|
|
||||||||
1-adrenergic receptor-mediated vasoconstriction and ligand binding in pulmonary artery
1Department of Pediatrics, University of Colorado Health Science Center, Denver, Colorado; and Departments of 2Medicine and 3Pediatrics, Duke University Medical Center, Durham, North Carolina
Submitted 27 May 2005 ; accepted in final form 23 August 2005
| ABSTRACT |
|---|
|
|
|---|
1-adrenergic GPCR system to inhibit pulmonary vasoconstriction by receptor agonists, e.g., phenylephrine (PE). Rat pulmonary artery rings were pretreated for 30 min with and without an S-nitrosothiol, either GSNO or S-nitrosocysteine (CSNO), and constricted with sequential concentrations of PE (108106 M). Effective cGMP-dependence was tested in rings pretreated with soluble guanylate cyclase inhibitors {either 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one (ODQ) or LY-83583} or G kinase inhibitor (KT-5823), and a thiol reductant [dithiothreitol (DTT)] was used to test reversibility of S-nitrosylation. Both S-nitrosothiols attenuated the PE dose response. The GSNO effect was not prevented by LY-83583, ODQ, or KT-5823, indicating cGMP independence. GSNO inhibition was reversed by DTT, consistent with S-nitrosylation or other GSNO-mediated cysteine modifications. In CSNO-treated lung protein, the
1-adrenergic receptor was shown to undergo S-nitrosylation in vitro using a biotin switch assay. Studies of
1-adrenergic receptor subtype expression and receptor density by saturation binding with 125I-HEAT showed that GSNO decreased
1-adrenergic receptor density but did not alter affinity for antagonist or agonist. These data demonstrate a novel cGMP-independent mechanism of reversible
1-adrenergic receptor inhibition by S-nitrosothiols. nitric oxide; S-nitrosylation; G protein-coupled receptor; guanosine 3',5'-cyclic monophosphate
We previously demonstrated that pulmonary vasoconstriction elicited by serotonin (5-hydroxytryptamine, 5-HT) is reversibly inhibited by the NO donor S-nitrosoglutathione (GSNO) (23). In addition, use of nitric oxide synthase (NOS) inhibitors or removal of the endothelium to decrease NO release has been shown to augment phenylephrine (PE)-mediated vasoconstriction in isolated rat pulmonary artery (28). Administration of an NO donor blocks both PE and norepinephrine (NE)-mediated vasoconstriction in rat aorta (17). Persistent inhibition of NE vasoconstriction is a characteristic of S-nitrosothiols, including GSNO, not found in other classes of NO donors such as 3-morpholinosydnonimine (SIN-1) or 2-(N,N-diethylamino)-diazenolate-2-oxide (DEA-NO) (3, 4). Furthermore, S-nitrosothiol inhibition of PE-mediated vasoconstriction is independent of the actions of NO on cGMP (17). 5-HT and catecholamines effect intracellular signaling via activation of G protein-coupled receptors (GPCR). The mechanism(s) by which NO inhibits these GPCR responses is not well understood.
GPCR activity and expression are regulated by multiple mechanisms and are involved in numerous physiological processes. Several studies have reported effects of NO on signaling through the
2-adrenergic receptors, as well as bradykinin, M2 muscarinic, thromboxane, and AT1 GPCRs. (2, 5, 22, 31, 32). The effects of NO on these GPCR-mediated responses have both cGMP-dependent and -independent components that, together with our data on the 5-HT GPCR system (23), raise the issue of cGMP-independent regulation of GPCRs by S-nitrosothiols as a general physiological mechanism for the regulation of vascular signal transduction. Thus we hypothesized that S-nitrosothiols inhibit catecholamine-induced vasoconstriction by thiol modification of the
1-adrenergic receptor in the pulmonary vasculature. In this study, we demonstrate that GSNO inhibits PE-mediated vasoconstriction in isolated rat pulmonary arteries (PA) through a reversible, cGMP-independent pathway. We further demonstrate that GSNO decreases
1-adrenergic receptor-ligand binding, thereby localizing the target of NO to the receptor-G protein complex.
| METHODS |
|---|
|
|
|---|
PA ring preparation.
Animal studies were approved by the Duke University Institutional Animal Use and Care Committee. Rat PA rings (2 mm) from the first branch PA were removed immediately from 300-g male Sprague-Dawley rats killed by halothane overdose and aortic transection. The rings were mounted in 15-ml tissue baths containing Krebs-Henseleit buffer and bubbled with 21% O2/5% CO2/balance N2 at 37°C. Krebs-Henseleit contained sodium chloride (82.8 mM), potassium chloride (4.7 mM), monobasic potassium phosphate (2.4 mM), sodium bicarbonate (25 mM), magnesium sulfate (1.2 mM), calcium chloride (2.7 mM), and dextrose (11.1 mM) at pH 7.4. All rings were suspended at baseline tension levels of
1 g, and isometric tension was monitored continuously with a calibrated force transducer. In all rings, a dose-response curve using cumulative concentrations of the specific
1-adrenergic receptor agonist PE (108106 M) was performed to test vascular responsiveness and to establish each rings baseline maximum control PE response. Acetylcholine (105 M) was administered to PE-constricted vessels to confirm endothelial integrity, shown by >30% decrease in ring tension.
PA ring protocol. A dose-response curve to PE (108106 M) was examined before and after a 30-min treatment with GSNO. Three concentrations of GSNO (2.5, 25, and 250 µM) were tested, and subsequent ring experiments were performed with 250 µM GSNO. Between interventions, the ring baths were rinsed thoroughly with three volumes of fresh buffer, which removed all GSNO before the second PE dose-response curve. Other rings were treated with a second S-nitrosothiol, CSNO (250 µM). The responses were expressed as percent change in isometric tension from the initial maximal tension induced by PE (106 M). Control experiments were performed with reduced GSH and oxidized GSSG (250 µM) before the PE dose-response curve.
To determine whether NO inhibited the PE dose-response curve through cGMP-independent mechanism(s), we repeated PE dose-response curves (108106 M) in the presence or absence of 250 µM GSNO after administering either a guanylate cyclase inhibitor or a protein kinase G (PKG) inhibitor to block the cGMP signaling pathway. The ring baths were rinsed with fresh buffer to remove the inhibitor and GSNO before administration of PE. Two guanylate cyclase inhibitors, 6-anilinoquinoline-5,8-quinone (LY-83583, 10 µM) and 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one (ODQ, 10 and 1 µM), were tested. LY-83583 blocks cGMP-dependent signaling by several mechanisms including inhibition of endothelial NO release and production of superoxide, both which decrease guanylate cyclase activity. ODQ irreversibly inhibits guanylate cyclase as well as other heme-containing proteins (10). We confirmed the ability of LY-83583 and ODQ to block cGMP production by GSNO by measuring cGMP levels with a commercially available radioimmunoassay kit (Amersham Biosciences, Piscataway, NJ). PA were incubated for 30 min at 37°C in the oxygenated bioassay chambers in buffer containing 300 µM IBMX to prevent degradation of cGMP, and cGMP levels were measured according to the kits instructions. Ring responses were also tested in the presence of the PKG inhibitor KT-5823 (C29H25N3O5, 10 µM).
The thiol reducing agent dithiothreitol (DTT, 10 mM) was administered to GSNO-treated rings to determine whether PE constriction could be restored. In one series, DTT (10 mM) was added alone or with GSNO (250 µM) to the ring bath 30 min before PE. We removed DTT and GSNO by rinsing the baths, and the sequential concentrations of PE were tested. The responses were expressed as percent change in isometric tension from the initial maximum PE dose of 106 M. In the next series, DTT (10 mM) was added to the bath after the administration of PE (106 M) in both control rings and GSNO-treated rings. The responses in the presence of PE (106 M) and DTT were expressed as the percent change in ring tension from the initial maximum 106 M PE dose.
The dose-response curve to PE was also compared before and after treatment with
1-adrenergic receptor antagonists. Rings were treated for 30 min with RS-100329 (Tocris Cookson, Ballwin, MO), which has a high specificity for the
1A-adrenergic receptor subtype, or BMY-7378 (Tocris Cookson), which selectively targets
1D-adrenergic receptors at low concentrations (14). Three different concentrations of antagonists (108, 107, 106 M) were tested for each PE dose-response curve. Five concentrations of PE between 108 M and 106 M PE were studied to construct dose-response curves for statistical analysis of antagonist studies.
Biotin switch assay.
S-nitrosylation of the
1-adrenergic receptor was determined by the biotin switch method (15). In brief, rat lung was treated with or without 100 µM freshly made CSNO or GSNO for 30 min at 4°C. The sample was treated with methyl methanethiosulfonate (20 mM) and 25% SDS at 50°C for 20 min, and protein was acetone-precipitated. The pellet was resuspended in buffer and treated for 1 h at room temperature with 400 µM N-[6-(biotinamido)hexyl]-3'-(2'-pyridyldithio) propionamide (biotin HPDP, Pierce) and 5 mM ascorbic acid. One aliquot of CSNO-treated protein was incubated with the biotin-HPDP without ascorbic acid as a negative control. Other negative controls were performed with DTT (10 mM) plus nitrosothiol to prevent S-nitrosylation of the receptor and nitrosothiol (CSNO) plus ODQ (10 µM) to confirm cGMP independence of S-nitrosylation. The samples were again acetone-precipitated, resuspended in buffer, and incubated overnight with 30 µl of streptavidin-agarose beads at 4°C. The beads were washed, and biotin-labeled protein was eluted with 45 µl of buffer containing 20 mM HEPES, 100 mM NaCl, 1 mM EDTA, and 100 mM 2-mercaptoethanol. Proteins were separated on gradient SDS-PAGE gels (Bio-Rad, Hercules, CA), transferred to polyvinylidene difluoride membranes, and blocked with 1% nonfat dried milk. The
1-adrenergic receptor was detected by Western blot analysis with a polyclonal rabbit anti-rat
1-adrenergic receptor antibody (1:400; Affinity Bioreagents, Golden, CO) followed by a secondary horseradish peroxidase-conjugated anti-rabbit antibody and developed with enhanced chemiluminescence.
Radioligand binding assays.
Saturation binding isotherms were performed on rat lung membranes treated with and without GSNO (250 µM) at 37°C for 30 min. Membranes (515 µg protein) were incubated with 20200 pM 125I-HEAT {2-[
-(4-hydroxyphenyl)-ethyl-aminomethyl] tetralone; New England Nuclear, Boston, MA}, a highly selective
1-adrenergic receptor antagonist, in a 0.25-ml final volume of assay buffer for 30 min at 25°C. Nonspecific binding was determined by incubation with 125I-HEAT in the presence of 10 µM phentolamine. The assay was terminated by rapid vacuum filtration on Whatman GF/C filters using a Brandel harvester, and each filter was counted for 2 min in a liquid scintillation counter (Beckman LS 1800). Filters were presoaked with 0.3% polyethyleneimine for 2 h to minimize nonspecific binding, which at the highest HEAT concentrations was roughly one-third of total binding. All assays were performed in triplicate. The
1-adrenergic receptor density (Bmax, fmol/mg protein) and affinity (Kd, nM) were calculated with Graphpad Prism (Graphpad, San Diego, AC). Bmax represents the membrane binding sites in the presence of saturating concentrations of 125I-HEAT, whereas Kd represents the concentration of ligand that half-maximally occupies the receptor. A PE competition curve (108104 M) was performed with 20 pM 125I-HEAT to calculate Ki, the concentration of competitor (PE) that half-maximally displaces 125I-HEAT. Binding to 20 pM 125I-HEAT also was compared following 30-min incubations at each of three concentrations of GSNO (2.5, 25, and 250 µM).
RT-PCR.
For RT-PCR, total RNA from rat pulmonary artery was isolated with TRIzol reagent (Life Technologies, Gaithersburg, MD). RNA was reverse-transcribed (M-MLV Reverse transcriptase, Life Technologies), and the resultant cDNA was amplified for up to 32 cycles using gene-specific primers for
1-adrenergic receptor subtypes as previously described (9). Control reactions were performed without reverse transcription. Adrenergic receptor amplicons were normalized to 18S ribosomal RNA as an external standard.
Statistics. PA ring data were analyzed by analysis of variance (ANOVA) followed by the Fisher protected least-square-difference test using commercially available software (Statview512+; Brain Power, Calabasas, CA). Data were expressed as means ± SE. Receptor binding studies were fit by nonlinear regression and compared by F-test (Prism 4; Graphpad, San Diego, CA). The Bmax and Kd were also compared by t-test (Statview512+). The P values have been provided where statistical tests were performed.
| RESULTS |
|---|
|
|
|---|
|
|
|
1-adrenergic receptor.
S-nitrosylated proteins were separated from total lung protein by biotin switch and probed for
1-adrenergic receptor (Fig. 4). Endogenous S-nitrosylation of
1-adrenergic receptors was not detectable in protein from control rat lungs, but receptor S-nitrosylation was demonstrated by the biotin switch assay after a 30-min treatment with CSNO. CSNO was used for its rapid and robust effects compared with GSNO (100 µM), which nitrosylated the receptor weakly. However, the addition of reduced cysteine (10 µM) to GSNO produced a strong signal similar to CSNO. A negative control (lane 4 of the immunoblot) showed no signal in lung protein treated with CSNO but without ascorbic acid, which is required to replace NO with biotin. Furthermore, DTT (10 mM), which blocked the nitrosothiol effect in rings, also inhibited receptor S-nitrosylation. ODQ did not prevent S-nitrosylation, indicating that the reaction was cGMP independent. These findings were reproduced in three to five separate experiments, and Fig. 4 shows an example of a representative blot.
|
1-adrenergic receptors.
Total and nonspecific HEAT saturation binding to
1-adrenergic receptors was measured in rat lung membranes with and without GSNO treatment. Specific HEAT binding was determined as the difference between total and nonspecific binding (20200 pM) in the presence of 10 µM phentolamine. GSNO (250 µM) decreased receptor-ligand binding (P < 0.01 by nonlinear regression analysis) (Fig. 5A). Bmax and Kd for HEAT binding were determined from individual saturation binding curves of control (n = 9) and GSNO-treated (n = 8) membranes.
1-Adrenergic receptor density (Bmax, fmol/mg protein) was decreased in GSNO-treated membranes (200 ± 10 fmol/mg protein in control vs. 133 ± 10 fmol/mg protein in treated, P < 0.05), whereas affinity for the ligand (Kd, nM) was not altered (33 ± 6 pM in control vs. 24 ± 10 7 pM in treated, P > 0.05). An agonist competition curve with PE (108104 M) indicated only the low-affinity binding state in lung membranes. GSNO-treated lungs demonstrated a small but definite increase in calculated Ki for HEAT binding (Ki = 6.1 x 109 M in control membranes compared with Ki = 8.3 x 109 M in GSNO-treated membranes, n = 3 each, P < 0.05) (Fig. 5B). The decrease in HEAT binding was similar after treatment with 2.5, 25, or 250 µM GSNO (n = 2, data not shown).
|
1-adrenergic receptor subtypes.
Rat PA expressed mRNA for all three subtypes of the
1-adrenergic receptor family,
1A,
1B, and
1D adrenergic receptor. mRNA expression was determined by RT-PCR using subtype-specific primers. The (-)RT controls showed no band, confirming that the signal did not represent genomic DNA (Fig. 6A). Two distinct subtype-selective antagonists showed concentration-dependent inhibition of the PE dose-response curve using five PE doses between 108 and 106 M. BMY-7378 (
1D-selective antagonist) and RS-100329 (
1A-selective antagonist) significantly decreased PE potency (n = 4 for each group, *P < 0.05 vs. control rings) (Fig. 6, B and C).
|
| DISCUSSION |
|---|
|
|
|---|
1-adrenergic receptor agonist-induced vasoconstriction and
1-adrenergic receptor-ligand binding in pulmonary vessels and found novel evidence of cGMP-independent inhibition of
1-adrenergic receptor activity.
1-Adrenergic receptors contribute to pulmonary vascular regulation by mediating vasoconstriction via activation of Gq linked to phospholipase C, which catalyzes the hydrolysis of polyphosphoinositide, liberating both diacylglycerol and inositol triphosphate. Three different
1-adrenergic receptor subtypes have been identified and cloned (
1A,
1B, and
1D), each with specific tissue distributions (8, 27). The lung expresses mRNA for all three
1-adrenergic receptor isoforms, and more than one
1-adrenergic receptor subtype mediates the contraction to PE in the pulmonary vasculature (14, 25). Our data confirm both the presence of all three
1-adrenergic receptor subtypes and the contribution of more than one
1-adrenergic receptor subtype in PE-induced constriction in rat PA.
In this study, PE-induced vasoconstriction in PA rings was inhibited by pretreatment with either of two S-nitrosothiols, GSNO and CSNO. This observation is consistent with earlier studies examining the role of NO in response to
1-adrenergic receptor agonists in different vascular beds. In rat PA rings, NOS inhibitors or removal of endothelium increased the constrictor response to PE, indicating that endogenous NO attenuates vasoconstriction by catecholamines (28). Furthermore, a nonselective NOS inhibitor, but not the relatively selective inhibitor of inducible NOS (NOS II) aminoguanidine, augmented PE constriction in PA rings, implicating other NOS isoforms (e.g., NOS III) as the source of NO responsible for this effect (26). Use of the NOS inhibitor, L-NAME, or removal of the endothelium also augmented
1-adrenergic vasoconstriction to NE in rabbit bronchial arteries (34). Pretreatment with S-nitrosylating agents including S-nitroso-N-acetylpenicillamine and GSNO blocked vasoconstriction to
1-adrenergic receptor agonists in rat aorta, similar to our findings with GSNO in rat PA (3, 4, 17); however, neither a role for cGMP nor DTT reversibility was determined in those studies.
To better define these effects, we examined whether inhibition of PE-induced vasoconstriction by S-nitrosothiols required cGMP generation. The guanylate cyclase antagonists LY-83583 and ODQ blocked cGMP production with GSNO but did not prevent GSNO-mediated inhibition of PE vasoconstriction. The PKG antagonist KT-5823, which blocks downstream effects of cGMP, produced similar results, further supporting a cGMP-independent mechanism of action by GSNO. The interpretation of the ODQ experiments is complicated by dose-dependent effects of ODQ on PE vasoconstriction without concomitant changes in cGMP content. In addition to its effects on guanylate cyclase, ODQ can alter vascular tone by inhibiting other heme-containing enzymes like NOS (10). This effect could account for the leftward shift in the PE response curve similar to that of inhibiting endogenous NO with L-NAME treatment. The decreased PE response with low-dose ODQ + GSNO-treated rings compared with ODQ alone indicates a cGMP-independent component of GSNO activity and is consistent with the results of experiments with LY-83583 and KT-5823. Moreover, the effects of S-nitrosothiols on both PE vasoconstriction in thoracic artery and 5-HT vasoconstriction in PA persist after inhibition of guanylate cyclase (17, 23), further indicating cGMP-independence of vasoactivity of (S)NO. Thus, in addition to the established NO-mediated vasorelaxation via the cGMP/PKG pathway, we observed that GSNO, by inhibiting PE vasoconstriction, can maintain low vascular tone through a cGMP-independent mechanism.
Several important lines of evidence indicate that S-nitrosothiols modify PE responses through thiol nitrosylation of the
1-adrenergic receptor system. S-nitrosothiols, including GSNO and CSNO, can facilitate transnitrosylation of other protein thiols (13, 20). The strong physiological effect of GSNO is not duplicated by the corresponding reduced thiol, GSH or its oxidized form, GSSG, indicating that the effect was specific to the S-nitrosothiol. The inhibitory effects of GSNO on PE vasoconstriction were fully reversed by the thiol reductant DTT, which can reverse disulfides, S-thiolation, and S-nitrosylation. The detection of S-nitrosylated
1-adrenergic receptor after S-nitrosothiol treatment supports the conclusion that the cysteine modification is S-nitrosylation. That GSNO-mediated S-nitrosylation was facilitated by reduced cysteine supports the idea that transnitrosylation is important for NO modification of proteins, and the ability of CSNO to enter the cell via specific amino acid transporters enhances its ability to react intracellularly (20). It is unlikely that GSNO S-thiolates directly or leads to thiol oxidation since GSH, GSSG, or H2O2 did not affect vascular responses. The reactive nitrogen species peroxynitrite also promotes S-glutathionylation of the sarco/endoplasmic reticulum calcium ATPase in the presence of GSH to mediate aortic relaxation (1). Recent data also indicate that glutathione sulfoxide, a GSNO decomposition product, rather than GSNO, can promote mixed disulfide formation (29). We did not exclude the possibility that these other cysteine modifications may also occur in the pulmonary circulation (11, 16).
We have shown
1-adrenergic receptor S-nitrosylation but not yet the specific S-nitrosylation sites responsible for the S-nitrosylation effect because rat
1-adrenergic receptors contain as many as 14 cysteine residues that are potential targets of S-nitrosylation (27). Several cysteines are highly conserved in the
1-adrenergic receptor and are located in extracellular and intracellular regions of the
1-adrenergic receptor responsible for receptor-ligand binding and receptor activation (National Center for Biotechnology Information Entrez Protein, ClustalW Multiple Sequence Alignment).
S-nitrosylation covalently modifies cysteine residues to regulate function of many proteins including hemoglobin, caspase-3, GAPDH, calcium release channel/ryanodine receptor, and the N-methyl-D-aspartate receptor (13). Aortic rings exposed to S-nitrosylating agents including GSNO, but not DEA-NO, contain a higher NO content and increased cysteine-NO immunostaining in association with sustained effects on ring tension (3). A monomeric G protein, p21ras, changes both protein conformation and activity when S-nitrosylated (19). S-nitrosylation can be reversed to restore enzyme activity of caspase-3 and tissue transglutaminase (18). The ability to regulate vascular tone through S-nitrosylation and denitrosylation provides further evidence of an important physiological function for this NO mechanism. Whereas other posttranslational protein modifications including phosphorylation are well-accepted mechanisms regulating GPCRs, the potential role of S-nitrosylation in the regulation of GPCR signaling has not been established.
We further tested the effects of GSNO on
1-adrenergic receptor by comparing 125I-HEAT saturation binding isotherms in lung membranes in the presence and absence of GSNO. GSNO treatment decreased total
1-adrenergic receptor binding sites (Bmax) in the lung, while it had no significant effect on binding affinity (Kd). This is the first direct evidence that nitrosothiols alter
1-adrenergic receptor binding. The ready explanation for the decrease in
1-adrenergic receptor binding sites with GSNO is that ligand binding is disrupted by NO modification of the receptor. Binding sites can also be decreased if the receptor is internalized or downregulated, but this is unlikely to occur after the relatively short 30-min incubation period with GSNO. It is also unlikely that GSNO treatment damaged the receptor sufficiently to disrupt binding sites. The effects of GSNO were reversed by DTT, restoring constrictor effects of PE and indicating receptor integrity after GSNO, which argues against receptor internalization or damage. The competition curve with PE did not reveal the high-affinity state of the receptor, which was likely due to experimental limitations of using actual lung tissue. Therefore, it was not possible to determine the effects of GSNO on G protein coupling in this system. Cell culture studies have demonstrated effects of NO on G protein uncoupling from other receptors. For example, SIN-1, which releases NO and superoxide simultaneously, promoted uncoupling of Gs from
2-adrenergic receptor expressed in HEK-293 cells. (2). S-nitrosothiols also decreased coupling of M2 muscarinic, bradykinin, and AT1 GPCR to their respective G proteins (5, 7) (22). NO does have other effects on GPCR signaling, including desensitization. NOS inhibition for example, promotes tachyphylaxis to the vasodilator effects of the
-adrenergic receptor agonist, isoproterenol, and pituitary adenylate cyclase activating polypeptide receptor agonist, PACAP-27 (30, 32, 33).
In conclusion, GSNO inhibits PE vasoconstrictor response in PA rings through a reversible cGMP-independent mechanism that decreases
1-adrenergic receptor binding sites. This represents a novel mechanism of NO regulation of pulmonary vascular tone in which S-nitrosylation of the
1-adrenergic receptor decreases the vasoconstrictor response to agonists. Future studies will locate specific molecular target(s) of NO within the
1-adrenergic receptor-G protein complex and determine the associations with altered SNO levels and vascular disease.
| GRANTS |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
W. J. Perkins, M. Taniguchi, D. O. Warner, E. N. Chini, and K. A. Jones Reduction in soluble guanylyl cyclase-specific activity following prolonged treatment of porcine pulmonary artery with nitric oxide Am J Physiol Lung Cell Mol Physiol, July 1, 2007; 293(1): L84 - L95. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |