Mechanisms Of Synaptic Plasticity In The Adult And Developing Nervous System

National Institute Of Environmental Health Sciences

Investigators

Linked publications & trials

Abstract

A longstanding issue in environmental health is the need to understand the role the environment plays in human brain development. The brain of the neonate is particularly susceptible to disruption of the sensory environment, which can have profound effects on its physiology and morphology. Such susceptibility of the developing brain to environmental influence by sensory manipulation or to environmental toxicants is particularly pronounced during defined sensitive periods of postnatal life. On the one hand, the plasticity of the connections between neurons, or synapses, is critical for refining brain circuitry during postnatal development. However, on the other hand, it is this very plasticity that makes the developing brain particularly vulnerable to various insults. In addition, similar mechanisms for changing synapses are likely to serve the basis for learning in the adult. Our primary interest, therefore, has been to determine the molecular basis of regulating long-lasting synaptic plasticity. Toward our goal of learning how neuronal activity can induce lasting modifications in neurons, we use a diverse collection of molecular, biochemical, electrophysiological, and imaging techniques. Most of our studies are performed using neonatal and adult mice. To study synaptic plasticity, we use techniques that include whole-cell patch clamp recordings from hippocampal slices maintained in vitro, and we can stimulate them either electrically or pharmacologically to induce long-term potentiation (LTP) or long-term depression (LTD). In addition, we now use molecular and biochemical methods and in vivo recording and tracing techniques to address the role of the different hippocampal subregions in specific behaviors. One approach that we have taken to gain insight into the mechanisms of regulating synaptic plasticity has been to compare highly plastic brain areas, such as the CA1 area of hippocampus, with less plastic areas. From the expression pattern of some genes, we predicted, and found, that one area of the hippocampus, the CA2, would be more resistant to synaptic plasticity such as LTP and LTD, even though we found that synaptic responses in CA2 were very similar to those in the neighboring CA1 and CA3 areas. We later established that dendritic spines in CA2 have very different calcium dynamics from spines in CA1 and CA3 in that both calcium buffering capacity and rates of calcium extrusion were higher in CA2 spines when compared with those in the neighboring regions. RNAseq experiments to measure mRNA from specific cellular compartments of tissue isolated by laser capture microscopy (LCM) revealed that CA2 neurons express an unusually large number of genes related to mitochondrial calcium regulation. In addition, we have found that a regulator of G-protein signaling (RGS-14), and a number of G-protein coupled receptors including adenosine A1 receptors (A1Rs) and receptors for the social neuropeptides oxytocin and vasopressin are highly expressed in CA2 neurons and regulate plasticity there. Consistent with this cell signaling is recent data showing that CA2 may play an important role in social behavior such as aggression. Most recently, we have found that a specialized extracellular matrix, called perineuronal nets (PNNs), plays a role in limiting synaptic plasticity in area CA2. We found that in mouse CA2, PNNs surround pyramidal (excitatory) neurons and their excitatory synapses on dendritic spines. We also found that the PNNs develop postnatally in CA2 in mice, and treatment of hippocampal slices with an enzyme that disrupts PNNs enables synaptic potentiation in CA2 pyramidal neurons. Importantly, these results suggested the existence of a critical period for synaptic plasticity in CA2, much like sensitive periods in other brain areas such as visual cortex. If PNNs are critical to restricting synaptic plasticity though, then LTP would be predicted to be induced in CA2 neurons prior to PNN development, which begins around postnatal day (PND)14. We found that indeed, LTP could be induced in CA2 around PND10, a time when PNNs are not yet expressed. Because accumulating evidence suggests a role for CA2 in social recognition memory, we tested whether PNNs in CA2 were affected in a mouse model of Rett Syndrome, a developmental disorder often accompanied by impairments in social cognition, among other abnormalities. Rett Syndrome (RTT) is caused by dysfunction of the methyl CpG binding protein 2 (MeCP2) and staining for PNNs had been reported to be increased in brain tissue isolated from individuals with Rett Syndrome. We found that staining for PNNs was dramatically increased in CA2 of Mecp2 knockout mice, and that they developed precociously, with a significant difference apparent by PND11. In addition, this early window of plasticity in CA2 closed prematurely in Mecp2 null mice and the capacity for LTP was restored in the slices from Mecp2 null mice with ChABC treatment. One of the co-morbid symptoms of Rett Syndrome is debilitating epilepsy, and the hyperactivity of neuronal circuits has been observed even when tissue from Mecp2 null mice is cultured in vitro. Therefore, to determine whether the observed increase in CA2 PNN expression was due to aberrant pyramidal neuronal activity or other Mecp2-dependent factors, such as regulation of gene expression, we took two independent strategies. First, we took advantage of chemogenetics to directly increase or decrease neuronal activity in CA2. Mice expressing either excitatory or inhibitory Designer Receptors Exclusively Activated by Designer Drugs (DREADDs) were treated daily with clozapine-N-oxide (CNO), the ligand for DREADDs, and PNNs measured 5 days later. Second, we examined PNN expression level in a mouse model of developmental onset epilepsy Kv1.1; Knca1 knockout) in which seizure onset age closely approximates PNN development. The results of our enrichment studies and the excessive deposition of PNNs in the Mecp2 null mice led us to predict that PNN development would be accelerated due to the abnormally high neuronal activity. However, we found exactly the opposite of what we had expected; staining for PNNs was more intense in mice expressing inhibitory DREADDs and less intense in those expressing excitatory DREADDs. Furthermore, as in the case with the excitatory DREADD, we observed decreased PNN staining in CA2 of Knca1 knock-out mice. These data indicated that PNN deposition and/or maintenance are inversely regulated by neuronal activity and suggest that enhanced neuronal activity is unlikely to be driving early development of PNNs in TRR. One explanation for our results could be that PNN deposition is countered by matrix metalloproteinases (MMPs), which themselves have been shown to be induced by neuronal activity. Given these findings of an inverse relationship between CA2 activity and PNN staining, we speculate that hypoactivity of CA2 neurons may be one contributor to the seizure activity in Rett Syndrome, given that prolonged CA2 silencing has been shown to result in CA1 hyperactivity. Another puzzling finding was that both Mecp2 knockout and environmental enrichment lead to increased CA2 PNN expression. We note, however, that environmental enrichment does not appear to enhance PNN development at earlier ages (<PND14), a stage at which PNNs are prematurely developed in the RTT mice. Together these studies have provided counterintuitive insight into the role of neuronal activity in regulating PNNs at excitatory synapses in a population of excitatory neurons with relevance to developmental disorders such as Rett Syndrome.