Increased Adhesion Molecules Expression and Production of Reactive Oxygen Species in Leukocytes of Sleep Apnea Patients LARISSA DYUGOVSKAYA, PERETZ LAVIE, and LENA LAVIE Unit of Anatomy and Cell Biology and Sleep Laboratory, Technion-Israel Institute of Technology, The Bruce Rappaport Faculty of Medicine, Haifa, Israel
Obstructive sleep apnea (OSA) is associated with increased cardiovascular morbidity and mortality. Free radicals and adhesion molecules were implicated in the pathogenesis of atherosclerosis leading to cardiovascular disorders. Therefore, we investigated the link between CD15, CD11c, CD11b, and CD64 expression on leukocytes and their ability to generate reactive oxygen species (ROS) in patients with OSA and control volunteers. We also studied the effects of hypoxia in vitro on monocytes from control subjects and the ability of monocytes from both groups to adhere to human endothelial cells in culture. The effect of nasal continuous positive airway pressure (nCPAP) treatment was studied as well. We found that OSA was associated with increased expression of adhesion molecules CD15 and CD11c on monocytes, increased adherence of monocytes in culture to human endothelial cells, increased intracellular ROS production in some monocyte and granulocyte subpopulations, and upregulation of CD15 expression due to hypoxia in vitro in monocytes of control subjects. Furthermore, nCPAP treatment was associated with downregulation of CD15 and CD11c monocyte expression and decreased basal ROS production in CD11c monocytes. Monocyte adherence to endothelial cells decreased as well. Our findings provide one of the possible mechanisms for explaining the high rate of cardiovascular morbidity in patients with sleep apnea. Keywords: adhesion molecules; monocytes; granulocytes; reactive oxygen species; nasal continuous positive airway pressure
Obstructive sleep apnea (OSA) syndrome has emerged in recent years as an important risk factor for cardiovascular morbidity. Association has been reported between sleep apnea and systemic hypertension (1, 2), pulmonary hypertension (3), ischemic heart disease (4, 5), and stroke (6, 7). Although the underlying mechanisms have not been fully elucidated, so far attempts to explain this relationship have mainly focused on hypoxia-related augmented sympathetic activation (8, 9) or on modification of basic cardiovascular physiologic regulatory mechanisms (10, 11). Several lines of evidence implicate oxidative stress in the pathogenesis of cardiovascular morbidity. Inflammatory leukocytes are one of the well-characterized pathways of free radical formation. Leukocyte accumulation and adhesion via
(Received in original form April 26, 2001; accepted in final form December 17, 2001) This study was supported in part by a grant from the Israeli Academy of Sciences to L. Lavie and P. Lavie and, in part, by a joint grant from the Center for Absorption in Science of the Ministry of Immigration Absorption and the Committee for Planning and Budgeting of the Council for Higher Education under the framework of the KAMEA program. Correspondence and requests for reprints should be addressed to Lena Lavie, Unit of Anatomy and Cell Biology, The Bruce Rappaport Faculty of Medicine, Technion, POB 4649, 31096, Haifa, Israel. E-mail:
[email protected] This article has an online data supplement, which is accessible from this issue’s table of contents online at www.atsjournals.org Am J Respir Crit Care Med Vol 165. pp 934–939, 2002 DOI: 10.1164/rccm.2104126 Internet address: www.atsjournals.org
appropriate receptors on the endothelium and initiation of leukocyte/endothelial cell interactions may critically impair endothelial cell function and propagate atherogenic processes (12). Recent reports on endothelial dysfunction in OSA suggested that facilitated atherogenic processes may play a role in the pathophysiology of cardiovascular morbidity in this syndrome. Thus, elevated plasma levels of the adhesion molecules intercellular adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1), L-selectin, and sE-selectin, suggesting an activated endothelium, were shown in sleep apnea patients (13, 14). Also, the inflammatory cytokine tumor necrosis factor-alpha (TNF-) that was shown to greatly increase ICAM-1 and VCAM-1 synthesis, was increased in patients with OSA compared with normal control subjects (15). Sleep apnea patients were reported to have higher levels of platelet aggregation and activation, as measured by P-selectin expression, than when treated with nasal continuous positive airway pressure (nCPAP) (16). More recently, increased neutrophils’ superoxide production was reported in OSA; this was fully reversed by nCPAP therapy (17). Because hypoxia per se enhances the production of adhesion molecules in a variety of cells, including endothelium and leukocytes (18), it is possible that the repetitive nocturnal hypoxemia in OSA may initiate atherogenic processes, which eventually lead to cardiovascular morbidity. In the present study, we characterized the cellular phenotypes of peripheral whole blood monocytes (PBMs) and granulocytes in patients with OSA in comparison with normal control subjects. These cellular phenotypes, which included identification of CD64, CD11c, CD11b, and CD15 positive cells, were correlated with oxidative metabolism and respiratory burst activity of monocytes and granulocytes. In addition, we compared the ability of monocytes of patients with OSA and normal control subjects to adhere to human coronary artery endothelial cells (HCAECs) or to human umbilical vein endothelial cells (HUVECs) in vitro. We also studied the effects of hypoxic conditions in vitro on the expression of CD15 by monocytes obtained from control subjects. Besides, we provide data demonstrating the effects of nCPAP treatment on some of these parameters. We used whole blood to minimize procedure-related changes in surface receptor expression (19).
METHODS Patients and Control Subjects Five milliliters of blood were withdrawn in the morning after overnight fasting from a total of 26 moderate-to-severe patients with OSA and 31 healthy volunteers. Of these, in 18 patients with OSA and 26 control subjects we studied the expression of adhesion molecules and oxidative metabolism in monocytes. Eight patients with OSA were studied more than 6 mo after the initiation of nCPAP treatment for two consecutive days, with and without nCPAP. Eleven normal control subjects participated only in the study on the effects of hypoxia in vitro on whole blood CD15 expression in monocytes. The clinical and demographic data of each of the groups are summarized in Table 1.
935
Dyugovskaya, Lavie, and Lavie: Oxidative Stress in Sleep Apnea Twelve of the 26 control subjects who participated in the adhesion molecules and oxidative metabolism experiments were recorded in the sleep laboratory and were found to be free of any significant breathing disorders in sleep (mean respiratory disturbance index [RDI] 6.5). In the other 14, as well as the additional control subjects participating in the hypoxia in vitro experiment, OSA was ruled out by a detailed clinical interview. Because there were no significant differences between control subjects who were recorded in the laboratory and those who were only interviewed, their data were combined for statistical analysis. Except for one participant with a history of ischemic heart disease, none of the control individuals had any history of cardiovascular, pulmonary, or metabolic diseases. Comparison of the demographic data between patients with OSA and control subjects revealed that patients had significantly higher body mass index (BMI) (p 0.01). The protocol was approved by the local human rights committee and all participants signed an informed consent before their enrollment.
Detection of Reactive Oxygen Species and Fluorescence Staining by Flow Cytometry An expanded version of the technical details of the methodologies used can be found in the online data supplement. Oxidative metabolism was assessed by flow cytometry (20–22). For all patients with OSA and control subjects, blood samples were processed for reactive oxygen species (ROS) production within 30 min of blood collection. Each of the experiments was conducted on groups of 8 to 15 patients with OSA and 9 to 17 control subjects.
Cultured Endothelial Cell Monolayers Adherence of leukocytes was studied either on HCAECs or on HUVECs. Cryopreserved HCAEC 5279 were obtained from Clonetics (Normal Human Cell Systems; BioWhittaker, A Cambrex Company, Walkersville, MD) and cultivated according to the manufacturer’s instructions and used from passage 2 to 5. Freshly prepared HUVECs were kindly provided by Dr. N. Lanir from Rambam Medical Center and used only in the first and second passages. Endothelial cells (HCAECs or HUVECs) were pretreated for 1 h at 37 C with antiCD62E/CD62P or anti-CD54 antibodies (Serotec, Oxford, UK), to inhibit adhesion to endothelium via E-selectin/P-selectin or ICAM-1, respectively, after which leukocytes derived from patients with OSA and control subjects were added.
Adherence Assay Mononuclear cells (MNCs) were isolated from peripheral blood of patients with OSA and control subjects using gradient 1.077 (Sigma Diagnostics, Israel) and radiolabeled by incubation with 51Cr. Radioactivity in the adherent cells was determined by an auto-gamma scintillation counter (LKB-Wallac, Turku, Finland) in triplicates. The percentage of cells adhering to HCAECs or to HUVECs was calculated TABLE 1. DEMOGRAPHIC, SLEEP, AND CLINICAL DATA OF PATIENTS WITH OSA AND NORMAL CONTROL SUBJECTS PARTICIPATING IN THE DIFFERENT EXPERIMENTS
Age, yr Male/female BMI, kg/m2 RDI, events/h MinOxy, % HTN, n IHD, n DM, n
OSA Patients (n 18)
Control Subjects (n 31)
CPAP-OSA Patients (n 8)
49.3 11.0 17/1 28.3 4.0 35.4 19.2 82.7 9.6 2/18 3/18 2/18
43.1 11.1 31/1 25.6 2.7 6.5 3.1* 91.3 5.1* 2/31 3/31 1/31
49.4 6.7 7/1 32.0 7.6 55.6 13.2 73.63 11.8 5/8 1/8 2/8
Definition of abbreviations: BMI body mass index; DM diabetes mellitus; HTN hypertension; IHD ischemic heart disease; MinOxy minimum oxygen saturation; RDI respiratory disturbance index. * Data are based on 12 normal control subjects investigated in the sleep laboratory. Note that 26 out of 31 of the normal control subjects participated in the adhesion molecules and in ROS production experiments. Five of the 31 control subjects participated only in the hypoxia in vitro experiment; these were only interviewed to exclude OSA.
as follows: Adherence (Mean cpm in Lysate)/(Mean cpm in Overlaid cells) 100% (23).
CD15 Expression under Hypoxic Conditions In Vitro Freshly obtained blood from 11 healthy control subjects was exposed to hypoxic conditions in a Modular Incubator Gas Chamber (Hot Box system, Del Mar, CA). Blood samples were exposed to hypoxia by purging with 1% O2, 5% CO2, 94% N2, and placed in a 37 C incubator for an overnight period or for an overnight period of hypoxia followed by 6 h reoxygenation (21% O2, 5% CO2). Control tubes were identically treated, except that they were kept under normoxic conditions (21% O2, 5% CO2) for the same durations. The percentage of CD15 monocytes in each individual was determined before and after exposure to hypoxic and normoxic conditions.
nCPAP Treatment The following parameters were compared between the nCPAP and no nCPAP conditions: expression of %CD15 (n 8) and %CD11c (n 8), %CD11c monocytes participating in basal ROS production (n 8), and monocytes’ ability to adhere to endothelial cells in culture (n 7). Three of the patients were studied with HCAECs and four with HUVECs. Generally, monocytes adhered more avidly to HUVECs than to HCAECs, but this did not affect the differences between patients with OSA and control subjects.
Statistical Analysis All data were expressed as mean SD. All differences between OSA and control groups were evaluated by nonparametric analysis of covariance (ANCOVA) after adjusting for BMI. Paired t tests were used to compare between variables before and after hypoxia or normoxia treatments and before and after nCPAP treatment. A nonparametric matched pairs Wilcoxon test was used for variables not normally distributed.
RESULTS Expression of CD64 and Adhesion Molecules CD11b, CD11c, and CD15 on Monocytes and Granulocytes
Monocyte, granulocyte, and lymphocyte subgroups were distinguished by side scatter and fluorescence characteristics (24). For each subpopulation, percentage of positive cells was obtained using a gate that included 1% of cells of the mouse isotypic negative control. The percentages of CD64, CD11b, CD11c, and CD15 positive cells in PBMs of patients with OSA and control subjects are depicted in Figure 1. Patients with OSA exhibited a 1.9-fold increase in the percentage of CD11c monocytes (p 0.004) and a 9-fold increase in the percentage of CD15 (p 0.0005) monocytes, but percentage of CD64 (Fc RI) monocytes was significantly lower than control subjects (p 0.03). The value for CD64 monocytes in healthy control subjects was reported to be 81.3 6.8%, which is in agreement with our results (25). There was no significant difference in the percentage of CD11b between groups. Similar values (83.7 8%) were reported by Sacks and coworkers for CD11b monocytes in healthy individuals (25). Also, the expression of CD14 markers, used to identify monocytes, did not Figure 1. Expression (%) of CD64, CD11b, CD11c, and CD15 molecules on PBMs from patients with OSA and control subjects. Blood was processed for flow cytometry as described under METHODS. Error bars represent SD. (n 9 to 17 control subjects and 9 to 15 patients.) *p 0.03, **p 0.004, ***p 0.0005. Control; OSA.
936
AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE
differ significantly between patients with OSA and controls, 55.2 6.8% (48 to 64%) versus 67.7 14.9% (51 to 82%). Nonactivated granulocytes of patients with OSA and control subjects were practically devoid of CD64 and CD11c expression (ranging from 0.8 to 8.0%). On the other hand, CD11b expression was very high in granulocytes of both groups (from 92 to 99.5%), but no differences between the groups were observed. Measurement of ROS Production in Whole Blood Monocytes
ROS production was determined in monocytes and granulocytes from the same preparation, without further purification or manipulation of the cells in order to avoid activation (22, 24). By using monoclonal antibodies (mAbs) for identification of CD64, CD11b, CD11c, and CD15 subpopulations of monocytes and granulocytes, we could specifically analyze the relationship between each subpopulation of cells bearing a specific surface marker and its ability to produce ROS either under nonactivated state representing basal ROS production or in response to phorbol myristate acetate (PMA). The intensity of ROS production by each individual cell, represented by the mean fluorescence intensity (MFI)/cell and the percentage of cells that express a specific surface molecule and oxidize dihydroethidium (DHE), was combined to an index ([MFI/ cell] % cells that oxidize DHE) representing total ROS production of a specific subpopulation. Significant differences were detected in CD11c monocytes between patients with OSA and control subjects. The percentage of CD11c cells participating in basal oxidation of DHE was higher in patients with OSA than in control subjects (38.8 17.2% versus 24.1 10.6%, p 0.05). Likewise, basal ROS production in OSA patients was 2.5-fold higher than in control individuals (106.2 98.7 versus 42.7 11 MFI/cell, p 0.0003). Similarly, in response to PMA stimulation, both the percentage of CD11c monocytes producing ROS and the intensity of ROS production in MFI/cell were significantly higher in patients with OSA (73.8 12.5% versus 57.8 10.8%, p 0.03 and 385.5 158.8 versus 184 112.7 MFI/ cell, p 0.01, respectively). Total ROS production, represented by index, of nonactivated and PMA-activated CD11c monocytes is depicted in Figure 2A. As expected, the ROS production indices by CD11c monocytes were significantly
Figure 2. Total ROS production without and with PMA stimulation by CD11c and CD64 monocytes. (A) Index ([MFI/cell] % of CD11c cells participating in DHE oxidation) of nonactivated and PMA-activated CD11c monocytes. (B) Index ([MFI/cell] % of CD64 cells participating in DHE oxidation) of nonactivated and PMA-activated CD64 monocytes. Whole blood samples were processed immediately as described under METHODS. MFI mean fluorescence intensity. Error bars represent SD. (n 11 to 14 control subjects and 8 to 10 patients) *p 0.003, **p 0.005, ***p 0.0004. Control; OSA.
VOL 165
2002
higher in patients with OSA than control subjects (p 0.003 and p 0.0005 for nonactivated and PMA-activated, respectively, Figure 2A). In contrast, ROS production of the small subpopulation of CD15 monocytes did not differ significantly between patients with OSA and control subjects neither in basal conditions nor after PMA stimulation, but MFI/cell was high in control as well as in OSA (data not shown). Analysis of the covariance revealed borderline statistically significant differences between groups in the percentages of nonactivated CD64 monocytes participating in the DHE oxidation (52 18.4% versus 42.8 16.4%, p 0.08, and 82.8 46.3 versus 53.9 30.9 MFI/cell, p 0.08 in OSA and control, respectively). However, after PMA stimulation the percentage of CD64 monocytes participating in DHE oxidation increased by 1.6-fold in comparison with basal conditions in patients with OSA (52.0 18.4% versus 82.03 16.6%, p 0.003) but did not change significantly in control subjects (42.8 16.4% versus 50.2 15.3%). Although the percentage of nonactivated CD64 monocytes did not differ significantly between patients and control subjects, percentage of activated CD64 monocytes as well as ROS production (MFI/cell) by PMA-activated monocytes were significantly higher in patients with OSA (p 0.002 and p 0.009, respectively). Thus, index for total ROS production (Figure 2B) by PMA-activated CD64 monocytes of patients with OSA was increased by more than 5-fold as compared with control subjects (p 0.0004). Despite the high level of CD11b expression on monocytes, only 2 to 7% of CD11b monocytes participated in DHE oxidation at rest. Also, the intensity of ROS production (MFI/cell) did not vary between OSA and control groups, and no differences were detected in this subpopulation due to PMA activation. Neither percentages of monocytes participating in DHE oxidation nor the MFI/cell varied between the groups (36.8 10.5% versus 40.0 9.4% and 113.8 30 versus 112 40.8 MFI/cell in patients with OSA and control subjects, respectively). Measurement of ROS Production in Whole Blood Granulocytes
As mentioned previously, the percentage of CD11c and CD64 nonactivated granulocytes was low (0.8 to 8%) and similar in patients and control individuals. Basal ROS production in CD11c and CD64 granulocytes was variable and significantly different between groups (p 0.05) only in CD64 cells (ROS index: 610 528 versus 446 462, for patients with OSA and control subjects, respectively). Thus, PMA stimulation of whole blood granulocytes increased both the number of CD11c and CD64 participating granulocytes (85 to 96%) and fluorescence intensity. Generally, PMA-stimulated granulocytes of patients with OSA exhibited higher ROS production as compared with control subjects. This is illustrated in Figure 3 by depicting the index ([MFI/cell] % cells) for CD11c (p 0.04) and CD64 granulocytes (p 0.08).
Figure 3. Total ROS production with PMA stimulation by CD11c and CD64 granulocytes. Index ([MFI/ cell] % cells participating in DHE oxidation) of PMA-activated CD11c and CD64 granulocytes. Blood samples were processed as in Figure 2. MFI mean fluorescence intensity. Error bars represent SD. (n 10 to 12 control subjects and 8 patients.) *p 0.04, **p 0.08. Control; OSA.
937
Dyugovskaya, Lavie, and Lavie: Oxidative Stress in Sleep Apnea Adhesion of Monocytes from Patients with OSA and Control Subjects to HCAECs and HUVECs in Culture
The ability of monocytes from patients with OSA (n 8) and control subjects (n 8) to adhere to endothelial cells in culture was assessed. In each experiment, patients and control subjects were compared by using the same batches of endothelial cell monolayers under the same experimental conditions. Monocytes from patients with OSA adhered significantly more avidly to both HCAEC and HUVEC monolayers than monocytes of control subjects (in both at least p 0.05) as summarized in Table 2. Monocyte adhesion of normal subjects to HCAEC or HUVEC in culture was unaffected by antibodies to CD62E/CD62P, that block binding through selectins. Conversely, in OSA monocytes, adhesion to HCAECs and HUVECs was inhibited by 31% and 38.8% (in both at least p 0.05), respectively. Antibodies to CD54 (binding ICAM-1 molecules) slightly and nonsignificantly inhibited monocyte HCAECs (by 19% and 14.4% for control subjects and patients with OSA, respectively) (Table 2A). However, adhesion of OSA monocytes to HUVECs was significantly decreased by 26% (p 0.05, Table 2B) after treatment with Abs to CD54. Effect of Hypoxia In Vitro on CD15 Expression in Monocytes of Control Subjects
To investigate the effects of hypoxemia on the expression of CD15 molecules, whole blood monocytes of 11 control subjects were exposed in vitro to hypoxia (1% O2, 5% CO2, 94% N2) or normoxia (21% O2, 5% CO2), for an overnight period. In five of the control subjects, hypoxia was followed by 6 h of reoxygenation. As shown in Table 3, the initial values obtained for %CD15 monocytes were low as shown earlier in control subjects (Figure 1). However, after overnight hypoxia, percentage of CD15 monocytes increased to comparable values observed in patients with OSA (12.8 5.7% versus 13.9 7.2%, respectively). This increase was significantly higher than the increase under normoxic conditions (12.8 5.7% versus 4.0 1.4%, p 0.001). The expression of CD15 was only slightly increased in control monocytes kept overnight under normoxia. Moreover, hypoxic treatment followed by 6 h of reoxygenation resulted in a further increase in percentage of
TABLE 2. ADHESION INDICES OF MONOCYTES FROM PATIENTS WITH OSA AND CONTROL SUBJECTS TO HCAECs AND TO HUVECs IN CULTURE WITH AND WITHOUT ANTIBODIES AGAINST CD62E/CD62P AND CD54* Antibodies Isotype Control
CD62E/CD62P
CD54
Control, n 4 Control relative, % OSA, n 4 OSA relative, %
2.59 0.28† 100 10.8 4.22 0.72†‡ 100 17
2.41 0.62 93.05 24 2.90 1.04‡ 68.7 24.6
2.10 0.27 81.08 10.4 3.61 1.17 85.6 27
Control, n 4 Control relative, % OSA, n 4 OSA relative, %
5.2 1.8† 100 34 9.8 0.48†‡§ 100 4.9
5.8 3.2 111.5 61 6.0 2.05‡ 61.2 20.9
5.4 2.6 103.8 50 7.2 1.64§ 73.5 16
A.
B.
* Results are expressed as adhesion index of monocytes to HCAECs (A) or to HUVECs monolayers (B). Endothelial cells were pretreated either with isotype control Abs (mouse IgG1) or with Abs against CD62E/CD62P or CD54 to block adherence via E-selectin/ P-selectin or ICAM-1, respectively. In each experiment, monocytes from a patient with OSA and a control subject were studied. For each individual, values were obtained by averaging 3 to 5 wells. Values are presented as mean SD; all significance levels are p 0.05 or better. † Between OSA and controls; ‡ and § between OSA with and without CD62E/CD62P and CD54, respectively.
TABLE 3. EXPRESSION OF CD15 MOLECULES ON PERIPHERAL BLOOD MONOCYTES OF CONTROL SUBJECTS TREATED IN VITRO WITH HYPOXIA AND HYPOXIA/REOXYGENATION IN COMPARISON WITH NORMOXIC CONDITIONS* Treatment Baseline, n 11 Normoxia, n 11 Hypoxia, n 11 Norm/reoxy, n 5 Hypox/reoxy, n 5
CD15 (%)
CD15 (MFI/cell)
Index ([MFI/cell] %)
1.83 1.12 3.96 1.41 12.79 5.67† 3.79 1.58 17.12 3.03†
58.8 19.5 74.09 24.0 60.45 24.4 80.6 30.1 84.2 17.9
83.3 43.1 278.5 92.8 759 371‡ 332 225 1,419 319‡
* Results are depicted as % monocytes expressing CD15, mean fluorescence intensity (MFI)/cell, and by the combined index (MFI % cells expressing CD15), as analyzed by FACS. In each experiment, whole blood monocytes of control subjects were treated as follows: immediately after blood collection (baseline), after overnight in normoxia, after overnight in hypoxia, after overnight in normoxia followed by 6 h reoxygenation (norm/reoxy), after overnight hypoxia followed by 6 h reoxygenation (hypox/reoxy). n number of subjects. Values are presented as mean SD. † Significantly different than normoxia/reoxyg, p 0.001 or better; ‡ Significantly different than normoxia p 0.001 or better.
CD15 monocytes, without any further increase in the expression under normoxia for the comparable period (17.1 3.0% versus 3.8 1.6%, p 0.02). Similar results were obtained with the index (MFI % cells expressing CD15). Figure 4 illustrates the expression of CD15 monocytes of a control subject by fluorescent-activated cell sorter (FACS) analysis under the various treatments described. Effects of nCPAP Treatment
Treatment with nCPAP decreased RDI (55.6 13.2 versus 15.1 9.4, p 0.008) and the percent time below 90% arterial oxygen saturation (20.4 18.7% versus 0.75 1.2%, p 0.02). The percentage of CD15 (n 8) and CD11c (n 8) monocytes was significantly decreased after treatment (20.9 10.4% versus 7.0 4.0%, p 0.02 and 71.2 22.7% versus 58.6 20.2%, p 0.04, respectively). Of note, %CD15 did not decrease in one patient who remained hypoxic during treatment. Also the percentage of CD11c monocytes participating in basal ROS production (n 8) was decreased after treatment (72.6 19.9% versus 36.2 23.9%, p 0.02). Monocytes’ adhesion index to HCAECs or HUVECs was significantly decreased after treatment as well (n 3, 4.3 0.8 versus 1.4 7, p 0.01 and n 4, 9.8 4.8 versus 5.6 4.1, p 0.01, respectively).
DISCUSSION This study provides strong support to the relevance of monocyte adhesive properties and oxidative metabolism to the pathogenesis of cardiovascular morbidity in sleep apnea syndrome. The following is a summary of our major findings. We showed that: (1) Sleep apnea syndrome is associated with increased expression of two adhesion molecules, CD15 and CD11c in monocytes. (2) Monocytes of patients with OSA adhered significantly more avidly to endothelial cells in culture than those of control subjects. (3) Treatment with hypoxia in vitro increased CD15 expression in monocytes of control subjects to percentages comparable to those of patients. (4) Patients with OSA had increased intracellular ROS production in some monocyte and granulocyte subpopulations. (5) nCPAP treatment downregulated adhesion molecule expression (CD15 and CD11c on monocytes), lowered basal ROS production by CD11 monocytes, and decreased monocyte adherence to human endothelial cells in culture. Activated leukocytes play a key role in the inflammatory response to injury caused by hypoxia/reoxygenation that may
938
AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE
Figure 4. Fluorescence histograms of CD15 expression in a control subject obtained by flow cytometry. Monocytes in whole blood before in vitro treatment (fresh), after overnight normoxia, after overnight hypoxia, and after overnight hypoxia followed by 6 h reoxygenation. Dotted lines show the background with isotype matched control antibody. A representative control out of 11 analyzed.
initiate atherogenic processes. Their interaction with the endothelium is mediated and closely regulated by adhesion molecules (26). The selectin binding CD15, which is primarily expressed on neutrophils and on a small subgroup of monocytes, and CD11c, which binds to ICAM-1 on activated or dysfunctional endothelial cells, were increased in patients with OSA. These observations clearly indicate that monocytes of patients with OSA are activated (19, 27). Upregulation of CD15 expression was shown in lipopolysaccharide-activated monocytes (22). In addition, hypoxia/reoxygenation stimulated the expression of ICAM-1 and VCAM-1 on human endothelial cells (28), upregulated CD15 expression, and activated leukocytes in vivo and in vitro (29, 30). Moreover, exposing HUVECs to anoxia/reoxygenation in culture, increased neutrophil/HUVECs adherence (18). Soluble selectin and ICAM-1 in sera of patients with OSA were also shown to increase, which was attributed to surges in apneas’ related sympathetic activity (13, 14). However, in our study, exposure of control monocytes to hypoxia in vitro increased the percentage of CD15 monocytes to comparable levels of patients with OSA. Subsequent exposure of these cells to reoxygenation further increased CD15 expression. Although the duration and magnitude of our in vitro hypoxemia are different than the intermittent nocturnal hypoxemia experienced by patients with OSA, our findings provide preliminary support to the notion that the apnea’s related hypoxia is largely responsible for the increased CD15 expression. Further in vitro studies using gradually increasing hypoxemia levels are needed to conclusively prove this association. The functional significance of the increased expression of these molecules was evident from the findings that monocytes of patients with OSA adhered significantly more avidly to HCAECs or HUVECs as compared with controls. Because mAbs against CD62E/CD62P (anti-E-selectin/P-selectin) inhibited the binding of OSA monocytes, this suggests that selectins may be primarily involved in this process (31, 32). Employing mAbs against CD54 (anti-ICAM-1) significantly inhibited the binding of OSA monocytes only to endothelial cells from umbilical vein origin, emphasizing the unique characteristics of endothelial cells from different origins. The increased expression of adhesion molecules on circulating monocytes from patients with OSA was accompanied by increased intracellular ROS production. This was evident both at rest and after PMA stimulation. Total basal ROS production of CD11c monocytes was increased by 2.5-fold in patients with OSA. Similarly, after PMA stimulation, total ROS production of CD11 monocytes was 2.7-fold greater in OSA patients. In contrast to the increased expression of CD11c monocytes in OSA, in inflammatory conditions CD11b expression was shown to be upregulated (25, 33, 34). In our
VOL 165
2002
study, monocytes’ expression of CD11b did not differ between patients with OSA and healthy control subjects. Although basal total ROS production of CD64 monocytes did not differ significantly between patients and control subjects, it was 1.9-fold higher than control subjects. Thus, PMA stimulation of this subpopulation resulted in more than 5-fold increase in total ROS production as compared with control individuals. These results as well point to the possibility that CD64 monocytes of OSA patients are preactivated. Fc RI (CD64) mediates many cellular responses crucial in host defense, including phagocytosis, cell cytotoxicity, and production and secretion of proinflammatory cytokines such as interleukin (IL)-1 , IL-6, and TNF- (35, 36). A finding of high concentrations of TNF- was reported in sleep apnea patients, which is in agreement with these results (15). In granulocytes, there were no significant between-group differences in total basal ROS production of CD11c cells. In CD64 granulocytes, however, basal ROS production was significantly higher in patients with OSA. In the PMA-activated cells, ROS production of CD11c and CD64 granulocytes was significantly and nearly significantly higher in patients with OSA, respectively. These results are in agreement with Schulz and coworkers (17) who reported on an enhanced release of superoxide from polymorphonuclear neutrophils of patients with OSA. The effects of nCPAP treatment on adhesion molecule expression and ROS production were studied in monocytes, by omitting nCPAP for a single night in patients treated for more than 6 mo. The %CD11c and %CD15 monocytes were relatively low on treatment night and were significantly increased after omitting treatment. %CD11c monocytes participating in basal ROS production was significantly increased as well. In seven out of eight patients, CD15 monocyte expression was decreased by at least 3-fold. In one patient, in whom %CD15 cells remained elevated, nCPAP treatment was not effective. Most importantly, in seven out of seven patients tested, treatment decreased monocytes adhesion index to endothelial cells. Collectively these data implicate hypoxemia in the upregulation of CD15 and CD11c expression. Our in vitro observations on increased adherence of monocytes from patients with OSA to endothelial cells and the decreased adherence of OSA monocytes as a result of nCPAP treatment clearly indicate that patients with OSA are exposed to atherogenic insult nightly. In vivo, this may cause endothelial dysfunction, which constitutes the initial step in the chain of events leading to cardiovascular morbidity. Indeed, there is recent evidence of endothelial dysfunction in normotensive patients with OSA free of any known cardiovascular disease (37, 38). In conclusion, we postulate that the repeated apneic-related hypoxic events, in a similar manner to hypoxia/reoxygenation injury, result in endothelial and monocyte activation. This was evident by the increased expression of CD15 and CD11c receptors and the increased oxidative metabolism. These primed monocytes adhere to the endothelium and become fully activated under hypoxic conditions. Subsequently, they emigrate into the subendothelial space and release lytic enzymes and oxygen radicals that injure the endothelium. The functionally impaired endothelium may initiate pathogenesis of cardiovascular morbidity in patients with sleep apnea. References 1. Lavie P, Herer P, Hoffstein V. Obstructive sleep apnoea syndrome as a risk factor for hypertension: population study. Br Med J 2000;320: 479–482. 2. Peppard PE, Young T, Palta M, Skatrud J. Prospective study of the asso-
Dyugovskaya, Lavie, and Lavie: Oxidative Stress in Sleep Apnea
3.
4.
5.
6.
7.
8.
9. 10.
11. 12. 13.
14.
15.
16.
17.
18.
19.
20.
ciation between sleep-disordered breathing and hypertension. N Engl J Med 2000;342:1378–1384. Sajkov D, Cowie RJ, Thornton AT, Espinoza HA, McEvoy RD. Pulmonary hypertension and hypoxemia in obstructive sleep apnea syndrome. Am J Respir Crit Care Med 1994;149:416–422. Peker Y, Kraiczi H, Hedner J, Loth S, Johansson A, Bende M. An independent association between obstructive sleep apnoea and coronary artery disease. Eur Respir J 1999;14:179–184. Schafer H, Koehler U, Ewig S, Hasper E, Tasci S, Luderitz B. Obstructive sleep apnea as a risk marker in coronary artery disease. Cardiology 1999;92:79–84. Bassetti C, Aldrich MS, Quint D. Sleep-disordered breathing in patients with acute supra-and infratentorial strokes: a prospective study of 39 patients. Stroke 1997;28:1765–1772. Dyken ME, Somers VK, Yamada T, Ren ZY, Zimmerman MB. Investigating the relationship between stroke and obstructive sleep apnea. Stroke 1996;27:401–407. Carlson JT, Hedner J, Elam M, Ejnell H, Sellgren J, Wallin BG. Augmented resting sympathetic activity in awake patients with obstructive sleep apnea. Chest 1993;103:1763–1768. Somers VK, Dyken ME, Clary MP, Abboud FM. Sympathetic neural mechanisms in obstructive sleep apnea. J Clin Invest 1995;96:1897–1904. Narkiewicz K, Pesek CA, Kato M, Phillips BG, Davidson DE, Somers VK. Baroreflex control of sympathetic nerve activity and heart rate in obstructive sleep apnea. Hypertension 1998;32:1039–1043. Kraiczi H, Hedner J, Peker Y, Carlson J. Increased vasoconstrictor sensitivity in obstructive sleep apnea. J Appl Physiol 2000;89:493–498. Price DT, Loscalzo J. Cellular adhesion molecules and atherogenesis. Am J Med 1999;107:85–97. Ohga E, Nagase T, Tomita T, Teramoto S, Matsuse T, Katayama H, Ouchi Y. Increased levels of circulating ICAM-1, VCAM-1, and L-selectin in obstructive sleep apnea syndrome. J Appl Physiol 1999;87:10–14. Chin K, Nakamura T, Shimizu K, Mishima M, Nakamura T, Miyasaka M, Ohi M. Effects of nasal continuous positive airway pressure on soluble cell adhesion molecules in patients with obstructive sleep apnea syndrome. Am J Med 2000;109:562–567. Vgontzas AN, Bixler EO, Papanivolaou D, Kales A, Chrousos GP. Plasma concentration of tumor necrosis factor alpha (TNF), interleukin-6 (IL-6) and leptin are elevated in sleep apnea independent of obesity (abstract). Sleep 1999;22:S331. Bokinsky G, Miller M, Ault K, Husband P, Mitchell J. Spontaneous platelet activation and aggregation during obstructive sleep apnea and its response to therapy with nasal continuous positive airway pressure: a preliminary investigation. Chest 1995;108:625–630. Schulz R, Mahmoudi S, Hattar K, Sibelius U, Olschewski H, Mayer K, Seeger W, Grimminger F. Enhanced release of superoxide from polymorphonuclear neutrophils in obstructive sleep apnea. Am J Respir Crit Care Med 2000;162:566–570. Ichikawa H, Flores S, Kvietys PR, Wolf RE, Yoshikawa T, Granger DN, Aw TY. Molecular mechanisms of anoxia/reoxygenation-induced neutrophil adherence to cultured endothelial cells. Circ Res 1997;81: 922–931. Elbim C, Hakim J, Gougerot-Pocidalo MA. Heterogeneity in Lewis-X and sialyl-Lewis-X antigen expression on monocytes in whole blood: relation to stimulus-induced oxidative burst. Am J Pathol 1998;152: 1081–1090. Toft P, Neilsen CH, Tonnesen E, Hansen TG, Hokland M. Changes in adhesion molecule expression and oxidative burst activity of granulocytes and monocytes during open-heart surgery with cardiopulmonary bypass compared with abdominal surgery. Eur J Anaesthesiol 1998;15: 345–353.
939 21. Kobzik L, Godleski JJ, Brain JD. Oxidative metabolism in the alveolar macrophage: analysis by flow cytometry. J Leukoc Biol 1990;47:295–303. 22. Himmelfarb J, Hakim RM, Holbrock DG, Leeber DA, Ault KA. Detection of granulocyte reactive oxygen species formation in whole blood using flow cytometry. Cytometry 1992;13:83–89. 23. Fadlon E, Vordermeier S, Pearson TC, Mire-Sluis AR, Dumonde DC, Phillips J, Fishlock K, Brown KA. Blood polymorphonuclear leukocytes from the majority of sickle cell patients in the crisis phase of the disease show enhanced adhesion to vascular endothelium and increased expression of CD64. Blood 1998;91:266–274. 24. Emmendorffer A, Hecht M, Lohmann-Matthes ML, Roesler J. A fast and easy method to determine the production of reactive oxygen intermediates by human and murine phagocytes using dihydrorhodamine 123. J Immunol Methods 1990;131:269–275. 25. Sacks GP, Studena K, Sargent K, Redman CW. Normal pregnancy and preeclampsia both produce inflammatory changes in peripheral blood leukocytes akin to those of sepsis. Am J Obstet Gynecol 1998;179:80–86. 26. Lefer AM. Role of the beta2-integrins and immunoglobulin superfamily members in myocardial ischemia-reperfusion. Ann Thorax Surg 1999; 68:1920–1923. 27. Siu KL, Golenbock DT, Sass PM, Maskati A, Xu H, Silverstein RL. Engagement of the Lewis X antigen (CD15) results in monocyte activation. Blood 1997;89:307–314. 28. Willam C, Schindler R, Frei U, Eckardt K-U. Increases in oxygen tension stimulate expression of ICAM-1 and VCAM-1 on human endothelial cells. Am J Physiol 1999;276:H2044–H2052. 29. Seko Y, Enokawa Y, Tamatani T, Kannagi R, Yagita H, Okumura K, Yazaki Y. Expression of sialyl Lewis (X) in rat heart with ischaemia/ reperfusion and reduction of myocardial reperfusion injury by a monoclonal antibody against sialyl Lewis. J Pathol 1996;180:305–310. 30. Seko Y, Enokawa Y, Tamatani T, Kannagi R, Yagita H, Okumura K, Yazaki Y. Induction of sialyl Lewis (X) on the surface of cultured rat vascular endothelial cells and cardiac myocytes by hypoxia/reoxygenation in vitro. J Pathol 1996;180:300–304. 31. Fassbender K, Kaptur S, Becker P, Groschl J, Hennerici M. Adhesion molecules in tissue injury: kinetics of expression and shedding and association with cytokine release in humans. Clin Immunol Immunopathol 1998;89:54–60. 32. Faller DV. Endothelial cell responses to hypoxic stress. Clin Exp Pharmacol Physiol 1999;26:74–84. 33. Muller-Kobold AC, Kallenberg CG, Tervaert JW. Monocyte activation in patients with Wegener’s granulomatosis. Ann Rheum Diseases 1999; 58:237–245. 34. Liote F, Boval-Boizard B, Weill D, Kuntz D, Wautier JL. Blood monocyte activation in rheumatoid arthritis: increased monocyte adhesiveness, integrin expression, and cytokine release. Clin Exp Immunol 1996;106:13–19. 35. Sanchez-Mejorada G, Rosales C. Signal transduction by immunoglobulin Fc receptors. J Leukoc Biol 1998;63:521–533. 36. Ericson SG, Zhao Y, Gao H, Miller KL, Gibson LF, Lynch JP, Landreth KS. Interleukin-6 production by human neutrophils after Fc-receptor cross-linking or exposure to granulocyte colony-stimulating factor. Blood 1998;91:2099–2107. 37. Kato M, Roberts-Thomson P, Phillips BG, Haynes WG, Winnicki M, Accurso V, Somers VK. Impairment of endothelium-dependent vasodilation of resistance vessels in patients with obstructive sleep apnea. Circulation 2000;102:2607–2610. 38. Kraiczi H, Caidahl K, Samuelsson A, Yuksel P, Hedner J. Impairment of vascular endothelial function and left ventricular filling: association with the severity of sleep-induced hypoxemia during sleep. Chest 2001;119; 1085–1091.
