of development had not been previously demonstrated.
A key question remained: what limits DGC activation with development? While VSDI and electrophysiology measurements suggest equivalent activation of PP, are DGCs less likely and randomly activated with development, or is this also controlled by inhibition? Using a modification of MCI to allow greater temporal resolution, the authors investigated the probability and latency of DGC activation across development. The authors found that activation of DGCs was not random. That is, if a DGC is activated once, it is likely to be always activated in future, regardless of age. Thus, random activation does not contribute to alterations in DGC activation. In immature rodents, a mixed population of events was detected; most were fast, but a small population was prolonged. These prolonged events were abolished by PTX with no change in the latency of fast events. In adult rodents, only fast latency events were detected and their latency was accelerated by PTX. This led to several important conclusions, though with underlying assumptions, about the role and developmental impact of GABAergic innervation. First, feed-back inhibition underlies the longer latency events in immature animals; these events are excitatory (due to immature chloride gradients, but not explicitly shown) to result in DGC activation. This excitatory effect of feedback inhibition would not be evident in mature animals. Feed-forward inhibition prolongs the latency of events in mature rodents, however this is not present in immature rodents. This suggests that feed-forward inhibition, not present in immature rodents, underlies reduced DGC activation in mature rodents. This important observation requires further study.
Finally, the authors looked at frequency dependent DGC activation across development using MCI. As might have been expected, immature DGCs showed no frequency selectivity for PP activation. However, mature DGCs showed 2- to 5-fold recruitment at higher PP activation frequencies.
The authors therefore conclude that the DG, during a relatively protracted development (not unlike your millennial) compared with other brain regions, critically depends on the development of local inhibitory circuits to achieve mature function. Mature DG function implies sparse, temporally precise and frequency selective DGC activation. Again, this may not be unlike your millennial who may need fewer distractions and timely and selectively activation to finally leave your basement. Though assumed, it was not shown that individual DGCs that respond to low frequencies also respond at higher frequencies. It was not stated if the lack of frequency selectivity was also present in the suprapyramidal blade of immature rodents where GABAergic inhibition is stronger or if this could be modulated by manipulations that enhance inhibition, such as bumetanide or benzodiazepines. Further, it was not shown how inhibition impacted propagation through the hippocampal circuit. Can selective modulation of feed-forward versus feedback inhibition affect propagation? Nevertheless, these studies establish an essential starting point to determine how inhibition develops in the DG. Further, the differences in DG regions (supratemporal vs infratemporal blades) are likely equally important in DG function as the dorsal/septal vs ventral/temporal poles (5, 6). Interestingly, the development of inhibitory circuits depends on the birth or population of DGCs. While this is intuitively obvious (there must be something to inhibit for the circuits to work), this must be related to the unique feature of the DG where DGCs are continually born throughout the lifespan (2, 7). Finally, the protracted development of the DG must be considered in developmental models of seizures and epilepsy, particularly those that may impact cognition. Thus, your millennial should remain very dear, even when they leave your basement.