Oxygen sensing and K+‐Cl− cotransport

Abstract

Oxygen modulates membrane transport in erythrocytes of a number of vertebrate species. In red cells of horse, sheep, human and trout, raising oxygen tension enhances K+-Cl− cotransport (KCC). Swelling, among other treatments, activates this transporter. Swelling-activated KCC promotes KCl efflux, water follows by obligatory osmotic coupling, and cell volume is decreased. In this manner regulation of cell volume is promoted. Direct activation of KCC is probably by a serine/threonine phosphatase. Regulation is afforded by a volume-sensitive kinase: cell swelling inhibits the kinase, allowing the phosphatase to activate KCC (see Jennings, 1999, for references). Since O2 stimulates KCC, there must be a sensor of O2 that signals KCC. The most obvious sensor for oxygen is haemoglobin (Hb). See Gibson et al. (2000) for a recent review. In this volume of The Journal of Physiology, Berenbrink et al. (2000) present studies on O2 modulation of K+ transport in trout red cells; most of the measured K+ transport was KCC. Changes in KCC were measured as functions of O2 partial pressure (PO2) and pH, varied independently. At physiological pH, raising PO2 to atmospheric pressure increased K+ transport several-fold. At lower pH, stimulation of K+ flux by O2 was reduced. At pH 7.9, the PO2 for half-maximal stimulation of K+ transport, P50, was ∼10-fold higher than the P50 for equilibrium binding of O2 to Hb. Stimulation of K+ flux by O2 was not cooperative, whereas the curve for Hb-O2 binding was sigmoid, with a Hill coefficient of ∼3. These results are strong evidence that bulk Hb is not the O2 sensor in trout red cells. Whatever the O2 sensor, it has a low O2 affinity and a high pH sensitivity (Berenbrink et al. 2000). Oxygen affects a diverse array of processes in all cells. There are sensing mechanisms based on haem proteins shared by most if not all cells (Bunn & Boyton, 1996). In cells such as erythropoietin-producing hepatocytes and type I cells of the carotid body, O2 sensing is their central function. Type I cells are a pertinent example because changes in PO2 provoke changes in membrane transport. The current view is that low PO2 is sensed by a haem protein which causes closing of a class of K+ channel. The resultant depolarization opens Ca2+ channels; the increased Ca2+ activity causes secretion of neurotransmitter. The post-synaptic neurons transmit signals to the brain stem where respiration is regulated (Gonzalez et al. 1994). The identity of the haem sensor is uncertain; the non-mitochondrial cytochrome b558 was proposed (Cross et al. 1990). There are several possible mechanisms of signalling from the sensor, including direct coupling of the sensor to the channel and action of a reactive O2 species, e.g. superoxide, produced from O2 interacting with the sensor. The signals may alter the function of a protein kinase and/or phosphatase. Serine/threonine phosphatases types 1 and 2A are metalloproteins (Barford, 1996) and could be targets of superoxide. A protein kinase regulated by haem has been reported (Chen & London, 1995). In these terms, a signalling pathway in carotid body type 1 cells can be envisioned. O2 modulation of KCC in trout as well as mammalian red cells may operate by similar signalling mechanisms. Trout red cells are nucleated and have mitochondria, and probably have the full complement of haem proteins as possible O2 sensors. In mammalian red cells, as in trout red cells, there is evidence that Hb is not the O2 sensor. Activation of KCC in horse red cells requires 70 % saturation of Hb with O2 (Honess et al. 1996), similar to the situation in trout red cells (Berenbrink et al. 2000): low affinity of the sensor for O2. There are no known haem proteins in mammalian red cells other than Hb that are likely candidates for an O2 sensor. It would make sense if the sensor were bound to the membrane to afford access to the cotransporter. It turns out that Hb binds to band 3, the membrane anion exchanger, from which it could interact with KCC. Deoxy-Hb binds preferentially (Chetrite & Cassoly, 1985). With an elevated fraction of deoxy-Hb over oxy-Hb at most PO2 levels, bound Hb would have a lower O2 affinity than bulk Hb. This low affinity is a characteristic of the O2 sensor regulating KCC (Honess et al. 1996; Berenbrink et al. 2000). The lack of cooperativity of O2 activation of KCC in trout red cells still makes it unlikely that Hb is the sensor in these cells. However, the O2 activation curve of KCC in horse red cells is strongly sigmoid (Speake et al. 1997), so Hb, membrane-bound, remains a candidate O2 sensor in mammalian red cells.