Daniela Bianchi

Computational modeling of the effects of amyloid-beta on release probability at hippocampal synapses.

The role of amyloid beta (Aβ) in brain function and in the pathogenesis of Alzheimers disease (AD) remains elusive. Recent publications reported that an increase in Aβ concentration perturbs pre-synaptic release in hippocampal neurons. In particular, it was shown in vitro that Aβ is an endogenous regulator of synaptic transmission at the CA3-CA1 synapse, enhancing its release probability. How this synaptic modulator influences neuronal output during physiological stimulation patterns, such as those elicited in vivo, is still unknown. Using a realistic model of hippocampal CA1 pyramidal neurons, we first implemented this Aβ-induced enhancement of release probability and validated the model by reproducing the experimental findings. We then demonstrated that this synaptic modification can significantly alter synaptic integration properties in a wide range of physiologically relevant input frequencies (from 5 to 200 Hz). Finally, we used natural input patterns, obtained from CA3 pyramidal neurons in vivo during free exploration of rats in an open field, to investigate the effects of enhanced Aβ on synaptic release under physiological conditions. The model shows that the CA1 neuronal response to these natural patterns is altered in the increased-Aβ condition, especially for frequencies in the theta and gamma ranges. These results suggest that the perturbation of release probability induced by increased Aβ can significantly alter the spike probability of CA1 pyramidal neurons and thus contribute to abnormal hippocampal function during AD.

On the mechanisms underlying the depolarization block in the spiking dynamics of CA1 pyramidal neurons

Under sustained input current of increasing strength neurons eventually stop firing, entering a depolarization block. This is a robust effect that is not usually explored in experiments or explicitly implemented or tested in models. However, the range of current strength needed for a depolarization block could be easily reached with a random background activity of only a few hundred excitatory synapses. Depolarization block may thus be an important property of neurons that should be better characterized in experiments and explicitly taken into account in models at all implementation scales. Here we analyze the spiking dynamics of CA1 pyramidal neuron models using the same set of ionic currents on both an accurate morphological reconstruction and on its reduction to a single-compartment. The results show the specific ion channel properties and kinetics that are needed to reproduce the experimental findings, and how their interplay can drastically modulate the neuronal dynamics and the input current range leading to a depolarization block. We suggest that this can be one of the rate-limiting mechanisms protecting a CA1 neuron from excessive spiking activity.

Effects of increasing CREB-dependent transcription on the storage and recall processes in a hippocampal CA1 microcircuit.

The involvement of the hippocampus in learning processes and major brain diseases makes it an ideal candidate to investigate possible ways to devise effective therapies for memory‐related pathologies like Alzheimers Disease (AD). It has been previously reported that augmenting CREB activity increases the synaptic Long‐Term Potentiation (LTP) magnitude in CA1 pyramidal neurons and their intrinsic excitability in healthy rodents. It has also been suggested that hippocampal CREB signaling is likely to be down‐regulated during AD, possibly degrading memory functions. Therefore, the concept of CREB‐based memory enhancers, i.e. drugs that would boost memory by activation of CREB, has emerged. Here, using a model of a CA1 microcircuit, we investigate whether hippocampal CA1 pyramidal neuron properties altered by increasing CREB activity may contribute to improve memory storage and recall. With a set of patterns presented to a network, we find that the pattern recall quality under AD‐like conditions is significantly better when boosting CREB function with respect to control. The results are robust and consistent upon increasing the synaptic damage expected by AD progression, supporting the idea that the use of CREB‐based therapies could provide a new approach to treat AD.

Kohonen neural networks and genetic classification

We discuss the property of a.e. and in mean convergence of the Kohonen algorithm considered as a stochastic process. The various conditions ensuring a.e. convergence are described and the connection with the rate decay of the learning parameter is analyzed. The rate of convergence is discussed for different choices of learning parameters. We prove rigorously that the rate of decay of the learning parameter which is most used in the applications is a sufficient condition for a.e. convergence and we check it numerically. The aim of the paper is also to clarify the state of the art on the convergence property of the algorithm in view of the growing number of applications of the Kohonen neural networks. We apply our theorem and considerations to the case of genetic classification which is a rapidly developing field.

Identifying short motifs by means of extreme value analysis

The problem of detecting a binding site —a substring of DNA where transcription factors attach— on a long DNA sequence requires the recognition of a small pattern in a large background. For short binding sites, the matching probability can display large fluctuations from one putative binding site to another. Here we use a self-consistent statistical procedure that accounts correctly for the large deviations of the matching probability to predict the location of short binding sites. We apply it in two distinct situations: a) the detection of the binding sites for three specific transcription factors on a set of 134 estrogen-regulated genes; b) the identification, in a set of 138 possible transcription factors, of the ones binding a specific set of nine genes. In both instances, experimental findings are reproduced (when available) and the number of false positives is significantly reduced with respect to the other methods commonly employed.

Storage and retrieval of ultrametric patterns in a network of CA1 neurons of the hippocampus

We explore the possibility of storage and retrieval of ultrametrically organized patterns in hippocampus, the part of the brain devoted to the memory processes. The ultrametric structure has been chosen for having a good representation of the categories of memory. The storage and retrieval process is the one typical of the hippocampus and it is based on the dynamic of the CA1 neurons under the input from the neurons of the Enthorinal cortex and the Ca3 system. We explore if this real system of neurons exhibits the property of associative memory introduced since a long time in the artificial neural networks. We study how the performance is dependent on the deviation of the system of patterns from ultrametricity. The evolution of the system is simulated by means of a parallel computer and the statistics of storage and retrieval is investigated.