The striatum comprises a network characterized by a highly shared feedforward inhibition (FFI) mediated by fast-spiking interneurons (FSI), which constitute only 1% of the striatal population. We investigated the dynamical consequences of this extensively shared FFI beyond inducing synchrony in a local striatal microcircuit. Our findings reveal that increased FFI sharing enhances the across-trial variability of striatal responses, activity of medium spiny neurons (MSNs), to cortical inputs, and endows the striatal network with the capacity to modulate output correlations in a bidirectional manner. Specifically, weakly shared cortical inputs become more correlated, whereas strongly shared cortical inputs are decorrelated in the presence of FSIs. These dynamic modulatory effects on MSNs, in turn, substantially alter the spiking statistics of downstream neurons in the globus pallidus, regarding across-trial variability and burstiness.

Biological neural networks exhibit remarkable diversity, where each neuron type is genetically encoded with certain chemical specificity and morphological properties. These neuron-level specifications, when put together into a network, give rise to various microcircuits in the brain. This thesis investigates how these single neuron properties determine microcircuit connectivity. In turn, the network dynamics emerge from and are constrained by the underlying connectivity.

The chemical specificity is reflected in the neuronal compositions and type-specific connection probabilities. In the cortical microcircuit, pyramidal neurons and three major classes of inhibitory interneurons have a preference for presynaptic sources and postsynaptic targets. At the population level, it gives rise to redundancy in rate coding and non-linearity to the neuronal transfer function. In turn, it supports flexible modulations of across-trial variability by tuning the input distribution across trials. In the striatal microcircuit, fast-spiking interneurons (FSI), despite constituting only around 1\% of the striatal population, exert powerful feedforward inhibition due to their dense and strong axonal projections. The connectivity of FSIs leads to a significant sharing of feedforward inhibition in the major striatal population, medium spiny neurons (MSN). In turn, this shared inhibition results in variable population responses across trials and bidirectional control of the correlation transfer from cortical stimuli to striatal activities.

The morphology of individual neurons is deeply related to the network connectivity as well. In Parkinson’s disease (PD), striatal microcircuits undergo structural changes: FSIs exhibit axonal sprouting, amplifying shared inhibition and synchrony, while MSNs display dendritic atrophy, weakening corticostriatal drive. Compensatory mechanisms, such as synaptic scaling or rewiring, could potentially restore functional balance. In a more general setting, spatial constraints shape higher-order structures of microcircuit connectivity. The arrangement of neurons and the shape of their neurites induce connection preferences in terms of overlaps between axons, dendrites, and axon/dendrite. Such overlaps represent the second-order approximation of connections. A spatially coherent pattern of overlap would generate spatiotemporal activity patterns as an implementation of cell assemblies.

A combination of chemical specificity and morphological properties leads to a complete theoretical framework from single neurons to networks. Experimental observations of type-specific connection rules (both transcriptomic and morphological) would enable the abstraction of distinct brain regions into concrete microcircuits. These microcircuits serve as a biologically plausible baseline of neural network models.

Biologiska neurala nätverk uppvisar en anmärkningsvärd mångfald, där varje neuronsubtyp är genetiskt kodad med en viss kemisk specificitet och morfologiska egenskaper. Dessa egenskaper på neuronivå, när de sammanförs i ett nätverk, ger upphov till olika mikrokretsar i hjärnan. Denna avhandling undersöker hur dessa egenskaper hos enskilda neuroner bestämmer mikrokretsarnas kopplingsmönster. I sin tur uppstår nätverksdynamiken från och begränsas av den underliggande kopplingsstrukturen.

Den kemiska specificiteten återspeglas i neuronsammansättningen och typ-specifika sannolikheter för anslutningar. I den kortikala mikrokretsen har pyramidalneuroner och tre huvudklasser av hämmande interneuroner en preferens för presynaptiska källor och postsynaptiska mål. På populationsnivå leder detta till redundans i frekvenskodning och icke-linearitet i den neurala överföringsfunktionen. Detta stödjer i sin tur flexibla modulationer av variabilitet mellan försök genom att justera inputfördelningen över försök. I den striatala mikrokretsen utövar snabbspikande interneuroner (FSI), trots att de endast utgör cirka 1\% av den striatala populationen, kraftfull feedforward-hämning på grund av sina täta och starka axonala utskott. Kopplingsmönstret hos FSIs leder till en betydande delning av feedforward-hämning i den största striatala populationen, medelstora taggade neuroner (MSN). Denna delade hämning resulterar i variabla populationssvar över försök och en dubbelriktad kontroll av korrelationsöverföringen från kortikala stimuli till striatal aktivitet.

Morfologin hos enskilda neuroner är också starkt kopplad till nätverkets kopplingsmönster. Vid Parkinsons sjukdom (PD) genomgår striatala mikronätverk strukturella förändringar: FSI-neuroner (fast-spiking interneurons) uppvisar axonell utväxt, vilket förstärker gemensam inhibition och synkronitet, medan mediospinyneuroner visar dendritisk atrofi, vilket försvagar den kortikostriatala drivningen. Kompensatoriska mekanismer, såsom synaptisk skalning eller omkoppling, skulle potentiellt kunna återställa den funktionella balansen. I en mer generell kontext formar rumsliga begränsningar högre ordningens strukturer i mikronätverkets kopplingsmönster. Arrangemanget av neuroner och formen på deras neutriter skapar preferenser för kopplingar i termer av överlappningar mellan axoner, dendriter och axon/dendrit. Sådana överlappningar representerar en andraordningsapproximation av kopplingar. Ett rumsligt sammanhängande mönster av överlappningar skulle generera spatiotemporala aktivitetsmönster som en implementering av cellassemblér.

En kombination av kemisk specificitet och morfologiska egenskaper leder till ett komplett teoretiskt ramverk från enskilda neuroner till nätverk. Experimentella observationer av typ-specifika kopplingsregler (både transkriptomiska och morfologiska) skulle möjliggöra abstraktion av olika hjärnregioner till konkreta mikrokretsar. Dessa mikrokretsar fungerar som en biologiskt trovärdig baslinje för neurala nätverksmodeller.

This in silico study predicts the impact that the single-cell neuronal morphological alterations will have on the striatal microcircuit connectivity. We find that the richness in the topological striatal motifs is significantly reduced in Parkinson's disease (PD), highlighting that just measuring the pairwise connectivity between neurons gives an incomplete description of network connectivity. Moreover, we predict how the resulting electrophysiological changes of striatal projection neuron excitability together with their reduced number of dendritic branches affect their response to the glutamatergic drive from the cortex and thalamus. We find that the effective glutamatergic drive is likely significantly increased in PD, in accordance with the hyperglutamatergic hypothesis.

Trial-by-trial variability is a ubiquitous property of neuronal activity in vivo which shapes the stimulus response. Computational models have revealed how local network structure and feedforward inputs shape the trial-by-trial variability. However, the role of input statistics and different interneuron subtypes in this process is less understood. To address this, we investigate the dynamics of stimulus response in a cortical microcircuit model with one excitatory and three inhibitory interneuron populations (PV, SST, VIP). Our findings demonstrate that the balance of inputs to different neuron populations and input covariances are the primary determinants of output trial-by-trial variability. The effect of input covariances is contingent on the input balances. In general, the network exhibits smaller output trial-by-trial variability in a PV-dominated regime than in an SST-dominated regime. Importantly, our work reveals mechanisms by which output trial-by-trial variability can be controlled in a context, state, and task-dependent manner.

The striatum features a network characterized by a high degree of shared feedforward inhibition (FFI) from fast-spiking interneurons (FSI), which constitute only 1\% of the striatal population. We investigated the potential roles of this extensively shared FFI in striatal function beyond inducing synchrony in a local striatal circuit. Our findings reveal that the sharing of FSI increases across-trial variability of striatal responses to cortical stimuli, and gives the striatum an ability to bidirectionally modulate correlation transfer, i.e., a weakly (strongly) correlated signal is correlated (decorrelated) in the presence of FSI. We show that such bidirectional modulation of correlation transfer is only possible due to the high sharing of FSI projections. These results suggest that FSIs may play a critical role in the learning phase for exploration and modulate the integration level of cortical inputs.

Striatum, the input stage of the basal ganglia, is important for sensory-motor integration, initiation and selection of behaviour, as well as reward learning. Striatum receives glutamatergic inputs from mainly cortex and thalamus. In rodents, the striatal projection neurons (SPNs), giving rise to the direct and the indirect pathway (dSPNs and iSPNs, respectively), account for 95% of the neurons and the remaining 5% are GABAergic and cholinergic interneurons. Interneuron axon terminals as well as local dSPN and iSPN axon collaterals form an intricate striatal network. Following chronic dopamine depletion as in Parkinson’s disease (PD), both morphological and electrophysiological striatal neuronal features are altered. Our goal with this \textit{in-silico} study is twofold: a) to predict and quantify how the intrastriatal network connectivity structure becomes altered as a consequence of the morphological changes reported at the single neuron level, and b) to investigate how the effective glutamatergic drive to the SPNs would need to be altered to account for the activity level seen in SPNs during PD. In summary we find that the richness of the connectivity motifs is significantly decreased during PD, while at the same time a substantial enhancement of the effective glutamatergic drive to striatum is present.