OUP user menu

Induction of nitric oxide synthase in human vascular smooth muscle: interactions between proinflammatory cytokines

Adrian H. Chester , Julie A.A. Borland , Lee D.K. Buttery , Jane A. Mitchell , Deirdre A. Cunningham , Sassan Hafizi , Ginette S. Hoare , David R. Springall , Julia M. Polak , Magdi H. Yacoub
DOI: http://dx.doi.org/10.1016/S0008-6363(98)00054-6 814-821 First published online: 1 June 1998


Objective: We have attempted to demonstrate the induction of inducible nitric oxide synthase in human vascular tissue and define the capacity of different cytokines to induce this enzyme. Methods: Segments of human arteries were stimulated with lipopolysaccharide (10 μg/ml), interleukin-1β (5 U/ml), tumor necrosis factor-α (10 U/ml), and interferon-γ (200 U/ml). Cytokines were either used alone or in certain combinations, as well as in the presence of l-NG-monomethyl-arginine (100 μmol/l) or cycloheximide (1 μmol/l). Induction was assessed by measurement of mRNA expression, immunocytochemical localisation of the expressed protein, nitric oxide synthase activity and levels of nitrite, a product of nitric oxide formation. Results: PCR analysis showed the presence of mRNA for iNOS in stimulated samples which could be inhibited by cycloheximide. There was positive staining with an antibody against human iNOS in the media of stimulated vessel segments. Stimulated segments were also shown to contain Ca2+-independent nitric oxide synthase activity. The cytokines and lipopolysaccharide together gave a significant rise in levels of nitrite in the medium after 36 and 48 h, which was inhibited by l-NG-monomethyl-arginine and cycloheximide. Only interferon-γ incubated alone was capable of increasing nitrite levels. This effect was enhanced by co-incubation with either interleukin-1β, tumor necrosis factor-α or lipopolysaccharide. Conclusion: We have shown that increased production of nitrite by human vascular tissue in response to cytokines is associated with induction of iNOS as shown at the molecular and protein levels, and further supported by the presence of increased Ca2+-independent nitric oxide synthase activity following cytokine stimulation.

  • Nitric oxide
  • Septic shock
  • Atherosclerosis
  • Cytokines
  • Human

Time for primary review 50 days.

1 Introduction

Nitric oxide plays an important role in the biological function of vascular tissue as well as many other cells and organ systems [1–3]. The molecule is formed enzymatically from l-arginine through the action of the enzyme nitric oxide synthase (NOS). NOS exists in several isoforms, including constitutive isoforms which are present in endothelial cells (eNOS) and some nerves (nNOS), and requires the presence of extracellular calcium for the binding of calmodulin [4–6]. In addition, an inducible isoform of the enzyme (iNOS) is expressed in vitro and in vivo following administration of certain cytokines and/or lipopolysaccharide (LPS) [1]. In contrast to constitutive isoforms, iNOS is able to bind calmodulin at very low levels of free extracellular calcium [7–9].

Recent evidence has suggested a possible role for the inducible form of the enzyme in atherosclerosis, heart failure, septic shock and transplant rejection [10–15]. However, there have been mixed reports on the ability of inflammatory agents to induce expression of iNOS in human cells, with significant species variation between animal and human cells [16, 17]. Studies with human tissue have suggested that the effects of pro-inflammatory agents on the vessel wall may not be mediated by iNOS induction [18–20]. Effects that have been reported include up-regulation of co-factors for eNOS in endothelial cells, direct effects on the vessel wall independent of nitric oxide or prostanoids and depression of endothelium-dependent relaxation which can subsequently recover, with no effect on iNOS expression or endothelial cell morphology. However, there is data showing modification of functional vasoactive responses in human blood vessels by cytokines and/or LPS [21, 22], and localisation of the enzyme in pathological specimens [23]. The potential agents and molecular mechanisms responsible for iNOS expression have not been defined in human tissues. The aim of the present study was to examine the ability of intact segments of human vascular smooth muscle to express iNOS using both molecular and biochemical techniques, and attempt to define the capacity of different pro-inflammatory cytokines, either alone or in combination to induce this enzyme.

2 Methods

2.1 Tissue harvesting

Unused distal ends of human internal mammary artery (IMA) were harvested during coronary artery bypass, from 12 patients, while samples of aorta were collected from 5 organ donors at the time of cardiac transplantation. Samples of IMA were used for characterisation of the inducibility of iNOS, while segments of aorta were used to study the interactions between the different cytokines. After harvesting, samples were placed in modified Tyrode's solution at 4°C, composition (mmol/l) NaCl 136.9, NaHCO3 11.9, KCl 2.7, NaH2PO4 0.4, MgCl2 2.5, CaCl2 2.5, glucose 11.1 and disodium EDTA 0.04. The tissue was transported to the laboratory immediately, dissected free of surrounding connective tissue under a surgical microscope and cut into segments 3 to 6 mm in length, depending on the total length of the original specimen. Segments of IMA or aorta were then placed in tissue culture medium (Dulbecco's modified Eagle's medium) containing 10% human AB serum, l-arginine (500 μmol/l), penicillin (100 U/ml) and streptomycin (100 μg/ml). Vessel segments were washed in fresh tissue culture medium for 5 minutes before commencing the experiment.

2.2 Tissue incubation

Vessel segments were placed in 2 ml of tissue culture medium in a 24 well plate. Tissues were incubated with tissue culture medium only, or with a combination of LPS (10 μg/ml), interleukin-1β (IL-1β) (5 U/ml), tumor necrosis factor-α (TNF-α) (10 U/ml) and interferon-γ (IFN-γ) (200 U/ml) (all from Sigma Chemical, UK). (The combination of all four agents together is referred to in the text as ‘cytomix’). In some experiments l-NG-monomethyl-arginine (l-NMMA) (100 μmol/l) (gift from Dr. Salvador Moncada, Cruciform Project, University College London, UK), a nitric oxide synthesis inhibitor, or cycloheximide, (1 μmol/l) (Sigma Chemical, UK), an inhibitor of protein synthesis, were added to the tissue samples prior to the incubation with LPS and/or the cytokines. Concentrations of cytokines used were determined from preliminary studies. In some experiments higher concentrations of IL-1β (100 U/ml), TNF-α (1000 U/ml) and of IFN-γ (400 U/ml) were tested and shown to give the same degree of induction as those employed in the study. The degree of iNOS induction, with respect to nitrite release, was similar in segments of IMA and aorta.

The 24 well plate was then placed in a 95% air/5% CO2 incubator at 37°C for 48 h. Tissue culture medium samples of 200 μl were removed 12, 24, 36 and finally 48 h after starting the incubation, placed in a sterile cryovial and frozen in liquid nitrogen until analysed. In some experiments vessel segments were fixed for immunocytochemistry or stored in liquid nitrogen for either mRNA extraction or measurement of NOS activity.

2.3 RNA extraction and preparation of cDNA

Total RNA was extracted from IMA samples according to the method of Chomczynski and Sacchi [24]. cDNA was prepared by reverse transcription of total RNA with AMV reverse transcriptase enzyme using Oligo (dT) as primer (Promega). After transcription at 42°C for 60 min the reaction mixture was denatured at 95°C for 5 min, chilled on ice and stored at −20°C until required for PCR amplification.

2.4 PCR analysis

A 2.5-μl aliquot of the cDNA mix was added to a 50 μl PCR reaction containing 20 μmol/l of each dNTP, 100 pmol/l of each primer, 1×PCR buffer and 1–1.5 mmol/l MgCl2 (1 mmol/l:-Glucose-6-phosphate dehydrogenase (GAPDH) and iNOS, 1.5 mmol/l eNOS). The MgCl2 concentration was titrated for each primer set. Samples were denatured at 94°C for 1 min, annealed at 58°C for 1 min and extended at 72°C for 1 min. This cycle was repeated 40 times and analyzed by staining with ethidium bromide after agarose gel electrophoresis of a 10-μl aliquot of the amplified product. Negative controls consisted of PCR reactions from which the cDNA was omitted and cDNA reactions from which reverse transcriptase enzyme was omitted. The identity of each product was verified by size. Size markers were the 100 base pairs (bp) ladder purchased from Life Technologies, Paisley, Scotland.

2.5 Oligonucleotide primers and probes

The oligonucleotide primers and probes used are listed below. Primers for iNOS, and GAPDH were synthesized in house from the published sequences [25–27].

iNOS (human hepatocyte): 5′ Primer 5′-ATT GAT CAG AAG CTG TCC C-3′; 3′ Primer 5′-GTA GAT TCT GCC GAG ATT TG-3′. The expected product size was 308 bp in length.

GAPDH (human liver): 5′ Primer 5′-TCA CCA TCT TCC AGG AGC GA-3′; 3′ Primer 5′-TCC TTG GAG GCC ATG TGG GC-3′. The expected product size was 763 bp in length.

2.6 Immunocytochemistry

Tissues were fixed in a solution of 1% paraformaldehyde in phosphate-buffered saline (PBS: 10 mmol/l phosphate buffer, pH 7.2–7.4, 150 mmol/l NaCl) for 4–6 h. After washing at 4°C in PBS containing 450 mmol/l sucrose, tissues were frozen and sectioned (10 μm thickness) for immunostaining. Tissue sections were stained by the avidin-biotinylated complex-peroxidase method. Endogenous peroxidase was blocked by immersing slides in 0.03% hydrogen peroxide in methanol for 20 min. Sections were then blotted to remove excess serum and incubated overnight with peptide antiserum diluted 1:1000 in PBS containing 0.05%(w/v) bovine serum albumin (BSA) and 0.1% sodium azide. Sections were washed in PBS and then successively incubated with biotinylated goat antiserum to rabbit IgG (Vector Laboratories, Burlingame, USA) diluted 1:100 in PBS:BSA and freshly prepared avidin-biotinylated complex (Vectastain, Vector Laboratories, Burlingame, USA) for 30 and 60 min respectively. Peroxidase activity was revealed using the glucose oxidase diaminobenzidine, with nickel enhancement method. Slides were dehydrated, cleared and mounted in Pertex Mounting Media (CellPath, Hemel Hempstead, UK). As a control, serial sections were stained as above, but either omitting primary antiserum or replacing it with pre-immune serum.

2.7 Antiserum to inducible nitric oxide synthase peptide

A polyclonal antiserum was raised in rabbits to a synthetic 25-residue peptide, based upon the deduced amino acid sequences of cDNA encoding the human hepatic inducible NOS. The peptide sequence was QNESPQPLVETGKKSPESLVKLDAT-C corresponding to amino acids 53–77 of the deduced amino acid sequence, with a cysteine residue at the C-terminus to assist coupling to the carrier. The peptide was coupled to maleimide-activated key-hole limpet haemocyanin (Pierce Chemical, USA) and used to immunise rabbits. To confirm localisation of iNOS staining, adjacent sections were immunostained with an antibody against human smooth muscle α-actin (Sigma Chemical, UK). It had previously been shown that this antibody reacted with a band of approximately 130 kDa (the known molecular weight for iNOS) in a Western blot of a crude homogenate of atherosclerotic human aorta [23].

2.8 Enzyme activity assay

NOS activity was measured, as previously described, by the ability of tissue homogenates to convert [3H]-l-arginine to [3H]-l-citrulline, the co-product of nitric oxide (NO) formation [28]. After stimulation of the vessel segments for 0, 24 or 48 h, they were frozen and stored in liquid nitrogen until NO synthase activity was measured. Tissues (0.035–0.1 g wet weight) were homogenized on ice in 500 μl of Tris buffer (50 mmol/l, pH 7.4) containing phenylmethylsulphonyl fluoride (1 mmol/l). Aliquots of these tissue homogenates were then incubated in the presence of l-arginine/[3H]-l-arginine (10 μmol/l, 150 000 disintegrations per min/tube), NADPH (1 mmol/l), tetrahydrobiopterin (5 μmol/l), calmodulin (300 U/ml) and calcium (2 mmol/l). The specificity of the assay was increased by including l-valine (50 mmol/l), which inhibits the conversion of l-arginine to l-citrulline by arginase [29]. In separate experiments designed to compare the contribution of calcium-dependent and calcium-independent isoforms of NOS, calcium was omitted from incubations and replaced with EGTA (1 mmol/l). All reactions were continued for 30 min at room temperature and stopped by the addition of ice cold HEPES (20 mmol/l, pH 5.4). The newly formed l-citrulline was separated from l-arginine by passing over dowex columns (0.5 ml), and the eluted material was measured using a Beckman scintillation counter. The protein content of each sample was assayed by using the Bradford assay.

2.9 Determination of nitrite

Nitrite concentrations were measured using a highly sensitive chemiluminescence nitric oxide analyser which has been shown to be linear for the detection of nitric oxide over the range 10–1000 pmoles (Sievers model 270, CO, USA). Samples were defrosted for 30 min, then a 100 μl aliquot of each sample was injected into a vessel containing 1% w/v potassium iodide in glacial acetic acid which was purged continuously with nitrogen. Under these conditions the nitrite in the sample is reduced to nitric oxide gas. The nitric oxide is then carried on the stream of nitrogen into the detector and a chemiluminescent signal emitted due to the reaction of the nitric oxide gas with ozone (which is generated by the analyser). Experimental samples were compared with a range of standard concentrations of sodium nitrite.

In a preliminary series of experiments accumulation of nitrite was compared to that of nitrate in response to stimulation of segments of internal mammary artery with cytokines and LPS. Nitrate was measured using the same chemiluminescence technique except that the reducing solution consisted of 100 mmol/l VCl3 in 1 mol/l HCl, which was heated to 95°C. The gas given off was passed through a 1 mol/l NaOH trap to remove any acidic vapours prior to detection of nitric oxide by the analyser. Calculation of increases in nitrite and nitrate concentrations demonstrated that nitrite gave a 1266±363% (n=4) rise above control values after 48 h incubation, while nitrate only rose by 208±16% (n=4) above control. While measurement of nitrite will only reflect an undefined proportion of nitric oxide released by each vessel segment, it is a more sensitive index of nitric oxide release than measurement of nitrate in this system.

2.10 Data analysis and statistics

Values in the results are means±standard error of the mean (S.E.M). Amounts of nitrite detected by chemiluminescence were normalised to the wet weight of tissue, while enzyme activity experiments were expressed per mg of protein. Values for n refer to the number of patients from which samples of internal mammary artery or aorta were obtained. Statistical analysis was performed using a one way analysis of variance test (ANOVA) followed by t-tests for equal or unequal variance, as appropriate.

3 Results

3.1 Characterisation of iNOS induction

3.1.1 PCR analysis

Vessel samples that had been stimulated with the cytomix for 48 h showed the presence of mRNA, as assessed by the generation of cDNA by the primers synthesized specific to human hepatocyte iNOS (Fig. 1a). This band was absent in the control and cycloheximide treated segments. All vessel segments showed the presence of a band for the house-keeping gene GAPDH (Fig. 1b)

Fig. 1

PCR amplification of (A) iNOS (B) GAPDH from cDNA prepared from vessel segments of IMA. Segments were incubated with tissue culture medium alone (lane 2); cytomix (lane 3); cytomix+l-NMMA (100 μmol/l) (lane 4); cytomix+cycloheximide (1 μmol/l) (lane 5). Lane 1 contains the amplification product when reverse transcriptase enzyme was omitted from the cDNA synthesis reaction (PCR negative control). Lane 6, (A) and (B), contains positive control cDNA from smooth muscle cells incubated with cytomix amplified for iNOS and GAPDH, respectively. Markers (M) were 100 bp DNA ladder (Life Technologies, Paisley, Scotland). The cytomix contained LPS (10 μg/ml), IL-1β (5 U/ml), TNF-α (10 U/ml), and IFN-γ (200 U/ml). Figures are representative of n=4 individual experiments.

3.1.2 Immunocytochemistry

Segments of IMA that had been stimulated for 48 h with cytomix, and then fixed, showed positive staining with the antibody against the human hepatocyte iNOS (Fig. 2a–d). This staining was present in the smooth muscle cells throughout the media of the vessel wall, as judged by a similar distribution of staining for smooth muscle α-actin (data not shown). A small amount of staining was evident in some endothelial cells (Fig. 2b) but was absent in control segments (Fig. 2a), and those that had been co-incubated with cycloheximide (Fig. 2d). Sections from which the primary antiserum had been omitted or when it was replaced with pre-immune serum also showed no staining. In contrast, those segments that had been incubated with l-NMMA retained positive staining which was comparable to the stimulated vessel segments (Fig. 2c). All the specimens that were examined appeared healthy with no sign of atherosclerosis or inflammatory infiltration by blood elements.

Fig. 2

Photomicrographs (×50) showing staining for the antibody raised against human iNOS. The IMA segment exposed to control conditions (A) shows an absence of any positive staining, while the segment stimulated with cytomix (B) shows a positive staining throughout the medial smooth muscle cells. Co-incubation of l-NMMA (C) failed to inhibit positive staining, while incubation with cycloheximide (D) inhibited the staining. M=the media of the vessel wall and L↑= the direction of the lumen. Figures are representative of n=4 individual experiments.

3.1.3 Nitric oxide synthase activity

NOS activity could be detected in segments of IMA which had been stimulated with cytomix for 24 and 48 h (Fig. 3). This activity was unaffected by removal of calcium, but could be inhibited by co-incubation with 100 μmol/l l-nitro-arginine-methyl-ester (l-NAME). Samples that had not been stimulated with cytomix showed no significant nitric oxide synthase activity after either 24 or 48 h incubation (Fig. 4).

Fig. 4

Nitrite accumulation in tissue culture medium following 12, 24, 36 and 48 h incubation of segments of IMA with either vehicle (open bars), cytomix (filled bars), cytomix and l-NMMA (100 μmol/l) (hatched bars) or cytomix and cycloheximide (1 μmol/l) (cross-hatched bars). The cytomix contained LPS (10 μg/ml), IL-1β (5 U/ml), TNF-α (10 U/ml), and IFN-γ (200 U/ml) (*=p<0.05, **=p<0.001 compared to control; n=8)

Fig. 3

NOS activity measured in homogenates of IMA after 0, 24, and 48 h of exposure to either vehicle (control samples) or cytomix (stimulated samples). NOS activity was determined in each group in the presence of Ca2+(open bars), the absence of Ca2+ (filled bars) or in the presence of Ca2+ and l-NAME (100 μmol/l) (cross-hatched bars). The cytomix contained LPS (10 μg/ml), IL-1β (5 U/ml), TNF-α (10 U/ml), and IFN-γ (200 U/ml). (*=p<0.05 compared to activity at time 0; n=4).

3.1.4 Chemiluminescent detection of nitrite

Samples of tissue culture medium removed from vessel segments of IMA stimulated with LPS and all 3 cytokines in combination showed a progressive increase in the amount of nitrite with time. Aliquots taken after 36 and 48 h incubation contained significantly higher levels of nitrite in stimulated samples as compared to control (Fig. 4). The addition of either l-NMMA or cycloheximide abolished the increase in levels of nitrite in the media at both 36 and 48 h.

Removal of the endothelium had no significant effect on the accumulation of nitrite. Vessel segments with endothelial cells present gave an increase in nitrite of 1766±747% above control, while those from which the endothelium had been removed yielded an increase in nitrite levels of 1827±935% after 48 h incubation with the cytomix. Exclusion of serum from the tissue culture medium inhibited the increased levels of nitrite in segments of IMA stimulated with cytomix. The percentage increase in nitrite released into the tissue culture medium after 48 h dropped from 2381±888% in the presence of serum to 96±24% in the absence of serum (p<0.05).

3.2 Cytokine and LPS interactions

3.2.1 Interactions between IFN-γ, IL-1β, TNF-α and LPS

Only IFN-γ alone was capable of inducing a rise in nitrite concentration above control levels after 48 h incubation with segments of aorta (p<0.05), but this effect was significantly less than that of the cytomix (p<0.05) (Fig. 5). Both IL-1β and TNF-α gave no significant rise in the levels of nitrite over the same time period (Fig. 5). When IFN-γ was co-incubated with either IL-1β, TNF-α or LPS for 48 h there was an enhancement in nitrite levels, which was not significantly different from the effect of all four agents together (Fig. 5). Combinations of IL-1β and TNF-α, IL-1β and LPS or TNF-α and LPS were identical to the effect of these agents alone and to control values.

Fig. 5

Nitrite accumulation from segments of aorta following 48 h stimulation by different combinations of cytokines and LPS. The cytokines were present at the following concentration: IL-1β (5 U/ml), TNF-α (10 U/ml), IFN-γ (200 U/ml), and LPS (10 μl/ml). The cytomix contained all four agents together, while control represents medium from a vessel segment incubated for 48 h in the absence of any of the cytokines or LPS. (*=p<0.05 compared to control, †=p<0.05 compared to cytomix, n=5).

4 Discussion

We have demonstrated the capacity of human vascular tissue to produce nitric oxide as a result of induction of iNOS in response to cytokines and LPS. This release is associated with expression of iNOS mRNA, enhanced activity of a calcium-independent NOS enzyme and expression of iNOS protein mainly in the smooth muscle cells of the media. In addition, we have demonstrated a central requirement for IFN-γ in the induction of iNOS in human blood vessels, and an interaction between this and other cytokines.

Our results extend findings in human IMA that suggest iNOS may mediate changes in vascular reactivity in response to cytokines and LPS [21, 22]. LPS has been shown to induce depressions in phenylepherine-induced contractions which are inhibitable by l-NAME and methylene blue [22]. However, injection of cytokines into dorsal hand veins has been shown to induce changes in vasomotor tone which are not attributable to iNOS induction [18, 20]. Similarly, in cultured human umbilical vein endothelial cells, pro-inflammatory cytokines enhance NOS activity via up-regulation of GTP cyclohydrolase I, the rate limiting enzyme for tetrahydrobiopterin synthesis, rather than by iNOS induction [19]. We have clearly demonstrated that increased nitrite production in response to cytokines and LPS stimulation is due to expression and function of iNOS. In addition, we have shown that this effect is not dependent upon the endothelium, but occurs mainly due to expression of the enzyme in the media of the vessel wall.

Induction of iNOS is dependent upon de-novo protein synthesis, which explains the inhibitory effect of cycloheximide on the expression of mRNA. The requirement for de-novo synthesis of transcription factors during induction may explain this effect. Potential DNA regulatory sequences within the promoter regions of both the mouse and human iNOS genes have been suggested and include NF-κB, AP-1, NF-IL-6, TNF and interferon response elements [30–32]. The interferon regulatory factor-1 has been shown to play an essential role in the induction of the iNOS gene in murine macrophages [33]. The gene for interferon regulatory factor-1 is itself induced by TNF-α and IL-1β as well as interferons [34, 35]. The requirement of multiple cytokines for iNOS induction in human smooth muscle suggests that a number of transcription factors are required for full activation of the iNOS gene, and that IFN-γ is obligatory. Further studies are required to examine the molecular mechanisms that regulate iNOS induction in human smooth muscle cells, and how these relate to the interaction between IFN-γ and the other agents that can amplify its effect.

Maximal induction of iNOS appears to rely on an interaction involving two or more agents, with IFN-γ playing a central role. It has been shown that IFN-γ can give rise to nitrite generation by human monocytes/macrophages, an effect which is augmented by TNF-α [36]. Other studies have also demonstrated that synergy exists between particular combinations of cytokines in the induction of iNOS [37]. The requirement of serum for iNOS induction suggests that LPS may play an important role since serum may be providing both an LPS-binding protein and the soluble LPS receptor (CD14). Further studies are required to examine whether this mechanism can contribute to nitrite production in arterial tissue. There is evidence to suggest that there may be a difference in the action of inflammatory mediators in venous compared to arterial tissue [22, 38]. Thorin-Trescases and colleagues reported that cytokines induced depression of vessel wall contraction in the saphenous vein, but these could not be blocked by l-NMMA [22]. The significance and rationale for choosing the IMA and aorta for this study was to use them to examine the capacity, as well as the mechanism involved, of human arteries to express iNOS in general.

Observations made with the different cytokine interactions relied on the measurement of nitrite levels accumulated in the bathing medium of vessel segments. This proved to be a reliable index of nitric oxide production and iNOS induction in our system. Experiments utilising PCR, immunocytochemistry and the enzyme activity assay always correlated with the increases observed in nitrite concentrations. The precise time-course of induction for iNOS in human samples is difficult to judge from the present experiments. Accumulation of nitrite in the bathing medium acts only as an indirect index of nitric oxide release. Our experiments only measure undefined percentages of the total nitric oxide released and are not intended to reflect the total nitric oxide produced by each vessel segment. Further studies are required using a more sensitive marker of nitric oxide production such as accumulation of cGMP or production of mRNA for iNOS at a range of time points.

The contribution of iNOS to inflammatory responses has recently been more clearly defined using transgenic animals. It has been shown that mice deficient in the iNOS gene are resistant to the lethal effects of endotoxin-induced hypotension, and have impaired tumoricidal and bacteriocidal capabilities rendering them more sensitive to micro-organisms such as Listeria and Leishmania [39, 40]. The ability of human smooth muscle cells to express this enzyme demonstrates that human tissues have the capacity to release enhanced levels of nitric oxide. This mechanism may be important in pathophysiological conditions, such as septic shock, heart failure and transplant rejection, all of which have elevated levels of pro-inflammatory cytokines [41–43]. In models of transplant rejection, inhibition of iNOS by aminoguanidine can increase graft survival, suggesting a role for nitric oxide in this process [14]. Several studies have also suggested a role for elevated production of nitric oxide in heart failure [10, 11, 44–48]. Katz and colleagues observed an elevated level of TNF-α in patients with severe heart failure and found that this correlated with an enhanced capacity of acetylcholine to increase forearm blood flow [44]. In addition, TNF-α staining has been co-localised with areas of iNOS immunoreactivity in hearts of patients with dilated cardiomyopathy [45]. Activity of the iNOS enzyme has also been detected in the hearts of patients with dilated and postpartum cardiomyopathy as well as myocarditis [46, 47]. It has also been shown that l-NMMA can restore the positive inotropic response to β-adrenergic agonists in patients with left ventricular dysfunction [48], and that administration of l-NMMA can be of benefit to patients with septic shock [49].

Our study demonstrates induction of iNOS in human blood vessels by measuring the product of NO production, the presence of mRNA as well as the activity and localisation of the enzyme. In addition, we have provided evidence for how different cytokines may interact in induction of the enzyme. Documentation of the conditions required to induce iNOS in human vascular tissue will allow future studies to examine the regulation of this important pathway in the human vasculature. It is hoped that these findings will help in further understanding the role of iNOS and proinflammatory cytokines in different disease processes, and possibly lead to new avenues of treatment.


View Abstract