Effects of AMPK deletion on the response to hypoxia
Udoh, Utibe–Abasi Sunday
MetadataShow full item record
The enzyme adenosine monophosphate activated protein kinase (AMPK), a critical regulator of energy metabolism in the body, is activated by a rise in the cellular AMP: ATP ratio in response to metabolic stresses such as hypoxia. The work in this thesis arises from the recent characterization by Mahmoud, A. (PhD thesis, University of Edinburgh, 2015) of the response to hypoxia of mice lacking the α1 and α2 isoforms of the catalytic subunit of the AMPK molecule. Targeted conditional deletion of the genes encoding the α1 and/or α2 subunits of AMPK in catecholaminergic cells (including cells in the carotid body and the brain) was achieved by crossing mice expressing Cre-recombinase under the control of the tyrosine hydroxylase (TH) promoter, with mice in which either or both α subunits of AMPK were flanked by loxP sequences. AMPKα1/α2-/- mice showed a profoundly abnormal ventilatory response to hypoxia, compared to AMPKα1/α2fl/fl controls. Interestingly however, in vitro recordings from the CSN in isolated carotid body preparations from AMPKα1/α2-/- mice showed that the carotid body afferent response to hypoxia was completely normal in these mice. The abnormal response to hypoxia in AMPKα1/α2-/- mice appears therefore to be due to a deficit in the central, catecholaminergic brainstem neurons involved in respiratory control, where a lack of AMPK activation appears to inhibit the normal hypoxia-induced hyperventilation. While the importance of the peripheral carotid body chemoreceptors in oxygen-sensing has long been established, these findings indicate that the synergistic activation of AMPK in central brainstem neurons by the hypoxic metabolic stress, is also required for the normal response to hypoxia. In this thesis, the responses to hypoxia of AMPKα1/α2-/- mice, AMPKα2-/- mice and AMPKα1/α2fl/fl controls were studied using whole-body plethysmography. The findings showed that AMPKα1/α2-/- mice displayed a respiratory phenotype of longer and increased number of apnoeas coupled with hypoventilation in comparison to both the AMPKα2-/- mice and AMPKα1/α2fl/fl controls confirming the earlier results of Mahmoud (2015). In addition TH immunostaining in the carotid bodies of AMPKα1/α2-/- mice and AMPKα1/α2fl/fl controls was compared to determine if there was any change in the number or density of TH-positive cells in the AMPKα1/α2-/- animals, and the gross result showed a 2-fold decrease in the number of TH- immunopositive cells in the AMPKα1/α2-/- mice as compared to the AMPKα1/α2fl/fl. Intriguingly, if this observation is statistically confirm coupled with the unattenuation of the normal afferent discharge from the carotid bodies of AMPKα1/α2-/- mice then it is plausible that a certain degree of redundancy operates in the physiology of the carotid body with regards to oxygen sensing or the glomus cells type lost may be those not involved in mediating the response to hypoxia. In the main part of this work, c-fos and TH immunohistochemistry were used to investigate the activation of brainstem catecholaminergic neurons by hypoxia, in AMPKα1/α2-/- mice, AMPKα2-/- mice and AMPKα1/α2fl/fl controls. Significant differences in c-fos immunostaining of TH+ve neurons were observed in the SubP region of the NTS, and the C2 region and the A1 region of the ventral respiratory group, implicating these specific regions in the abnormal hypoxic ventilatory response in AMPKα1/α2-/- animals. Catecholaminergic neurons in these brainstem regions are known to play key role in the control of breathing as a loss of these neurons or decrease in their catecholamine content results in severe respiratory abnormalities including respiratory arrhythmias and apnoeas as seen in the Rett syndrome. A significant hyperplasia of the brainstem stem catecholaminergic neurons was also observed in both the AMPKα1/α2-/-and AMPKα2-/- mice consistent with the known inhibitory effects of AMPK activation on cell growth and proliferation. Finally a pilot study was carried out to determine if the respiratory phenotype observed in AMPKα1/α2-/- mice could be replicated using a viral vector to deliver Cre-recombinase to targeted areas in the brainstem in AMPKα1/α2fl/fl mice, to knock out AMPK in specific neuronal subgroups. Although the data obtained from this set of experiments were encouraging, the animals did not show the respiratory phenotype as observed in the conditional knockout mice maybe due to poor targeting of specific brainstem neuronal populations including the SubP region or inadequate transfection of these cells due to low viral titre. One advantage of this pilot work was that responses due to compensatory mechanisms as may be the case in the conditional knock out animals were eliminated. These findings are of interest in understanding the neural control of respiration in hypoxia and acclimatization to altitude, and may also suggest new avenues for therapeutic intervention in breathing disorders such as non-obstructive sleep apnoea. It also raises the possibility that AMPK may be a useful therapeutic tool for disease conditions whose etiology is based on cellular proliferation, such as various forms of cancer and even atherosclerosis.