Brian E. McLaughlin

Does carbon monoxide have a physiological function

Recently endothelium-derived relaxing factor (EDRF) has been identified as nitric oxide. The source of the nitric oxide is L-arginine, and the L-arginine-nitric oxide pathway has been proposed to function as a widespread transduction mechanism for the regulation of cell function and communication. Gerald Marks and colleagues suggest that carbon monoxide, which is formed endogenously from heme catabolism and which shares some of the chemical and biological properties of nitric oxide, may play a similar role. This would be achieved by carbon monoxide binding to the iron atom of the heme moiety of soluble guanylyl cyclase and to the iron-sulfur centers of macrophage enzymes.

Carbon monoxide formation in the ductus arteriosus in the lamb: implications for the regulation of muscle tone.

1 We have previously shown that carbon monoxide (CO) potently relaxes the lamb ductus arteriosus and have ascribed this response to inhibition of a cytochrome P450‐based mono‐oxygenase reaction controlling the formation of endothelin‐1 (ET‐1). In the present study, we have examined whether CO is formed naturally in the vessel. 2 The CO‐forming enzyme, haem oxygenase (HO), was identified in ductal tissue in its constitutive (HO‐2) and inducible (HO‐1) isoforms by Western immunoblotting and immunological staining procedures (both light and electron microscopy). HO‐1 was localized to endothelial and muscle cells, while HO‐2 was found only in muscle cells. Inside the muscle cells, HO‐1 and HO‐2 immunoreactivity was limited to the perinuclear region, and the Golgi apparatus in particular. However, upon exposure to endotoxin, HO‐1 became more abundant, and both HO isoforms migrated towards the outer region of the cytoplasm close to the sarcolemma. 3 CO was formed enzymatically from added substrate (hemin, 50 μm) in the 10,000 g supernatant of the ductus and its formation was inhibited by zinc protoporphyrin IX (ZnPP, 200 μM). 4 ZnPP (10 μM) had no effect on the tone of the ductus under normal conditions (2.5 to 95% O2), but it contracted the endotoxin‐treated ductus (at 2.5% O2). At the same concentration, ZnPP also tended to contract the hypoxic vessel (zero O2). 5 ZnPP (10 μm) curtailed the relaxant response of the oxygen (30%)/indomethacin (2.8 μM)‐contracted ductus to bradykinin (35 nM), while it left the sodium nitroprusside (35 nM) relaxation unchanged. 6 We conclude that CO is formed in the ductus and may exert a relaxing influence when its synthesis is upregulated by an appropriate stimulus.

Quantitation of nitric oxide formation from nitrovasodilator drugs by chemiluminescence analysis of headspace gas

A rapid and reliable method has been developed for the quantitation of nitric oxide by chemiluminescence analysis of headspace gas. Aqueous nitric oxide standards are used to calibrate the method. There is a linear relationship between the amount of nitric oxide and the redox chemiluminescence detector response over the range of 52-2445 pmol of nitric oxide in 2.0 mL of deionized water contained in a sealed 6.2-mL flask. The intra-day and inter-day coefficient of variation values of the method do not exceed 4% and 9%, respectively. The lower limit of quantitative sensitivity and the lower limit of qualitative detection are 52 pmol and 26 pmol of nitric oxide, respectively. This method has been used to measure nitric oxide formation during the incubation of glyceryl trinitrate or sodium nitroprusside in the presence of cysteine. This method has two major advantages over the currently available procedures for the quantitation of nitric oxide, viz., no artifactual formation of nitric oxide during sample preparation and decreased instrumental contamination.

Measurement of endogenous carbon monoxide formation in biological systems.

Endogenous carbon monoxide (CO) formation has been measured in different biological systems using a variety of analytical procedures. The methods include gas chromatography-reduction gas detection, gas chromatography-mass spectroscopic detection, laser sensor-infrared absorption, UV-visible spectrophotometric measurement of CO-hemoglobin or CO-myoglobin complex, and formation of (14)CO from (14)C-heme formed following [2-(14)C]glycine administration. CO formation ranged from a low of 0.029 nmol/mg of protein/h in chorionic villi of term human placenta to a high of 0.28 nmol/mg of protein/h in rat olfactory receptor neurons in culture and rat liver perfusate.

Heme oxygenase and nitric oxide synthase in the placenta of the guinea-pig during gestation

Nitric oxide (NO) and carbon monoxide (CO) are novel gaseous chemical messengers that play key roles in cell function and cell-cell communication in many organ systems, including the cardiovascular system. Although the presence of NO synthase (NOS) in the placenta and its role in the regulation of fetoplacental and uteroplacental blood flow are well established, little is known about placental expression and activity of heme oxygenase (HO), the enzyme that catalyses the oxidation of heme to CO, biliverdin and iron, during gestation. The objectives of this study were to elucidate the localization of HO-1 and HO-2 isoforms relative to NOS III protein, and to determine the enzymatic activity of HO in the placenta of the guinea-pig during gestation. Placentae were obtained from pregnant guinea-pigs at gestational day (GD) 34, 50, 62 and full term (term, about GD 68). Immunohistochemical localization of HO-1, HO-2 and NOS III protein was conducted using selective polyclonal antibodies. HO activity was determined by using a gas chromatographic method to measure the rate of formation of CO from heme. Faint staining for HO-1 was observed in the adventitial layer of larger fetal blood vessels of the placenta at GD 34. The intensity of this staining was higher at GD 50 and GD 62, and decreased at full term. Similar areas in serial sections of placentae obtained at these selected times during gestation exhibited lower staining intensity when incubated with anti-HO-2 antiserum. Placental HO activity was significantly increased (P<0.05) at GD 62 compared with GD 34, GD 50 and full term. NOS III (endothelial constitutive NOS) staining was highest at GD 34, decreasing thereafter, and was localized mostly to trophoblast lining maternal channels. The data demonstrate that, in the guinea-pig, placental HO and NOS differ in tissue localization during the second half of gestation, with expression of HO protein and its catalytic activity being higher during near-term pregnancy. In a preliminary immunohistochemical investigation of the full-term human placenta, HO-1 protein was localized primarily in the adventitial region of fetal blood vessels of stem chorionic villi. In view of the vasodilator action of CO and NO, the HO and NOS systems may play key roles in the regulation of placental haemodynamics.

Heme oxygenase activity and immunohistochemical localization in bovine pulmonary artery and vein.

Recent studies suggest that carbon monoxide (CO) derived from heme oxygenase (HO)-catalyzed metabolism of heme plays a role in the regulation of cell function and communication. In blood vessels, CO may regulate vascular smooth-muscle tone through the activation of soluble guanylyl cyclase, in a manner similar to that of nitric oxide. The objective of this study was to determine the relation between HO enzymatic activity and localization of HO protein in bovine pulmonary blood vessels. HO enzymatic activity was determined by quantitating the rate of CO formation in the microsomal fraction of homogenates of bovine pulmonary artery (BPA) and vein (BPV). HO protein was localized by immunohistochemical analysis of paraformaldehyde-fixed tissue by using polyclonal antibodies to inducible HO (HO-1) and noninducible HO (HO-2). HO enzymatic activity was measured in BPA and BPV, which correlated with the presence of HO protein. In BPA, HO enzymatic activity was found in the adventitia and medial layer; HO protein was localized in the nerves and vasa vasorum of the adventitia and was found throughout the smooth-muscle cells in the medial layer. The data clearly demonstrate the presence of HO enzymatic activity for the formation of CO in blood vessels that contain HO protein.

Chronic prenatal ethanol exposure and increased concentration of fatty acid ethyl esters in meconium of term fetal Guinea pig.

In humans, the occurrence of prenatal exposure to ethanol is difficult to validate objectively. Increased concentration of fatty acid ethyl esters (FAEE) in the meconium of the newborn may be a biomarker of prenatal ethanol exposure. The validity of this proposed biomarker was tested in pregnant guinea pigs that received chronic oral administration of 4 g ethanol/kg maternal body weight/day (n=8), isocaloric-sucrose/pair-feeding (n=8) or water (n=2) throughout gestation. At gestational day 65 (term, gestational day 66 to 69), each dam and her offspring were euthanized, and meconium was collected from the term fetal large intestine. Eight individual FAEE (lauric, myristic, palmitic, palmitoleic, stearic, oleic, linolenic and arachidonic AEE) were measured by gas chromatography––flame ionization detection and confirmed by gas chromatography––mass spectrometry. The chronic maternal ethanol regimen decreased fetal body weight and brain weight. There was virtually no measurable FAEE in the meconium for the water group (n=3 fetuses). For meconium of the ethanol offspring (n=25 fetuses) compared with the sucrose offspring (n=23 fetuses), the total FAEE concentration was 8-fold higher; and lauric, palmitic, stearic and oleic AEE concentrations were at least 5-fold higher for the ethanol group. The data indicate that fetal meconium FAEE constitute a biomarker of prenatal ethanol exposure for a maternal ethanol regimen that restricts fetal development, with an inverse relationship between meconium total FAEE concentration and both body weight and brain weight.

Heme Oxygenase Expression in Selected Regions of Term Human Placenta

Carbon monoxide (CO), formed during heme oxygenase (HO)-catalyzed oxidation of heme, has been proposed to play a complementary role with nitric oxide in the regulation of placental hemodynamics. The objective of this study was to elucidate HO enzymatic activity and HO-1 (inducible) and HO-2 (constitutive) protein content in the microsomal subcellular fraction of homogenate of selected regions of placenta from normotensive and mild pre-eclamptic pregnancies. HO enzymatic activity was measured under optimized conditions by gas chromatography using CO formation as an index of activity, and HO-1 and HO-2 protein content were determined by Western immunoblot analysis. Microsomal HO activity in each of the four placental regions was not different between normotensive and mild pre-eclamptic pregnancies. Microsomal HO-2 protein content was not different between normotensive and mild pre-eclamptic pregnancies, whereas there was increased expression of microsomal HO-1 protein in chorionic villi and fetal membranes from pre-eclamptic pregnancy compared with normotensive pregnancy. Microsomal HO enzymatic activity correlated with HO-2, but not HO-1, protein content.

Biotransformation of glyceryl trinitrate and elevation of cyclic GMP precede glyceryl trinitrate-induced vasodilation

In this study, we examined glyceryl trinitrate (GTN) biotransformation and cyclic GMP elevation in vascular smooth muscle before onset of GTN-induced relaxation. Isolated rabbit aortic strips (RAS) and strips of bovine pulmonary artery (BPA) and bovine pulmonary vein (BPV) were contracted submaximally and incubated with [3H]GTN. Before onset of GTN-induced vasodilation, the tissues were freeze-clamped and then analyzed for GTN, glyceryl-l, 2-dinitrate (1, 2-GDN), and glyceryl-1, 3-dinitrate (1, 3-GDN) and for cyclic GMP. Before onset of relaxation of RAS, BPA, and BPV, there was significant biotransformation of GTN to GDN and significant elevation of cyclic GMP. There was significantly greater biotransformation of GTN and elevation of cyclic GMP by BPV than by BPA incubated with the same concentration of GTN, which was temporally related with the more rapid onset of relaxation induced in BPV than in BPA. These results are consistent with the hypothesis that the magnitude of GTN biotransformation before vasodilation is the important determinant of subsequent tissue relaxation. In GTN biotransformation before vasodilation, there was preferential formation of 1, 2-GDN. These data indicate that the mechanism of GTN biotransformation to 1, 2-GDN is related to elevation of cyclic GMP and subsequent vasodilation.

Structural analysis and identification of PhuS as a heme-degrading enzyme from Pseudomonas aeruginosa.

Bacterial pathogens require iron for proliferation and pathogenesis. Pseudomonas aeruginosa is a prevalent Gram-negative opportunistic human pathogen that takes advantage of immunocompromised hosts and encodes a number of proteins for uptake and utilization of iron. Here we report the crystal structures of PhuS, previously known as the cytoplasmic heme-trafficking protein from P. aeruginosa, in both the apo- and the holo-forms. In comparison to its homologue ChuS from Escherichia coli O157:H7, the heme orientation is rotated 180° across the α-γ axis, which may account for some of the unique functional properties of PhuS. In contrast to previous findings, heme binding does not result in an overall conformational change of PhuS. We employed spectroscopic analysis and CO measurement by gas chromatography to analyze heme degradation, demonstrating that PhuS is capable of degrading heme using ascorbic acid or cytochrome P450 reductase-NADPH as an electron donor and produces five times more CO than ChuS. Addition of catalase slows down but does not stop PhuS-catalyzed heme degradation. Through spectroscopic and mass spectrometry analysis, we identified the enzymatic product of heme degradation to be verdoheme. These data taken together suggest that PhuS is a potent heme-degrading enzyme, in addition to its proposed heme-trafficking function.