Abstract
The number of red blood cells (RBC) is tightly controlled to maintain tissue oxygen delivery. During hypoxia, RBC mass is increased by erythropoietin (EPO), which is regulated by hypoxia-inducible transcription factors (HIFs). Neocytolysis is a pathophysiological mechanism that overcorrects the high RBC mass generated during chronic hypoxemia by preferential destruction of young RBCs after normoxia is restored. We recently showed that neocytolysis is caused by excessive mitochondrial-derived reactive oxygen species (ROS) in reticulocytes mediated by downregulation of HIF-regulated Bnip3L (Bnip3L regulates mitophagy) and a decrease in catalase in hypoxia-generated erythrons. Catalase changes are due to hypoxia-induced miR-21 that downregulates catalase. We showed that removal of ROS in plasma by PEG-catalase attenuated neocytolysis, indicating that excessive ROS in plasma from reticulocytes diffuses into RBCs, resulting in preferential destruction of hypoxia-born RBCs with low catalase (Song J, et al. J Mol. Med., 2015 93:857-66). Neocytolysis is well described in astronauts and mountaineers, and here we report this phenomenon in obstructive sleep apnea (OSA).
OSA results in chronic intermittent hypoxia (CIH), which is associated with increased incidence of cardiovascular disease, insulin resistance, and cancer. However, the mechanisms of CIH-induced morbidities are still unknown. Despite CIH, polycythemia is rarely present in OSA; our analysis of 190 consecutive OSA patients at the University of Utah showed that only 4% of OSA patients had polycythemia that corrects after therapy with continuous positive airway pressure therapy (CPAP). Several studies reported that CIH also increases ROS. We hypothesize that neocytolysis might explain why most patients with OSA do not have polycythemia, and that increased ROS in blood cells during OSA may account for oxidant related injury seen with OSA associated diseases.
We analyzed rate of erythropoiesis by EPO levels, HIFs (by HIF-regulated genes transcripts), catalase and miR-21, mitochondrial mass and Bnip3L transcripts, and ROS in blood cells in 30 OSA patients without polycythemia and without chronic cardiopulmonary disease before and after 3 months of CPAP. Patients ranged in OSA severity according to an apnea-hypopnea index from mild to severe. An average of 99.0 minutes was spent with oxygen saturations of 89% or lower, with a 4% oxygen desaturation index of 20.5/hr. The average lowest oxygen saturation value was 78.0%. EPO was higher before CPAP (p=0.0054), indicating that uncorrected OSA augments erythropoiesis in spite of normal hematocrit. We then reasoned that CIH of OSA might cause overcorrection of increased erythropoiesis by neocytolysis. We first tested HIFs changes during CIH indirectly by measuring selected HIF-regulated gene transcripts (GLUT1, HK1, PDK1, NOX2, SOD2, TFRC, and EGLN1) in blood cells. We found that CIH-mediated downregulation of HIF-regulated gene transcripts is consistent with intermittent rapid abrogation of apnea cycles. Further, blunted HIFs levels were differently regulated in reticulocytes, platelets and granulocytes. ROS were found to be increased during OSA in B-cells, T-cells, monocytes, granulocytes, and mature RBCs, but not in platelets. The elevated ROS corrected after CPAP treatment. These changes of ROS in blood cells coexisted with changes in mitochondrial mass and BNIP3L transcript levels; CPAP treatment decreased mitochondrial mass in reticulocytes and granulocytes but not in lymphocytes or monocytes. Catalase levels in granulocytes, platelets and reticulocytes were lower in uncorrected OSA and corrected after CPAP treatment; reverse correlation with miR-21 was found.
These results indicate that neocytolysis occurs in uncorrected OSA resulting in increased mitochondrial mass in reticulocytes (and presumably also in erythroid progenitors) and granulocytes and that excessive ROS generated by increased mitochondrial mass preferentially destroy RBCs with low catalase,resulting in an absence of polycythemia in untreated OSA patients, supporting the central role of HIFs in OSA pathophysiology. Furthermore, increased ROS production may be causally related to other pathologies in OSA such as increased sympathetic tone due to stimulation of carotid bodies by ROS.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.