Normal production of inflammatory cytokines in chronic fatigue and fibromyalgia syndromes determined by intracellular cytokine staining in short-term cultured blood mononuclear cells.
Division of Immunology, A Floor West Block, Queen's Medical Centre, Nottingham NG7 2UH, UK.
E-mail: ian.todd@nottingham.ac.uk
Keywords
Article abstract
It has been proposed that cytokines play a role in the pathogenesis of chronic fatigue syndrome (CFS) and fibromyalgia syndrome (FMS). However, different studies have reported conflicting results using enzyme-linked immunosorbent assay or polymerase chain reaction to detect cytokines in these conditions. In the present study, for the first time, the production of inflammatory [interleukin (IL)-1alpha, IL-6, and TNF-alpha] and anti-inflammatory (IL-10) cytokines by CD14+ and CD14- peripheral blood mononuclear cells (PBMC) from chronic fatigue syndrome (CFS) and fibromyalgia syndrome (FMS) patients and sex- and age-matched normal subjects was investigated at the level of individual cells using the technique of intracellular cytokine staining and flow cytometry. Cultures were carried out in the presence of polymyxin B to inhibit the effect of endotoxins on cytokine production by monocytes. The mean intensity of fluorescence (MIF) and percentage of CD14+ (monocytes) and CD14- (lymphocytes) cytokine-producing mononuclear cells were comparable in patients and controls in either unstimulated or IFN-gamma-stimulated conditions. Our study indicates that dysregulation of cytokine production by circulating monocytes or non-monocytic cells (lymphocytes) is not a dominant factor in the pathogenesis of CFS/FMS.
Article content
INTRODUCTION
The currently preferred terms for patients with unexplained fatigue and/or chronic pain are chronic fatigue syndrome (CFS) and fibromyalgia syndrome (FMS). To fulfil the criteria for FMS, an individual must have both chronic widespread pain and the presence of ‘tender points’ on examination. The current definition of CFS requires that the affected individual displays severe persistent fatigue without a defined cause, as well as the presence of four of the eight following symptoms: myalgia, arthralgia, sore throat, tender nodes, cognitive difficulty, headache, postexertional malaise or sleep disturbances [1].
There is a significant body of opinion which believes that CFS and FMS are part of a spectrum of conditions classified as neuroendocrine immune dysfunction [2]. It is likely that these chronic widespread fatigue/pain syndromes arise from interactions between the autonomic central nervous system [3,4], hypothalamic–pituitary–adrenal (HPA) axis [5–7] and the immune system [8].
Cytokines are considered to play a role in the pathogenesis and clinical manifestations of CFS/FMS. Specifically, tumour necrosis factor (TNF)-α, interleukin (IL)-1 and IL-6 are important regulators of inflammation, and may also be involved in the regulation of the HPA and the sympathetic nervous system [9]. Investigation into circulating cytokines levels has not indicated changes in CFS or FMS compared to controls [10,11]. However, the published data on the levels of cytokines secreted by cultured cells are contradictory. In fact, studies have reported increased [11–14], normal [5,10] or even decreased [15] production of inflammatory cytokines in the stimulated or unstimulated culture supernatants of peripheral blood mononuclear cells (PBMC) from CFS/FMS patients using enzyme-linked immunosorbent assay (ELISA) or polymerase chain reaction (PCR).
To analyse the quantity and quality of inflammatory and anti-inflammatory cytokine production by CD14+ (monocytes) and CD14− PBMC (mainly lymphocytes) from CFS/FMS patients, we employed intracellular cytokine staining and flow cytometric analysis to facilitate the detection of cytokine production by individual cells. This is the first time this technique has been applied to the analysis of cytokine production in CFS/FMS.
MATERIALS AND METHODS
Study subjects
The present study comprised 22 CFS/FMS patients (six males/16 females, aged 18–55 years) and 19 age- and sex-matched healthy controls (five males/14 females, aged 18–55 years). Controls and patients were divided into two groups based on whether their cells were cultured in presence or absence of polymyxin B. Three patients and four controls were analysed for cytokine production in both the presence and absence of polymyxin B. The study protocol was approved by the Ethics Committee of University Hospital Nottingham and all subjects gave informed consent. The CFS/FMS outpatients were recruited in a specialist CFS/FMS clinic at University Hospital, Nottingham, UK. All patients fulfilled either the American College of Rheumatology criteria for the classification of FMS [16] or the Oxford criteria for CFS [17], and their clinical profiles were assessed by comprehensive semistructural biopsychosocial interview conducted by D.S. Twenty-two patients were examined; two patients presented the symptoms of CFS only, four patients presented with the symptoms of FMS, while 14 patients presented with the symptoms for both CFS and FMS. This is consistent with a recently reported study in which 58% of female FMS and 80% of male FMS patients met the full criteria for CFS [18].
The controls were 19 healthy adults, with negative past, present or family history for psychiatric disorders, and all were free of any medications known to affect the immune system.
Cell separation and culture
Peripheral blood mononuclear cells (PBMC) were isolated from heparinized whole blood of normal healthy donors and CFS/FMS patients by standard density gradient centrifugation with Histopaque 1077 (Sigma, Poole, UK). PBMC were harvested from the interface, washed once with Hanks's balanced salt solution (Sigma) and once with RPMI-1640 medium (Invitrogen, Paisley, UK), centrifuging at 400 g for 10 min. The PBMC were resuspended at 1 × 106 cells/ml in RPMI-1640 medium supplemented with 10% heat-inactivated fetal calf serum (FCS) (Harlan, Loughborough, UK), 10 mm HEPES buffer (Sigma), 100 U/ml penicillin, 100 µg/ml streptomycin and 2 mm l-glutamine (all from Invitrogen).
The above cell suspensions (1-ml aliquots) were dispensed into 5 ml sterile tubes (Elkay, Hampshire, UK) with or without 1500 U/ml human recombinant (hr) IFN-γ (R&D Systems, Oxon, UK) for 6 h at 37°C in 5% CO2 with tubes at a 5° slant from the horizontal position in the presence of 10 µg/ml brefeldin A (Sigma) and with or without 50 µg/ml polymyxin B sulphate (Sigma). Cells without hrIFN-γ were cultured in the presence of brefeldin A only and considered as unstimulated cells. At the end of the incubation, 20 µl of 100 mm EDTA were added to each tube and incubated for 10 min at room temperature. Cells were washed with RPMI-1640/2% FCS, resuspended in 0·5 ml of 0·5% formaldehyde in azide-free balanced electrolyte solution (Beckman Coulter, High Wycombe, UK) and stored at 4°C overnight. We found that 6 h stimulation was the optimal time for the detection of cytokines within CD14+/CD14− cells, as culturing of cells for more than 6 h resulted in severe down-regulation of CD14 expression. The percentage of CD14+ cells was greatly reduced by 12 and 24 h of culture without stimulation in the presence of brefeldin A only. Reduction of the monocyte population was also observed in the light scatter cytogram, indicating that cell death rather than CD14 marker down-regulation was responsible for this reduction. The same results were obtained with PBMC from the patient group (data not shown).
Antibodies
The following monoclonal antibodies were used: phycoerythin (PE)-rat antihuman IL-10, fluorescein isothiocyanate (FITC)-rat antihuman IL-6 and FITC-mouse antihuman IL-1α (all from Becton Dickinson, Oxford, UK); PE-mouse antihuman TNF-α and allophycocyanin (APC)-mouse antihuman CD14 (Beckman Coulter); mouse IgG1-PE and mouse IgG1-FITC (Beckman Coulter) were used as isotype controls.
Intracellular cytokine staining
Intracellular cytokine staining was performed as described previously [19]. Briefly, the fixed cells were washed three times with cold PBA [phosphate buffered saline (PBS) with 0·5% bovine serum albumin and 0·1% sodium azide], saponin buffer [PBA/0·04% saponin (Sigma)] and 10% FCS in saponin buffer, respectively, at 300 g for 5 min at 4°C. Fluorochrome-conjugated antibodies were mixed and added to the cells after the last wash, followed by incubation at 4°C in the dark for 2 h with occasional shaking. Fluorochrome-conjugated, isotype-matched IgG1 was used as a control for detecting non-specific binding. Cells were then washed three times with saponin buffer and fixed with 0·5 ml of 0·5% formaldehyde in azide-free balanced electrolyte solution (Beckman Coulter).
Flow cytometric analysis
Cells were analysed using a Coulter EPICS Altra flow cytometer (Beckman Coulter). For each tube, 150 000 events were collected in a gate created around the viable lymphocyte and monocyte populations by forward and side scatter characteristics. Dot plots of CD14 versus cytokine staining (TNF-α, IL-1α, IL-6 and IL-10) were created. Quadrants were applied to the isotype control dot plots and medium-control dot plots to exclude nonspecific staining of cells. The same quadrants were used for the dot plots of the stimulated cells. The reproducibility of the method was confirmed by analysing cells from the same individual on separate occasions.
Statistical analysis
Comparisons between two sample populations were made with the non-parametric Mann–Whitney U-test. Comparison within one sample population was assessed by Wilcoxon signed rank test. Correlation between cytokine-producing cells and disease activity were tested using Spearman's correlation analysis.
RESULTS
Viable PBMCs were gated electronically in the flow cytometer and cytokine production by CD14+ and CD14− cells was ascertained in CFS/FMS patients and normal individuals. Preliminary experiments revealed that when cells were cultured in medium supplemented with 10% FCS in the presence of brefeldin A, a high proportion of intracellular cytokine-positive CD14+ cells were detected in both patients and controls (Table 1). This was particularly the case for expression of TNF-α and IL-1α, with lower expression of IL-6 and IL-10. Culturing the cells in presence of polymyxin B dramatically reduced the proportion of cytokine positive cells and the mean intensity of fluorescence (MIF) of staining in both patients and controls, particularly the production of inflammatory cytokines (IL-1α, TNF-α and IL-6) in the CD14+ monocyte population (Table 2). Following 6 h culture with polymyxin B, the median percentage of cytokine-positive cells and (MIF) for intracellular TNF-α, IL-1α and IL-6 produced by CD14+ cells reduced from 43% (327), 31% (254) and 3·1% (126) to 1·2% (77), 0·18% (122) and non-detectable levels in the normal control group; and from 44% (194), 44% (225), and 2·5% (98) to 1·4% (76), 0·22% (146) and non-detectable levels in the patient group, respectively (Tables 1 and 2). However, the MIF and percentage of CD14+ and CD14− cytokine-producing cells were comparable in patients and controls when cells were cultured for 6 h without stimulation either in presence or absence of polymyxin B (P > 0·05) (Tables 1–3).
Stimulation with phorbol myristate acetate and ionomycin (PMA/I) resulted in severe cell death and down-regulation of CD14 expression (data not shown). Therefore, a more physiological stimulator, namely hrIFN-γ, was chosen. A summary of the production of inflammatory and anti-inflammatory cytokines by CD14+ and CD14− cells in CFS/FMS patients and controls following 6 h stimulation with hrIFN-γ in the presence of polymyxin B is shown in Tables 2 and and3,3, respectively. MIF and percentage of intracellular TNF-α and IL-1α-producing monocytes increased significantly after 6 h stimulation in both patients and controls (P < 0·05) (Table 2). However, hrIFN-γ stimulation significantly increased production of only TNF-α, but not IL-1α, in the CD14− cells of both patients and controls (Table 3). Intracellular IL-6 was undetectable in either stimulated or unstimulated CD14+ cells of the patient and control groups in presence of polymyxin B and IL-6 production by CD14− cells was very low (Tables 2 and 3). Examination of the IL-10 response of CD14+ and CD14− cells to hrIFN-γ showed that cells did not respond in either CFS/FMS patients or controls (P > 0·05). The MIF and percentage of cytokine-producing CD14+ and CD14− cells were comparable between patients and controls for all the cytokines measured in hrIFN-γ stimulated conditions (P > 0·05).
There was no significant correlation between the frequency of cytokine-producing cells and fatigue/pain severity or the impact of disease within the patient group, which were measured by the Chalder Fatigue Scale [20], the Brief Pain Inventory [21] and the Work and Social Adjustmant Scale [22].
DISCUSSION
In the present study, using flow cytometry, the percentage of cells and MIF of inflammatory and anti-inflammatory cytokines from CFS/FMS patients and healthy controls were determined and compared at the individual cell level. The main findings are that the MIF and percentage of CD14+ (monocytes) and CD14− (lymphocytes) cytokine-producing cells were comparable between patients and controls when cells were cultured for 6 h with or without hrIFN-γ stimulation and either in the presence or absence of polymyxin B.
It has been hypothesized that abnormal production of cytokines may play a role in the pathogenesis of CFS/FMS. However, several studies have reported conflicting results with respect to the levels of certain cytokines in the circulation and in culture supernatants using ELISA or PCR. Chao et al. [13] reported an increase in LPS-stimulated release of IL-1β, IL-6 and TNF-α from PBMC, but not in serum, from CFS patients versus control subjects. In contrast, Swanink et al. [15] showed that endotoxin-stimulated ex vivo production of TNF-α and IL-1β was significantly lower in CFS; however, there was no obvious difference in circulating cytokines and ex vivo production of IL-1α. Gupta et al. [14] investigated the levels of cytokines in the supernatant of adherent (monocytes) and non-adherent (lymphocytes) mononuclear cells cultured in the presence or absence of LPS or phytohaemagglutinin (PHA). The levels of IL-6 produced spontaneously by both monocytes and lymphocytes from CFS patients were significantly increased, while those of IL-10 decreased. Furthermore, a recent report from the National Institute of Health on 13 FMS patients and eight healthy controls showed no differences in IL-6 serum levels [5]. Finally, no sign of monocyte activation, as a main source of inflammatory cytokines, was reported in FMS patients by measuring the levels of neopterin [23].
Although these conflicting finding may have resulted from differences in experimental approaches (e.g. differences in capture antibodies or effects of serum-derived blocking agents in ELISA) or differences in patients’ conditions, patients consistently show normal levels of cytokines in the circulation compared to PBMC culture supernatants. In fact, discrepancies always arise between data derived from the levels of cytokines produced by cultured mononuclear cells either in stimulated or unstimulated conditions [10–12]. Overall, a review of the literature suggests that in vivo levels of proinflammatory cytokines in CFS/FMS are normal, but that the PBMC of these patients can, in some circumstances, produce increased levels of these cytokines in vitro. This could be a result of other factors associated with the pathophysiology of CFS/FMS affecting PBMC, but in a way that becomes manifest only in the artificial in vitro culture environment. Because our protocol involved a very brief ex vivo culture of 6 h, it is not surprising that we found normal levels of cytokine production by PBMC of CFS/FMS patients, as observed by others for circulating cytokines in vivo.
Monocytes/macrophages are the main source of inflammatory cytokines IL-1, IL-6 and TNF-α. We found that culturing monocyte for more than 6 h dramatically reduced the frequency of these cells as a result of apoptosis. Therefore, an accurate evaluation of increased or decreased levels of cytokines in PBMC cultured for more than 6 h is highly problematical.
The dramatic reduction of MIF and percentage of intracellular cytokine positive cells (particularly for TNF-α and IL-1α) in the monocytes of both patients and controls in the presence of polymyxin B indicates that LPS contaminated the cultures: this was despite all possible efforts to ensure that the cell cultures were performed in endotoxin-free conditions. However, it is notable that LPS at very low concentrations is able to bind to CD14 on the surface of monocytes [24] within 1 min of exposure, and is internalized within 5 min. Binding is dependent on the presence of LPS-binding protein, which is a normal plasma component. Therefore, 5–10 min of exposure to physiologically relevant concentrations of LPS are sufficient to trigger maximal TNF-alpha release by monocytes [25]. Polymyxin B, an inhibitor of LPS-mediated activation, essentially abrogated the LPS-binding protein- and CD14-dependent binding of LPS to monocytes. Our data suggest that the total removal of LPS from our system was not possible and, by considering the predominant effect of LPS on monocytes, culturing of monocytes in the absence of polymyxin B may have resulted to false positive cytokine production. This was magnified by the highly sensitive nature of the intracellular cytokine staining technique, which is able to detect very low quantities of cytokines within individual cells.
Although some CFS/FMS symptoms may be due directly to inflammatory cytokines [26], our study indicates that dysregulation of cytokines derived from circulating monocytes or nonmonocytic cells (lymphocytes) is not a dominant factor in the pathogenesis of these syndromes. However, it does not exclude totally a role for cytokines in the pathogenesis of CFS/FMS for the following reasons: first, the results obtained in either ex vivo or in vitro studies do not necessarily represent the in vivo conditions. Secondly, some neuropsychiatric symptoms in patients with CFS/FMS could be related more closely to disordered cytokine production within tissues, e.g. by glial cells within the CNS [27], than to production of cytokines in the circulation.