Carbon monoxide-induced arterial hypoxemia.

Inhalation of carbon monoxide produces an increase in the alveolar to arterial oxygen gradient in the presence of veno-arterial shunts or ventilation-perfusion imbalance but has no such effect in normal subjects. The increase in the alveolar to arterial oxygen gradient with rising concentrations of carboxyhemoglobin results from changes induced by carbon monoxide in the shape of the oxyhemoglobin dissociation curve.

Maximal erythrocyte and hemoglobin catabolism.

A series of experiments were performed on anesthetized dogs in which varying quantities of normal canine erythrocytes damaged by incubation with N-ethylmaleimide were injected into the circulation. Cell sequestration and catabolism of hemoglobin to carbon monoxide remained normal after injections that contained from 0.058 to 0.154 g of hemoglobin per kg of body weight. After administration of larger quantities of these cells, from 0.154 to 0.364 g of hemoglobin per kg of body weight, sequestration remained normal but the rate of catabolism of hemoglobin to carbon monoxide reached a maximum. Large quantities of hemoglobin entered the plasma in these experiments at a time when cell sequestration in the reticuloendothelial system appeared to be virtually complete. After injection of even larger quantities of damaged erythrocytes, 0.545 and 0.552 g of hemoglobin per kg, sequestration became slightly delayed, but was complete. These data appear to indicate that (a) the maximal rate of hemoglobin catabolism in normal anesthetized dogs averages approximately 0.07 g/kg of body weight per hr; (b) hemoglobinemia can result from "overloading" the reticuloendothelial system with damaged sequestered cells and, therefore, may not always indicate "intravascular" hemolysis; and (c) the sequestering function of the reticuloendothelial system appears not to limit the maximal rate of catabolism of hemoglobin. The limiting parameter or parameters were not defined in these studies.

The production of carbon monoxide from hemoglobin in vivo.

Dogs anesthetized with pentobarbital were shown to produce carbon monoxide at an average rate of 0.21 +/- (SD) 0.05 ml per hour. After intravenous injection of erythrocytes damaged by incubation with N-ethylmaleimide, CO was produced in excess of base-line production for 3 to 4 hours with an average yield of 0.89 +/- (SE) 0.046 mumole of carbon monoxide to 1 mumole of heme degraded. After intravenous injection of N-ethylmaleimide (NEM)-treated erythrocytes containing hemoglobin labeled with (14)carbon, (14)CO was produced. Its specific activity was approximately one-eighth that of the injected heme. It was also produced after intravenous injection of solutions of hemoglobin-(14)C and of reconstituted methemoglobin containing hemin-(14)C, but not after injections of methemoglobin containing globin-(14)C. The average yields of (14)CO from metabolized heme in the experiments with damaged erythrocytes and hemoglobin solutions were 89 +/- (SE) 4.6 and 97 +/- (SE) 17.0%, respectively. These results demonstrate that the CO produced during hemoglobin degradation arises from the heme moiety. The yield of (14)CO after injection of hemoglobin-(14)C solutions decreased significantly to values of 35 and 42% in two experiments when exogenous CO was added to the body stores, resulting in blood carboxyhemoglobin levels of 11.3 and 13.2% saturation. This finding suggests that oxidative metabolism is required during catabolism of hemoglobin to CO and that carboxy-hemoglobin levels in this range are sufficient to cause inhibition. After intravenous injection of either hemin-(14)C or protoporphyrin-(14)C, (14)CO was also produced. After injection of protoporphyrin-(14)C labeled bilirubin was isolated from gall bladder bile, and labeled hemin was isolated from the liver. It is thus very likely that protoporphyrin is converted to heme before the formation of CO. There was a large difference between the maximal rates of catabolism of hemoglobin to CO observed after injection of damaged erythrocytes and hemoglobin solutions. The limiting parameters in these processes are not yet clear.

Carbon monoxide production associated with ineffective erythropoiesis.

The rate of endogenous carbon monoxide production ( Vco), determined by the closed rebreathing system technique, was elevated above the normal range in four of five patients studied with ineffective erythropoiesis (four patients with primary refractory anemia, one with thalassemia). The mean molar ratio of Vco to Vheme (rate of circulating heme catabolism, determined from (51)Cr red cell survival curves) was 3.0 +/- 0.6 (SE), indicating that most of the CO originated from sources other than circulating erythrocyte hemoglobin, in contrast to previous findings in patients with hemolytic anemia, where Vco paralleled Vheme closely.After administration of glycine-2-(14)C to these patients, endogenous CO was isolated by washout of body CO stores at high pO(2) or by reacting peripheral venous blood samples with ferricyanide. The CO was then oxidized to CO(2) by palladium chloride and trapped for counting in a liquid scintillation spectrometer. "Early labeled" peaks of (14)CO were demonstrated which paralleled "early labeled" peaks of stercobilin and preceded maximal labeling of circulating heme. Production of "early labeled" (14)CO in patients with ineffective erythropoiesis was greatly increased, up to 14 times that found in a normal subject. The increased Vco and "early (14)CO" production shown by these patients are presumably related mainly to heme catabolism in the marrow. The possibility exists that hepatic heme and porphyrin compounds may also contribute significantly to Vco, as suggested by the finding of a high Vco in an additional patient with porphyria cutanea tarda.

Calcium dependence of prostaglandin release from the guinea pig Taenia coli.

We studied the calcium dependency of the stimulation of prostaglandin synthesis which occurs when perfusing strips of guinea pig Taenia coli with potassium-free media. Stimulation was rapidly reversed by removal of extracellular Ca from the bathing solution. The Ca ionophore A23187 markedly stimulated prostaglandin E2 synthesis, an effect that is dependent on the presence of extracellular Ca. Prostaglandin E2 production in strips in potassium-deficient media was also sensitive to increases in extracellular Ca, and was augmented at concentrations of 7-15 mM. In strips which had been incubated with [3H]arachidonic acid, exposure to potassium-free media caused an increased release of [3H]arachidonic acid and [3H]prostaglandin E2. Release of these labeled compounds with the strips in potassium-free media was further augmented by increasing extracellular [Ca2+] from 2.5 to 10 mM. Treatment with the Ca antagonist agent verapamil did not influence activation of prostaglandin synthesis by potassium-deficient media. The presence of Mn2+ of Ba2+ had similar effects on prostaglandin synthesis, although they had opposite effects on mechanical activity. We conclude that a plasma membrane associated Ca pool is involved in activation of phospholipid metabolism which results in release of esterified arachidonic acid and subsequent prostaglandin synthesis. This Ca pool is in rapid equilibrium with extracellular Ca, is not influenced by cytoplasmic Ca, and is not related to Ca involved in Ca gating in the surface membrane. These data also indicate dissociation between processes involved in muscle contraction and activation of prostaglandin synthesis.

Zero extracellular K+ and prostaglandin E release in the guinea-pig taenia coli.

Prostaglandin E release rates from isolated strips of guinea-pig taenia coli increased during exposure to zero K+ bathing fluid, from control values of 0.78 +/- 0.11 ng/g per min to levels as high as 29.2 ng/per min. Release rates increased for 40-50 min and then remained constant or fell despite progressive increases in intracellular sodium [Nai+] or fall in intracellular potassium [Ki+]. Readmittance of K+ to the bathing solution resulted in rapid reversal of elevated prostaglandin E release rates. [Nai+] and [Ki+] were markedly more abnormal in strips exposed to zero K+ for 70-201 min compared to 30-min exposures. Upon the readdition of K+ after long zero K+ exposure, the rate of prostaglandin E release fell long before [Nai+] and [Ki+] returned to control levels. After K+ was readded to the bathing solution, the ion concentration of tissues exposed to zero K+ for 30 min returned to normal much more quickly than did those of tissues exposed for the longer time periods, yet the exponential rate constants for fall of prostaglandin E release rate after K+ was added were not significantly different after short or long zero K+ exposure. Thus there was a dissociation between the return of [Nai+] and [Ki+] and the fall of prostaglandin E release rate to control levels. Ouabain augmented prostaglandin E release under conditions where [Ki+] could not fall. Addition of known neurotransmitters present in this tissue to the bathing fluid did not augment prostaglandin E release. Guinea-pig taenia coli strips that had been incubated with [3H]arachidonic acid, constantly released [3H]arachidonic acid and [3H]prostaglandin E and a prostaglandin which cochromatographed with prostaglandin E but could not be converted to prostaglandin B by alkali and was shown to be 6-ketoprostaglandin F1 alpha. Release of [3H]arachidonic acid and [3H]prostaglandin E plus 6-[3H]ketoprostaglandin F1 alpha was increased when strips were exposed to zero K+. Data obtained in this study suggest the augmented prostaglandin E release seen during zero K+ or ouabain is related to increased availability of unbound arachidonic acid at the site of cyclooxygenase in the cell. Augmented prostaglandin E release is apparently not related to alterations in intracellular electrolyte concentrations or release of known neurotransmitters.

Smooth muscle aldolase C-bound inositol 1,4,5-trisphosphate studied in vitro under physiological conditions.

Our goal was to quantitate inositol 1,4,5-trisphosphate (Ins(1,4,5)P3) binding to aldolase C tetramer (aldolase4) and its displacement by inositol 1,3,4-trisphosphate (Ins(1,3,4)P3) under conditions which approximated the in vivo state. Anions were found to have major effects. Decreasing [KCl] from 100 to 10mM, at 0 degrees C and pH 7.0, increased maximal Ins(1,4,5)P3 binding to 1.0 to 2.4mol per mol aldolase4. At 10 and 30mEq/l [Cl-], an additional high affinity site was detected (Kds = 0.43 and 0.86 microM, respectively). Increasing concentrations of other anions (SO42-, propanoate-, HCO3-, acetate-) also inhibited binding, but effects would be minimal at concentrations of these anions present in the cytoplasm of living cells. Ins(1,3,4)P3 displacement of aldolase C-bound Ins(1,4,5)P3 was sensitive to [Cl-]; at 30mEq/l [Cl-] and 37 degrees C, Ins(1,3,4)P3 released 20% of bound Ins(1,4,5)P3 at concentrations of 100nM. Changing temperature from 0 to 37 degrees C increased Kds for Ins(1,4,5)P3 binding. Changes in free [Ca2+], [Mg2+], [Na+] and [K+] and changes in osmolality had no effect on Ins(1,4,5)P3 binding to aldolase C. In vivo Ins(1,4,5)P3-aldolase4 binding at 30mEq/l [Cl-] and 37 degrees C were calculated for different [Ins(1,4,5)P3]free over the range 0.2 to 1.0 microM. For different cytoplasmic [Ins(1,4,5)P3]free. Ins(1,4,5)P3 binding to aldolase4 was sufficient, if acutely released, to nearly double cytoplasmic [Ins(1,4,5)P3]free. We proposed a schema whereby release of aldolase C-bound Ins(1,4,5)P3 evoked by Ins(1,3,4)P3 amplifies effects of phospholipase C-formed Ins(1,4,5)P3.

Effect of incubation of guinea-pig taenia coli in potassium-free media on arachidonate release and lipid metabolism.

We studied the effects of immersion of guinea-pig taenia coli strips in potassium-free media on arachidonate stores and other lipid fractions. Control studies obtained with the strips in Krebs solution showed that greater than 97% of arachidonate was found esterified in phospholipid with the following distribution: phosphatidylethanolamine greater than phosphatidylcholine greater than phosphatidylserine plus phosphatidylinositol. 30 min incubation of the strips with [3H]arachidonate complexed to albumin resulted in incorporation of this isotope into phospholipid and neutral lipid fractions, phosphatidylcholine greater than neutral lipid greater than phosphatidylserine plus phosphatidylinositol greater than phosphatidylethanolamine. 30 min incubations with 32PO4(2-)-resulted in an isotope incorporation into phospholipids, phosphatidylcholine greater than phosphatidylserine plus phosphatidylinositol greater than phosphatidylethanolamine. After 'loading' with [3H]arachidonate and 32P, placing the strips in potassium-free media caused the following: there was an increased release of [3H]arachidonate from the tissue into the bathing solution. [3H]Arachidonate and 32P radioactivity in phosphatidylinositol fell without a change in phosphatidylinositol content. [3H]Arachidonate and 32P radioactivity in other phospholipid fractions was unchanged. Arachidonate specific activity fell and arachidonate content increased in the phosphatidylserine plus phosphatidylinositol fraction. [3]Arachidonate in neutral lipid did not change significantly. We conclude that exposure of taenia coli to potassium-free media activates turnover of phosphatidylinositol, which results in release of arachidonate.

Morphological features of spiking and nonspiking cells in the paratracheal ganglion of the ferret.

The present series of experiments was designed to study details of the morphology and connectivity of functionally identified cells located in the paratracheal ganglia of the ferret. The morphology of 11 spiking (AH cells) and seven nonspiking (type B cells) ganglion cells was examined. Intra-axonally injected horseradish peroxidase (HRP) was used as the label. Each spiking and nonspiking cell was identified by intracellular recording prior to the HRP injection. "Whole mount preparations" were processed for HRP histochemistry with diaminobenzidine as the chromogen. HRP-labeled cell bodies of both the spiking AH and nonspiking type B neurons demonstrated similar morphological features. Both types of ganglion cells showed axons arising from a small, ill-defined axon hillock which exited from the cell as single or multiple branches of equal diameter and coursed unidirectionally through the interganglionic nerve trunk to an adjacent ganglion; short, fine, tapering processes (presumptive dendrites) in the immediate vicinity of the injected cell; and processes extending out of the ganglion cell perpendicular to the interganglionic nerve trunk which could be followed into the smooth muscle. Extraperikaryal injections of HRP into a ganglion retrogradely labeled perikarya in the adjacent ganglia. These results demonstrate that in airway ganglia the morphology of spiking and nonspiking neurons is remarkably similar despite electrophysiological differences. In addition it appears that ganglion cells project to adjacent ganglia and to smooth muscle by means of independent axonal processes. These morphological features of the ganglion cells in airways and the trajectories of their axons correspond to known features of their physiology: i.e., the axon of a ganglion cell travels unidirectionally toward the adjacent ganglion and arborizes there, providing anatomical evidence of communication between ganglia via the interganglionic nerve trunk; and the spiking and nonspiking neurons possess similar morphological features that are typical of ganglion cells described in other systems, such as in the myenteric plexus.

Stretch-induced membrane depolarization in ferret trachealis smooth muscle cells.

We determined the effects of increasing the length of the ferret trachealis muscle on smooth muscle membrane potentials recorded on successive impalements by microelectrodes. The preparation included the paratracheal ganglion nerve plexus as well as trachealis muscle. With sustained increases in muscle length over the range 0.5-0.8 to 1.2 maximal length (Lmax), depolarization occurred, which was related to the amplitude of the length increase. Membrane depolarizations were also evoked after stretching to lengths approximately 1.1 Lmax and returning to the control length. Stretch-induced membrane depolarizations developed after the stretch maneuver was complete; were slowly reversible; were not influenced by tetrodotoxin or atropine; were related to stretch rather than to maintained increase in muscle length; were not transmitted to adjacent nonstretched segments of the trachea; and were often associated with slow waves which appear to be secondary to membrane depolarization rather than stretch per se.

Oxygen sensors in vascular smooth muscle.

Inhibition or activation of cellular function due to acute decreases in PO2 can be considered in terms of two different processes: 1) a sensor that monitors PO2 decreases and 2) transduction systems directed from the O2 sensor to reactions that control cellular function. We used the norepinephrine-contracted aortic smooth muscle model to study the nature of the O2 sensor and transduction system during decreased PO2-evoked relaxations. The phosphorylation potential, a measurement of kinetic energy required for ATP hydrolysis, was decreased to 30% of control at the onset of relaxation and progressively fell as muscle relaxed. The free inorganic phosphate intracellular concentration ([Pi]) was experimentally increased approximately 0.6 mM during transients that followed a rapid decrease in PO2. Relaxations to 80% of maximal force were more rapid under conditions of an augmented [Pi] than in control rings, and they occurred at a higher phosphocreatine concentration and phosphocreatine-to-free creatine ratio but at the same phosphorylation potential. Results support the operation of a cytochrome aa3 O2 sensor in the mechanism of decreased PO2-evoked relaxations and implicate an increase in [Pi] and a decrease in kinetic energy in the transduction mechanism directed at rate-limiting reactions that control force.

Mechanical and biochemical events during hypoxia-induced relaxations of rabbit aorta.

Hypoxic relaxation of norepinephrine contractions of isolated rabbit aorta is rapid, whereas relaxation of KCl contractions is slower and blunted. The data given here suggest that with receptor-evoked contractions of rabbit aorta, the energy-limitation of ATP-dependent K+ channels and other sarcolemmal channels, myosin light chain kinase, and actin-activated myosin ATPase are probably not involved in oxidative energy-contraction coupling. The data strongly support the hypothesis that the rate limiting, energy-dependent step is upstream to myosin light chain kinase, which is 50% inhibited at an ATP concentration of about 0.5 mM. This energy-dependent step may be in the inositol phospholipid transduction system, as we have previously postulated (Coburn et al., 1988). In contrast the energy-limited reaction during KCl contractions appears to be the actin-activated myosin ATPase which is 50% inhibited at a mean ATP concentration of about 0.1 mM.