Modulation of intrinsic circuits in developing neocortex.

Abstract

Numerous experimental studies have indicated that the developing rat brain is very susceptible to seizures. Immature rats have a low threshold for flurothyland kainic acid-induced seizures. Additionally, prolonged epileptiform activity is observed in slices of immature rat brain exposed to convulsant drugs in vitro. The mechanisms responsible for this seizure-prone period in development are poorly understood. Developmental changes in expression of metabotropic glutamate receptors (mGluRs) and local circuit abnormalities may contribute to the increase in seizure susceptibility of the immature brain. We have hypothesized that because other neuromodulators display marked developmental changes in receptor expression, these systems also could contribute to epileptogenesis. The intricate neural circuits of the mammalian neocortex develop as a result of a series of complex, precisely timed events. These events include cell differentiation, migration, synapse formation, and stabilization. The cerebral cortex is innervated by monoaminergic afferents at early stages of development, suggesting that these systems play a prominent role in synaptogenesis and circuit formation. In particular, disruptions of the serotonin (5-HT) system during embryonic or early postnatal development are known to produce, in adults, profound disturbances of sleep patterns, learning, and behavior. 5-HT has been implicated in drug-induced psychoses and a number of psychiatric disorders, major depression and schizophrenia in particular. Increasing 5-HT levels in neonatal rats cause impairments in performance on passive avoidance learning paradigms when these rats are tested 1-2 months later. Decreasing neonatal 5-HT levels have been associated with adult deficits in attention, spatial memory, and learning. Although these manipulations are not physiologic, environmental enrichment, a treatment known to augment neuronal plasticity, also has been shown to alter 5-HT mRNA expression and binding in rat brain. Abnormal 5-HT levels during development also have been implicated in a number of human neurologic disorders such as schizophrenia, Rett syndrome, and autism. In addition to involvement with mechanisms of brain development, serotonin has been recently shown to be an endogenous modulator of epileptiform activity in the rat hippocampus (1). The 5-HT reuptake inhibitor fluoxetine has anticonvulsant effects as an add-on therapy in patients (2). Multiple 5-HT-receptor subtypes are expressed in the cerebral cortex ( 3 ) . 5-HT, receptors are principally expressed in inhibitory neurons (4), whereas 5-HT2, receptors are heavily expressed in pyramidal cells and to a lesser extent in inhibitory neurons. Based on experiments using in vivo microiontophoretic methods, the predominant effect of 5-HT on cerebral cortical pyramidal neurons was believed to be inhibition of spontaneous spiking, but the underlying mechanisms was not clear (see 5 for review). Intracellular studies in vitro have reported varying postsynaptic effects of 5-HT. In neocortical neurons, 5-HT has been reported to have no effect on input resistance and membrane potential (6) or to both hyperpolarize and to depolarize neurons (6,7). These multiple effects likely result from the diversity of ion channels that 5-HT modulates and the multiplicity of 5-HT receptors. Because there is a transient expression of a dense innervation of 5-HT receptors in sensory areas in the immature neocortex (8,9), we compared the effects of 5-HT on immature and mature neocortex. To examine the effects of 5-HT on activation of local circuits in the neocortex, optical imaging of voltagesensitive dye signals was used. This technique allows measurement of the spatial temporal pattern of excitation of a large area of cortex, Because we have shown that 5-HT enhances the activity of inhibitory interneurons and increases GABA-mediated synaptic activity in mature neocortex (Zhou and Hablitz, unpublished observations), it was reasoned that 5-HT-receptor activation should decrease voltage-sensitive dye signals. Stimulation was applied in deeper cortical layers. As shown in Fig. 1A (left), dye signal responses were initially recorded in the region of cortex overlying the stimulation site. This activity then spread laterally over a distance of 5 1,000 Fm. The area of cortex excited and the spread of activity was reduced by application of 5-HT (Fig. lA, right). These effects were reversible on washing. These