Simple neural networks for the amplification and utilization of small changes in neuron firing rates

I describe physiologically plausible “voter-coincidence” neural networks such that secondary “coincidence” neurons fire on the simultaneous receipt of sufficiently large sets of input pulses from primary sets of neurons. The networks operate such that the firing rate of the secondary, output neurons increases (or decreases) sharply when the mean firing rate of primary neurons increases (or decreases) to a much smaller degree. In certain sensory systems, signals that are generally smaller than the noise levels of individual primary detectors, are manifest in very small increases in the firing rates of sets of afferent neurons. For such systems, this kind of network can act to generate relatively large changes in the firing rate of secondary “coincidence” neurons. These differential amplification systems can be cascaded to generate sharp, “yes–no” spike signals that can direct behavioral responses.

Specific infection of CD4+ target cells by recombinant rabies virus pseudotypes carrying the HIV-1 envelope spike protein.

A recombinant rabies virus (RV) mutant deficient for the surface spike glycoprotein (G) gene was used to study the incorporation of envelope proteins from HIV-1 expressed from transfected plasmids. A hybrid HIV-1 protein in which the cytoplasmic domain was replaced with that of RV G was incorporated into the virus envelope and rescued the infectivity of the RV mutant. The RV(HIV-1) pseudotype viruses could infect only CD4+ cells, and their infectivity was neutralized specifically by anti-HIV-1 sera. In contrast to the chimeric protein, wild-type HIV-1 envelope protein or mutants with truncated cytoplasmic domains failed to produce pseudotyped particles. This indicates the presence of a specific signal in the RV G cytoplasmic domain, allowing correct incorporation of a spike protein into the envelope of rhabdovirus particles. The possibility of directing the cell tropism of RV by replacement of the RV G with proteins of defined receptor specificity should prove useful for future development of targetable gene delivery vectors.

Unifying theory of hypoxia tolerance: molecular/metabolic defense and rescue mechanisms for surviving oxygen lack.

We develop a unifying theory of hypoxia tolerance based on information from two cell level models (brain cortical cells and isolated hepatocytes) from the highly anoxia tolerant aquatic turtle and from other more hypoxia sensitive systems. We propose that the response of hypoxia tolerant systems to oxygen lack occurs in two phases (defense and rescue). The first lines of defense against hypoxia include a balanced suppression of ATP-demand and ATP-supply pathways; this regulation stabilizes (adenylates) at new steady-state levels even while ATP turnover rates greatly decline. The ATP demands of ion pumping are down-regulated by generalized "channel" arrest in hepatocytes and by "spike" arrest in neurons. Hypoxic ATP demands of protein synthesis are down-regulated probably by translational arrest. In hypoxia sensitive cells this translational arrest seems irreversible, but hypoxia-tolerant systems activate "rescue" mechanisms if the period of oxygen lack is extended by preferentially regulating the expression of several proteins. In these cells, a cascade of processes underpinning hypoxia rescue and defense begins with an oxygen sensor (a heme protein) and a signal-transduction pathway, which leads to significant gene-based metabolic reprogramming-the rescue process-with maintained down-regulation of energy-demand and energy-supply pathways in metabolism throughout the hypoxic period. This recent work begins to clarify how normoxic maintenance ATP turnover rates can be drastically (10-fold) down-regulated to a new hypometabolic steady state, which is prerequisite for surviving prolonged hypoxia or anoxia. The implications of these developments are extensive in biology and medicine.

Pattern and inhibition-dependent invasion of pyramidal cell dendrites by fast spikes in the hippocampus in vivo.

The invasion of sodium spikes from the soma into dendrites was studied in hippocampal pyramidal cells by simultaneous extracellular and intracellular recordings in anesthetized rats and by simultaneous extracellular recordings of the somatic and dendritic potentials in freely behaving animals. During complex-spike patterns, recorded in the immobile or sleeping animal, dendritic invasion of successive spikes was substantially attenuated. Complex-spike bursts occurred in association with population discharge of CA3-CA1 pyramidal cells (sharp wave field events). Synaptic inhibition reduced the amplitude of sodium spikes in the dendrites and prevented the occurrence of calcium spikes. These findings indicate that (i) the voltage-dependent calcium influx into the dendrites is under the control of inhibitory neurons and (ii) the temporal coincidence of synaptic depolarization and activation of voltage-dependent calcium conductances by the backpropagating spikes during sharp wave bursts may be critical for synaptic plasticity in the intact hippocampus.

Deficits in memory and hippocampal long-term potentiation in mice with reduced calbindin D28K expression.

The influx of calcium into the postsynaptic neuron is likely to be an important event in memory formation. Among the mechanisms that nerve cells may use to alter the time course or size of a spike of intracellular calcium are cytosolic calcium binding or "buffering" proteins. To consider the role in memory formation of one of these proteins, calbindin D28K, which is abundant in many neurons, including the CA1 pyramidal cells of the hippocampus, transgenic mice deficient in calbindin D28K have been created. These mice show selective impairments in spatial learning paradigms and fail to maintain long-term potentiation. These results suggest a role for calbindin D28K protein in temporally extending a neuronal calcium signal, allowing the activation of calcium-dependent intracellular signaling pathways underlying memory function.

Mice deficient for prion protein exhibit normal neuronal excitability and synaptic transmission in the hippocampus.

We recorded in the CA1 region from hippocampal slices of prion protein (PrP) gene knockout mice to investigate whether the loss of the normal form of prion protein (PrPC) affects neuronal excitability as well as synaptic transmission in the central nervous system. No deficit in synaptic inhibition was found using field potential recordings because (i) responses induced by stimulation in stratum radiatum consisted of a single population spike in PrP gene knockout mice similar to that recorded from control mice and (ii) the plot of field excitatory postsynaptic potential slope versus the population spike amplitude showed no difference between the two groups of mice. Intracellular recordings also failed to detect any difference in cell excitability and the reversal potential for inhibitory postsynaptic potentials. Analysis of the kinetics of inhibitory postsynaptic current revealed no modification. Finally, we examined whether synaptic plasticity was altered and found no difference in long-term potentiation between control and PrP gene knockout mice. On the basis of our findings, we propose that the loss of the normal form of prion protein does not alter the physiology of the CA1 region of the hippocampus.