Nitroxyl anion (HNO) donors are currently being assessed for their therapeutic utility in several cardiovascular disorders including heart failure. Here, we examine their effect on factors that precede atherosclerosis including endothelial cell and monocyte activation, leucocyte adhesion to the endothelium and macrophage polarization. Similar to the NO donor glyceryl trinitrate (GTN), the HNO donors Angeli's salt (AS) and isopropylamine NONOate (IPA/NO) decreased leucocyte adhesion to activated human umbilical vein endothelial cells (HUVECs) and mouse isolated aorta. This reduction in adhesion was accompanied by a reduction in intercellular adhesion molecule-1 (ICAM-1) and the cytokines monocyte chemoattractant protein 1 (MCP-1) and interleukin 6 (IL-6) which was inhibitor of nuclear factor κB (NFκB) α (IκBα)- and subsequently NFκB-dependent. Intriguingly, the effects of AS on leucocyte adhesion, like those on vasodilation, were found to not be susceptible to pharmacological tolerance, unlike those observed with GTN. As well, HNO reduces monocyte activation and promotes polarization of M2 macrophages. Taken together, our data demonstrate that HNO donors can reduce factors that are associated with and which precede atherosclerosis and may thus be useful therapeutically. Furthermore, since the effects of the HNO donors were not subject to tolerance, this confers an additional advantage over NO donors.
HNO donors are currently being assessed clinically for their therapeutic utility in cardiovascular disorders such as heart failure. Therefore, we examined their effect on factors that precede atherosclerosis including endothelial cell and monocyte activation, leucocyte adhesion to the endothelium and macrophage polarization.
Similar to the NO donor GTN the HNO donors AS and IPA/NO decreased leucocyte adhesion to activated HUVECs and mouse isolated aorta. They also reduced monocyte activation and induced macrophage polarization into the anti-inflammatory M2 phenotype. Unlike GTN, the effects of AS on leucocyte adhesion, like those on vasodilation, were found to not be susceptible to pharmacological tolerance.
Taken together, our data demonstrate that HNO donors can reduce factors that are associated with and which precede atherosclerosis and that they are a promising prospect for use as a novel treatment for the chronic inflammation seen in cardiovascular diseases.
The one-electron reduced and protonated form of NO•, nitroxyl anion (HNO), is now recognized as a novel entity with distinct pharmacology and therapeutic advantages over its redox sibling [1,2]. Recently, the HNO donor CXL-1020 was reported to improve myocardial function without altering heart rate in patients with acute decompensated heart failure , supporting the notion that this novel class of drug has clinical potential. Indeed, numerous vasoprotective properties of HNO have been reported, including that it induces vasodilatation in rodent  and human vessels , inhibits platelet aggregation  and limits vascular smooth muscle cell (VSMC) proliferation  and neointimal hyperplasia . It is the ability of HNO to directly oxidize thiols which most distinguishes its actions from NO•. This is particularly evident in the heart, where the benefits of HNO in patients with heart failure are thought to be due their ability to activate cardiac thiol-containing proteins such as the sarcoplasmic ryanodine receptors  and sarcoplasmic/endoplasmic reticulum Ca2+-ATPase (SERCA)  to enhance Ca2+ cycling and increase myocardial contractility. Critically, HNO is resistant to scavenging by superoxide  and we have shown that HNO itself can limit reactive oxygen species (ROS) generation via the suppression of vascular and macrophage NAPDH oxidase 2 (NOX2) oxidase activity , an action independent of soluble guanylate cyclase (sGC)/cGMP signalling. These vasoprotective actions of HNO are thus preserved under disease conditions associated with oxidative stress and compromised NO• signalling .
An early feature of atherogenesis is endothelial activation, observed as an increased expression of endothelial adhesion molecules leading to monocyte–endothelial cell adhesion and transmigration of these monocytes into the subendothelial space. These monocytes differentiate into macrophages and, depending on the extracellular milieu, have been described to either polarize into macrophages demonstrating pro-inflammatory phenotype (loosely termed M1 macrophages) or those demonstrating a pro-fibrotic anti-inflammatory phenotype (M2 macrophages). Interestingly, although the HNO donor nitroxocyclohexyl acetate has previously been shown to block leukaemia inhibitory factor (LIF)-induced increases in intercellular adhesion molecule 1 (ICAM-1) expression via the inhibition of signal transducer and activator of transcription 3 (STAT3) phosphorylation , the HNO donor Angeli's salt (AS) has been reported to negate dexamethasone-induced neutrophil migration and ICAM-1 expression in a renal model of ischaemia .
In the present study, we explore the effect of HNO donors (compared with the NO• donor glyceryl trinitrate (GTN)) on monocyte–endothelial cell adhesion in real time and under physiological conditions, as well as the cellular mechanisms by which these effects occur. The influence of HNO donors on monocytic activation as well as on macrophage polarization was also examined. We show that HNO donors demonstrate cardio-protective effects beyond those observed with vascular function and myocardial contractility.
MATERIALS AND METHODS
All experiments conform to the National Health and Medical Research Council Animal Welfare Committee guidelines and were approved by the Alfred Medical Research and Education Precinct (AMREP) ethics committee (E/0946/2010/B). All animals were housed on a 12 h light/12 h dark cycle with food and water provided ad libitum. Mice were killed for tissue collection via CO2 asphyxiation. Aortae were harvested from 8- to 10-week-old male C57/Bl6 mice, cleared of connective tissue and were either untreated or treated as detailed below for 4 h.
Human umbilical vein endothelial cells (HUVECs, Lonza; three separate batches) were cultured to 80% confluency in endothelial basal medium-2 (EBM-2, Lonza) with endothelial growth medium supplements (EGM, Lonza) and 100 units/ml antibiotic/antimycotic (Life Technologies) and were used at passages 2–6. Thp-1 monocytes (A.T.C.C.) were cultured in RPMI 1640 medium (Life Technologies) with 10% fetal bovine serum. All HUVECs and Thp-1 cells were maintained at 37°C in 5% CO2.
Treatment protocols for HUVECs and excised aortae
To examine the anti-inflammatory potential of the nitroxyl donors AS or isopropylamine NONOate (IPA/NO) compared with GTN, HUVECs or aortae were incubated with either medium or Krebs solution alone, in the presence of the inflammatory stimulus tumour necrosis factor α (TNFα; 10 ng/ml, R&D Systems) and/or graded concentrations of AS or GTN (0.1 μM, 1 μM or 10 μM) or IPA/NO (10 μM) for 4 h. AS, IPA/NO and GTN treatment was administered at 0 h and again at 2 h of the incubation period.
To examine the intracellular signalling pathway, HUVECs or aortae were incubated with combinations of TNFα, GTN (10 μM), AS (10 μM), IPA/NO (10 μM) or the sGC stimulator BAY 41-2272 (10 μM) in the presence of the sGC inhibitor 1H-[1,2,3]oxadiazole[4,3-a]quinoxaline-1-one (ODQ, 10 μM), the NO• scavenger hydroxocobalamin (HXC, 100 μM) or the HNO scavenger L-cysteine (3 mM). ODQ, HXC and L-cysteine were added 15 min before TNFα, AS and GTN with AS and GTN re-added after 2 h of incubation. To examine whether the anti-inflammatory effects of AS or GTN were subject to tolerance, an additional set of aortae or cultured cells were treated with AS (10 μM) or GTN (10 μM) for 1 or 4 h respectively, washed with Krebs solution or PBS for 1 h and treated for a further 4 h with TNFα, AS or GTN as described above.
At the end of each treatment period, cells and vessels were collected and examined for inflammatory cytokine gene expression using real-time PCR, protein expression using antibody-directed flow cytometry and functional adhesion assays as detailed below.
Static adhesion assay
The HUVEC monolayer was cultured on a glass coverslip, treated as detailed above and following the 4 h treatment period, 3×105 Thp-1 monocytes were added to the incubation for a further 2 h. At the conclusion of the 2 h treatment period, the glass cover-slips were removed, washed to remove any unbound monocytes, fixed in 4% formalin, mounted and imaged using phase-contrast microscopy (×4 magnification, Olympus FSX100). The number of adhered monocytes was quantified in four fields of view per slide using ImageJ software (version 1.44p, NIH). Adhesion is presented as the percentage of TNFα-induced adhesion for each independent assay. Each ‘n’ represents the average of three triplicate slides per assay.
Quantification of endothelial inflammation in an intact vessel
Human blood was obtained via venopuncture from healthy volunteers who provided informed written consent. All procedures involving human were approved by the local Alfred Human Research and Ethics Unit (HREC, reference number 397-09) complying with the principles of the Declaration of Helsinki (2013) of the World Medical Association. Treated aortae were mounted on a vessel chamber  and perfused with whole human blood (0.12 ml/min) which was treated with Vybrant DiI (1:1000 dilution, Lonza) to fluorescently label leucocytes. Leucocyte–endothelial cell interactions were imaged using a Zeiss Discovery V.20 microscope (Zeiss) coupled to a high-definition camera (Hamamatsu) and AxioVision software (release 220.127.116.11, Zeiss). Fifteen second recordings of at least two fields of view were taken at each time point and adhered/stationary leucocytes were counted and averaged.
Following treatment, HUVECs were harvested and stained with anti-ICAM-1 (1:200 dilution, BD Biosciences), anti-monocyte chemoattractant protein 1 (MCP-1) (1:200 dilution, BD Biosciences), anti-interleukin (IL)-6 (1:200 dilution, BD Biosciences) or anti-inhibitor of nuclear factor κB (NF-κB) α (IκBα) (1:200 dilution, BD Biosciences). Protein expression was quantified using fluorescence-directed flow cytometry on the FACSCalibur II flow cytometer (BD Biosciences).
Gene expression analysis
RNA was extracted from treated HUVECs and aortae using TRIzol reagent (Life Technologies) as previously described [15,16]. RNA (1 μg) was DNAse-treated and reverse-transcribed into cDNA. Gene expression for inflammatory cytokines ICAM-1, MCP-1 and IL-6 (Geneworks) was assessed using real-time quantitative PCR (ABI Prism 7500, PerkinElmer, PE Biosystems) with SYBR Green chemistry (Roche Diagnostics). Gene expression was quantified using the ΔCT method as previously described .
Intracellular NFκB staining
HUVECs were cultured on glass coverslips and were treated as described above for 15 min. Following treatment, cells were fixed (10% formalin; Sigma–Aldrich), blocked with 0.3% Triton X-100 (Sigma–Aldrich) and 5% normal rabbit serum (Vector Laboratories), incubated with anti-NFκB-p65 (1:50, Cell Signaling Technology) for 24 h at 4°C and subsequently with a secondary antibody conjugated with Alexa Fluor 546 (1:750, Life Technologies) for 1 h. Cells were mounted in ProLong® Gold antifade reagent with DAPI for nuclear staining (Life Technologies) and were imaged using fluorescence microscopy (0.75 μm z stack sections, Nikon AR1 Confocal Microscope). DAPI-positive staining determined the nuclear area mask for each z stack which was used to quantify NFκB-positive staining in the nucleus and cytoplasm. NFκB staining was quantified as the fluorescence intensity/area for each z stack slice and is presented as the average fluorescence intensity per cell.
Fresh whole human blood was obtained via venopuncture from healthy volunteers and collected in sodium citrate (13 mM, Merck). Whole blood (200 μl) was pre-treated with AS (10 μM) or AS+ODQ (10 μM) or left untreated for 15 min at 37°C, prior to staining with anti-CD14 (1:100 dilution, BD Biosciences) and anti-CD11b (1:100 dilution, BD Biosciences) and activation with phorbol myristate acetate (PMA; 1 μM, Sigma–Aldrich) for 15 min at 37°C. Red blood cells were subsequently lysed and the fluorescence of the activation marker CD11b was determined in CD14+ monocytes using flow cytometry (FACSCalibur II flow cytometer, BD Biosciences).
Macrophage differentiation and polarization
Human monocytes were isolated from 60 ml of buffy coats (Australian Red Cross Blood Bank in accordance with Material Supply Agreement 13-05VIC-12) using magnetic-activated cell sorting (MACS), pan monocyte negative selection kit and LS MACS columns (Miltenyi Biotec). Isolated monocytes were differentiated into macrophages using a Complete RPMI 1640 medium (Life Technologies) and macrophage colony-stimulating factor (M-CSF) (100 ng/ml, R&D Systems) for 6 days. Macrophages were then cultured in medium alone (unpolarized, M0) or polarized into either an M1 phenotype [with lipopolysaccharide (LPS; 100 ng/ml)+interferon-γ (IFNγ; 20 ng/ml, R&D Systems)], or an M2 phenotype [with IL-4 (20 ng/ml, R&D Systems)] in the presence or absence of AS (10 μM) or AS+ODQ (10 μM). Stock solutions of AS (0.01 M) and subsequent dilutions were prepared in Krebs solution immediately prior to use.
The effect of AS on macrophage phenotype was assessed via mRNA gene expression and flow cytometric analysis of cell-surface marker expression. For gene expression experiments, macrophages received 6 h of polarization and treatment prior to harvesting in TRIzol reagent (Life Technologies). Concurrently, to examine protein expression, a separate pool of macrophages received polarization and treatment for 24 h prior to harvesting and staining with anti-hCD64 (1:50 dilution, R&D Systems), anti-hCD192 (1:100 dilution, BD Biosciences), anti-hCD200R (1:50 dilution, R&D Systems) and anti-CD206 (1:100 dilution, BD Biosciences) for flow cytometry experiments. AS was replenished three times over the 24 h treatment period. Expression of these surface markers was detected using fluorescence-directed flow cytometry (FACSCalibur II Flow Cytometer, BD Biosciences). Protein expression is presented as fold change from M0.
All results are expressed as means±S.E.M. For cell culture experiments each ‘n’ represents independent experiments performed in triplicate and for animal experiments each ‘n’ represents a separate animal. Statistical significance was accepted at P<0.05. All statistical analyses were performed using GraphPad Prism Version 6.0. Student's unpaired t test, one-way or two-way ANOVA was used to compare two or between three or more experimental groups respectively. A Tukey post-hoc analysis was performed when the ANOVA indicated statistical differences.
HNO reduces TNFα-induced leucocyte adhesion in HUVECs
The effects of HNO on activated endothelial cells was determined and compared with GTN using a static adhesion assay assessing the adhesion of Thp-1 monocytes to HUVECs. TNFα (10 ng/ml) caused an increase in leucocyte adhesion (control 286±59 compared with TNF-α 2036±213 cells/field; n=13; P<0.001). Both the NO• donor GTN (0.1–10 μM) and the HNO donor AS (0.1–10 μM) caused a concentration-dependent decrease in leucocyte adhesion to TNFα-stimulated HUVECs (P<0.05; Figures 1A–1D). The NO• scavenger HXC prevented the GTN-induced reduction in leucocyte adhesion, but had no effect on AS or the HNO donor IPA/NO-induced adhesion (Figures 1E–1G). Alternatively, the sGC inhibitor ODQ reduced the effect of GTN, AS and IPA/NO on leucocyte adhesion (Figures 1E–1G). The NO-independent sGC stimulator BAY 41-2272 (10 μM) also reduced monocyte adhesion to 74±4.4% of TNFα-induced adhesion (P<0.05; n=6).
The NO donor GTN and the HNO donors AS and IPA/NO reduce TNFα-induced leucocyte adhesion in HUVECs
HNO reduces adhesion molecule and cytokine level expression
We next sought to determine the effect of HNO (compared with GTN) on adhesion molecule and cytokine level expression. A 4 h incubation with TNFα increased ICAM-1, MCP-1 and IL-6 gene and protein expression in HUVECs (Figures 2A–2L). GTN had no effect on TNFα-induced increases in ICAM-1 but significantly reduced MCP-1 and IL-6 gene and protein expression (P<0.05; Figures 2A and 2B). AS, on the other hand, significantly reduced TNFα-induced ICAM-1, MCP-1 and IL-6 gene and protein expression. ODQ did not affect any of the GTN- or AS-induced reductions in adhesion molecule or cytokine levels (Figures 2A–2L).
GTN and AS reduce adhesion molecule and cytokine level expression
HNO induces IκBα expression which induces NFκB phosphorylation
The anti-inflammatory effects of NO• are reported to be dependent on its ability to inhibit the transcription factor NFκB and increase the expression of IκBα, the cytoplasmic inhibitor of NFκB . Activation of NFκB occurs after the phosphorylation and degradation of IκBα, allowing translocation of the unbound NFκB-p65 from the cytoplasm to the nucleus and transcription of adhesion molecule genes . Indeed, in TNFα-activated HUVECs both GTN and AS resulted in increased expression of IκBα protein expression as determined via flow cytometry (P<0.01; Figures 3A and 3B). This effect was abolished in the presence of HXC for GTN but not AS. Furthermore, using fluorescence microscopy, both GTN and AS reduced cytoplasmic and nuclear NFκB-p65 expression (P<0.01; Figures 3C–3G).
HNO and GTN induce IκBα expression and NFκB phosphorylation
HNO reduces monocyte activation, M1 inflammatory gene expression and promotes M2 polarization
We then sought to determine whether AS could reduce PMA-induced monocyte activation or influence macrophage polarization. PMA increased expression of the activation marker CD11b on CD14+ monocytes (P<0.001; Figure 4A). AS reduced the expression of CD11b, an effect which was not sGC-dependent. Macrophages were polarized to the M1 phenotype with LPS+IFNγ as evidenced by the increase in CD64 and CD192 surface expression (Figures 4B and 4C). Addition of AS had no effect on the expression of CD64 or CD192 (Figures 4B and 4C), indicating that HNO did not change polarization to the M1 phenotype. However, AS modulated the inflammatory gene expression profile of M1 macrophages, reducing TNFα gene expression (Figure 4D). Macrophages treated with IL-4 were polarized to the M2 phenotype as evidenced by the increased expression of CD200R and CD206 and co-incubation of AS during polarization increased CD200R and CD206 expression further (Figures 4E and 4F). The ability of AS to promote the M2 phenotype was unaffected by ODQ and was thus sGC-independent (Figures 4E and 4F). In support of these findings, mRNA gene expression profiles of CD206 and the anti-inflammatory gene scavenger receptor B1 (SR-B1) were also increased in macrophages treated with IL-4 (Figures 4G and 4H) compared with M0 and treatment with AS further increased gene expression.
HNO decreases monocyte activation, M1 inflammatory gene expression and promotes M2 polarization
HNO abolishes leucocyte adhesion and adhesion molecule and cytokine expression in mouse aorta
To see whether the anti-inflammatory effects observed in HUVECs translated to a functional effect in intact vessels we examined the influence of HNO on leucocyte adhesion and gene expression of ICAM-1 and IL-6 in mouse aorta. We show that a 4 h incubation with TNFα (10 ng/ml) led to an increase in leucocyte adhesion to the endothelium (Control; 4±1 compared with TNFα; 15±3 cells/field of view at 10 min; P<0.001; Figures 5A and 5B) in aortae isolated from C57BL6/J mice. Treatment with GTN (10 μM) significantly decreased TNFα-induced leucocyte adhesion compared with TNFα alone (P<0.001; Figure 5A). The effects of GTN were abolished by HXC but not ODQ, suggesting that the reduction in adhesion was via NO• but not dependent on sGC activation. Treatment with AS (10 μM) also attenuated the TNFα-stimulated increase in leucocyte adhesion (P<0.001; Figure 5B). Furthermore, the HNO scavenger L-cysteine and ODQ abolished the effects of AS, suggesting that the effects of AS were both HNO- and sGC-dependent. TNFα also increased mRNA expression of both ICAM-1 and IL-6 in mouse isolated aortae (Figures 5C–5F). Similar to the findings in HUVECs, GTN did not reduce ICAM-1 gene expression (Figure 5C) but AS did (Figure 5E), whereas both GTN and AS reduced IL-6 gene expression (Figures 5D and 5F). Consistent with the leucocyte adhesion findings, the effects of AS on both ICAM-1 and IL-6 were sGC-dependent, whereas the effects of GTN on IL-6 were not.
AS and GTN abolish leucocyte adhesion and adhesion molecule and cytokine expression in mouse aorta through sGC-dependent and sGC-independent mechanisms respectively
HNO-induced reduction in leucocyte adhesion is not subject to tolerance
The vasodilatory effects of GTN are subject to nitrate tolerance, where increased concentrations are required to produce the same effect . To examine whether the ability of GTN to reduce leucocyte adhesion is also subject to tolerance, both HUVECs and mouse aortae were pre-treated with GTN (both 10 μM), followed by a 1 h washout period before the re-addition of GTN. Concurrent studies were performed with AS. The reduction in TNFα-induced leucocyte adhesion by GTN was abolished in HUVECs and mouse aorta pre-treated with GTN (P<0.05; Figures 6A and 6B). The effect of AS on leucocyte adhesion, on the other hand, was unchanged in HUVECs or mouse aortae pre-treated with AS (Figures 6A and 6C) suggesting that the anti-inflammatory effects of HNO are resistant to tolerance.
GTN, but not AS, induces tolerance to leucocyte adhesion
We show for the first time that the HNO donor AS reduces adhesion molecule and cytokine expression by increasing IκBα, reducing NFκB translocation to the nucleus and decreasing leucocyte adhesion to HUVECs and mouse aortae. We also report that, similar to the reported vasodilatory actions of GTN, the effects of GTN on endothelial cell activation are subject to tolerance and those of AS are not. These findings confer additional cardio protective benefits of HNO donors beyond those previously described and above those observed with GTN.
The present study provides evidence that the HNO donors AS (10 μM) and IPA/NO (10 μM) were able to induce a decrease in endothelial cell–leucocyte adhesion in TNFα-stimulated HUVECs to a similar, if not greater, extent than the NO• donor GTN (10 μM). Furthermore, like NO•, HNO abolished leucocyte adhesion in TNFα-stimulated mouse aortae. In both HUVECs and mouse aortae, treatment with the NO scavenger HXC did not have any effect on the ability of HNO to reduce adhesion. Indeed, the ability of HXC to scavenge NO• and not HNO has been reported in several studies [20–22]. HXC was found to significantly diminish the anti-inflammatory properties of GTN in HUVECs and mouse aortae, suggesting that the responses to GTN are NO•-specific and that it is also unlikely that the effect of HNO was due to extracellular conversion of HNO into NO. Alternatively, the ability of AS to reduce leucocyte adhesion was abolished in the presence of L-cysteine. We and others have reported the sensitivity of HNO to thiols, such as L-cysteine [2,23,24], where in the presence of L-cysteine, vasorelaxation to HNO donors were significantly attenuated [25–27]. Conversely, L-cysteine has been shown to be ineffective in binding NO• [21,24,27]. Collectively, these findings suggest that the concentration-dependent reduction in monocyte adhesion by GTN and HNO was due to NO• and HNO respectively.
Atherosclerosis is an inflammatory disease, and an essential process of the inflammatory response is leucocyte transmigration through the endothelium to the site of injury to engulf lipids. This process, the leucocyte adhesion cascade, involves the capture, rolling, slow rolling, firm adhesion and transmigration of monocytes to an activated endothelium; disruption of any of these processes can significantly reduce leucocyte infiltration. The adhesion cascade is dependent upon the expression of adhesion molecules and chemokines such as ICAM-1, MCP-1 and IL-6. NO donors have been shown to modulate adhesion molecule and chemokine expression in endothelial cells, where they have been shown to reduce ICAM-1 , MCP-1  and IL-6 . In HUVECs, activation of the endothelium with TNFα increased expression of ICAM-1, MCP-1 and IL-6 but not P-selectin (results not shown), whereas in mouse aortae increases in ICAM-1 and IL-6 were also observed. Consistent with previous studies using different NO donors, treatment with GTN reduced MCP-1 and IL-6 gene and protein expression. However, GTN did not have any effect on ICAM-1 gene or protein expression in HUVECs or mouse aortae. Importantly, treatment with AS was able to inhibit the TNFα-induced expression of ICAM-1 and IL-6 in mouse aortae and ICAM-1, MCP-1 and IL-6 in HUVECs to levels similar to control in mouse aortae, suggesting additional protective effects of HNO compared with NO•. This is the first instance where the ability of HNO to reduce adhesion molecules and cytokines has been reported.
Central to adhesion molecule expression is the activity of the transcription factor NFκB. For NFκB to translocate to the nucleus, IκBα, the cytoplasmic inhibitor of NFκB, has to be phosphorylated and degraded. This degradation allows the translocation of the unbound NFκB to the nucleus, resulting in the transcription of adhesion molecule genes . Importantly, both endogenous NO  and NO• donors can inhibit the activation of NFκB by inducing the expression of IκBα  leading to a decrease in adhesion molecules such as ICAM-1 and vascular cell adhesion molecule-1 (VCAM-1) [32,33]. With this in mind, the effect of HNO on IκBα and NFκB expression was investigated. TNFα induced a decrease in IκBα and an increase in both cytoplasmic and nuclear NFκB expression, and both GTN and AS prevented the TNFα-induced decrease in IκBα and increase in NFκB.
NO• has been shown to exert its anti-inflammatory actions via both sGC/cGMP-dependent and -independent actions [28,34]. For example, the NO• donor DEA/NO and the sGC stimulator BAY 41-2272 decrease adhesion via an increase in cGMP and subsequently decrease the expression of the adhesion molecule P-selectin . Conversely, NO donors such as S-nitrosoglutathione and sodium nitroprusside can directly inhibit the activation of NFκB by inducing the expression of IκBα independently of sGC/cGMP signalling . The present study, using the sGC inhibitor ODQ, also demonstrates the ability of GTN to reduce leucocyte adhesion and adhesion molecule and cytokine expression via both sGC-dependent (HUVECs monocyte adhesion and IκBα expression) and sGC-independent (mouse aorta leucocyte adhesion, adhesion molecule expression and NFκB) mechanisms. Collectively these findings highlight the complexity surrounding the involvement of cGMP in the modulation of leucocyte adhesion and adhesion molecule expression by NO.
Like NO, HNO can target sGC to increase cGMP , but it also has a unique ability to directly modulate thiols and thiol-containing proteins . It is thought that many of the additional beneficial properties of HNO relate to this ability. The HNO-induced decrease in leucocyte adhesion in HUVECs and isolated mouse aorta was sensitive to ODQ and thus predominantly sGC/cGMP-dependent. This was also true in regard to NFκB expression, where the decrease in expression in the presence of AS was abolished by ODQ. This finding is supported by others that show that sGC activation can suppress the transcriptional activity of NFκB . Alternatively, the AS-induced reduction in adhesion molecule and cytokine levels in HUVECs was resistant to ODQ as was the AS-induced increase in IκBα expression. We hypothesize that HNO may directly target IκBα to decrease adhesion molecule expression. Indeed, IκBα contains a low-pH key cysteine residue at its activation loop at position 179, where modifications may affect function. However, further investigation of the nature of IκBα thiol modification by HNO is beyond the scope of the present study. More importantly, in a physiologically relevant setting (mouse aortae), we demonstrated that both the ability of AS to reduce leucocyte adhesion and expression of ICAM-1 and IL-6 was sGC-dependent. Collectively, these results suggest that HNO reduces endothelial inflammation, predominantly via sGC activation resulting in suppression of NFκB, but it appears likely that direct activation of IκBα is also involved (Figure 7).
Schematic representation of the proposed mechanism of action of HNO in endothelial cells stimulated with TNFα
Another important component of the leucocyte adhesion cascade is the activation of the monocytes themselves. Monocytes are activated by stimuli such as PMA and LPS that results in the expression of tissue factor (TF) and MAC-1 (CD11b) and the production of pro-inflammatory cytokines and chemokines. Once activated, monocytes play a crucial role in inflammation, particularly in conditions such as atherosclerosis, where they are involved in endothelial cell activation, the recruitment of leucocytes and the formation of unstable plaques. NO• is also known to decrease monocyte activation by reducing the expression of CD11b . Both anti-inflammatory and pro-inflammatory actions of NO• donors have been reported in monocytes. For instance, NO• donors have a biphasic effect on peripheral blood mononuclear cells, where low concentrations of NO• donors increased levels of IL-6 and higher concentrations decreased IL-6 levels . Interestingly, this effect was both cGMP- and NFκB-dependent. The effects of HNO on monocyte activation have not been reported. Therefore, the ability of HNO to modulate PMA-stimulated CD11b expression in human monocytes was investigated. Indeed, similar to the reported actions of NO, HNO was also able to induce a reduction in CD11b expression; however, unlike NO, it was unaffected by ODQ and is therefore not sGC-dependent.
With an inflammatory insult, activated monocytes transmigrate into the vessel wall where they differentiate into macrophages (M0). Depending on their microenvironment they have the ability to polarize into two subtypes, classically activated M1 or alternatively activated M2 macrophages. Macrophages are polarized into the M1 subtype by stimuli such as IFNγ, LPS and TNFα, and into the M2 subtype by IL-4, IL-13, glucocorticoids and hormones. Macrophages play a pivotal role in atherosclerosis and, although much work in this area is still warranted, it appears that macrophages of the M2 phenotype are found predominantly in stable plaques and M1 macrophages are dominant in rupture-prone areas of plaque [39,40]. Furthermore, treatments that reduce the progression of atherosclerosis can promote macrophages of the M2 subtype . AS increased the expression of the M2 subtype-specific markers CD200R and CD206 and the mRNA expression of CD206 and SR-B1. Alternatively, AS had no effect on the M1 subtype specific markers CD64 and CD192, but did decrease the gene expression of TNFα. Although the implications of these findings are difficult to fully appreciate at present, due to the dynamic nature of the literature surrounding macrophage biology, the current literature suggests that M2 macrophages dominate the regressing plaque, indicating an increase in the M2 phenotype is beneficial.
From a clinical perspective, NO• donors (i.e. GTN) are commonly prescribed in the treatment of angina, acute hypertensive crises and heart failure. However, the clinical utility of traditional NO• donors are limited due to their susceptibility to the development of nitrate tolerance (defined as diminished effectiveness with repeated use) and decreased vasodilatory and anti-platelet efficacy under conditions of oxidative stress . We have shown in several studies that the vasodilator effects of HNO are not susceptible to tolerance or cross-tolerance with GTN [2,5,42]. With these findings in mind we investigated whether the effects of GTN and AS on leucocyte adhesion were subject to tolerance. HUVECs and mouse aortae were pre-treated with GTN or AS, washed and then retreated with GTN and AS, and leucocyte adhesion was assessed. Similar to the vasodilatory tolerance development, GTN but not AS was subject to tolerance, whereby GTN was no longer effective in reducing leucocyte adhesion, whereas there was no difference in the effectiveness of AS. Tolerance to nitrates has been shown to be multifactorial and with GTN in particular, oxidative stress and the desensitization of the esterase activity of the enzyme mitochondrial dehydrogenase, which is crucial for the release of NO, is pivotal in its development . Since AS does not require enzymatic biotransformation, is able to suppress NADPH oxidase activity  and tolerance itself appears to be largely confined to nitrates, although extremely beneficial, it is not surprising that pre-treatment with AS does not induce tolerance.
The present study has advanced our knowledge in regard to the vasoprotective properties of HNO highlighting the ability of HNO to reduce endothelial and monocyte activation in the vasculature. To begin with, like GTN, the HNO donor AS has the ability to cause a concentration-dependent decrease in endothelial cell–leucocyte adhesion and adhesion molecule expression in HUVECs and mouse aortae and these effects are HNO-dependent. The actions of HNO appear to be mostly due to the activation of sGC, an increase in IκBα and a subsequent decrease in NFκB. Unlike the actions of GTN, which are subject to tolerance, the vasoprotective actions of HNO are not. Finally, HNO can reduce monocyte activation and increase the polarization of macrophages into the M2 phenotype. Given these findings and the fact that the HNO donor CXL-1020 has shown potential in the treatment of heart failure,  HNO donors are indeed a promising prospect for use as a novel treatment for the chronic inflammation seen in cardiovascular diseases.
Amanda Sampson, Jennifer Irvine, Waled Shihata, Danielle Michell, Natalie Lumsden, Chloe Lim, Olivier Huet and Karen Andrews performed the research and analysed the data. Karen Andrews, Amanda Sampson, Jennifer Irvine, Grant Drummond, Barbara Kemp-Harper and Jaye Chin-Dusting designed the research, interpreted the data and prepared and wrote the paper.
The present study was supported by the National Health and Medical Research Council of Australia [grant number APP10363652 (to J.C.D.)]; the National Heart Foundation of Australia [grant numbers PF 12M 6806 and PF 12M 6810 (to A.K.S. and J.C.I.)]; and the Victorian Government's OIS Program.
human umbilical vein endothelial cell
intercellular adhesion molecule-1
inhibitor of NFκB α; IL, interleukin
magnetic-activated cell sorting
monocyte chemoattractant protein 1
nuclear factor κB; ODQ, 1H-[1,2,3]oxadiazole[4,3-a]quinoxaline-1-one
phorbol myristate acetate
soluble guanylate cyclase
scavenger receptor B1
tumour necrosis factor α
These authors contributed equally to this work.