International Immunology, Vol. 26, No. 4, pp. 233–242 doi:10.1093/intimm/dxt068 Advance Access publication 16 December 2013
© The Japanese Society for Immunology. 2013. All rights reserved. For permissions, please e-mail: [email protected]
Role of adaptive and innate immune cells in chronic fatigue syndrome/myalgic encephalomyelitis Ekua Weba Brenu1,2*, Teilah K. Huth2, Sharni L. Hardcastle2, Kirsty Fuller2, Manprit Kaur2, Samantha Johnston2, Sandra B. Ramos2, Don R. Staines2,3 and Sonya M. Marshall-Gradisnik1,2 School of Medical Science, Griffith University, Gold Coast, QLD 4215, Australia. The National Centre for Neuroimmunology and Emerging Diseases, Griffith Health Institute, Griffith University, Gold Coast, QLD 4215, Australia. 3 Queensland Health, Gold Coast Public Health Unit, Robina, QLD 4230, Australia. 1 2
*Correspondence to: E. W. Brenu, Centre for Medicine and Oral Health, National Centre for Neuroimmunology and Emerging Diseases, Microbiology and Immunology Research Group, Griffith University, Southport, QLD 4215, Australia; E-mail: [email protected]
Abstract Perturbations in immune processes are a hallmark of a number of autoimmune and inflammatory disorders. Chronic fatigue syndrome/myalgic encephalomyelitis (CFS/ME) is an inflammatory disorder with possible autoimmune correlates, characterized by reduced NK cell activity, elevations in regulatory T cells (Tregs) and dysregulation in cytokine levels. The purpose of this article is to examine innate and adaptive immune cell phenotypes and functional characteristics that have not been previously examined in CFS/ME patients. Thirty patients with CFS/ME and 25 non-fatigued controls were recruited for this study. Whole blood samples were collected from all participants for the assessment of cell phenotypes, functional properties, receptors, adhesion molecules, antigens and intracellular proteins using flow cytometric protocols. The cells investigated included NK cells, dendritic cells, neutrophils, B cells, T cells, γδT cells and Tregs. Significant changes were observed in B-cell subsets, Tregs, CD4+CD73+CD39+ T cells, cytotoxic activity, granzyme B, neutrophil antigens, TNF-α and IFN-γ in the CFS/ME patients in comparison with the non-fatigued controls. Alterations in B cells, Tregs, NK cells and neutrophils suggest significant impairments in immune regulation in CFS/ME and these may have similarities to a number of autoimmune disorders. Keywords: γδT cells, B cells, degranulation, dendritic cells, NK cells
Introduction Chronic fatigue syndrome/myalgic encephalomyelitis (CFS/ ME) remains a controversial disorder in research and clinical medicine. Persistent debilitating fatigue and frequent episodes of flu-like symptoms that are not alleviated by rest or medication are among the hallmarks of this enervating disorder (1). Although some therapies are suggested to be successful in some patients, the vast majority of patients do not benefit from clinical interventions (2–7). Currently, six classification systems may be used to identify individuals as having CFS/ME and patients may be grouped into different subgroups based on the severity or incapacitating nature of their symptoms (1, 8–10). In spite of exhaustive research in recent decades, the cause(s) of CFS/ME is not yet completely understood, while improved prognostic indicators, more-efficient targeted treatment and better coping approaches continue to be elusive. Coordination between the innate and adaptive immune system is necessary for health and wellness as it ensures cell survival, maturation, proliferation and functionality. Disruptions in these synchronized processes are hallmarks
of most autoimmune and inflammatory disorders and may include decreases in NK cell lysis, shifts in cytokine profile, proliferation in certain T cells and abnormal cell functions (11). Failures in cellular processes significantly affect physiological homeostasis and have been confirmed in a number of CFS/ME patients (12–14). Importantly, in CFS/ME, reports on basic immune cell numbers are inconsistent and this may be related to a number of factors including the severity and cyclical nature of the disorder. Nonetheless, reduced NK cytotoxic activity remains a consistent finding in most CFS/ME patients, perhaps suggesting a possible breakdown in the cytotoxic mechanism of these cells in CFS/ME. Reductions in NK cell function and significant increases in regulatory T cells (Tregs) in CFS/ME patients (14) may suggest an increase in Treg suppression, which may explain the reduced cytotoxic ability of the NK cells in CFS/ME patients. Similarly, heightened Treg levels may substantially suppress antigen-presenting cells by binding to co-receptors CD80 and CD86 and MHCII molecules on the surface of dendritic cells (DCs) via ligands including cytotoxic T lymphocyte
Downloaded from http://intimm.oxfordjournals.org/ at Fudan University on May 8, 2015
Received 23 June 2013, accepted 25 November 2013
234 Immune cells in CFS/ME
Methods Participant recruitment Thirty CFS/ME patients (mean age = 51.15 ± 1.92 years) and 25 non-fatigued controls (mean age = 50.42 ± 1.76 years) were recruited for this study. Prior to participation in the study, each participant provided informed consent. All participants completed a questionnaire based on the Centre for Disease Prevention and Control (CDC 1994) criteria for CFS/ME (1, 8). CFS/ME patients qualified for the study upon meeting the CDC 1994 criteria for CFS/ME, while the non-fatigued group had no incidence of fatigue. An exclusionary criterion was pertinent to individuals with autoimmune disorders, psychosis, epilepsy, cardiac related disorders, pregnant or breastfeeding. Approval for this study was obtained from the Griffith University Human Research Ethics Committee. Sample collection Whole blood was collected from the antecubital vein of all participants into EDTA and lithium heparinized tubes. Routine full blood counts were performed to provide an indication of the levels of red blood cells, lymphocytes, granulocytes, monocytes and other blood parameters, using the Coulter ACT diff Analyzer (Beckman Coulter, High Wycombe, UK).
Cell phenotyping studies All cell phenotyping studies were performed using whole blood samples. Cell phenotyping studies were performed for the following cells: DCs [plasmacytoid (pDCs), myeloid (mDCs) and CD16+ DCs], NK cells (CD56dim and CD56bright), B cells (immature, memory and plasma), γδT cells (naive, central memory, effector memory and effector memory CD45RA), Tregs (FOXP3+Tregs) and CD4−T cells (CD4+CD73+CD39−, CD4+CD73+CD39+, and CD4+CD73−CD39+). Immunoflu orescent antibody staining was used to determine the phenotypes of each cell. The staining procedures were executed as previously described (32, 33). The mAbs (BD Biosciences, San Jose, CA, USA) used in determining cell phenotypes of interest included those recognizing CD3, CD19, CD21, CD20, CD27, HLA-DR, CD123, CD33, CD16, CD11c, CD62L, CD27, CD45RA, CD127, CD4, CD25, CD73, CD39, CD56, CD80, CD86, CD56, CD38, CD138 or CD10 (32, 33). Assessment of surface antigens, adhesion molecules and receptors Neutrophils were assessed based on the expression of HNAs including HNA-1 (CD16b), HNA-2 (CD177), HNA-4 (CD11b) and HNA-5 (CD11a). Whole blood samples were labelled with combinations of antibodies recognizing CD11a, CD62L, CD11b, CD16b, CD66abcd or CD177. NK receptors were measured following negative isolation of NK cells from whole blood using Rosettesep (StemCell Technologies, Vancouver, Canada). Isolated cells were then stained with CD3, CD16, CD56 and mAbs for CD158a/h (KIR2DL1/S1), CD158e (KIR3DL1), CD158b (KIR2DL2/DL3), CD158i (KIR2DS4) (Miltenyi Biotech, Bergisch Glabach, Germany) or NKG2D (BD Biosciences) (11). The migratory potential of γδT cells was examined using the CD11a and CD62L markers, while CD94 determined MHC recognition. Expression of co-stimulatory receptors CD80 and CD86 was used as a determinant of DC function. This was performed as previously described (34). Heparinized whole blood was diluted with RPMI at a ratio of 1:2 and stimulated with or without PMA (5 ng/ml)/ionomycin (1 μg/ml) and incubated for 6 h at 37°C with 5% CO2. Samples were then labelled with anti-HLA-DR, lineage cocktail 2 (CD3, CD14, CD19, CD20, CD56) and either CD80 or CD86 (BD Biosciences). Subsequently, samples were lysed, washed and fixed. All cells were then analysed on the flow cytometer. Intracellular staining of functional proteins Examination of NK proteins and FOXP3 was performed via intracellular staining (35). Briefly, PBMCs (1 × 107 cells/ml) were stained and incubated with fluorochrome-conjugated mAbs CD56, CD16, CD3, CD25, CD127 and CD4 (BD Biosciences). Samples were washed, resuspended and incubated in BD Cytofix (BD Biosciences) for 30 min, following which they were washed in diluted perm wash buffer (BD Biosciences). mAbs for the proteins of interest, perforin, granzyme A (GZA), granzyme B (GZB) and FOXP3 were then added to the appropriate samples, incubated, washed and analysed on the flow cytometer.
Downloaded from http://intimm.oxfordjournals.org/ at Fudan University on May 8, 2015
antigen-4 (CTLA-4) and lymphocyte activation gene-3 (LAG3), respectively (15). The suppressive functions of Tregs on DCs may be important in CFS/ME as a lack of mature DCs may affect the antigen-presenting function of DCs, particularly altering cytokine secretion by T helper (Th) cells (16). Interestingly, co-stimulatory molecules such as CD86 have been shown to be either decreased or increased on the surfaces of DCs in certain diseases (17, 18). In CFS/ME, little is known about the role of DCs. Given that DCs are central to a collapse in immune tolerance, it is imperative to determine their role in CFS/ME. DCs are known to incite the generation of autoreactive immune responses resulting in the generation of pathogenic autoantibodies and chronic inflammation (19–21). DCs also interact with other antigen-presenting cells including γδT cells and this is a potent mechanism for CD86 expression (22). In the periphery, γδT cells represent about 1–10% of lymphocytes and 2–5% of the total T-cell populace (23, 24). In humans, there are two subsets of γδT cells, Vδ1 and Vδ2. Vδ1 is predominant in the spleen and thymus, while Vδ2 is abundant in the periphery, skin, tonsils and lymph nodes (25, 26). γδT cells are functionally similar to cytotoxic cells, NK cells and CD8+T cells (27). These cells also secrete cytokines that inhibit CD4+T-cell proliferation, induce apoptosis and promote B-cell antibodies and have been implicated in a number of diseases (28–30), hence, they may have a role in CFS/ME. Peripheral γδT cells may decrease in the presence of high levels of FOXP3 (31). Direct confirmation of the role of these cells in regulating immune mechanisms in CFS/ME remains to be determined. Similarly, there are no reports on the role of human neutrophil antigens (HNAs) in CFS/ME patients. This is the first study to investigate the potential role of innate and adaptive immune cell phenotypes and functional proteins of γδT cells, DCs and HNAs in the CFS/ME symptom profile.
Immune cells in CFS/ME 235 NK cell cytotoxic activity, degranulation and IFN-γ
Th1/Th2/Th17 cytokines following mitogenic stimulation
NK cytotoxic activity and degranulation were examined as a measure of NK cell function. Cytotoxic activity was assessed as previously described (13, 14, 36). Briefly, PBMCs were isolated from whole blood samples following density gradient centrifugation. PBMCs (1 × 106 cells/ml) were labelled with 0.4% Paul Karl Horan (PKH)-26 fluorescent cell linker dye (Sigma-Aldrich, St Louis, USA). Cells were incubated with K562 tumour cells (1 × 105 cells/ml) for 4 h at three effector to target ratios: 12.5:1, 25:1 and 50:1. A control sample containing no PBMCs was also included as a determinant of cells undergoing spontaneous apoptosis. In advance of flow cytometric analysis, cells were stained with annexin V and 7-aminoactinomycin D (BD Pharmingen, San Diego, CA, USA). The percentage lysis was calculated for each sample at the different ratios as described previously (13, 14, 36). CD107a and IFN-γ production under two stimulatory conditions was used as a measure of NK cell degranulation. In short, cells were stained with CD107a, monensin (BD Biosciences) and Brefeldin A (Golgi Plug; BD Biosciences) with or without K562 cells (1 × 105/ml) or PMA/I [2.5 μg/ml of PMA and 1 µg/ml of ionomycin (Sigma–Aldrich)] (37, 38). Samples were incubated at 37°C with 5% CO2 for 6 h. Intracellular staining was then performed as described above for the detection of IFN-γ.
The levels of Th1/Th2/Th17 cytokines were examined via a cytometric bead array (CBA; BD Pharmingen) following mitogenic stimulation of PBMCs. Briefly, PBMCs isolated from whole blood were placed in cell culture for 72 h at a cell concentration of 1 × 106 cells/ml in RPMI culture media with or without 1 µg of PHA at 37°C with 5% CO2. After 72 h, cell supernatants were removed and stored at −80°C for later assessment. Th1, Th2 and Th17 cytokine profiles including IL-2, IL-4, IL-6, IL-10, TNF-α, INF-γ and IL-17A were determined using a CBA kit according to the manufacturer’s instructions (35, 39). Flow cytometric analysis
Table 1. Combinations of various antibodies used in examining the different cell attributes Cell type
B cells
γδT cells
Tregs CD4+T cells NK cells
Dendritic cells
Neutrophils
KIRexpressing cells
Antibody combinations Immature
Memory
Plasma cells
CD19-APC CD20-PerCP CD21-FITC CD10-PE γδ1T cells γδ1-TCR FITC CD3-PerCP CD45RA-APC CD27-PE CD4-FITC CD4-FITC Phenotypes CD3-PerCP CD56-PE CD16-FITC
CD19-APC CD20-PerCP CD21-FITC CD27-PE γδ2T cells γδ2-TCR-PE CD3-PerCP CD45RA-APC CD27-FITC CD25-APC CD39-APC Degranulation CD3-PerCP CD56-PE CD107a-FITC IFN-γ-APC pDC Lin2-FITC HLA-DR-PERCP CD123-PE
CD19-PerCP CD138-APC CD38-FITC CD27-PE CD3-PerCP γδTCR-APC CD11a-FITC CD62L-PE CD127-PerCP-CY-5.5 CD73-PE Cytotoxic activity PKH-26 7-AAD Annexin-V-FITC
FOXP3-PE CD127 PerCP-Cy 5.5
CD80 expression Lin2-FITC HLA-DR-PERCP CD80-PE
CD86 expression Lin2-FITC HLA-DR-PERCP CD86-PE
HNA-2 CD16-PE-Cy-5 CD62L-APC CD66-PE CD177-FITC KIR3DL1 CD158e-APC CD56-PE CD16-FITC CD3-PerCP
HNA-3 CD16-PE-Cy-5 CD62L-APC CD66-FITC CD11b/18-PE KIR2DL2/L3 CD158b-APC CD56-PE CD16-FITC CD3-PerCP
HNA-4 CD16-PE-Cy-5 CD62L-APC CD66-FITC C11a-PE KIR2DS4 CD158i-APC CD56-PE CD16-FITC CD3-PerCP
mDC Lin2-FITC HLA-DR-PERCP CD11c-PE CD33-APC HNA-1 CD16-PE-Cy-5 CD62L-APC CD66-FITC CD16b-PE KIR2DL1 CD158a-APC CD56-PE CD16-FITC CD3-PerCP
CD3-PerCP γδTCR-APC CD94-PE
NKG2D NKG2D-APC CD56-PE CD16-FITC CD3-PerCP
The table shows a representation of all the antibodies used in identifying cell phenotypes, receptors and function. AAD, aminoactinomycin D; APC, allophycocyanin; FITC, fluorescein isothiocyanate; KIR, killer cell immunoglobulin-like receptor; PE, phycoerythrin; PerCP, peridinin chlorophyll protein complex.
Downloaded from http://intimm.oxfordjournals.org/ at Fudan University on May 8, 2015
All experiments were analysed on a dual laser four colour flow cytometer, BD FACSCalibur (BD Biosciences). The specific combination of mAbs for each assay is presented in Table 1, while the gating strategies for cell phenotypes are presented in Supplementary Figures 1–6, available at International Immunology Online. The quantity for each phenotype was enumerated using the percentage of gated events from flow cytometric analysis and absolute lymphocyte count from haematological counts (32).
236 Immune cells in CFS/ME Statistical analysis Statistical analysis was performed using SPSS software version 16.0 (SPSS Inc., Chicago, IL, USA) and Graph Pad Prism software version 6.0 (Graph Pad Software, Inc., San Diego, CA, USA). Pairwise comparison using multivariate testing was used to perform comparative assessments of all data generated from all participants (CFS/ME and non-fatigued control subjects). The analysis of variance test and independent sample t-test were used to determine the significance of changes in the variables measured. The Bonferroni method was used as post-hoc analysis to assess changes in the data. A test for normality was performed using Kolmogorov– Smirnov and Shapiro–Wilk normality tests. Spearman’s rank correlation was the non-parametric test used to examine correlations in the data generated. Data are represented as mean ± standard error of the mean. All significant results had P values ≤ 0.05.
Patient blood characteristics Prior to all flow cytometric testing of immune cell phenotypes, all participants were assessed on blood characteristics (Table 2). There were significant differences between CFS/ ME patients and the non-fatigued controls on measures of blood parameters including platelets, haematocrit and erythrocyte sedimentation rate. All other parameters remained unchanged. All CFS/ME patients met the CDC 1994 criteria for CFS/ME. The duration of CFS/ME among the patients was over 2 years. Alterations in Tregs, B cells and DC subsets in CFS/ME The total number of lymphocytes, monocytes or neutrophils did not differ between the two population groups. Similarly, routine blood assessments demonstrated similarities in all blood characteristics measured (Table 2). A significant difference in cell subsets was observed in DCs, B-cell phenotypes and Tregs [Fig. 1(a–c)]. pDCs were significantly decreased in the CFS/ME group, while the mDCs although elevated in the CFS/ME group did not achieve statistical significance (Fig. 1a). CFS/ME patients showed a significant decrease in
Decreases in antigens with no significant changes in surface receptors in CFS/ME HNA-2 (CD177+) was significantly reduced on the surface of neutrophils from CFS/ME patients in comparison with the nonfatigued controls (Fig. 2). When we examined the expression of co-stimulatory markers, CD80 and CD86 pre- and poststimulation, we failed to observe any significant changes in the expression of these markers on the surfaces of DCs. Equally, there were no significant differences in the expression of killer cell immunoglobulin-like receptors (KIRs) or NKG2D on the surfaces of isolated NK cells between the two groups of participants. Similarly, there were no significant differences observed in the following cell surface molecules on γδT cells: CD62Lnegative, CD62Llow, CD62Lhigh, CD11alow, CD11ahigh, CD94+ and CD94−. Reduced NK cell function in CFS/ME Reduced cytotoxic activity may be a prominent hallmark of CFS/ME as it has consistently been reported in CFS/ME patients. In this study, we confirmed significant reductions in the ability of NK cells from CFS/ME patients to lyse K562 cells, at 25:1 and 50:1 effector:target ratios (Fig. 3a). Conversely, degranulation was increased in CFS/ME patients in PMA/I and K562 samples (Fig. 3b). Similarly, IFN-γ-producing CD3−CD56+NK cells were increased in the CFS/ME patients compared with the non-fatigued controls (Fig. 3d). As cytotoxic activity is routinely decreased in CFS/ME patients, we assessed the intracellular levels of these lytic proteins (perforin, GZA and GZB) in NK and T cells. There were no significant differences in the levels of perforin or GZA in the NK cells of both CFS/ME patients and the non-fatigued controls. However, GZB was significantly reduced, while IFN-γ levels in CD3−CD56+NK cells were increased in the CFS/ME patients (Fig. 3c).
Table 2. Comparative assessment of blood characteristics in CFS/ME patients and non-fatigued controls Blood Markers
CFS/ME Patients (n=30)
Controls (n=25)
P values
Haemoglobin (g/l) White Cell Count (×109/l) Platelets (×109/l) Haematocrit Red Cell Count (×1012/l) Mean Cell volume (fl) Neutrophils (×109/l) Lymphocytes (109/l) Monocytes (×109/l) Eosinophils (×109/l) Basophils (×109/l) Erythrocyte sedimentation rate (mm/Hr)
134.43 ± 2.41 6.41 ± 0.33 274.61 ± 13.11 0.41 ± 0.006 4.46 ± 0.08 91.86 ± 0.57 3.90 ± 0.27 1.92 ± 0.10 0.33 ± 0.018 0.16 ± 0.019 0.03 ± 0.004 17.22 ± 2051
140.67 ± 2.56 6.06 ± 0.31 238.38 ± 9.39 0.43 ± 0.007 4.65 ± 0.09 92.08 ± 0.51 3.84 ± 0.26 1.73 ± 0.09 0.32 ± 0.02 0.15 ± 0.02 0.03 ± 0.004 10.48 ± 1.70
0.08 0.51 0.05 0.05 0.12 0.65 0.95 0.20 0.64 0.69 0.53 0.05
The table shows representative data of all the blood characteristics obtained from full blood count analysis. The units of measurement are specified in parentheses. Data are represented as means ± standard error of the mean. P
© The Japanese Society for Immunology. 2013. All rights reserved. For permissions, please e-mail: [email protected]
Role of adaptive and innate immune cells in chronic fatigue syndrome/myalgic encephalomyelitis Ekua Weba Brenu1,2*, Teilah K. Huth2, Sharni L. Hardcastle2, Kirsty Fuller2, Manprit Kaur2, Samantha Johnston2, Sandra B. Ramos2, Don R. Staines2,3 and Sonya M. Marshall-Gradisnik1,2 School of Medical Science, Griffith University, Gold Coast, QLD 4215, Australia. The National Centre for Neuroimmunology and Emerging Diseases, Griffith Health Institute, Griffith University, Gold Coast, QLD 4215, Australia. 3 Queensland Health, Gold Coast Public Health Unit, Robina, QLD 4230, Australia. 1 2
*Correspondence to: E. W. Brenu, Centre for Medicine and Oral Health, National Centre for Neuroimmunology and Emerging Diseases, Microbiology and Immunology Research Group, Griffith University, Southport, QLD 4215, Australia; E-mail: [email protected]
Abstract Perturbations in immune processes are a hallmark of a number of autoimmune and inflammatory disorders. Chronic fatigue syndrome/myalgic encephalomyelitis (CFS/ME) is an inflammatory disorder with possible autoimmune correlates, characterized by reduced NK cell activity, elevations in regulatory T cells (Tregs) and dysregulation in cytokine levels. The purpose of this article is to examine innate and adaptive immune cell phenotypes and functional characteristics that have not been previously examined in CFS/ME patients. Thirty patients with CFS/ME and 25 non-fatigued controls were recruited for this study. Whole blood samples were collected from all participants for the assessment of cell phenotypes, functional properties, receptors, adhesion molecules, antigens and intracellular proteins using flow cytometric protocols. The cells investigated included NK cells, dendritic cells, neutrophils, B cells, T cells, γδT cells and Tregs. Significant changes were observed in B-cell subsets, Tregs, CD4+CD73+CD39+ T cells, cytotoxic activity, granzyme B, neutrophil antigens, TNF-α and IFN-γ in the CFS/ME patients in comparison with the non-fatigued controls. Alterations in B cells, Tregs, NK cells and neutrophils suggest significant impairments in immune regulation in CFS/ME and these may have similarities to a number of autoimmune disorders. Keywords: γδT cells, B cells, degranulation, dendritic cells, NK cells
Introduction Chronic fatigue syndrome/myalgic encephalomyelitis (CFS/ ME) remains a controversial disorder in research and clinical medicine. Persistent debilitating fatigue and frequent episodes of flu-like symptoms that are not alleviated by rest or medication are among the hallmarks of this enervating disorder (1). Although some therapies are suggested to be successful in some patients, the vast majority of patients do not benefit from clinical interventions (2–7). Currently, six classification systems may be used to identify individuals as having CFS/ME and patients may be grouped into different subgroups based on the severity or incapacitating nature of their symptoms (1, 8–10). In spite of exhaustive research in recent decades, the cause(s) of CFS/ME is not yet completely understood, while improved prognostic indicators, more-efficient targeted treatment and better coping approaches continue to be elusive. Coordination between the innate and adaptive immune system is necessary for health and wellness as it ensures cell survival, maturation, proliferation and functionality. Disruptions in these synchronized processes are hallmarks
of most autoimmune and inflammatory disorders and may include decreases in NK cell lysis, shifts in cytokine profile, proliferation in certain T cells and abnormal cell functions (11). Failures in cellular processes significantly affect physiological homeostasis and have been confirmed in a number of CFS/ME patients (12–14). Importantly, in CFS/ME, reports on basic immune cell numbers are inconsistent and this may be related to a number of factors including the severity and cyclical nature of the disorder. Nonetheless, reduced NK cytotoxic activity remains a consistent finding in most CFS/ME patients, perhaps suggesting a possible breakdown in the cytotoxic mechanism of these cells in CFS/ME. Reductions in NK cell function and significant increases in regulatory T cells (Tregs) in CFS/ME patients (14) may suggest an increase in Treg suppression, which may explain the reduced cytotoxic ability of the NK cells in CFS/ME patients. Similarly, heightened Treg levels may substantially suppress antigen-presenting cells by binding to co-receptors CD80 and CD86 and MHCII molecules on the surface of dendritic cells (DCs) via ligands including cytotoxic T lymphocyte
Downloaded from http://intimm.oxfordjournals.org/ at Fudan University on May 8, 2015
Received 23 June 2013, accepted 25 November 2013
234 Immune cells in CFS/ME
Methods Participant recruitment Thirty CFS/ME patients (mean age = 51.15 ± 1.92 years) and 25 non-fatigued controls (mean age = 50.42 ± 1.76 years) were recruited for this study. Prior to participation in the study, each participant provided informed consent. All participants completed a questionnaire based on the Centre for Disease Prevention and Control (CDC 1994) criteria for CFS/ME (1, 8). CFS/ME patients qualified for the study upon meeting the CDC 1994 criteria for CFS/ME, while the non-fatigued group had no incidence of fatigue. An exclusionary criterion was pertinent to individuals with autoimmune disorders, psychosis, epilepsy, cardiac related disorders, pregnant or breastfeeding. Approval for this study was obtained from the Griffith University Human Research Ethics Committee. Sample collection Whole blood was collected from the antecubital vein of all participants into EDTA and lithium heparinized tubes. Routine full blood counts were performed to provide an indication of the levels of red blood cells, lymphocytes, granulocytes, monocytes and other blood parameters, using the Coulter ACT diff Analyzer (Beckman Coulter, High Wycombe, UK).
Cell phenotyping studies All cell phenotyping studies were performed using whole blood samples. Cell phenotyping studies were performed for the following cells: DCs [plasmacytoid (pDCs), myeloid (mDCs) and CD16+ DCs], NK cells (CD56dim and CD56bright), B cells (immature, memory and plasma), γδT cells (naive, central memory, effector memory and effector memory CD45RA), Tregs (FOXP3+Tregs) and CD4−T cells (CD4+CD73+CD39−, CD4+CD73+CD39+, and CD4+CD73−CD39+). Immunoflu orescent antibody staining was used to determine the phenotypes of each cell. The staining procedures were executed as previously described (32, 33). The mAbs (BD Biosciences, San Jose, CA, USA) used in determining cell phenotypes of interest included those recognizing CD3, CD19, CD21, CD20, CD27, HLA-DR, CD123, CD33, CD16, CD11c, CD62L, CD27, CD45RA, CD127, CD4, CD25, CD73, CD39, CD56, CD80, CD86, CD56, CD38, CD138 or CD10 (32, 33). Assessment of surface antigens, adhesion molecules and receptors Neutrophils were assessed based on the expression of HNAs including HNA-1 (CD16b), HNA-2 (CD177), HNA-4 (CD11b) and HNA-5 (CD11a). Whole blood samples were labelled with combinations of antibodies recognizing CD11a, CD62L, CD11b, CD16b, CD66abcd or CD177. NK receptors were measured following negative isolation of NK cells from whole blood using Rosettesep (StemCell Technologies, Vancouver, Canada). Isolated cells were then stained with CD3, CD16, CD56 and mAbs for CD158a/h (KIR2DL1/S1), CD158e (KIR3DL1), CD158b (KIR2DL2/DL3), CD158i (KIR2DS4) (Miltenyi Biotech, Bergisch Glabach, Germany) or NKG2D (BD Biosciences) (11). The migratory potential of γδT cells was examined using the CD11a and CD62L markers, while CD94 determined MHC recognition. Expression of co-stimulatory receptors CD80 and CD86 was used as a determinant of DC function. This was performed as previously described (34). Heparinized whole blood was diluted with RPMI at a ratio of 1:2 and stimulated with or without PMA (5 ng/ml)/ionomycin (1 μg/ml) and incubated for 6 h at 37°C with 5% CO2. Samples were then labelled with anti-HLA-DR, lineage cocktail 2 (CD3, CD14, CD19, CD20, CD56) and either CD80 or CD86 (BD Biosciences). Subsequently, samples were lysed, washed and fixed. All cells were then analysed on the flow cytometer. Intracellular staining of functional proteins Examination of NK proteins and FOXP3 was performed via intracellular staining (35). Briefly, PBMCs (1 × 107 cells/ml) were stained and incubated with fluorochrome-conjugated mAbs CD56, CD16, CD3, CD25, CD127 and CD4 (BD Biosciences). Samples were washed, resuspended and incubated in BD Cytofix (BD Biosciences) for 30 min, following which they were washed in diluted perm wash buffer (BD Biosciences). mAbs for the proteins of interest, perforin, granzyme A (GZA), granzyme B (GZB) and FOXP3 were then added to the appropriate samples, incubated, washed and analysed on the flow cytometer.
Downloaded from http://intimm.oxfordjournals.org/ at Fudan University on May 8, 2015
antigen-4 (CTLA-4) and lymphocyte activation gene-3 (LAG3), respectively (15). The suppressive functions of Tregs on DCs may be important in CFS/ME as a lack of mature DCs may affect the antigen-presenting function of DCs, particularly altering cytokine secretion by T helper (Th) cells (16). Interestingly, co-stimulatory molecules such as CD86 have been shown to be either decreased or increased on the surfaces of DCs in certain diseases (17, 18). In CFS/ME, little is known about the role of DCs. Given that DCs are central to a collapse in immune tolerance, it is imperative to determine their role in CFS/ME. DCs are known to incite the generation of autoreactive immune responses resulting in the generation of pathogenic autoantibodies and chronic inflammation (19–21). DCs also interact with other antigen-presenting cells including γδT cells and this is a potent mechanism for CD86 expression (22). In the periphery, γδT cells represent about 1–10% of lymphocytes and 2–5% of the total T-cell populace (23, 24). In humans, there are two subsets of γδT cells, Vδ1 and Vδ2. Vδ1 is predominant in the spleen and thymus, while Vδ2 is abundant in the periphery, skin, tonsils and lymph nodes (25, 26). γδT cells are functionally similar to cytotoxic cells, NK cells and CD8+T cells (27). These cells also secrete cytokines that inhibit CD4+T-cell proliferation, induce apoptosis and promote B-cell antibodies and have been implicated in a number of diseases (28–30), hence, they may have a role in CFS/ME. Peripheral γδT cells may decrease in the presence of high levels of FOXP3 (31). Direct confirmation of the role of these cells in regulating immune mechanisms in CFS/ME remains to be determined. Similarly, there are no reports on the role of human neutrophil antigens (HNAs) in CFS/ME patients. This is the first study to investigate the potential role of innate and adaptive immune cell phenotypes and functional proteins of γδT cells, DCs and HNAs in the CFS/ME symptom profile.
Immune cells in CFS/ME 235 NK cell cytotoxic activity, degranulation and IFN-γ
Th1/Th2/Th17 cytokines following mitogenic stimulation
NK cytotoxic activity and degranulation were examined as a measure of NK cell function. Cytotoxic activity was assessed as previously described (13, 14, 36). Briefly, PBMCs were isolated from whole blood samples following density gradient centrifugation. PBMCs (1 × 106 cells/ml) were labelled with 0.4% Paul Karl Horan (PKH)-26 fluorescent cell linker dye (Sigma-Aldrich, St Louis, USA). Cells were incubated with K562 tumour cells (1 × 105 cells/ml) for 4 h at three effector to target ratios: 12.5:1, 25:1 and 50:1. A control sample containing no PBMCs was also included as a determinant of cells undergoing spontaneous apoptosis. In advance of flow cytometric analysis, cells were stained with annexin V and 7-aminoactinomycin D (BD Pharmingen, San Diego, CA, USA). The percentage lysis was calculated for each sample at the different ratios as described previously (13, 14, 36). CD107a and IFN-γ production under two stimulatory conditions was used as a measure of NK cell degranulation. In short, cells were stained with CD107a, monensin (BD Biosciences) and Brefeldin A (Golgi Plug; BD Biosciences) with or without K562 cells (1 × 105/ml) or PMA/I [2.5 μg/ml of PMA and 1 µg/ml of ionomycin (Sigma–Aldrich)] (37, 38). Samples were incubated at 37°C with 5% CO2 for 6 h. Intracellular staining was then performed as described above for the detection of IFN-γ.
The levels of Th1/Th2/Th17 cytokines were examined via a cytometric bead array (CBA; BD Pharmingen) following mitogenic stimulation of PBMCs. Briefly, PBMCs isolated from whole blood were placed in cell culture for 72 h at a cell concentration of 1 × 106 cells/ml in RPMI culture media with or without 1 µg of PHA at 37°C with 5% CO2. After 72 h, cell supernatants were removed and stored at −80°C for later assessment. Th1, Th2 and Th17 cytokine profiles including IL-2, IL-4, IL-6, IL-10, TNF-α, INF-γ and IL-17A were determined using a CBA kit according to the manufacturer’s instructions (35, 39). Flow cytometric analysis
Table 1. Combinations of various antibodies used in examining the different cell attributes Cell type
B cells
γδT cells
Tregs CD4+T cells NK cells
Dendritic cells
Neutrophils
KIRexpressing cells
Antibody combinations Immature
Memory
Plasma cells
CD19-APC CD20-PerCP CD21-FITC CD10-PE γδ1T cells γδ1-TCR FITC CD3-PerCP CD45RA-APC CD27-PE CD4-FITC CD4-FITC Phenotypes CD3-PerCP CD56-PE CD16-FITC
CD19-APC CD20-PerCP CD21-FITC CD27-PE γδ2T cells γδ2-TCR-PE CD3-PerCP CD45RA-APC CD27-FITC CD25-APC CD39-APC Degranulation CD3-PerCP CD56-PE CD107a-FITC IFN-γ-APC pDC Lin2-FITC HLA-DR-PERCP CD123-PE
CD19-PerCP CD138-APC CD38-FITC CD27-PE CD3-PerCP γδTCR-APC CD11a-FITC CD62L-PE CD127-PerCP-CY-5.5 CD73-PE Cytotoxic activity PKH-26 7-AAD Annexin-V-FITC
FOXP3-PE CD127 PerCP-Cy 5.5
CD80 expression Lin2-FITC HLA-DR-PERCP CD80-PE
CD86 expression Lin2-FITC HLA-DR-PERCP CD86-PE
HNA-2 CD16-PE-Cy-5 CD62L-APC CD66-PE CD177-FITC KIR3DL1 CD158e-APC CD56-PE CD16-FITC CD3-PerCP
HNA-3 CD16-PE-Cy-5 CD62L-APC CD66-FITC CD11b/18-PE KIR2DL2/L3 CD158b-APC CD56-PE CD16-FITC CD3-PerCP
HNA-4 CD16-PE-Cy-5 CD62L-APC CD66-FITC C11a-PE KIR2DS4 CD158i-APC CD56-PE CD16-FITC CD3-PerCP
mDC Lin2-FITC HLA-DR-PERCP CD11c-PE CD33-APC HNA-1 CD16-PE-Cy-5 CD62L-APC CD66-FITC CD16b-PE KIR2DL1 CD158a-APC CD56-PE CD16-FITC CD3-PerCP
CD3-PerCP γδTCR-APC CD94-PE
NKG2D NKG2D-APC CD56-PE CD16-FITC CD3-PerCP
The table shows a representation of all the antibodies used in identifying cell phenotypes, receptors and function. AAD, aminoactinomycin D; APC, allophycocyanin; FITC, fluorescein isothiocyanate; KIR, killer cell immunoglobulin-like receptor; PE, phycoerythrin; PerCP, peridinin chlorophyll protein complex.
Downloaded from http://intimm.oxfordjournals.org/ at Fudan University on May 8, 2015
All experiments were analysed on a dual laser four colour flow cytometer, BD FACSCalibur (BD Biosciences). The specific combination of mAbs for each assay is presented in Table 1, while the gating strategies for cell phenotypes are presented in Supplementary Figures 1–6, available at International Immunology Online. The quantity for each phenotype was enumerated using the percentage of gated events from flow cytometric analysis and absolute lymphocyte count from haematological counts (32).
236 Immune cells in CFS/ME Statistical analysis Statistical analysis was performed using SPSS software version 16.0 (SPSS Inc., Chicago, IL, USA) and Graph Pad Prism software version 6.0 (Graph Pad Software, Inc., San Diego, CA, USA). Pairwise comparison using multivariate testing was used to perform comparative assessments of all data generated from all participants (CFS/ME and non-fatigued control subjects). The analysis of variance test and independent sample t-test were used to determine the significance of changes in the variables measured. The Bonferroni method was used as post-hoc analysis to assess changes in the data. A test for normality was performed using Kolmogorov– Smirnov and Shapiro–Wilk normality tests. Spearman’s rank correlation was the non-parametric test used to examine correlations in the data generated. Data are represented as mean ± standard error of the mean. All significant results had P values ≤ 0.05.
Patient blood characteristics Prior to all flow cytometric testing of immune cell phenotypes, all participants were assessed on blood characteristics (Table 2). There were significant differences between CFS/ ME patients and the non-fatigued controls on measures of blood parameters including platelets, haematocrit and erythrocyte sedimentation rate. All other parameters remained unchanged. All CFS/ME patients met the CDC 1994 criteria for CFS/ME. The duration of CFS/ME among the patients was over 2 years. Alterations in Tregs, B cells and DC subsets in CFS/ME The total number of lymphocytes, monocytes or neutrophils did not differ between the two population groups. Similarly, routine blood assessments demonstrated similarities in all blood characteristics measured (Table 2). A significant difference in cell subsets was observed in DCs, B-cell phenotypes and Tregs [Fig. 1(a–c)]. pDCs were significantly decreased in the CFS/ME group, while the mDCs although elevated in the CFS/ME group did not achieve statistical significance (Fig. 1a). CFS/ME patients showed a significant decrease in
Decreases in antigens with no significant changes in surface receptors in CFS/ME HNA-2 (CD177+) was significantly reduced on the surface of neutrophils from CFS/ME patients in comparison with the nonfatigued controls (Fig. 2). When we examined the expression of co-stimulatory markers, CD80 and CD86 pre- and poststimulation, we failed to observe any significant changes in the expression of these markers on the surfaces of DCs. Equally, there were no significant differences in the expression of killer cell immunoglobulin-like receptors (KIRs) or NKG2D on the surfaces of isolated NK cells between the two groups of participants. Similarly, there were no significant differences observed in the following cell surface molecules on γδT cells: CD62Lnegative, CD62Llow, CD62Lhigh, CD11alow, CD11ahigh, CD94+ and CD94−. Reduced NK cell function in CFS/ME Reduced cytotoxic activity may be a prominent hallmark of CFS/ME as it has consistently been reported in CFS/ME patients. In this study, we confirmed significant reductions in the ability of NK cells from CFS/ME patients to lyse K562 cells, at 25:1 and 50:1 effector:target ratios (Fig. 3a). Conversely, degranulation was increased in CFS/ME patients in PMA/I and K562 samples (Fig. 3b). Similarly, IFN-γ-producing CD3−CD56+NK cells were increased in the CFS/ME patients compared with the non-fatigued controls (Fig. 3d). As cytotoxic activity is routinely decreased in CFS/ME patients, we assessed the intracellular levels of these lytic proteins (perforin, GZA and GZB) in NK and T cells. There were no significant differences in the levels of perforin or GZA in the NK cells of both CFS/ME patients and the non-fatigued controls. However, GZB was significantly reduced, while IFN-γ levels in CD3−CD56+NK cells were increased in the CFS/ME patients (Fig. 3c).
Table 2. Comparative assessment of blood characteristics in CFS/ME patients and non-fatigued controls Blood Markers
CFS/ME Patients (n=30)
Controls (n=25)
P values
Haemoglobin (g/l) White Cell Count (×109/l) Platelets (×109/l) Haematocrit Red Cell Count (×1012/l) Mean Cell volume (fl) Neutrophils (×109/l) Lymphocytes (109/l) Monocytes (×109/l) Eosinophils (×109/l) Basophils (×109/l) Erythrocyte sedimentation rate (mm/Hr)
134.43 ± 2.41 6.41 ± 0.33 274.61 ± 13.11 0.41 ± 0.006 4.46 ± 0.08 91.86 ± 0.57 3.90 ± 0.27 1.92 ± 0.10 0.33 ± 0.018 0.16 ± 0.019 0.03 ± 0.004 17.22 ± 2051
140.67 ± 2.56 6.06 ± 0.31 238.38 ± 9.39 0.43 ± 0.007 4.65 ± 0.09 92.08 ± 0.51 3.84 ± 0.26 1.73 ± 0.09 0.32 ± 0.02 0.15 ± 0.02 0.03 ± 0.004 10.48 ± 1.70
0.08 0.51 0.05 0.05 0.12 0.65 0.95 0.20 0.64 0.69 0.53 0.05
The table shows representative data of all the blood characteristics obtained from full blood count analysis. The units of measurement are specified in parentheses. Data are represented as means ± standard error of the mean. P
