|Home | About | Journals | Submit | Contact Us | Français|
Nitric Oxide (NO) plays a critical role in diverse physiological and pathological processes. We show that a hypomorphic mouse model of argininosuccinate lyase (Asl) deficiency exhibits a distinct phenotype manifest by multi-organ dysfunction and NO deficiency. Loss of Asl leads to reduced NO synthesis due to decreased endogenous arginine synthesis as well as reduced utilization of extracellular arginine for NO production in both humans and mice. Hence, ASL as seen in other species through evolution has a structural function in addition to its catalytic activity. Importantly, therapy with nitrite rescued the tissue autonomous NO deficiency in hypomorphic Asl mice, while a NOS independent NO donor restored NO-dependent vascular reactivity in subjects with ASL deficiency. Our data demonstrate a previously unappreciated role for ASL in NOS function and NO homeostasis. Hence, ASL may serve as a target for manipulating NO production in experimental models, as well as treatment of NO-related diseases.
L-Arginine is the natural substrate of nitric oxide synthases (NOS) for generating nitric oxide (NO). As a by-product of the NOS reaction, L-citrulline is formed from L-arginine. Within the cell, citrulline can be recycled back to arginine by the cytoplasmic enzymes argininosuccinate synthetase (ASS) and argininosuccinate lyase (ASL), constituting the citrulline-NO cycle (Supplementary Fig. 1). The availability of intracellular arginine is potentially a rate-limiting factor in cellular NO production in spite of the fact that the extracellular and exogenous sources should theoretically be able to replenish its deficiency. It has been hypothesized that compartmentalization and intracellular metabolite channeling underlies the “arginine paradox” in which extracellular and intracellular pools of arginine are distinguishable. However, this has yet to be proven in vivo and the mechanistic basis is unknown.
Many tissues and cell types contain the cytoplasmic enzymes, ASS and ASL, providing a cell-autonomous mechanism for generating arginine (Supplementary Fig. 1). Arginine serves as the precursor for the synthesis of urea, NO, polyamines, proline, glutamate, creatine and agmatine1. The regulation of L-arginine availability for these intracellular pathways is cell-type and end-product specific, e.g., arginine generated by ASL in hepatocytes is mostly directed to urea production, while in other cell types its metabolic fate is context dependent2.
Argininosuccinic aciduria (ASA) (MIM 207900) is the second most common human urea cycle disorder (UCD) and is caused by deficiency of ASL. Subjects with ASA disease cannot generate arginine from citrulline. It is noteworthy that despite early treatment and adequate metabolic control of hyperammonemia, subjects with ASA disease can exhibit persistent intellectual impairment, delayed motor skills3,4 and progressive hepatic disease5–7. Importantly, intellectual impairment and liver cirrhosis are seen even in those with early initiation of treatment and few, if any documented episodes of hyperammonemia. Recent experience also suggests that patients are at risk for the development of systemic hypertension4,8,9. The mechanism behind these unique clinical features that are not observed in other UCDs is open to speculation, although a reduction of NO production secondary to a localized deficiency of L-arginine is one intriguing possibility. An association between urea cycle function and NO production has been previously suggested by the association of genetic variants in urea cycle genes with NO-related disease processes10,11. Recently, NO synthesis was evaluated in patients with UCD12. However, no mechanism was demonstrated that could explain potential differences amongst or between patients and controls. Such a study emphasizes the difficulties in studying NO metabolism at the level of the whole organism in non-steady state conditions.
We hypothesized that a cell autonomous deficiency of ASL would lead to systemic NO deficiency. In addition, because of the inability of supplemental arginine to prevent long-term complications in ASA patients, we investigated whether ASL plays a more central role in cellular arginine utilization for NO synthesis beyond intracellular recycling of citrulline into arginine. We tested our hypothesis in vivo in a hypomorphic mouse model of Asl as well as in ASA subjects with absent enzyme activity. Our results were complemented with in vitro studies in human ASA fibroblasts, in primary cell lines with ASL knockdown, and in cells over expressing ASL mutants that are enzymatically inactive but structurally intact.
We hypothesized that loss of Asl would cause deficiency of endogenous arginine production resulting in reduced NO synthesis. Similar to human neonates with ASA, complete loss of function of Asl in mice leads to neonatal hyperammonemia and lethality13. Therefore, we generated a conditional hypomorphic allele by introducing a Neomycin (Neo) selection cassette into intron 9 of the mouse Asl gene (Supplementary Fig. 2a–c). As predicted, quantitative RT-PCR, Western blot analysis and Asl enzymatic activity of mice homozygous for the Neo insertion, AslNeo/Neo, demonstrated significant reduction in gene and protein expression with 25% residual RNA, 25% residual protein and 16% residual enzyme activity, thus confirming the hypomorphic nature of this allele (Supplementary Fig. 3a–c). Biochemically, plasma amino acid analysis also showed a profile consistent with Asl deficiency with elevation of citrulline and argininosuccinic acid, two precursor metabolites upstream of the enzymatic block, and reduction of arginine, a product metabolite downstream of the enzymatic block (Supplementary Fig. 3d). The hypomorphic nature of this allele was confirmed to be due to the Neo cassette insertion, as its Frt-mediated deletion resulted in phenotypically normal mice (AslFlox/Flox) (Supplementary Fig. 4a).
Asl hypomorphic mice were born at the expected Mendelian ratio but die within the first 3–4 weeks of life from multi-organ failure (Fig. 1a,b), providing an opportunity to explore the consequences of impaired endogenous arginine synthesis. During the early postnatal period, the dietary supply of arginine (from breast milk) is insufficient and thus, there is dependency on endogenous arginine synthesis14.
While AslNeo/Neo mice were indistinguishable from wild type littermates at birth, they showed abnormal hair patterning and significant growth restriction by two weeks (Fig 1a,c). AslNeo/Neo mice also displayed evidence of multi-organ dysfunction characterized by elevation of liver transaminases, decreased renal creatinine clearance, and elevated systolic and diastolic blood pressures as compared to WT animals (Fig. 1d–f). Histological analyses revealed multi organ involvement, including the immune, hematopoietic, renal and cardiovascular systems, features that are not attributable to hyperammonemia15 (Supplementary Figure 5a–f).
We noted that some features of this phenotype would be consistent with a systemic disruption of NO homeostasis, since NO has been shown to play an important role in diverse pathological processes including systemic hypertension, immune dysfunction, renal disease and hepatic fibrosis16,17.
To determine whether the physiological features observed in the Asl hypomorphic mice were attributable to impaired NO synthesis or altered NO homeostasis, we examined the effect of a genetic interaction between Asl and Nos3 on systemic blood pressure. While neither the Nos3+/− or the AslNeo/+ mice are hypertensive, the double heterozygous mice, Nos3+/−;AslNeo/+ had significant hypertension (Figure 1g) confirming an epistatic effect.
While measurement of NO oxidative products (nitrite, nitrate) can reflect the status of NO production, we additionally assayed the tissue and plasma nitrosothiols (RSNO) levels as a reflection of NO production and signaling. We found reduced NO production in tissues of AslNeo/Neo mice as reflected by a significant decrease in S-nitrosylation and/or nitrite in heart, and other tissues (Fig. 2a,b). To address whether the loss of Asl led to secondary changes within the NO pathway that might alternatively account for reduced NO synthesis, we measured RNA and protein expression levels of Ass (upstream of Asl), arginine transporter Cat-1, and Nos3. We found that their expression in mutant lung and liver was either unchanged or even up-regulated in the face of Asl loss of function (Fig. 2c, Supplementary Fig. 6a–c). Similarly, the levels of the endogenous inhibitor of L-arginine, asymmetric dimethyl arginine (ADMA), were also comparable in mutant and WT mice (Data not shown).
Hence, the multi-organ dysfunction observed in Asl mutant mice correlated with evidence of decreased systemic NO resulting specifically from the primary deficiency of Asl activity. This specificity was further supported by the finding of normal tissue RSNO levels in the AslFlox/Flox mice in which the Neo cassette had been excised and Asl expression normalized (Supplementary Fig. 4b,c).
Recent studies show that nitrate and nitrite can be recycled to form NO as an alternative to the classical L-arginine-NOS pathway18,19. If some of the phenotypic consequences of Asl deficiency were due to decreased NO production, normalization of NO status by supplementation with sodium nitrite should partially correct the phenotype. However, in this model, hyperammonemia (secondary to an inability to clear waste nitrogen by urea synthesis in the liver) complicates the postnatal picture and contributes to the lethality in AslNeo/Neo mice. Therefore, we first treated Asl mutant animals with either sodium benzoate (250 mg kg−1 per day) or with L-arginine (100 mg kg−1 per day). These medications are the standard of care in ASA patients for prevention of hyperammonemia. The former works by stimulating alternative disposal of glycine containing nitrogen via conjugation, and the latter works by priming the urea cycle to generate more ASA as a nitrogen sink that is cleared via urinary excretion. Not surprisingly, we found significantly improved survival in AslNeo/Neo mutant mice treated with these drugs suggesting that, similar to human patients, hyperammonemia is indeed partially responsible for early lethality in this model (Fig. 3a). Survival improved more with arginine treatment than sodium benzoate treatment suggesting that exogenous arginine may also have a salutary effect on NO deficiency state in addition to its effects on nitrogen clearance. However, the predominant contribution of NO deficiency to the phenotype was best evidenced by the observation that treatment with sodium nitrite, which can be metabolized to form NO18,19, produced the greatest survival among all treated groups (Fig. 3a, Supplementary Table 2). Nitrite supplementation produced comparable survival to the traditional treatment of sodium benzoate plus arginine; however, weight gain was only evident in the nitrite group (Fig. 3b,c). Importantly, combination triple therapy with sodium benzoate, arginine, and sodium nitrite, produced the greatest survival and weight gain (Fig. 3b,c). Moreover, triple therapy was associated with correction of liver protein nitrosylation to levels higher than those observed in the WT mice and led to normalization of the blood pressures in mutant mice (Fig. 3d,e).
To evaluate for NO insufficiency at a tissue level in AslNeo/Neo mice, we performed classic aortic ring relaxation measurements. In contrast to WT mice, aorta from AslNeo/Neo mice exhibited evidence of significant endothelial dysfunction as shown by an inability of preconstricted aortic rings to relax in response to acetylcholine (Fig. 3f,g). Furthermore, neither preconstricted aortic rings from WT nor AslNeo/Neo mice responded to arginine treatment. A lack of response in WT rings ex vivo was expected as the vascular tissues are saturated with arginine to levels higher than the Km values of NOS for NO production20–22. However, this result was surprising in the AslNeo/Neo mouse aortas given their low intracellular levels of arginine. In contrast, AslNeo/Neo preconstricted aortic rings relaxed in response to sodium nitroprusside, a NOS independent NO donor, demonstrating the integrity of the vascular preparation and a functional signaling pathway downstream of NO (Fig. 3g).
To further evaluate the requirement of Asl for NO synthesis in a cell specific manner, we knocked down its expression using RNAi in primary piglet endothelial cells (Figure 4a,b). We found that decreased levels of ASL led to significantly lower levels of nitrite production in response to inducers such as bradykinin and L-arginine (Figure 4c,d).
Together, these data underscore the requirement for Asl in the production of NO at the organism, tissue, and cellular levels.
Our findings from the hypomorphic mouse model and in vitro studies predict that human ASA patients should also be deficient for markers of NO synthesis in spite of normal or even elevated levels of extracellular plasma arginine from therapeutic arginine supplementation provided as “standard of care”. To assess the role of ASL in NO synthesis in humans, we studied fibroblasts from human subjects with ASA disease with null ASL enzymatic activity. First, we verified that primary human fibroblasts express both NOS3 and NOS2 (Supplementary Fig. 7a). Although the expression levels of both NOS2 and NOS3 are lower than those seen in endothelial cells, both isoforms were detected at comparable levels in cultured human fibroblasts (Supplementary Fig. 7a). In addition, as previously published23–25, we established that NOS is functional in primary fibroblasts and that it can be activated by BH4 and inhibited by a NOS inhibitor (Supplementary Fig. 7b). As predicted from our mouse and cell data, addition of either L-arginine (1000µM) or L-citrulline to the media of primary fibroblasts from control subjects for either 30 minutes or 24 hours resulted in significant increases in nitrite and cGMP production reflecting an increase in NO synthesis from NOS. In contrast, nitrite production, nitrosylation, and cGMP production in fibroblasts from subjects with ASA did not increase in response to L-arginine treatment despite similar expression of NO synthetic proteins (Fig. 5a,b and Supplementary Fig. 7c–f). The viability of cells from both controls and subjects with ASA disease cells was comparable and intracellular arginine levels were effectively increased in both cell types by supplementation (Data not shown) and yet, normal NO synthesis did not occur in the absence of ASL.
In spite of elevations in plasma arginine (as a consequence of therapeutic supplementation) and of citrulline (that accumulates upstream of the block), subjects with ASL deficiency had significantly decreased levels of plasma RSNO and decreased nitrite as compared to healthy control subjects (Fig. 5c). In order to directly measure NO production and trace the contribution of extracellular arginine to NO conversion, we performed stable isotopic flux measurements in control and ASA subjects on a steady state nitrite and protein restricted diet. Using a two-day multi-tracer protocol, we found increased arginine and urea appearance (flux) rates in ASA subjects that are attributable to pharmacological treatment with L-arginine (Fig. 5d-Left Panel). We also found increased citrulline flux and fractional transfer of 15N label from 15N-glutamine to 15N-citrulline that is explained by the accumulation of this metabolite upstream of the block in ASL deficiency. However, consistent with the decreased plasma RSNO levels and the mouse and cellular data, we found dramatically decreased 15N transfer from infused 15N2-guanidino-labelled-arginine to 15N-citrulline, a surrogate marker for NO production (Fig. 5d-Right Panel). Hence, dynamic measurements of metabolite fluxes reveal that in spite of high extracellular arginine and citrulline fluxes, ASA subjects have dramatically decreased transfer of guanidino nitrogen from arginine to citrulline, a marker of NO production from arginine. Our data suggest that the decrease in NO production is attributable to an inability both to recycle intracellular citrulline into arginine for endogenous synthesis, and to channel extracellular plasma arginine for NO production. These complementary human data further support our finding, that in the absence of ASL, extracellular arginine (the source of the isotopic label) is inefficiently converted to citrulline to form NO. Hence, ASL deficiency is a model of deficient NO synthesis from both intracellular and extracellular sources of arginine.
To evaluate the functional consequences of this observation in vivo, we performed vessel reactivity assays in human subjects with ASA. We examined the vasodilatory response of the brachial artery via Doppler ultrasound following transient occlusion of flow as a measure of NOS-dependent vascular relaxation, and then the vasodilator response following sublingual nitroglycerin as a measure of NOS-independent vascular relaxation (Fig. 5e). In healthy controls, vascular dilatation following both steps were equivalent and the magnitude of response was similar to previously reported responses26. In contrast, subjects with ASA disease failed to show any NOS-dependent vascular relaxation after release of vessel occlusion. However, they responded similarly to controls after administration of sublingual nitroglycerin (Fig. 5e) showing normal response to an exogenous, NOS-independent source of NO. These results are significant as the near absence of flow-mediated relaxation, a measure of NO mediated dilation, has not been reported in other pediatric diseases. Importantly, these results in human subjects corroborate with the aortic ring experiments in the hypomorphic Asl mouse model and support that while the signaling downstream of NO is intact, loss of ASL limits NOS-dependent NO production, leading to loss of NOS-dependent vascular reactivity.
Direct biochemical interaction within a compartmentalized NOS complex may explain metabolite channeling27,28. The three key protein components responsible for recycling citrulline into NO are ASS, ASL, and NOS and their expressions are coordinately regulated29. Several studies have shown either interaction or colocalization amongst these three proteins30,33. Interestingly, an interaction between NOS3 and the cationic amino acid transporter CAT-1 responsible for arginine transport, was also recently described in endothelial cells30.
We first verified the interaction between the proteins involved in the complex by performing a mass spectrometry analysis of immunoprecipitate using an antibody to ASS on LPS induced RAW246.7 cells, which confirmed the existence of this complex for Nos2 (Supplementary Table 1). We expanded upon these findings by testing the existence of a complex between the protein components involved in NO synthesis by performing immunoprecipitation using antibody to ASS in lung and brain tissues from WT mice. Our results support the existence of a NOS-containing multi-protein complex for both NOS isoforms i.e., Nos3 in lung and Nos1 in brain (Fig. 6a, Supplementary Figure 8).
To determine the effects of partial loss of Asl on this protein complex, we quantified the component proteins co-immunoprecipitated by antibody to ASS in WT vs. AslNeo/Neo mice. Since ASA patients have hypertension and intellectual delays that are unique and independent of hyperammonemic episodes, we focused our studies on the effect of Asl deficiency on its complex formation with Nos3 and Nos1. AslNeo/Neo mouse lung and brain showed significantly decreased quantities of both Nos3 and Nos1 isoforms, respectively, as well as other proteins involved in the complex (Fig. 6a, Supplementary Fig.8); this is despite increased abundance of the individual components (other than Asl) in tissues of AslNeo/Neo mice compared to WT on Western analysis (Figure 2c, Supplementary Fig.8). These data support a central requirement for Asl in the formation of this complex.
To dissect the structural requirement of ASL for NOS complex formation from its catalytic activity, we tested NOS complex assembly using human ASL mutants in its catalytic site R236W31 (Fig. 6b). This mutation abolishes the enzymatic activity of ASL, i.e., the recycling of intracellular citrulline via the cleavage of ASA, but does not abolish its tertiary structure. Indeed, in vitro over-expression of mutated ASL in COS7 cells did not prevent NOS protein assembly with ASS as shown by IP using antibodies against ASS or NOS. However, consistent with our finding in the AslNeo/Neo mouse, absence of ASL prevented efficient immunoprecipitation of the proteins involved in the NOS complex by either one of these antibodies (Fig. 6b).
To test whether this structural requirement for ASL in NOS complex formation is in fact required for utilization of extracellular arginine for NO production, we transduced ASA mutant fibroblasts null for ASL activity with lentivirus expressing either wild type human ASL or with human ASL mutated in the catalytic domain R113Q31. As extensively studied by others, the R113Q similarly abolishes the catalytic activity of ASL without affecting ASL protein stability31,32. If the structural requirement of this NOS complex is required for channeling of extracellular arginine to NOS, the catalytic site mutation would be expected to restore NO production in response to extracellular arginine in ASA mutant cells. Indeed, when we overexpressed human ASL with the R113Q mutation in vitro, ASA fibroblasts were able to generate nitrite at a level comparable to control cells when supplemented with arginine (Figure 6c).
This structural requirement for an ASL-dependent NOS complex begins to explain our physiological and cellular observations, suggesting that in the absence of ASL, there is a less efficient formation of a NOS multi-protein complex leading to decrease NO production from both endogenously synthesized and exogenously channeled arginine (Fig. 6d).
Our findings support a requirement for ASL not only to synthesize intracellular arginine, but also to utilize extracellular arginine for NOS-dependent NO synthesis. We demonstrate an intracytosolic complex of proteins important for nitric oxide synthesis to explain the structural basis for metabolite channeling that reconciles the phenotypes observed in human and mouse models of ASL deficiency. While likely not the only explanation, the data support that decreased ASL levels leads to loss of NOS complex formation that is associated with NO deficiency at the whole organism, tissue and cellular levels in both humans and mice.
This conclusion is based on a combination of mouse and human studies that were prompted by clinical observations in ASA patients. First, the natural history of ASL-deficient humans with argininosuccinic aciduria suggested the presence of organ dysfunction and complications that were independent of hyperammonemia caused by hepatic urea cycle deficiency. This led us to hypothesize alternative mechanisms of injury including deficiency of NO secondary to loss of citrulline recycling or endogenous arginine production. However, this could not alone explain the phenotypic complexity because ASL patients are replete with extracellular arginine due to pharmacological supplementation.
This apparent conundrum was evident in the hypomorphic mouse model of Asl deficiency, where we observed histological evidence of multi-organ dysfunction that correlated with biochemical evidence of systemic NO deficiency. This was further supported by the finding of decreased markers of NO production (plasma RSNO and nitrite) in ASA patients. Importantly, ASA patients, ASA fibroblasts, and primary endothelial cells made deficient of Asl by siRNA knockdown, were also unable to efficiently generate NO after extracellular arginine supplementation. In humans, this was demonstrated via dynamic measurement of arginine to citrulline flux; while in cells, it was demonstrated via measurement of nitrite and/or cGMP. This NO deficiency was found to be tissue autonomous as preconstricted aortic rings from mutant mice relaxed in response to NO donors but not arginine, while ASA subjects exhibited abnormal flow-mediated vascular relaxation that was restored by nitroglycerin. On the whole organism level, provision of an NOS independent source of NO in the form of nitrite therapy significantly prolonged survival of hypomorphic mice while also restoring tissue nitrosylation and normalizing blood pressure.
The NO deficiency was caused in part by the inability of patients and cells to efficiently generate intracellular arginine or to utilize extracellular arginine for NO production in the face of ASL deficiency. This was in spite of adequate expression of all other protein components necessary for NO production including the arginine transporter CAT-1, ASS, HSP90, and NOS. This observation correlated with the existence of a NOS complex that depends on the structural, but not enzymatic function of ASL. Loss of ASL was associated with decreased abundance of this complex, decreased utilization of arginine for NO production, and functional consequences of NO deficiency at the organ and cellular levels.
This structural requirement was supported by the ability of specific mutations in the ASL catalytic domain to participate in the NOS complex while the complete absence of ASL prevented efficient complex formation. Together, these studies distinguish two essential roles for ASL: the recycling of citrulline in the cell for cell autonomous arginine synthesis, and the maintenance of a NOS complex that is required for efficient NO production from extracellular sources of arginine. The former depends on the catalytic function of ASL, while the latter requires its structural integrity. Interestingly, this is consistent with the distinct evolutionary roles of ASL in other species33. Since in cell, mouse, and patient models, the NO deficiency are evidenced in the face of excess arginine, it is likely that the primary dysfunction is at the level of the NO metabolon where insufficient channeling of arginine due to loss of the ASL-NOS complex leads to secondary decreased NO production. An important question for future study is whether the loss of arginine channeling leads to NOS uncoupling and consequent increase in free radical stress.
Clinically, these data suggest that the vascular dysfunction and intellectual delay seen in ASA patients may be partly due to NO insufficiency, and hence, treatment with a NOS-independent source of NO, e.g., sodium nitrite, or NO donors, would be beneficial in the long term. Moreover, the clinical variability seen in ASA patients may depend on the differential effects of specific mutations on catalytic vs. structural functions.
These data also have broader implications for NO biology and disease. Mechanistically, they support intracellular compartmentalization as an explanation for the “arginine paradox”, i.e., the increased production of NO with the addition of extracellular arginine despite apparently saturating intracellular arginine levels. Moreover, they suggest an explanation as to why the arginine paradox is not observed with ASL deficiency. As such, ASL may serve as the linchpin in NO production. Hence, inhibition of ASL in a cell-specific fashion may be an effective way to probe NO function in vivo, independent of potential NOS redundancy. Similarly, it may serve as a novel target for manipulating NO production in a cell autonomous fashion in human disease processes.
This work was supported by the NIH (DK54450, RR19453, RR00188, GM90310 to B. Lee, GM07526 and DK081735 to A. Erez, HL75511 to J. Aschner, RR024173 to J. Marini, and DK37175 to W. Mitch). We acknowledge and thank the clinical efforts of M. Mullins, S. Carter, A. Tran, J. Stuff, W. Martinek and the TCH General Clinical Research Center nursing staff. The mutated hASL plasmids were kindly provided by L. Salviati from the Department of Pediatrics, University of Padova, Padova, Italy.
We thank the subject families for their kind participation. We greatly appreciate the technical contributions of M. Jiang, E. Munivez, B. Dawson, G. Sule, and J. Zhang. We thank L. Castillo for helpful discussions. Figures were produced using Servier Medical Art. A. Erez, O. Shchelochkov, SC. Nagamani are awardees of the National Urea Cycle Disorder Foundation Research Fellowship. O. Shchelochkov is an awardee of the O’Malley Fellowship of the Urea Cycle Disorders Rare Disease Clinical Research Network. N.Bryan is supported by the American Heart Association-National 0735042N. P. Campeau is an awardee of the CIHR (Canadian Institute of Health Research) clinician-scientist training award.
Author contributionA.E.- Generated the mouse models, performed most of the experiments and wrote the manuscript
S.C.S.N.- Performed the human in vivo and in vitro experiments
O.A.S.- Conducted the mouse therapy experiments
M.H.P.- Conducted the immunopercipitation experiments
P.M.C.- Helped conducting the mutated ASL experiments
Y.C.- Generated the mouse models
H. K.G.- Analyzed the NOx data and helped with the ELISA experiments
L.L.- Performed western experiments
A.M.- Conducted the patients’ studies
T.K.B.- Conducted the RT-PCR experiments
J.O.B.- Performed the histological analysis
H.Z.- Performed western and immunopercipitation experiments.
Y.T.- Conducted the vascular ring experiments
A.K.R.- Helped with the blood pressure analysis
M.S.- Contributed to the conception of the hypothesis
W.E.O.- Analyzed the biochemical data
D.G.H.- Contributed to our understanding regarding NOS function
W.E.M.- Conducted the vessel reactivity assay, helped with assessing creatinine clearance
J.C.M.- Performed the labeled isotope studies
J.L.A.- Supervised the experiments performed on the Asl hypomorphic lungs, helped in critical analysis and in revising of the manuscript
N.S.B. – Helped with the NOx and NOS data analysis and supervised related experiments
B.L.- Led and supervised the project through all stages