Introduction

Obstructive sleep apnea (OSA) is a condition characterized by sleep disruption and cycles of intermittent hypoxemia, and it is a known risk factor for hypertension and cardiovascular disease [1–6]. Previous studies have demonstrated an independent relationship between the severity of OSA and the carotid artery intima-media thickness [7, 8], a marker for atherosclerosis, but the mechanisms underlying this association are not entirely understood. Immune response alterations in atherosclerosis have led it to be redefined as an inflammatory condition, a pathway operating in other chronic diseases [9, 10].

Activated nuclear factor kappaB (NF-kappaB) is a key transcription factor in chronic inflammatory diseases [11, 12]. Recently, exposure of transformed cervical cells (HeLa cells) to intermittent hypoxia was demonstrated to result in the activation of NF-kappaB [13]. This observation seemed potentially relevant to reports in OSA populations of increased circulating cytokines and proteins in the downstream pathways from NF-kappaB activation, including plasma levels of tumor necrosis factor alpha (TNF-α) and interleukin (IL)-6 [14, 15], adhesion molecule-1 and vascular cell adhesion molecule-1 [16], and increased matrix metalloproteinases [17].

Our hypothesis was that circulating blood monocytes, a key player in the inflammatory theories for atherosclerosis, would exhibit NF-kappaB activation in OSA patients. We report nuclear fraction p65 levels, one marker of NF-kappaB activation, comparing OSA patients to healthy controls, and both p65 levels and TNF-α production in peripheral blood monocytes before and after one night of treatment of OSA with nasal continuous positive airway pressure (CPAP) to test this hypothesis.

Materials and methods

Subjects

Over a 12-month period of time, nine nonsmoking men, age <50 years, with OSA symptoms but without known hypertension or presence of a morning blood pressure elevation (systolic >140 and diastolic >90 mmHg) were recruited from approximately 100 patients referred to the sleep clinic for suspected sleep apnea. Other exclusion criteria were a known or suspected diagnosis of asthma, chronic obstructive pulmonary disease, arthritis, autoimmune diseases, stroke, rhinitis, and coronary artery disease. The primary symptoms in the OSA group were excessive daytime sleepiness and/or snoring and disturbed sleep, and OSA was confirmed by full standard polysomnography (PSG; see below). Seven male nonsmoking healthy volunteers with similar recruitment and exclusion criteria were recruited as control subjects from the greater hospital community and entered the study on the basis of absence of symptoms of snoring, disturbed sleep, or observed apneas. No patient or healthy subject was on prescribed medications. Before this study, we explained its purpose and the procedures involved to the subjects, and we obtained written consent, as approved by the Human Subjects Committee of Nara Medical University.

Sleep study

PSG was performed using a computerized polysomnography system (Alice 4; Respironics; Pittsburgh, PA, USA). Data acquisition started from 9:00 p.m. and continued until 6:00 a.m. on the following morning. Apnea was defined as cessation of airflow for 10 s or more, while hypopnea was defined as a decrease in airflow by at least 50% for a minimum of 10 s or a clear decrease in airflow (>20%) followed by either oxygen desaturation of more than 3% or signs of physiological arousal. The apnea–hypopnea index (AHI) was calculated as the number of apnea–hypopnea events per hour of total sleeping time. To assess the severity of hypoxia induced by apnea–hypopnea events, we also measured the duration for which the oxygen saturation was below 90% and expressed this as a percentage of the total sleeping time (T90). We also calculated the oxygen desaturation index (ODI), defined as the number of dips in oxygen saturation by >3% per hour of sleep.

CPAP

On the night after the diagnosis of OSA by PSG, patients underwent a second PSG during which apneas were largely reduced or eliminated with nasal CPAP (REMstar Auto; Respironics).

Isolation of peripheral blood monocytes for nuclear p65 levels and TNF-α production

Heparinized whole blood (50 ml) was collected from the OSA patients in the early morning after the diagnostic PSG and also in the early morning after the CPAP treatment PSG. In the control subjects, blood was collected once in the early morning after the PSG night. From the collected blood, peripheral blood monocytes were isolated using standard methods [18]. Greater than 90% of the cells thus obtained expressed CD14, which was detected by the fluorescent antibody method with a fluorescein isothiocyanate-labeled anti-CD14 antibody. In the OSA group, the remainder was placed in culture for TNF-α production. For measurement of NF-kappaB activation, nuclear fractions were prepared from 0.5 × 106 monocytes using a Nuclear Extract Kit (Active Motif, Carlsbad, CA, USA). Levels of nuclear p65 concentrations were determined by an enzyme-linked immunosorbent assay (ELISA) kit (Active Motif). Measurement of TNF-α production was determined from short-term culture of isolated monocytes [18]. Two hundred microliters of the suspension (2.5 × 106 monocytes/ml) was poured into each well of a 96-well TC plate (3799 Corning, Corning). Lipopolysaccharide (E. coli 055:B5, Calbiochem, San Diego, CA, USA) was added to the cell suspension at a final concentration of 10 μg/ml. After the plate was cultured for 24 h at 37°C under a 5% CO2 atmosphere, the culture supernatant was collected and stored at −80°C until analysis. The TNF-α concentration was determined with a Human TNF-α ELISA kit (R&D System, Minneapolis, MN, USA).

Statistical analysis

The power of the study was projected using previous studies of TNF-α measurement in OSA populations, with consideration of the number of subjects that could be recruited given the inclusion and exclusion criteria. Data are presented as medians and interquartile range (IQR). Comparison of continuous variables between OSA patients and control subjects was performed by the Mann–Whitney test. To investigate correlations between NF-kappaB and AHI, ODI and T90, we performed Spearman’s correlation analysis. For statistical evaluation of changes in the variables in the morning before and after CPAP, the Wilcoxon signed rank test was used. A probability value of less than 0.05 was considered to indicate significance. Statistical analysis was done with SPSS software for Windows (version 10, SPSS, Chicago, IL).

Results

Values indicative of OSA (AHI, ODI, T90, and the Epworth Sleepiness Scale) and BMI were significantly different between groups. Age was not statistically different (Table 1).

Table 1 Subject’s characteristics
Full size table

Morning p65 concentration after the diagnostic PSG was significantly higher in the OSA group than in the control group [median, 0.037 ng/μl (IQR, 0.034 to 0.051) vs 0.019 ng/μl (IQR, 0.013 to 0.032); p = 0.008] and was not affected by adjustment for BMI and age (p < 0.001; Fig. 1). Using values from the OSA group alone, p65 concentration adjusted for BMI and age were found to be significantly correlated with the AHI (r = 0.77), ODI (r = 0.80), and T90 (r = 0.88) during the preceding PSG (Fig. 2).

Fig. 1
figure 1

Monocyte NF-kappaB activity as measured by p65 concentration of OSA patients and control subjects in the morning. a Unadjusted p65 concentration, b p65 concentration adjusted for BMI and age

Full size image
Fig. 2
figure 2

Scatterplots of p65 concentration adjusted for BMI and age vs apnea–hypopnea index (AHI; left), oxygen desaturation index (ODI; middle), and percentage of time spent with SpO2 < 90% (T90; right). p65 Concentration were significantly correlated with AHI (r = 0.77), ODI (r = 0.80), and T90 (r = 0.88) by Spearman’s correlation analysis

Full size image

In OSA patients on CPAP the next night, the AHI was significantly lower on CPAP [62.3 (IQR, 21.7 to 72.5) to 4.0 (IQR, 3.0 to 4.8); p < 0.001], as was ODI [57.3 (IQR, 19.7 to 62.7) to 3.9 (IQR, 3.0 to 4.5); p < 0.001]. T90 was also significantly lower [19.8% (IQR, 0.3 to 25.3) to 0% (IQR, 0 to 0); p < 0.001]. In this group, p65 concentrations upon awakening before CPAP were significantly higher in the morning compared to those after one night of CPAP morning values ∼24 h later [0.037 ng/μl (IQR, 0.034 to 0.051) to 0.020 ng/μl (IQR, 0.010 to 0.036); p = 0.03]. The post-CPAP level of p65 concentration was no different from that in control subjects the morning after the PSG. In OSA patients, TNF-α production was lower after one night of CPAP [16.26 ng/μl (IQR, 7.75 to 24.85) to 7.59 ng/μl (IQR, 5.19 to 12.95); p = 0.01; Fig. 3).

Fig. 3
figure 3

a Level of p65 concentration (nuclear fraction) in monocytes in the morning before and in the morning after CPAP. b Level of TNF-α production by cultured monocytes from the morning before and in the morning after CPAP. Symbols connected by lines indicate results from an individual patient

Full size image

Discussion

The present study shows that there is an activation of NF-kappaB in peripheral blood monocytes, as a representative transcription factor involved in regulating inflammation, and there is evidence showing that this activation is functionally reversed by one night of CPAP titration.

Of the potential mechanisms for this activation of NF-kappaB, our results implicate that events (AHI, ODI, and T90) in sleep are different from healthy subjects and reversed by one night of CPAP. Recent studies have proposed the importance of the oxidative stress in OSA in the development of cardiovascular diseases [19–25]. Reactive oxygen species (ROS) are produced during reoxygenation [26, 27], which occurs after episodes of sleep apnea, and ROS were implicated in the cell culture demonstration of NF-kappaB activation by cycles of hypoxia and reoxygenation [11, 13]. Therefore, our findings indicate this effect. CPAP acts to improve oxygenation but could have other conceivable but less likely effects, for instance effects on sleep continuity or sleep architecture.

With regard to the relevance of the association of cardiovascular disease and OSA, various pro-inflammatory cytokines, including TNF-α, IL-6, and adhesion molecules, induce the monocyte migration into the endothelial cells, after which the monocytes become foam cells and contribute to the progression of atherosclerosis [9]. NF-kappaB regulates the genes of pro-inflammatory cytokines, chemokines that attract inflammatory cells to sites of inflammation. This suggests that higher monocyte and circulating levels of inflammatory cytokines and adhesion molecules would contribute, if not initiate, atherosclerosis in those with OSA.

There are potential limitations of our study. First, the BMI of the OSA patients was higher than that of the control subjects. Although we used statistical adjustments in an attempt to eliminate the effect of BMI, we cannot completely exclude its influence. One of the issues effecting this was that we had a relatively small sample size, which handicapped our ability to control for the effect of BMI. However, we also found a decline of NF-kappaB activation and a decrease in the production of TNF-α by monocytes after only one session of CPAP, suggesting that obesity, per se, has only a small effect in this population. Unfortunately, it remains unknown what the effect of CPAP on NF-kappaB would be in clinically normal controls. It may be the case that CPAP lowers the levels of NF-kappaB in both those with OSA and those without. Second, p65 is only one marker of NF-kappaB activity and TNF-α production by monocytes is not solely dependent on NF-kappaB. Values of p65 and the impact of CPAP varied even in this highly selected population, suggesting that variations on this relationship may exist in clinical populations, including OSA in women, children, or the elderly. Third, it is known that TNF-α itself activates NF-kappaB [11]. Thus, it is still unclear what the causal relationship between a decline of NF-kappaB activation and a decrease in the production of TNF-α by monocytes after CPAP is.

Unlike better characterized diseases like asthma where atopy is a known trigger, in OSA the triggers are related to recurrent apneas during sleep, and monocyte NF-kappaB activity is modified by a CPAP treatment. These results suggest that the modification of inflammation is a specific target for treatments of OSA. The presence or development of obesity, hypertension, or hyperlipidemia may amplify the effects of sleep apnea, and, hence, the process of atherosclerosis.