Geometric Analysis of Population Rhythms in Synaptically Coupled Neuronal Networks

We develop geometric dynamical systems methods to determine how various components contribute to a neuronal network's emergent population behaviors. The results clarify the multiple roles inhibition can play in producing different rhythms. Which rhythms arise depends on how inhibition interacts with intrinsic properties of the neurons; the nature of these interactions depends on the underlying architecture of the network. Our analysis demonstrates that fast inhibitory coupling may lead to synchronized rhythms if either the cells within the network or the architecture of the network is sufficiently complicated. This cannot occur in mutually coupled networks with basic cells; the geometric approach helps explain how additional network complexity allows for synchronized rhythms in the presence of fast inhibitory coupling. The networks and issues considered are motivated by recent models for thalamic oscillations. The analysis helps clarify the roles of various biophysical features, such as fast and slow inhibition, cortical inputs, and ionic conductances, in producing network behavior associated with the spindle sleep rhythm and with paroxysmal discharge rhythms. Transitions between these rhythms are also discussed.

Neonatal Seizures

Definitions and Epidemiology A seizure is a sudden, paroxysmal discharge of a population of neurons that causes a transient alteration in neurologic function. This alteration may involve abnormal motor activity, sensory symptoms, a change in the level of alertness, alteration in autonomic function, or any combination of these. When a seizure occurs in the neonatal period, several special considerations arise (Table 1). A seizure in a newborn almost always reflects significant nervous system pathology, and recognizing and treating seizures properly may prevent subsequent chronic neurologic impairment. It is important not to confuse seizures with epilepsy. A seizure is a single event and may be due to a transient abnormality that will not recur (eg, hypoglycemia). Epilepsy is the condition of unprovoked recurrent seizures. Many neonatal seizures are transient events that will not progress to epilepsy. Ictal refers to clinical (based on visual observation) or electrical (based on the electroencephalogram [EEG]) activity occurring during a seizure (Table 2). Seizures occur in about 0.2% to 1.4% of all newborns. However, the incidence is much higher among certain high-risk groups. About 20% of newborns whose birthweight is less than 2500 g have seizures, with a higher incidence among sicker preterm infants. Incidence also depends on etiology; as many as 50% of newborns who have severe hypoxic-ischemic encephalopathy (HIE) develop neonatal seizures.

Potassium and calcium concentrations in interstitial fluid of hippocampal formation during paroxysmal responses

The concentration of potassium ([K+]o) and of calcium ([Ca2+]o) in interstitial fluid of the hippocampal formation of rats anesthetized with urethan was recorded with double-barreled ion-selective microelectrodes. The ipsilateral angular bundle was stimulated with trains of repetitive pulses. [K+]o increased during angular bundle stimulation in both dendritic and cell body layers of the fascia dentata. When stimulation was frequent and intense enough to provoke intercurrent paroxysmal discharge (IPaD), [K+]o in the granule cell body layer rose much above the level it attained during previous, nonparoxysmal activation. No similar excess increase of [K+]o related to paroxysmal firing was observed in the dendritic layer. It is concluded that tonic paroxysmal discharge of the granule cells is associated with an outflow of K ions from the cell somata, but not the dendrites. Extracellular sustained potential (SP) shifts and responses of [K+]o associated with paroxysmal firing showed no consistent correlation in fascia dentata. It is concluded that paroxysmal SP shifts in fascia dentata (unlike in spinal cord and cerebral neocortex) are dominated by the extracellular currents generated by granule cells, not by neuroglia. In the postparoxysmal phase, however, a small residual SP shift was observed in both soma and dendrite layers, which had characteristics compatible with its being generated by glial cells. Responses of [Ca2+]o varied from rat to rat. During nonparoxysmal excitation [Ca2+]o increased, decreased, or remained unchanged. During paroxysmal firing [Ca2+]o always decreased in the granule cell body layer, but the magnitude of the response varied greatly. In the dendritic layer a similar but smaller decrease was observed in some but not all cases. Probable reasons for the unpredictability of the responses of [Ca2+]o are discussed. The responses of [Ca2+]o recorded in fascia dentata of urethan-anesthetized rats that have previously been kindled were not detectably different from those of control animals. Leao's spreading depression (LD) was associated with large increase of [K+]o, decrease of [Ca2+ )o, and intense negative SP shift in both dendritic and cell body layers of fascia dentata, as well as in CA1 zone of hippocampus. It is concluded that LD in hippocampal formation is associated with more widespread depolarization of pyramidal and granule cells than in cerebral neocortex and cerebellar cortex where changes of [K+]o are limited to the more superficial layers.