Browsing by study line "Neurovetenskap"

Background: Despite the well-established link between diabetes and dementia risk, the impact of prediabetes and diabetes on the prodromal dementia phase remains controversial. In this study, we investigated whether prediabetes and diabetes increase the risk of cognitive impairment–no dementia (CIND) and accelerate its progression to dementia, as well as the possible underlying mechanisms. Methods: In the Swedish National Study on Aging and Care-Kungsholmen (SNAC-K), one cohort of cognitively-intact individuals (n=1,837) and one cohort of individuals with CIND (n=671) aged ≥60 years were followed for up to 15 years. At baseline and each follow-up (every 3 or 6 years), a neuropsychological test battery was administered, and the domains of episodic memory, processing speed, executive function, visuospatial abilities, and language were derived. CIND was defined as having no dementia and cognitive performance ≤1.5 SDs below age group-specific means in at least one cognitive domain. Dementia was diagnosed according to DSM-IV criteria. Diabetes (controlled and poorly-controlled) was diagnosed by physicians through medical assessment, clinical records, and glycated hemoglobin (HbA1c) ≥6.5%. Prediabetes was identified as HbA1c 5.7-6.4% in diabetes-free participants. Clinicians diagnosed heart disease and collected blood samples used to measure C-reactive protein (CRP). Data were analyzed with Cox regression models adjusted for possible confounders. Results: At baseline, in the cognitively-intact cohort, 133 (7%) participants had diabetes and 615 (34%) had prediabetes. During follow-up (mean 9.2 ± 3.0 years [range=2.2-15.5 years]), 544 (30%) individuals in the cognitively-intact cohort developed CIND. Poorly-controlled diabetes (HbA1c ≥7.5%) was associated with 2-times higher risk of CIND (HR 2.0, 95% CI:1.11-3.48) than diabetes-free participants. In the CIND cohort, 84 (13%) had diabetes and 238 (36%) prediabetes. During follow-up (mean 7.7 ± 4.0 years [range=0.2-15.2 years]), 132 (20%) individuals progressed to dementia. Poorly-controlled diabetes was associated with 3-times higher risk of dementia progression (HR 3.3, 95% CI: 1.29-8.33). Furthermore, comorbid heart disease and diabetes was associated with 2.5-times higher risk of progression to dementia (HR 2.5, 95% CI: 1.17-5.47), particularly if the diabetes was poorly-controlled (HR 5.8, 95% CI: 1.72-19.3). Similarly, having elevated CRP levels and diabetes was associated with increased risk of progression to dementia (HR 4.1, 95% CI: 1.15-14.2), especially in participants with poorly-controlled diabetes (HR 13.6, 95% CI: 1.89-98). No associations between prediabetes and CIND were detected in either cohort. Conclusions: Diabetes, especially if poorly-controlled, increases the risk of cognitive impairment and accelerates its progression to dementia. The diabetes-associated progression from CIND to dementia is further exacerbated by the presence of heart disease and elevated levels of systemic inflammation.

Kainate receptors (KARs) act as prominent regulators of neuronal excitability, network activity as well as neurotransmitter release in the developing brain. In the neonatal hippocampus the GluK1 subunit containing KARs take part in regulating the activity of the CA3 interneurons and hence the maintenance of early synchronous network oscillations, which are thought to be vital for developing connections. In the interneurons of the CA3 subfield this regulatory activity is likely performed through a noncanonical, G-protein mediated inhibition of a Ca2+ sensitive medium-duration afterhyperpolarizing current (ImAHP). As in various central neurons the ImAHP has been shown to be regulated by voltage-gated calcium channels (VGCCs) and as the activity of the voltage-gated calcium channels has been in turn shown to be modulated by G-protein coupled signaling of GluK1 KARs, we went on to investigate whether a direct link between GluK1 KARs and VGCCs could be detected in the CA3 stratum radiatum interneurons of neonatal hippocampus. Here we show that the pharmacological inhibition of GluK1 KARs does not affect the amplitude of Ca2+ influx through VGCCs in the CA3 stratum radiatum interneurons of acute hippocampal slices from neonatal mice. As G-protein mediated signaling has been shown to induce alterations in the voltage-dependence of the VGCC-mediated currents, we similarly investigated the effects of GluK1 inhibition on the current-voltage relationship of Ca2+ currents in CA3 interneurons during the first postnatal week as well as during the second postnatal week, since GluK1 subunit is known to undergo developmental changes in its expression during this time. No significant effect was however detected in either of the age groups. Although in our experiments the GluK1-KAR inhibition seemed to induce no statistically significant changes in the Ca2+ current amplitudes or in the voltage-dependence of VGCC-mediated currents in the CA3 interneurons, further, more specific studies should be encouraged to investigate the phenomenon in specific interneuron subtypes and in distinct calcium channel families.

Beta frequency (15-25 Hz) oscillations in the extracellular field potential recorded by cortical EEG and depth electrodes have been connected to stopping. Especially short increases in beta power, so called beta bursts, occur more frequently close to stopping an ongoing movement or when cancelling a planned action. However, there are discrepancies about the causal role of these beta bursts on stopping. Although some studies indicate causality, in others the bursts occur too late for being causal or their number does not increase prior to stopping. One explanation to the disagreement may lie in the behavioral task commonly used to study the neural correlates of action inhibition, the stop signal task. In this task the movement is cancelled before it starts, and actual stopping is thus hidden from the experimenter. Instead, an estimated stop signal reaction time is mathematically modelled. It is likely that this reaction time varies trial by trial, which causes inaccuracy in the results. We were able to define an exact stopping time using head fixed rats running on a treadmill. This enabled us to align brain activity precisely with stopping. With this task, we showed that the number of transient beta bursts increases just prior to stopping. Moreover, the increase correlates with the velocity. These results indicate that beta bursts are causal to stopping. Beta bursts have been noted to be disturbed in Parkinson’s disease and our results may open new doors for early diagnoses or treatments.

Sensory systems display a topographical organization, and in the murine somatosensory system there is oneto-one correspondence between individual whiskers and individual cortical columns called barrels. Functional connectivity in the whisker-to-barrel system is formed prenatally and refined after birth, guided by both spontaneous and whisker-evoked activity. GABAergic connectivity emerges already prenatally and includes transient circuits, but the exact role of GABAergic signalling in early development is elusive. The neuronal, major chloride extruder, potassium-chloride cotransporter (KCC2) is heavily upregulated in the cortex during the first two postnatal weeks resulting in the emergence of hyperpolarizing inhibition. However, in cortical interneurons (INs) KCC2 expression can be detected already at the time of birth. The role of this early interneuronal KCC2 expression is unclear. The aim of this thesis was to study the role of KCC2 in the network activity of cortical INs during the perinatal period. Transgenic mice with conditional inactivation of Kcc2 gene, and expression of the calcium indicator GCaMP6f in GAD2+ neurons (INs) were used to image cortical Ca2+ activity. Transcranial widefield Ca2+ imaging in awake head-fixed mice was performed at the day of birth (P0) and showed that spontaneous, but not evoked, activity was significantly reduced in the knock-out animals. Moreover, immunostaining for the activity-induced transcription factor Egr1 showed that thalamic network activity was significantly decreased in the knock-out and heterozygous animals, suggesting involvement of subcortical areas in the decreased cortical activity. Additional experiments are needed to elucidate the role of other mechanisms contributing to the observed change in activity.

Neurodegenerative disorders are globally the most common cause of disability leading up to 10 million deaths every year, but the mechanisms underlying neurodegeneration are not understood well. Methylation of messenger RNAs (mRNA) at adenosine base position N6 (m6A) by a methyltransferase-complex is a modification that regulates gene expression by influencing mRNA stability, transport, translation and degradation. The mRNA m6A levels are decreased in many neurodegenerative diseases. Our group has shown that dopamine neurons can be rescued, by unknown mechanisms, by activating the methyltransferase-like 3 (Mettl3) enzyme, which increases mRNA m6A levels. The aim of this study was to understand the mechanisms underlying mRNA m6A-induced neuronal survival. We investigated with reverse transcription quantitative polymerase chain reaction (RT-qPCR) how the changes in mRNA m6A modifications affect gene expression. We validated with female rat striatum samples that Mettl3 activation upregulates Neurexophilin-3 (Nxhp3), an important protein for neurotransmitter release and motor functions. Pellino 1 (Peli1), a E3 ubiquiting ligase, was upregulated in nucleus accumbens region of the same rats. Finally, we found with in vitro mouse cortical neurons that Mettl3 inhibition induces endoplasmic reticulum (ER) stress and an unfolded protein response. We saw the activation of inositol-requiring enzyme-1 (Ire1-alpha) and PKR-like ER kinase (Perk) signalling pathways. ER stress is a hallmark of neurodegenerative diseases. This is the first study to show a connection between Mettl3 inhibition and ER stress but the mechanism is still unknow. Further studies need to be performed in order to see if Mettl3 inhibition regulates Ire1-alpha and Perk activity directly, or does Mettl3 inhibition induce ER stress indirectly. Our results highlight how crucial mRNA m6A modification is for neuronal survival.

While weeks of continuous treatment is required for conventional antidepressant drugs (e.g. fluoxetine) to bring their full therapeutic effects, a subanesthetic dose of ketamine alleviates the core symptoms of depression (anhedonia, depressed mood) and suicidal thinking within just few hours and the effects may last for days. Nitrous oxide (N2O, “laughing gas”), another NMDAR antagonist, has recently been shown to have similar rapid antidepressant effects in treatment-resistant depressed patients (Nagele et al. 2015). We previously found using naïve mice ketamine and N2O treatment to upregulate five mRNAs related to the MAPK pathway and synaptic plasticity, both implicated as being important in the action of rapid-acting antidepressants. In the current study, these shared mechanisms were further investigated in C57BL/6JHsd mice, using behavioral test batteries to study depressive-like behaviour and RT-qPCR for biochemical analyses. We first aimed to demonstrate behavioral differences between naïve mice and a chronic corticosterone-induced animal model of depression, and to use this model to investigate antidepressant-like effects of ketamine and N2O. According to the results, chronic corticosterone produced some behaviors typical of a depressive-like phenotype, namely significant worsening of coat state and decreased saccharin consumption in the saccharin preference test. Both ketamine and N2O exhibited antidepressant-like effects by reverting decreased saccharin preference. We then aimed to elucidate the effects of ketamine and N2O on five previously found shared mRNAs: Arc, Dusp1, Dusp5, Dusp6 and Nr4a1. N2O significantly upregulated all targets in the vmPFC, except Dusp5, two hours after beginning of N2O treatment. Neither ketamine nor sole chronic corticosterone produced any significant changes. The results of this study suggest that N2O is a potential candidate for rapid alleviation of depressive symptoms. We suggest that the action of rapid-acting antidepressants, more specifically N2O, is based on a homeostatic response of the brain to a presented challenge. Here this challenge would be cortical excitation previously been shown to be caused by N2O, which leads to activation of pathways such as MAPK and subsequent Arc, Dusp and Nr4a1 signaling. The level of expression of these markers would then depend on which phase this response is in and hence, the differences in time between treatment and brain sample dissection could be a reason for conflicting results to previous research. Future studies would benefit from detailed investigation of the chronic corticosterone-induced model due to its potential in controlling for behavioral variability, thus reducing the number of animals needed for preclinical research. Overall the preliminary findings of this study could be one of the first steps in the search for the mechanisms underlying the potential antidepressant effect of N2O, how these molecular markers are related to its action and how it differs from the action of ketamine.

Schizophrenia (SCZ) is a chronic neuropsychiatric disorder believed to arise from the intricate interplay between genetic predisposition and environmental factors. Though the aetiology of SCZ is unknown many findings support an excessive synaptic pruning hypothesis. Maternal immune activation (MIA), encompassing prenatal infection and systemic inflammation, constitutes a significant environmental risk factor implicated in SCZ onset (Patterson, 2009; Brown, 2012). MIA induces persistent alterations in the microglia of offspring termed microglial priming, characterized by heightened reactivity to inflammatory stimuli (Choudhury and Lennox, 2021). Notably, studies have reported increased sensitivity to activation, elevated expression of inflammatory markers, and an increase in the total number of microglia (Perry and Holmes, 2014; Choudhury and Lennox, 2021). Primed microglia may contribute to excessive synaptic pruning, thereby compromising neuronal connectivity and potentially leading to the onset of SCZ. This thesis investigated the impact of microglia on neurons and explored the microglial tendency for hyperactivation in the context of SCZ predisposition. It utilized induced pluripotent stem cell (iPSC) technology to create a rat astrocyte/unaffected control human iPSC-derived neuron/induced microglia-like cell (iMGL) tri-culture model. Uniquely, iMGLs were differentiated from a library of monozygotic twin lines discordant for SCZ, and unaffected controls. This allows for exploration of the differences between iMGLs from unaffected twins with a genetic predisposition for SCZ, affected twins with clinical manifestation of SCZ, and unaffected controls without a known genetic predisposition for SCZ. The tri-culture system was subjected to lipopolysaccharide (LPS) and polyinosinic:polycytidylic acid (poly(I:C)) treatments to activate iMGLs, and differences in cytokine release, synapse pruning, and neuronal activity were assessed. The principal outcomes of our investigation revealed enhanced cytokine release from SCZ-derived iMGLs when exposed to inflammatory stimuli, alongside increased network connectivity among samples containing genetically predisposed iMGLs. While most of the results did not reach significance, they suggest a potential link between SCZ pathophysiology and hyperactive microglia. Future research will focus on enlarging the study cohort, establishing tri-culture models featuring neurons and iMGLs derived from the iPSCs of the same patient, conducting CBA analysis to confirm the elevated cytokines finding, and scrutinizing iMGL morphology.