A genetically encoded system for oxygen generation in living cells

Oxygen plays a key role in supporting life on our planet. It is particularly important in higher eukaryotes where it boosts bioenergetics as a thermodynamically favorable terminal electron acceptor and has important roles in cell signaling and development. Many human diseases stem from either insufficient or excessive oxygen. Despite its fundamental importance, we lack methods with which to manipulate the supply of oxygen with high spatiotemporal resolution in cells and in organisms. Here, we introduce a genetic system, SupplemeNtal Oxygen Released from ChLorite (SNORCL), for on-demand local generation of molecular oxygen in living cells, by harnessing prokaryotic chlorite O₂-lyase (Cld) enzymes that convert chlorite (ClO₂⁻) into molecular oxygen (O₂) and chloride (Cl⁻). We show that active Cld enzymes can be targeted to either the cytosol or mitochondria of human cells, and that coexpressing a chlorite transporter results in molecular oxygen production inside cells in response to externally added chlorite. This first-generation system allows fine temporal and spatial control of oxygen production, with immediate research applications. In the future, we anticipate that technologies based on SNORCL will have additional widespread applications in research, biotechnology, and medicine.

Oxygen is vital for life and is one of the most widely used substrates in all of biochemistry (1). One of the most important events for life on our planet was the great oxygenation event (GOE), some 2.1 to 2.4 billion years ago (2), which changed our environment and spawned aerobic life. Oxygen provides a thermodynamically favorable terminal electron acceptor that helps to power metabolism and has been proposed as a prerequisite for the emergence of complex forms of animal life (3). Since oxygen is a di-radical and can be toxic, numerous mechanisms evolved to allow organisms to safely wield its thermodynamic potential (4). In addition, oxygen plays a key role in signaling (5, 6) and contributes to cell differentiation and development (7). Humans have an absolute requirement for oxygen, being only able to survive minutes in complete anoxia. At the other extreme, hyperoxia can also be toxic, leading to seizures, pulmonary toxicity, and retinopathy (8).Blood oxygen levels are routinely monitored in clinical medicine, and when required, we have facile means of delivering supplemental oxygen through nasal cannula, face masks, mechanical ventilation, and even extracorporeal membrane oxygenation. In contrast, we only have few ways of providing supplemental oxygen within cells. In the research setting, cells and organisms of course can be grown in chambers in which the ambient oxygen is regulated with gas mixtures (8). However, the poor solubility of oxygen in biofluids, its continuous exchange with the atmosphere, and its active consumption by mitochondrial respiration, make it challenging to manipulate intracellular oxygen levels with high spatiotemporal precision. Ideally, we would have an easy-to-use, genetically encoded system capable of delivering on-demand, localized production of molecular oxygen in living cells.Here we sought to develop such a tool by harnessing naturally occurring enzymes that generate molecular oxygen. While genetic tools exist for generating reactive oxygen species such as singlet oxygen in living cells (9), none have been described that generate molecular oxygen in its more familiar and stable triplet state. Enzymatic formation of the O-O bond is extremely rare. The most well appreciated and studied example is the water-splitting oxygen evolving complex (OEC) of photosystem II, which is central to oxygenic photosynthesis. However, the OEC contains numerous cofactors including chlorophyll, quinones, and a unique manganese cluster (10). Oxygen can also be produced from methane-oxidizing bacteria (11), though the mechanism is not well studied. Another enzyme, called chlorite O₂-lyase or chlorite dismutase (Cld), converts chlorite (ClO₂⁻) to oxygen (O₂) and chloride (Cl⁻) in numerous bacterial and archaeal species [reviewed in ref. (12)].We chose to focus on the Cld family of oxidoreductases as a chassis for a simple-to-use oxygen generator given that its substrate is bioorthogonal to eukaryotic metabolism. We show that when expressed in human cells, Cld enzymes exhibit high activity, and that we can coexpress plasma membrane transporters that promote uptake of sodium chlorite for its subsequent intracellular conversion to oxygen. In this way we are able to successfully deploy a genetic system for SupplemeNtal Oxygen Released from ChLorite (SNORCL).