Participation of Peripheral Polymorphonuclear Leukocytes in the Oxidative Stress and Inflammation in Patients with Essential Hypertension

Abstract

Oxidative stress and inflammation have recently been linked to endothelial damage in essential hypertension (EH). Activated peripheral polymorphonuclear leukocytes (PMN) damage surrounding tissue by releasing reactive oxygen species (ROS) and proteolytic enzymes before self-necrosis. PMN necrosis further exacerbates inflammation and promotes chemotaxis and PMN recruitment. The number and properties of PMN from untreated EH patients is the focus of the present study. Oxidative stress was assessed by measuring the rate of superoxide anion release from separated, phorbol ester-stimulated PMN and the redox state of plasma glutathione. Inflammation was estimated indirectly by determining PMN number and their in vitro survival.

PMN from EH patients (n = 37) released superoxide anion faster (P < .0001) than those of normotensives (NC, n = 37), 17.7 ± 1.14 v 9.54 ± 0.51 nmol/10 min/10⁶cells. The redox state of glutathione was twofold higher in EH plasma (P < .02) indicating systemic oxidative stress. PMN survival in vitro correlates linearly with the rate of superoxide release (r² = 0.60, P < .02) and PMN count of EH patients, although in the normal range, were significantly higher (P < .0001), indicating necrosis and recruitment. Hypertensive plasma significantly reduced NC PMN viability, whereas normal plasma significantly increased EH PMN viability. What our studies show is that EH is accompanied by a primed state PMN that does not correlate with the levels of blood pressure. PMN priming in EH patients reflects an in vivo exposure to a constant stimulus ending in oxidative stress, increased self-necrosis, and cell recruitment. Oxidative stress and inflammation will result in endothelial damage and atherosclerosis in the long run.

Hypertension is a well established and independent risk factor for the development and progression of atherosclerosis.^1,2 The mechanisms that predispose hypertensive subjects to organ injury and atherosclerosis are multifactorial. Abnormalities of endothelium function and morphology appear to have a central role in the pathogenesis of hypertension-induced atherosclerosis.^3,–5 Among the mechanisms causing endothelial dysfunction that have recently been implicated in essential hypertension (EH) are oxidative stress and inflammation.^6,–8 Oxidative stress may result from either excessive production of reactive oxygen species (ROS) or from reduced antioxidant levels. In hypertensive patients decreased concentrations of antioxidants, such as vitamin E, vitamin C, and superoxide dismutase (SOD), have been demonstrated.^9,–11 Glutathione (GSH) is a ubiquitous intra- and extracellular antioxidant.^12,13 The plasma level of GSH may be a useful marker for the degree of oxidation injury.¹⁴ An increase in plasma oxidized glutathione (GSSG), produced by ROS, would reflect GSH consumption and a systemic oxidation stress,¹⁵

The peripheral polymorphonuclear leukocyte (PMN) is one of the main inflammatory cells. Once activated, PMN release ROS and mediators of proteolytic tissue degradation, contributing to oxidative stress, inflammation, and endothelial damage.^16,17 It is well established that activated or primed PMN are committing suicide and die by self-necrosis. PMN necrosis further adds to the inflammation, promotes chemotaxis and PMN recruitment. Some studies had demonstrated that increased leukocyte count is a risk factor for coronary artery disease.^18,19 Friedman et al²⁰ had found that the leukocytes count is in the highest quartile of white blood cells (WBC) in EH and can even serve as a simple cheap test to predict EH. Using animal models of hypertension, Schmid-Schönbein and colleagues observed that the activation and number of PMN increased. They postulated that PMN are involved in the development of organ injury and vascular complications.^21,22 Few recent studies have also noted that peripheral PMN are activated in EH patients, although the results are conflicting.^{10,11,23,–25}

In the light of these findings, this study evaluates the properties and number of peripheral PMN in EH patients possibly contributing to oxidative stress and to inflammation. Evaluation of oxidative stress was carried out by measuring superoxide release from PMN and the plasma redox state of glutathione. Evidence for the inflammatory process was correlated to PMN number and their necrosis, the latter being measured by PMN survival in vitro.

Materials and Methods

Subjects

Thirty-seven untreated EH patients with mild to moderate hypertension and 37 age- and sex-matched controls were enrolled in the study (Table 1). The selection of the control participants (NC) was based on a clinical examination with laboratory confirmation. The hypertensive patients had sitting diastolic blood pressure >90 mm Hg, sitting systolic blood pressure >140 mm Hg (the average of measurements made on three outpatient visits), and no clinical evidence of other systemic diseases or secondary causes of hypertension. Hypertension was uncomplicated in all patients with no evidence of target organ damage. All the subjects had normal fasting (>14 h) serum cholesterol and glucose levels with normal kidney and liver function (Table 1). Subjects with evidence of infection, receiving medication, or smoking were excluded. All subjects signed an informed consent form.

Blood withdrawal and Cell Count

After an overnight fast, 30 mL of blood were drawn from each subject and placed into heparinized tubes (50 U/mL blood). The blood sample was centrifuged for plasma and the isolation of the PMN. An additional blood sample was collected in EDTA tubes for total and differential white blood cell count (Coulter STKS, Miami, FL).

PMN and Plasma Separation

For isolation of PMN, the blood fractionation was done in a laminar flow hood, immediately after blood withdrawal, using the technique as described by Klebanoff and Clark.²⁶ Briefly, a 20-mL blood sample was mixed with 1:1 v/v dextran (T-500, Pharmacia Biotech, Upssala, Sweden) and left standing for 30 min. During this period, the majority of the red blood cells (RBC) settled on the bottom of the tube. The upper layer containing PMN was centrifuged for 5 min at 350 g at 10°C. The pellet was resuspended in cold Hank’s balanced salt solution (HBSS, Biological Industries, Beit Haemek, Israel) and overlayed on cold 1.077g/mL Histopaque (Sigma Chemical Co., St. Louis, MO). The tubes were centrifuged at 1000 g for 20 min at 10°C, resulting in a pellet containing both RBC and PMN. The pellet was resuspended with cold hypotonic (0.2%) NaCl for 25 sec after which an equal volume of hypertonic (1.6%) NaCl was added to restore isotonicity and centrifuged for 5 min at 350 g at 10°C. The lysis step was repeated if the cell pellet was red. The separated PMN (>98% pure) were resuspended in a minimal volume of HBSS, immediately counted, and diluted according to different experimental needs.

The remaining blood sample (10 to 20 mL) was separated for its plasma to be used in the incubation mixtures at a concentration of 25% v/v by dilution in HBSS.

Oxidative Stress

Determination of Superoxide Anion Release from Activated PMN

The release of superoxide anion (O₂⁻) from isolated PMN was measured to assess the oxidative metabolism of PMN. Superoxide release was assayed under basal (resting) conditions and after 0.32 × 10⁻⁷ mol/L phorbol 12-myristate 13-acetate stimulation (PMA; Sigma). The assay is based on SOD (Sigma) inhibitable reduction of ferricytochrome C (cyt. C; Sigma) to its ferrous form²⁷ and monitors changes in absorbance at 549 nm in the presence and absence of 0.02 mg/mL SOD. The changes in O₂⁻ release were monitored continuously for at least 90 min in resting cells and in PMA-stimulated cells for up to 50 min. The rate of O₂⁻ release was determined in the presence of 0.08 mmol/L cyt.C by 10⁶ PMN in 1 mL of HBSS. The changes were expressed as nanomoles per 10 min per 10⁶ cells, using the extinction coefficient (21.1 × 10⁴ mol/L/cm).²⁷

To eliminate the confounding effects of heparin on PMN (because heparin has been shown specifically to bind to neutrophils,²⁸ to modulate superoxide release,^29,30 and to induce apoptosis in these cells³¹), we also withdrew blood into EDTA before PMN separation in a randomly chosen group of 5 NC and 5 EH patients. The rate of superoxide release was determined and compared to the rate when the blood was withdrawn into heparinized tubes. The type of anticoagulant did not influence the rates of superoxide release from separated PMN in NC and EH patients.

Determination of Plasma Glutathione Levels

The determination of plasma glutathione using glutathione reductase was done according to Adams et al¹⁵ and Griffith.³² Two separate measurements were done: the total plasma glutathione level (oxidized and reduced glutathione) and the oxidized form (GSSG). The reduced glutathione (GSH) was calculated as the difference between total glutathione and the oxidized form. Total glutathione concentration {[GSH] + 2[GSSG]} was determined using 10 mmol/L DTNB (5,5′-dithio-bis-2-nitrobenzoic acid), after reduction of the GSSG by GSSG reductase (from baker’s yeast; Sigma). The resulting amount of GSH was determined according to a standard curve (0 to 2.5 nmoles GSSG), recorded in a spectrophotometer (Spectronic 1201, Milton Roy, Rochester, NY) at 412 nm and expressed as the change in absorbance per minute during the reduction with GSH reductase. To measure the oxidized glutathione (GSSG) in plasma, alkylation of SH groups was carried out with freshly prepared 10 mmol/L N-ethylmaleimide (NEM). To avoid the influence of NEM on the enzymatic reaction used to determine the GSSG concentration, the excess of NEM was removed by separation on Sep-Pak (C18) column (Sigma). To avoid spontaneous oxidation of GSH in plasma, the plasma was separated within 4 min of blood withdrawal, and acidified (1 mL plasma with 50 μL of 50% 5-sulfosalicylic acid).³³ At this point acidified plasma was frozen for a maximum of 3 weeks. Before the determination of GSH, the plasma was thawed and back-titrated to pH 7 with 0.2 mol/L NaOH. The redox state of the plasma expressed as the ratio [GSSG]/[GSH] provides a sensitive measure for oxidative stress.³⁴

Inflammation

In Vitro Survival of PMN: An Indirect Measure for Necrosis

Nonstimulated, separated PMN (10⁷ cells/mL) from NC and EH subjects were incubated for 90 min at 37°C in heparinized plasma (25% v/v dilution in HBSS), using plasma from the normotensive or hypertensive subjects. In previous studies³¹ we have shown that apoptosis under the identical in vitro conditions is 2% to 4%. PMN can die either by necrosis or apoptosis. Hence, one can infer that a reduction in the number of PMN >4% is attributable to necrosis. PMN from NC and EH subjects were incubated in autologous and heterologous plasma. Survival count was performed as mentioned above by Coulter counter. In each sample counted, viability was >97% by trypan blue (0.1% w/v) exclusion. The reduction of PMN number after incubation was expressed as the percent difference between the number of cells counted after 90 min of incubation and the cells counted at the beginning of incubation.

The Effect of SOD on Survival of PMA-Stimulated PMN

Once PMN are activated they release O₂⁻ and are committed to die. Therefore, it is possible that their death is attributable to O₂⁻ cytotoxicity. Thus, in the presence of SOD, which dismutates O₂⁻ and abolishes its cytotoxicity, one could anticipate that PMN survival would increase. The exposure of isolated stimulated PMN to SOD could be used to distinguish between the commitment of the stimulated PMN to die and the cytotoxic action of O₂⁻. PMA-stimulated PMN (10⁷ cells/mL) in the presence and absence of 0.02 mg/mL SOD were incubated in HBSS at room temperature for 5, 10, 15, 20, 30, and 50 min. PMN count was performed as mentioned above by Coulter counter. The time needed for a 50% reduction in the number of PMA-stimulated cells was defined as half-life of survival (Figure 3). This extrapolated half-life of survival for each sample was correlated with its matching rate of superoxide anion release by linear regression analysis.

Effect of superoxide dismutase (SOD) on phorbol 12-myristate 13-acetate (PMA)-stimulated peripheral polymorphonuclear leukocyte (PMN) survival. PMN (10⁷ cells/mL), in the presence and absence of 0.02 mg/mL SOD, stimulated with 0.32 × 10⁻⁷. PMA were incubated in Hank’s balanced salt solution at room temperature for 5, 10, 15, and 30 min. Survival counts were performed by Coulter counter at the time points mentioned. The time needed for 50% disintegration of the PMA-stimulated cells is defined as half-life of survival.

Statistical Analysis

Data are expressed as means ± SEM. Differences between the study parameters of the two groups were compared by unpaired and paired Student’s t test. The relationships between blood pressure and the study parameters were correlated by linear regression analysis. P < .05 was considered statistically significant.

Results

Study Population

Table 1 shows that NC and EH groups were similar with regard to age and sex distribution, blood cholesterol, blood glucose, serum creatinine, blood urea nitrogen, and hepatic transaminase levels. As expected, the systolic, diastolic, and mean arterial blood pressures were higher in EH than in NC subjects. Our EH patients had significantly greater WBC and PMN counts, as well as higher plasma alkaline phosphatase activity (ALP) levels, although all values fell within the accepted normal range (Table 1).

Oxidative Stress

Superoxide Release

Superoxide release from nonstimulated PMN from either NC or EH subjects was undetectable for at least 90 min. Superoxide release from PMA-stimulated PMN from EH patients was significantly greater than superoxide release of PMA-stimulated PMN from NC subjects (Figure 1). This indicates that the PMN from EH patients are primed.

Rate of superoxide anion release by stimulated peripheral polymorphonuclear leukocytes (PMN) from normal subjects (NC, n = 37) and hypertensive patients (EH, n = 37). Superoxide was measured by superoxide dismutase inhibitable reduction of ferricytochrome C, after stimulation of 10⁶ separated polymorphonuclear leukocytes (PMN) by 0.32 × 10⁻⁷ mol/L phorbol 12-myristate 13-acetate (PMA), followed spectrophotometrically at 549 nm. Superoxide release is expressed as nmol/10 min/10⁶ cells. Data are means ± SE. *P = .0001 v the NC PMN.

Plasma Glutathione Levels

Total plasma glutathione levels are similar in NC and EH subjects (Table 2). In EH patients the plasma levels of oxidized glutathione were significantly higher than the NC subjects. Plasma GSH concentrations in EH patients tended to be lower than in control subjects. The redox state expressed as [GSSG]/[GSH] in EH patients was twice that of the NC patients (Table 2), indicating systemic oxidative stress.

Inflammation

Survival of PMN

The results are depicted in Figure 2. Incubation of NC PMN in autologous plasma showed no significant reduction in cell number. However, when NC PMN were incubated in heterologous, hypertensive plasma, a significant reduction (P = .0006) in NC PMN number occurred. Incubation of EH PMN in autologous EH plasma also significantly reduced (P = .03) the PMN number, whereas no noticeable reduction in EH PMN number was observed when EH PMN were incubated in NC plasma. We have interpreted these results to indicate that PMN survival is dependent on extracellular factors.

Percentage of reduction in peripheral polymorphonuclear leukocyte (PMN) numbers from normal subjects (NC, n = 37) and hypertensive patients (EH, n = 37) after 90 min of incubation in autologous and heterologous, normal or hypertensive plasma. Survival count was performed by Coulter counter. Reduction in cell number is expressed as the percent difference between the number of cells counted after 90 min of incubation and the cells counted at the beginning of incubation. *Indicates a significant decrease (P = .0006) of NC PMN number in EH heterologous plasma v incubation in autologous NC plasma. **Indicates a significant decrease (P = .03) of EH PMN number in EH autologous plasma.

Effect of SOD on PMA-Stimulated PMN Survival

Exposure of isolated PMA-stimulated PMN to SOD was used to distinguish between the commitment of the stimulated PMN to die and the cytotoxic action of O₂⁻. SOD had no effect on PMA-stimulated PMN survival rates (Figure 3), emphasizing that activation of PMN is an irreversible commitment to die.

Correlation Between Survival and Superoxide Release from PMA-Stimulated PMN

The definition of half-life of survival of PMN is depicted in Figure 3. The half-life of PMN survival, regardless the origin of the separated cells, either NC or EH, correlated significantly with the rate of their superoxide release (Figure 4). This indicates that once a PMN is stimulated (defined by the rate of superoxide release), it is committed to die by necrosis as the cell disintegrates and disappears (defined by the PMN count). As these PMN are isolated and SOD did not affect their survival (Figure 3), these data indicate that the rate of superoxide release by PMN and their survival are not dependent on the extracellular factors but rather on its intracellular properties.

Relationship between superoxide release and survival of peripheral polymorphonuclear leukocytes (PMN). PMN are not divided into normal subjects or hypertensive patients, but are taken at random according to their rate of superoxide release. Extrapolated half-life (t_1/2) of survival (Figure 3) is plotted against superoxide release of the same sample of cells.

Relationships Between Superoxide Release and Blood Pressure

To establish whether a relationship between the level of blood pressure and superoxide release exists, linear regression analysis was performed by correlating superoxide release and the mean arterial pressure (MAP) of each subject. No correlation between MAP and superoxide release was found (r² = 0.03, P = NS). No correlation could be shown either with systolic or diastolic blood pressure. This suggests no cause–effect correlation between increased superoxide release and the height of the blood pressure.

Discussion

Endothelial dysfunction has been hypothesized to underlie the development of atherosclerosis in hypertension.^3,–5 Oxidative stress and inflammation are recognized among the mechanism causing endothelial damage.^6,–8 The vascular endothelial and smooth muscle cells are sources of ROS in the cardiovascular system. Both cells possess a membrane-bound, extramitochondrial NADH/NADPH-dependent oxidase system. On the basis of observations in an animal model of hypertension, Suzuki et al³⁵ suggested enhancements of spontaneous oxidative stress in the microvascular wall in vivo. Rajagopalan and colleagues demonstrated in a similar animal model that circulating vasoactive compounds, such as angiotensin II, were proinflammatory because the peptide could activate this system to increase the extent of oxidative stress in the vessel wall.^36,37 The PMN also contains this identical ROS-generating system and if activated would be an additional systemic source of ROS. PMN are dormant inflammatory cells. When activated, ROS, proteolytic enzymes, and chemotactic substances leak out resulting in tissue damage, inflammation, and the recruitment of additional PMN. Thus, if some circulating PMN are already in a state of partial activation owing to extra- or intracellular factors, they may contribute to oxidative stress and inflammation leading to damage of the endothelial cell.

The objectives of this study focused on the properties and activity of PMN and their possible role in the contribution to oxidative stress and inflammation in EH patients. Oxidative stress results from either excessive production of ROS or reduced antioxidant levels. The redox state of the ubiquitous antioxidant glutathione is a useful plasma marker of oxidative stress.³⁴ Reduced glutathione (GSH) is an endogenous antioxidant playing an important role in the detoxification of ROS.^12,13 An increase in plasma GSSG reflects GSH consumption and whole body oxidation stress.¹⁴ Although the total plasma glutathione levels were not different between the two groups in the present study, we found that the plasma redox state of glutathione (GSSG/GSH) in EH patients was twice that of NC subjects. Our data indicate that this elevated redox state was primarily attributable to an increased plasma level of oxidized glutathione at the expense of reduced glutathione. To the best of our knowledge, this is the first report of altered plasma glutathione redox state in EH patients. In previous studies other markers of oxidative stress were reported in EH patients. For example, Kumar and Das¹¹ reported higher plasma lipid peroxides and lower red blood cell levels of vitamin E and SOD in EH patients compared to control volunteers. Together these findings indicate that there is a source of ROS in plasma of EH individuals causing GSH oxidation. Because superoxide reacts directly with GSH,¹³ its augmented release by EH PMN may well have contributed to the oxidation of GSH as well as to a significant alteration in the redox state of GSH in the plasma of these patients. Primed PMN may be one such source. Superoxide release by PMN in human hypertension has been the subject of several studies. Sagar et al,¹⁰ Kumar and Das,¹¹ and Pontremoli et al²³ have shown that superoxide anion and hydrogen peroxide production by PMN was higher in EH patients compared with normals. We also observed that superoxide release is faster in EH PMN than in NC individuals.

We have shown that the half-life survival of PMN correlated significantly with the rate of their superoxide release. Once activated, PMN are committed to die by necrosis. Hence, the life span of activated PMN will be shorter than that of nonstimulated PMN. Therefore, one would expect that the survival rates of PMN from EH patients would be lower than that of PMN from normotensives. We report such findings in the present study and interpret these data as indicative of PMN activation and necrosis in EH.

What primes PMN in EH? Our data indicate that both intracellular and extracellular factors are involved in the priming process. The experiments in which we describe a more rapid release of superoxide by separated PMN in an artificial environment and the absence of an effect by SOD indicates a cellular feature of the EH PMN. Our experiments in which we incubated PMN in autologous and heterologous plasma suggest that extracellular factors are involved in PMN priming. Our demonstration that a primed state of PMN does not correlate with the levels of blood pressure also warrants comment. Priming of PMN in EH merely reflects the response to a constant stimulus ending in oxidative stress. The identity or identities of these intracellular and extracellular priming factors are currently being sought. One can list prolonged exposure to elevated plasma levels of angiotensin-II as one such factor,^36,37 with the possible addition of a hereditary metabolic component.^10,11,22

In the inflammatory response, primed PMN die by self-necrosis and at the same time actively recruit more PMN. Our data invoke the coexistence of an inflammatory reaction with oxidative stress in EH. We have shown that superoxide release is augmented in EH PMN and correlates with a decrease cell number (Figure 3). The decrease in cell number supports necrosis as the mode of cell death as necrosis per se is characterized by rapid disappearance and disintegration of cells as opposed to the slower cell intact process of apoptosis. Because EH PMN are activated, one can also conclude that EH PMN are predisposed to a faster rate of necrosis than NC PMN. This may be interpreted to indicate PMN recruitment in response to the increased rate of necrosis. Of interest is our serendipitous finding that EH patients have an increased normal plasma ALP activity, with no significant change in alanine aminotransferase and aspartate aminotransferase, and other evidence of hepatic or bone disease. It has been shown that ALP activity can be elevated due to PMN degranulation, another feature of PMN activation.^22,38

The EH patients of this study have raised WBC and PMN counts that fell within the upper quartile of the normal range. Raised leukocyte and PMN counts have been already described by Schmid-Schönbein et al²¹ in animal models of hypertension, and by Friedman et al²⁰ in EH patients. Increase in circulating leukocytes and PMN count has been shown by others to be a potential cardiovascular risk,¹⁹ not in relation to increased blood pressure. Thus, increased WBC and PMN number in EH patients may be considered as an additional risk factor in EH.

In summary, our study shows that EH is accompanied by a primed state PMN that does not correlate with the levels of blood pressure. PMN priming in EH patients reflects an in vivo exposure to a constant stimulus ending in oxidative stress, increased self-necrosis, and cell recruitment. Oxidative stress and inflammation will result in endothelial damage and atherosclerosis in the long run.

Acknowledgments

We are grateful to Prof. A. Bomzon for critical review of the manuscript.

References