The anterior cingulate cortex (ACC) can influence emotional and motivational states in primates by its dense connections with many neocortical and subcortical regions. Pyramidal neurons serve as the basic building blocks of these neocortical circuits, which have been extensively studied in other brain regions, but their morphological and electrophysiological properties in the primate ACC are not well understood. In this study, we used whole-cell patch clamp and high-resolution laser scanning confocal microscopy to reveal the general electrophysiological properties and detailed morphological features of layer 2 and 3 pyramidal neurons in ACC (area 24/32) of the rhesus monkey. Neurons from both layers had similar passive membrane properties and action potential properties. Morphologically, dendrites of layer 3 ACC neurons were more complex than those of layer 2 neurons, by having dendrites with longer total dendritic lengths, more branch points and dendritic segments, spanning larger convex hull volumes. This difference in total dendritic morphology was mainly due to the apical dendrites. In contrast, the basal dendrites displayed mostly similar features between the two groups of neurons. However, while apical dendrites extend to the same layer (layer 1), the basal dendrites of layer 3 extended into deeper layers than layer 2 because of the difference in soma-pia distance. Thus, basal dendrites of the two groups of neurons receive different laminar inputs. Analysis of spines showed that more spines were found in neurons of layer 3 apical dendritic arbors than layer 2 neurons. However, the apical spine densities were similar between neurons in the two layers. Thus, while higher spine number suggests that layer 3 neurons receive more excitatory input than layer 2 neurons, the similar spine density suggests similar spatial and temporal summation of these inputs. The combined effects of increased number of excitatory input and higher dendritic complexity in layer 3 than in layer 2 ACC neurons suggest the additional information received by layer 3 neurons, especially in the apical dendrites, might undergo more complex integration.

The central nervous system (CNS) is protected by a vascular blood-brain barrier that prevents many types of pathogens from entering the brain. Still, some pathogens have evolved mechanisms to traverse this barrier and establish a long-term infection. The apicomplexan parasite, Toxoplasma gondii, is one such pathogen with the ability to infect the CNS in virtually all warm-blooded animals, including humans. Across the globe, an estimated 30% of the human population is infected with Toxoplasma, an infection for which mounting evidence suggests increases the risk for developing neurological and neuropsychiatric disorders, like seizures and schizophrenia. In my dissertation, I investigate the telencephalic neural circuit changes induced by long-term Toxoplasma infection in the mouse brain and the neuroimmune signaling role of the complement system in mediating microglial remodeling of neural circuits following parasitic infection. While there has been keen interest in investigating neural circuit changes in the amygdala – a region of the brain involved in fear response and which Toxoplasma infection alters in many species of infected hosts – the hippocampus and cortex have been less explored. These are brain regions for which Toxoplasma also has tropism, and moreover, are rich with fast-spiking parvalbumin perisomatic synapses, a type of GABAergic synapse whose dysfunction has been implicated in epilepsy and schizophrenia. By employing a range of visualization techniques to assess cell-to-cell connectivity and neuron-glia interactions (including immunohistochemistry, ultrastructural microscopy, and microglia-specific reporter mouse lines), I discovered that longterm Toxoplasma infection causes microglia to target and ensheath neuronal somata in these regions and subsequently phagocytose their perisomatic inhibitory synapses. These findings provide a novel model by which Toxoplasma infection within the brain can lead to seizure susceptibility and a wider range of behavioral and cognitive changes unrelated to fear response. In the Toxoplasma infected brain, microglia, along with monocytes recruited to the brain from the periphery, coordinate a neuroinflammatory response against pathogenic invasion. This is characterized by a widespread activation of these cells and their increased interaction with neurons and their synaptic inputs. Yet, whether T. gondii infection triggers microglia and monocytes (i.e. phagocytes) to target, ensheath, and remove perisomatic inhibitory synapses on neuronal somata indiscriminately, or whether specificity exists in this type of circuit remodeling, remained unclear. Through a comprehensive assessment of phagocyte interactions with cortical neuron subtypes, I demonstrate that phagocytes selectively target and ensheath excitatory pyramidal cells in long-term infection. Moreover, coupling of in situ hybridization with transgenic reporter lines and immunolabeling revealed that in addition to phagocytes, excitatory neurons also express complement component C3 following infection (while inhibitory interneurons do not). Lastly, by employing targeted deletion of complement components, C1q and C3, I show that complement is required for phagocyte ensheathment of excitatory cells and the subsequent removal of perisomatic inhibitory synapses on these cells (albeit not through the classical pathway). Together, these studies highlight a novel role for complement in mediating synapse-type and cell-type specific circuit remodeling in the Toxoplasma infected brain. / Doctor of Philosophy / Parasites are microorganisms that rely on other living organisms (called hosts) for their survival. Although some parasites only live on their hosts, others have developed ways to establish infections and obtain the nutrients that keep them alive from host cells. My Ph.D. research has focused on studying one of these parasites, Toxoplasma gondii (commonly referred to as Toxo), that has evolved the unique ability to establish brain infections in almost all animals around the world, from rodents to humans. Recent discoveries show that brain infection with this parasite can cause seizures, an imbalance in the way that specialized cells of the brain (called neurons) communicate with each other, causing harmful hyperactivity within the brain. Toxo infection can also cause behavioral and cognitive changes in infected animals, making them more susceptible to predation. In humans, infection with Toxo increases their risk for developing different types of mental illness, such as schizophrenia. The focus of my Ph.D. research has been in trying to understand, at the cellular and molecular level, how infection with this parasite can lead to seizures and behavioral changes, by using mice as a model. Mice have a similar brain structure to humans, and over the years, scientists have developed many tools that allow us to visualize and study the connections between neurons (called synapses). I'm interested in understanding how changes in these connections affect how neurons communicate with each other, and ultimately, how we behave and think. I have been studying a type of connection that, if lost or damaged, can lead to seizures and some types of mental illness. These connections are called 'perisomatic inhibitory synapses', and they form on many distinct types of neurons, but specifically on the cell bodies of these neurons. They act as a traffic light, informing neurons when and for how long to 'slow down' their activity. I discovered that after the parasite enters the brain, it causes another type of cell in the brain, called microglia, to extensively interact with neurons in the cortex and hippocampus (areas of your brain important for thinking, executing behavior, and learning). Microglia are immune cells of the brain that inspect the brain for anything damaged or that doesn't belong (like parasites) and removes them from the brain. By performing experiments where I delete individual immune molecules from mice, I found that one immune molecule, called 'complement component C3' acts as cue for microglia to find these cells, wrap around them, and permanently remove these important connections. Surprisingly, however, microglia don't remove these connections from all neurons, indiscriminately, they do so only on one specific cell type called 'excitatory pyramidal neurons,' and as the name implies, they're the ones who drive activity in the brain. My half-a-decade's worth of research helps us understand parasitic infections in the brain in a couple of ways: First, I have discovered one of the mechanisms by which neuronal connections are lost in the Toxo-infected brain (which is a mechanism that leads to loss of neuronal connections in the injured and aging brain as well). This is significant because it might provide insight into why some people who are infected with Toxo develop seizures or mental illness, while others don't. More importantly, Toxo-infection causes changes in the brain that are very specific, in terms of both the type of neuronal connection that is affected and the type of cell that is affected. Why these changes are so specific remain to be uncovered, but it suggests that Toxo can either a) trigger a unique immune response in the brain that leads to very precise changes in neuron-toneuron connections and signaling or b) the parasite, while hiding inside of neurons, may hijack the machinery of certain cell types in a way that helps them survive longer.

The sigma-1 receptor (sig-1R) is a unique endoplasmic reticulum (ER) chaperone protein that interacts with a variety of voltage- and ligand-gated ion channels, which are components of an intricate system that regulates neuronal functioning. While there is an extensive body of knowledge pertaining to the sig-1R, many questions remain. The first question this thesis addresses is how the sig-1R modulates the functioning of the N-methyl-D-aspartate receptor (NMDAR). Using a heterologous expression system, I provide evidence that the mechanism of modulation is likely not a direct interaction between sig-1R and NMDAR and that this is not affected by the presence or absence of the membrane-associated guanylate kinases (MAGUK) protein PSD-95. The next question addressed investigates the impact of sig-1R absence on the synaptic physiology and action potential firing of CA1 pyramidal neurons. It was found that there is not a significant difference in these parameters, suggesting a non-essential role of the sig- 1R under normal physiological conditions. The third topic covered in my studies explores the sig-1R KO mouse in the Aβ25-35 infusion model of Alzheimer’s disease (AD). Preliminary results suggest that there is a dysfunction in the action potential characteristics and after- hyperpolarization characteristics of challenged sig-1R KO mice. Overall my results provide the groundwork for future experiments that will lead to a better understanding of the sig-1R and its role in cellular and synaptic physiology.

付記する学位プログラム名: 京都大学卓越大学院プログラム「メディカルイノベーション大学院プログラム」 / 京都大学 / 新制・課程博士 / 博士(医学) / 甲第24509号 / 医博第4951号 / 新制||医||1064(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 林 康紀, 教授 渡邉 大, 教授 高橋 淳 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

Pesquisas in vivo e in vitro, têm estabelecido uma correlação entre alterações na morfologia dendrítica e a epilepsia. No entanto, ainda não se conhecem em detalhe as consequências dessas modificações, sobre a eletrofisiologia e o padrão de disparo. Também existe um fenômeno que não tem sido completamente explicado, conhecido como o paradoxo do dendrito epiléptico, no qual neurônios piramidais, mesmo com a diminuição dramática do principal lugar de inervação glutamatérgica (como consequência de, por exemplo, uma redução do diâmetro e comprimento das árvores dendríticas), inesperadamente apresentam um estado de hiperexcitabilidade crônica. Nesta pesquisa foram aproveitadas as vantagens de uma abordagem neurocomputacional, para induzir sistematicamente alterações na arquitetura dendrítica do mesmo tipo às observadas na epilepsia, e avaliar os seus efeitos sobre a eletrofisiologia e o padrão de disparo. Para isso foi construído um modelo computacional biologicamente realista, de um neurônio piramidal do neocórtex. O código-fonte do modelo está na linguagem do NEURON, e foi baseado em dados eletrofisiológicos (i.e. propriedades da membrana e condutâncias iônicas) e morfométricos, obtidos in vitro previamente por outros pesquisadores. A análise foi feita com base em parâmetros eletrofisiológicos do padrão de disparo. O nosso modelo sugere uma influencia muito forte da morfologia dendrítica sobre a eletrofisiologia, a geração de potencias de ação e o padrão de disparo. Os resultados obtidos mostram que, mesmo mantendo constantes todos parâmetros biofísicos (que têm a ver com as dinâmicas elétricas dos canais iônicos), é possível induzir um aumento grande no comportamento elétrico e na geração de potenciais de ação, a partir da redução do diâmetro e comprimento das ramificações das árvores dendríticas. Estes resultados, também permitem contribuir no fornecimento de uma explicação para o paradoxo mencionado. / In vivo and in vitro studies had found a correlation between dendritic morphology alterations and epilepsy. Nevertheless, it has not been established in detail the consequences of those modifications, over the electrophysiology and firing pattern. There is also a phenomenon still not completely understood, known as the epileptic dendrite paradox, in which pyramidal neurons with a dramatic reduction in the principal place of glutamatergic innervation (due to, for example, a loss in dendritic trees\' diameter and length), unexpectedly present a chronic hyperexcitable state. In this study we took advantage of a neurocomputational approach, to systematically induce dendritic alterations of the same type as observed in Epilepsy, and evaluate their effect over the electrophysiology and firing behavior. With that purpose in mind, we constructed a biologically realistic computational model of a pyramidal neuron of the neocortex. For this model, it was implemented the programming language (hoc) of the NEURON software, and was elaborated based on electrophysiological data (i.e. membrane properties and ionic conductances), and morphological measurements, taken in vitro previously by other investigators. The analysis was done from electrophysiological parameters of the firing pattern. Our model suggests a great influence of dendritic morphology over the electrophysiology, spike generation and firing pattern. The results obtained show that, even when all the biophysical parameters involved in ion channel dynamics are maintained constant, it is possible to induce a strong increase in electric behavior and spike firing, from a reduction in the length and diameter of the dendritic trees\' ramifications. These results, also contribute to a explanation of the mentioned paradox.

Over time astrocytes have been thought to function in an auxiliary manner, providing neurons with metabolic and structural support. However, recent research suggests they may play a fundamental role in the generation and propagation of focal epileptic seizures by causing synchronized electrical bursts in neurons. It would be helpful to have a simple mathematical model that represents this dynamic and incorporates these updated experimental results. We have created a two-compartment model of a typical neuron found in the hippocampal CA1 region, an area often thought to be the origin of these seizures. The focus is on properly modeling the astrocytic input to examine the pathological excitation of these neurons and subsequent transmission of the signals. In particular, we consider the intracellular astrocytic calcium fluctuations which are associated with slow inward currents in neighbouring neurons. Using our model, a variety of experimental results are reproduced, and comments are made about the potential differences between graded and “all-or-none” astrocytes.

Over time astrocytes have been thought to function in an auxiliary manner, providing neurons with metabolic and structural support. However, recent research suggests they may play a fundamental role in the generation and propagation of focal epileptic seizures by causing synchronized electrical bursts in neurons. It would be helpful to have a simple mathematical model that represents this dynamic and incorporates these updated experimental results. We have created a two-compartment model of a typical neuron found in the hippocampal CA1 region, an area often thought to be the origin of these seizures. The focus is on properly modeling the astrocytic input to examine the pathological excitation of these neurons and subsequent transmission of the signals. In particular, we consider the intracellular astrocytic calcium fluctuations which are associated with slow inward currents in neighbouring neurons. Using our model, a variety of experimental results are reproduced, and comments are made about the potential differences between graded and “all-or-none” astrocytes.

Pesquisas in vivo e in vitro, têm estabelecido uma correlação entre alterações na morfologia dendrítica e a epilepsia. No entanto, ainda não se conhecem em detalhe as consequências dessas modificações, sobre a eletrofisiologia e o padrão de disparo. Também existe um fenômeno que não tem sido completamente explicado, conhecido como o paradoxo do dendrito epiléptico, no qual neurônios piramidais, mesmo com a diminuição dramática do principal lugar de inervação glutamatérgica (como consequência de, por exemplo, uma redução do diâmetro e comprimento das árvores dendríticas), inesperadamente apresentam um estado de hiperexcitabilidade crônica. Nesta pesquisa foram aproveitadas as vantagens de uma abordagem neurocomputacional, para induzir sistematicamente alterações na arquitetura dendrítica do mesmo tipo às observadas na epilepsia, e avaliar os seus efeitos sobre a eletrofisiologia e o padrão de disparo. Para isso foi construído um modelo computacional biologicamente realista, de um neurônio piramidal do neocórtex. O código-fonte do modelo está na linguagem do NEURON, e foi baseado em dados eletrofisiológicos (i.e. propriedades da membrana e condutâncias iônicas) e morfométricos, obtidos in vitro previamente por outros pesquisadores. A análise foi feita com base em parâmetros eletrofisiológicos do padrão de disparo. O nosso modelo sugere uma influencia muito forte da morfologia dendrítica sobre a eletrofisiologia, a geração de potencias de ação e o padrão de disparo. Os resultados obtidos mostram que, mesmo mantendo constantes todos parâmetros biofísicos (que têm a ver com as dinâmicas elétricas dos canais iônicos), é possível induzir um aumento grande no comportamento elétrico e na geração de potenciais de ação, a partir da redução do diâmetro e comprimento das ramificações das árvores dendríticas. Estes resultados, também permitem contribuir no fornecimento de uma explicação para o paradoxo mencionado. / In vivo and in vitro studies had found a correlation between dendritic morphology alterations and epilepsy. Nevertheless, it has not been established in detail the consequences of those modifications, over the electrophysiology and firing pattern. There is also a phenomenon still not completely understood, known as the epileptic dendrite paradox, in which pyramidal neurons with a dramatic reduction in the principal place of glutamatergic innervation (due to, for example, a loss in dendritic trees\' diameter and length), unexpectedly present a chronic hyperexcitable state. In this study we took advantage of a neurocomputational approach, to systematically induce dendritic alterations of the same type as observed in Epilepsy, and evaluate their effect over the electrophysiology and firing behavior. With that purpose in mind, we constructed a biologically realistic computational model of a pyramidal neuron of the neocortex. For this model, it was implemented the programming language (hoc) of the NEURON software, and was elaborated based on electrophysiological data (i.e. membrane properties and ionic conductances), and morphological measurements, taken in vitro previously by other investigators. The analysis was done from electrophysiological parameters of the firing pattern. Our model suggests a great influence of dendritic morphology over the electrophysiology, spike generation and firing pattern. The results obtained show that, even when all the biophysical parameters involved in ion channel dynamics are maintained constant, it is possible to induce a strong increase in electric behavior and spike firing, from a reduction in the length and diameter of the dendritic trees\' ramifications. These results, also contribute to a explanation of the mentioned paradox.

Basische Helix-Loop-Helix (bHLH)-Proteine bilden eine diverse Gruppe evolutionär gut konservierter Transkriptionsfaktoren. Viele transaktivierende bHLH-Proteine werden zelltyp- oder gewebespezifisch exprimiert und fungieren als wichtige Schlüsselregulatoren zellulärer Determinations- und Differenzierungsprozesse. Die eng verwandten neuronalen bHLH-Gene Neurod1, Neurod2 und Neurod6 werden in differenzierenden Pyramidenneuronen des sich entwickelnden zerebralen Kortex exprimiert und stehen schon lange im Verdacht, deren Reifung zu steuern. In der Vergangenheit wurde jedes der drei Gene in Mäusen inaktiviert. Untersuchungen an den einfach-defizienten Tieren konnten jedoch keine wichtigen Funktionen in embryonalen Pyramidenneuronen identifizieren. Da die Aminosäuresequenzen und die Expressionsmuster der Faktoren sehr ähnlich sind, wurde angenommen, dass sie sich redundante Funktionalität teilen. Um dies zu überprüfen, habe ich Neurod2/6-doppel-defiziente Tiere gezüchtet und unter besonderer Berücksichtigung der Differenzierung von Pyramidenneuronen und der Konnektivität des zerebralen Kortex analysiert: Die Experimente zeigen, dass Neurod2 und Neurod6 tatsächlich mehrere bisher unbekannte gemeinsame Funktionen haben, wobei jeder Faktor für den Verlust des jeweils anderen kompensieren kann. Zumindest eines der beiden Gene ist notwendig für (1) die Kontrolle der radialen Migration eines Teils der Pyramidenneurone, (2) die frühe Regionalisierung des zerebralen Kortex und (3) die Bildung kortikaler Projektionen vom Neokortex zum Striatum, zum Thalamus und zur kontralateralen Hemisphäre. Callosale Axone bilden in Neurod2/6-doppel-defizienten Mäusen Faserbündel die tangential in den medialen Kortex einwachsen, aber noch vor Erreichen des ipsilateralen Cingulums und vor dem Kontakt mit der Mittellinie stoppen und defaszikulieren. Es resultiert eine neue Variante der callosalen Agenesie, die nahelegt, dass es bisher nicht identifizierte Wachstumssignale im medialen Kortex gibt. Die Expression von Neurod1, welche sich normalerweise auf die Subventrikularzone beschränkt, persistiert in radial migrierenden Pyramidenneuronen der Intermediärzone und der Kortikalplatte von Neurod2/6-doppel-defizienten Mäusen. Diese ektopische Neurod1-Expression kann dort den Verlust von Neurod2 und Neurod6 kompensieren. In einem weiteren Schritt habe ich konditionale Neurod1/2/6-tripel-defiziente Mäuse gezüchtet. In diesen Tieren wird das Neurod1-Gen durch selektive genetische Rekombination in all jenen Zellen, die über Neurod6-Promoteraktivität verfügen, irreversibel entfernt: Wie erwartet, teilt sich Neurod1 weitere gemeinsame Funktionen mit Neurod2 und Neurod6. Zumindest eines der drei Gene ist notwendig für die Differenzierung hippokampaler Pyramidenzellen und die Hemmung des programmierten Zelltods der unreifen Neuronen des Cornu Ammonis. Während die gemeinsame Inaktivierung von Neurod1/2/6 zur Aplasie des Hippocampus führt, überlebt ein Großteil der neokortikalen Pyramidenzellen. Die terminale neuronale Differenzierung ist jedoch auch im Neokortex gestört und die neokortikale Konnektivität sehr stark reduziert. Diese Arbeit zeigt, dass die Transkriptionsfaktoren der NeuroD-Familie gemeinsam die Differenzierung, das Überleben, die Migration und das axonale Wachstum von pyramidalen Neuronen des sich entwickelnden zerebralen Kortex steuern. Während der Embryonalentwicklung ergeben sich folgende, teils überschneidende Funktionen der NeuroD-Gene: Die Differenzierung und das Überleben von hippocampalen Körnerzellen ist abhängig von Neurod1. Die frühen Schritte der Differenzierung von hippocampalen Pyramidenneuronen und deren Überleben sind eine Funktion von wahlweise Neurod1, Neurod2 oder Neurod6. Spätere neuronale Differenzierungsschritte, die Regionalisierung des Neokortex und das gezielte Wachstum wichtiger neokortikaler Faserzüge basieren auf Funktionen von Neurod2 oder Neurod6, aber nicht von Neurod1. Der postnatale Umbau des somatosensorischen Kortex und die funktionale Integration thalamischer Afferenzen wurden bereits als strikt Neurod2-abhängig beschrieben.

Unter Spikelets versteht man kleine Depolarisationen mit einer Spike-ähnlichen Wellenform, die man in intrazellulären Ableitungen von verschiedenen Neuronentypen messen kann. In kortikalen Pyramidenzellen wurde ausgeprägte Spikelet-Aktivität nachgewiesen, die erheblich das Membranpotential beeinflussen kann (Crochet et al., 2004; Epsztein et al., 2010; Chorev and Brecht, 2012). Nichtsdestotrotz bleibt der Ursprung von Spikelets in diesen Neuronen unbekannt. In der vorgelegten Arbeit nutzte ich theoretische Modellierung um die Mechanismen von Spikelet-Erzeugung in Pyramidenzellen zu untersuchen. Zuerst sah ich die verschiedenen Hypothesen über den Ursprung von Spikelets durch. In der Literatur entdeckte ich zwei verschiedene Typen von Spikelets. Diese Arbeit konzentriert sich auf den häufiger vorkommenden Typ von Spikelets, welcher durch relativ große Amplituden gekennzeichnet ist. Die Eigenschaften dieser Spikelets passen am besten zu einem axonal Erzeugungsmechanismus. Im zweiten Kapitel widmete ich mich der Hypothese, dass somatische Spikelets axonalen Ursprungs mit somato-dendritischen Inputs hervorgerufen werden können. Ich identifizierte Bedingungen, die es erlauben ein Aktionspotential (AP) am Initialsegment vom Axon (AIS) zu initiieren, welches sich entlang des Axons ausbreitet, aber kein AP im Soma auslöst. Schließlich simulierte ich extrazelluläre Wellenformen von APs und Spikelets und verglich sie mit experimentellen Daten (Chorev and Brecht, 2012). Dieser Vergleich zeigte auf, dass die extrazellulären Wellenformen von Spikelets, die innerhalb einer Zellen am AIS erzeugt werden, gut zu den Daten passen. Zusammenfassend unterstützen meine Ergebnisse die Hypothese, dass Spikelets in Pyramidenzellen am AIS entstehen. Dieser Mechanismus könnte ein Mittel zum Energiesparen bei der Erzeugung von Output-APs sein. Außerdem könnte dadurch die dendritische Plastizität, die auf der Rückwärtspropagierung von APs beruht, reguliert werden. / Spikelets are transient spike-like depolarizations of small amplitudes that can be measured in somatic intracellular recordings of many neuron types. Pronounced spikelet activity has been demonstrated in cortical pyramidal neurons in vivo (Crochet et al., 2004; Epsztein et al., 2010; Chorev and Brecht, 2012), influencing membrane voltage dynamics including action potential initiation. Nevertheless, the origin of spikelets in these neurons remains elusive. In thi thesis, I used computational modeling to examine the mechanisms of spikelet generation in pyramidal neurons. First, I reviewed the hypotheses previously suggested to explain spikelet origin. I discovered two qualitatively different spikelet types described in the experimental literature. This thesis focuses on the more commonly reported spikelet type, characterized by relatively large amplitudes of up to 20 mV. I found that the properties of these spikelets fit best to an axonal generation mechanism. Second, I explored the hypothesis that somatic spikelets of axonal origin can be evoked with somato-dendritic inputs. I identified the conditions allowing these orthodromic inputs to trigger an action potential at the axon initial segment, which propagates along the axon to the postsynaptic targets, but fails to elicit an action potential in the soma and the dendrites. Third, I simulated extracellular waveforms of action potentials and spikelets and compared them to experimental data (Chorev and Brecht, 2012). This comparison demonstrated that the extracellular waveforms of single-cell spikelets of axonal origin are consistent with the data. Together, my results suggest that spikelets in pyramidal neurons might originate at the axon initial segment within a single cell. Such a mechanism might be a way of reducing the energetic costs associated with the generation of output action potentials. Moreover, it might allow to control the dendritic plasticity by backpropagating action potentials.