EFFECTS OF OXYGEN DYSREGULATION ON CENTRAL NERVOUS SYSTEM: A FOCUS ON COGNITIVE DETERIORATION

Introduction and aims. The brain is the more susceptible organ to changes in oxygen levels that could cause several pathologies in the central nervous system (CNS). In this context, maintaining physiological oxygen levels is crucial for brain health as well as studies on mechanisms that lead to oxygen dysregulations and cognitive deficits could be of interest to correct oxygen-based therapies or prevent derived cognitive decline. In this framework, obstructive sleep apnoea (OSAS), a sleep breathing disorder, seems to be related to mild cognitive impairment (MCI) insurgence, that often could progress into AD. OSAS is characterized by oscillation of oxygen partial pressure (pO2) levels in the blood, caused by upper airway collapse and obstruction during nighttime; this phenomenon leads to inconstant O2 levels in tissues causing Intermittent Hypoxia (IH) with consequent oxidative stress and damage in cells. This pathology is currently treated with surgical interventions on the upper airway or by the use of continuous positive air pressure (CPAP) to restore correct airflow. Recent studies demonstrated that CPAP could revert cognitive functions in OSAS patients. Parallelly, also supraphysiological oxygen levels (hyperoxia) could have deleterious effects on CNS. This condition occurring in pre-term infants is caused by mechanical ventilation. Indeed, previous studies demonstrated that even if higher oxygen percentages could reduce mortality rate, they are responsible for the insurgence of several complications, as retinopathy of prematurity (ROP) or cognitive deficits in adulthood. For these reasons, this research work was mainly focused on the study of mechanisms at the basis of insufficient or supraphysiological oxygen levels and cognitive decline. In this context, this research work had 3 main aims: (1) the development of a new IH in vitro model to reproduce oxygen oscillation on CNS cells to study mechanisms that link IH and cognitive impairment; (2) research of new possible biomarkers IH-related in plasma of OSAS patients to discriminate patients with MCI from those without comorbidity; (3) investigate possible effects of hyperoxia on microglial cells and vasculature in the hippocampus in brain by using a mouse model of hyperoxia, previously used as Oxygen-induced retinopathy (OIR) model. Results- Part I For the first aim of the project (1), an immortalized human microglia cell line (C20) was used, since microglial cells represent the first immune defence of CNS. After model set-up and studies of IH effects on microglial cells, the in vitro model was then used on a new direct co-culture of differentiated human neural progenitor cells (hNPCs) and microglia to assess IH effects on a more complete CNS in vitro model and to assess IH-primed microglial effects on neurons. The in vitro model is made of 4 elements connected each other: gas bottles as sources for N2, O2 and CO2, a gas mixer used to obtain desired gas mixture for experiments and two boxes for Multiwell plates or Petri dishes for cell cultures realized by a 3D printer; an incubator to maintain physiological temperature (37° C). Cells were subjected to IH by using the model described above and particular plates or dishes with a gas semi-permeable membrane to ensure adequate and rapid gas exchanges. Firstly, experimental conditions were set up and the hypoxic state and mitochondrial activity of cells were assessed after treatments compared to cells in normoxic conditions. Hypoxic state after treatments was evaluated by measuring EF5-derived hypoxic adducts formation, HIF-1α protein levels and oxidative stress (OS) by reactive oxygen species (ROS) production assessment. Mitochondrial activity was evaluated by MTS assay. Then, microglial inflammatory status was studied and protein levels or mRNA expression of some inflammatory markers (NF-B and IL-6) and membrane markers (CX3CR1, CD-86, HLA-DRα) were evaluated. Results showed that after IH treatment followed by a normoxia period (IH/N) cells presented an increase in “microglial priming” markers expression. For this reason, cells were then subjected to a new protocol of IH followed by inflammatory stimulus (i.e. administration of IL-1β) and inflammatory markers were measured. Results obtained support that after this protocol, cells had an exaggerated inflammatory response, and microglial cells assumed a “priming phenotype”. Moreover, after the in vitro model set-up and investigations of microglial response to IH, the same system was used on a new direct co-culture of differentiated human neural progenitor cells (hNPCs) and microglial cells. Qualitative immunofluorescence analysis of microglial (IBA-1) and neuronal markers (TUBB3) was performed to assess that two cell lines can be maintained in co-culture and metabolic activity was assessed by MTS assay. Then, neuronal cells’ number and neuritic outgrowth were assessed after IH/N by immunofluorescence images analysis. Alongside, astrocytic marker GFAP was evaluated by measuring mRNA and protein levels. Lastly, also axonal marker (Tau) mRNA and protein levels and its phosphorylated form levels were evaluated. More relevant preliminary data showed that primed microglial influences neuronal fate, and in particular neuritic outgrowth and axonal structure. Results-Part II For the second aim (2) of this project, 30 OSAS patients and 15 healthy controls were enrolled thanks to the collaboration with Sleep Centre of Neurologic Unit of Azienda Ospedaliera Universitaria Pisana. Patients and controls were evaluated by clinical and neuropsychological tests and blood sampling. Demographic or clinical parameters did not show significant differences between OSAS patients and healthy controls, while polygraphic parameters evaluated (ODI, AHI and T90) were increased in OSAS patients. Plasma was isolated from blood samples and was used to perform biochemical analysis of HIF-1α and p-Tau protein levels, as well as radical scavenging ability towards peroxynitrites (TOSC assay). Results obtained showed a positive correlation between HIF-1α and p-Tau with polygraphic parameter T90 (% of time spent with oxygen saturation below 90%); antioxidant scavenging ability toward peroxynitrites showed instead a negative correlation with T90. Considering OSAS patients with MCI, among polygraphic parameters, only T90 is higher compared to OSAS patients without MCI; in addition, HIF-1α and p-Tau levels increased in OSAS+MCI patients compared to OSAS-MCI. ROC curve analysis confirmed that these two proteins are able to discriminate between the two subgroups, suggesting their possible role as peripheral markers useful in the future for early MCI prediction with less invasive methods. Results-Part III In the end, for the last aim of this work (3), a mouse model of Oxygen induced retinopathy was used to study effects of hyperoxia on microglial cells and vasculature of hippocampus from P16 mice. More specifically, C57Bl6/J mice were subjected to hyperoxic conditions (85% v/v oxygen) from day 8 from birth (P8) to day 11 (P11). Then, mice were returning to normoxic conditions from P11 to P16. At P16, mice were culled, and brains were collected. Preliminary results obtained from hippocampus tissue analysis by qPCR and immunofluorescence from tissue slices, evidenced reduced mRNA and protein levels of microglial marker (IBA-1) in OIR mice compared to controls. Moreover, another microglial marker (F4/80) was evaluated but its levels did not show differences between OIR and control mice, thus suggesting that OIR reduced IBA-1 levels but not microglial cells’ number. Lastly, results obtained from immunofluorescence showed that high percentage of oxygen leads to a decrease in vessel length and their branching index in the hippocampus, alongside to increased vessel area in OIR-treated mice compared to controls. Preliminary results obtained suggests that an hyperoxic environment lead to morphological change in microglial cells and vessels. Conclusions In conclusion, this thesis work explored the IH and hyperoxia effects on the central nervous system and allowed to achieve significant results in the field of oxygen dysregulations causing cognitive decline. Then, starting from the link between OSAS-derived IH and MCI, promising data were obtained in the investigation of new peripheral biomarkers to discriminate OSAS+MCI patients from OSAS-MCI. This research project pointed out how the link between oxidative stress and cognitive decline is fundamental for new therapeutic strategies against neurodegeneration, in researching new tools for diagnosis and improving optimal concentration for oxygen-based treatments minimizing critical effects showed, for example, in pre-term infants.