ERK1/2, MEK1/2 and p38 downstream signalling molecules impaired in CD56dimCD16+ and CD56brightCD16dim/− natural killer cells in Chronic Fatigue Syndrome/Myalgic Encephalomyelitis patients

Background Natural Killer (NK) cell effector functions are dependent on phosphorylation of the mitogen-activated protein kinases (MAPK) pathway to produce an effective immune response for the clearance of target cells infected with viruses, bacteria or malignantly transformed cells. Intracellular signals activating NK cell cytokine production and cytotoxic activity are propagated through protein phosphorylation of MAPKs including MEK1/2, ERK1/2, p38 and JNK. Reduced NK cell cytotoxic activity is consistently reported in Chronic Fatigue Syndrome/Myalgic Encephalomyelitis (CFS/ME) patients and intracellular signalling by MAPK in NK cells remains to be investigated. Therefore, the purpose of this paper was to investigate MAPK downstream signalling molecules in NK cell phenotypes from CFS/ME patients. Methods Flow cytometric protocols were used to measure phosphorylation of the MAPK pathway in CD56brightCD16dim/− and CD56dimCD16+ NK cells following stimulation with K562 tumour cells or phorbol-12-myristate-13-acetate plus ionomycin. NK cell cytotoxic activity, degranulation, lytic proteins and cytokine production were also measured as markers for CD56brightCD16dim/− and CD56dimCD16+ NK cell function using flow cytometric protocols. Results CFS/ME patients (n = 14) had a significant decrease in ERK1/2 in CD56dimCD16+ NK cells compared to the non-fatigued controls (n = 11) after incubation with K562 cells. CD56brightCD16dim/− NK cells from CFS/ME patients had a significant increase in MEK1/2 and p38 following incubation with K562 cells. Conclusions This is the first study to report significant differences in MAPK intracellular signalling molecules in CD56dimCD16+ and CD56brightCD16dim/− NK cells from CFS/ME patients. The current results highlight the importance of intracellular signalling through the MAPK pathway for synergistic effector function of CD56dimCD16+ and CD56brightCD16dim/− NK cells to ensure efficient clearance of target cells. In CFS/ME patients, dysfunctional MAPK signalling may contribute to reduced NK cell cytotoxic activity. Electronic supplementary material The online version of this article (doi:10.1186/s12967-016-0859-z) contains supplementary material, which is available to authorized users.


Background
Natural Killer (NK) cells are innate immune cells which comprise approximately 10-15 % of lymphocytes circulating in the peripheral blood [1]. Two predominant NK cell phenotypes identified by the surface expression of cluster of differentiation (CD) 56 and CD16 and an absence of CD3 provide host immunity through the production of immunoregulatory cytokines and the cytotoxic lysis of target cells [2][3][4].
Longitudinal reports of significantly reduced NK cell cytotoxic activity in Chronic Fatigue Syndrome/Myalgic Encephalomyelitis (CFS/ME) patients suggests the presence of an NK cell functional deficiency which may contribute to the illness pathogenesis [26][27][28][29][30][31][32][33][34]. Current investigations into NK cell phenotypes, receptors and lytic proteins in CFS/ME patients have reported equivocal findings and importantly, intracellular signalling by MAPKs in NK cells remains to be examined [27,35,36]. Therefore, the purpose of the present study was to investigate NK cell phosphorylation of the MAPK signalling cascade, cytotoxic activity, degranulation, lytic proteins and cytokine production in CD56 dim CD16 + and CD56 bright CD16 dim/− NK cells from CFS/ME patients.

Participant recruitment and inclusion criteria
CFS/ME patients and non-fatigued controls (NFC) were recruited from a participant database at the National Centre for Neuroimmunology and Emerging Diseases, Menzies Health Institute Queensland. All participants completed an online questionnaire based on the 1994 Fukuda definition for fatigue and symptom presentation to determine suitability for study inclusion [37]. From the questionnaire responses, CFS/ ME patients meeting the 1994 Fukuda definition and NFC were included. All participants were screened for exclusionary conditions such as epilepsy, thyroid conditions, psychosis, diabetes, cardiac disorders, smoking, pregnant or breastfeeding and immunological, inflammatory or autoimmune diseases. This study was conducted with ethical approval from Griffith University Human Research Ethics Committee (MSC/18/13/ HREC) and prior to participation, all subjects provided written informed consent.

Blood collection and cell isolation
Forty millilitres of sodium heparin blood was collected by venepuncture from the antecubital vein of each participant. To avoid the influence of circadian variation, all blood samples were collected in the morning between 7:30-10 a.m. Laboratory analysis commenced within 4 h of blood collection to maintain cell viability. Routine blood parameters including a full blood count, erythrocyte sedimentation rate, electrolytes and high sensitivity C-reactive protein were assessed on each participant sample by Queensland Pathology. The whole blood samples were diluted with unsupplemented Roswell Park Memorial Institute medium (RPMI) 1640 media (Life Technologies, Carlsbad, USA) and peripheral blood mononuclear cells (PBMCs) were isolated by density gradient centrifugation with Ficoll-Hypaque (GE Health Care, Uppsala, UP).

NK cell cytotoxic activity
Flow cytometry was used to measure NK cell cytotoxic activity against the human chronic myelogenous leukaemia K562 cell line as previously described [29,41]. Briefly, K562 cells (Sigma-Aldrich, St Louis, USA) were cultured in RPMI 1640 media (Life Technologies, Carlsbad, USA) supplemented with 10 % fetal bovine serum (FBS) (Life Technologies, Carlsbad, USA). Following isolation, the PBMCs were stained with Paul Karl Horan-26 fluorescent cell linker dye (Sigma-Aldrich, St Louis, USA) and washed with RPMI supplemented with 10 % FBS. The concentrations of the PBMCs and K562 cells were adjusted to 2.5 × 10 6 cells/ml and 1 × 10 5 cells/ ml respectively and combined at three effector to target (E:T) ratios including 25:1, 12.5:1 and 6.25:1. A control sample of only K562 cells was also included to determine K562 cells undergoing apoptosis not induced by NK cell cytotoxic activity. The PBMCs and K562 cells were incubated for 4 h at 37 °C with 5 % CO 2 and then stained with fluorescein isothiocyanate (FITC) annexin V and 7-aminoactinomycin (Becton Dickinson [BD] Pharminogen, San Diego, USA) for flow cytometric analysis on a BD Calibur (BD Biosciences, San Diego, USA) dual laser four colour flow cytometer. NK cytotoxic activity was calculated as percent specific death of the K562 cells for the three E:T ratios as previously described [41].

NK cell degranulation
NK cell surface expression of CD107a and CD107b was measured as a marker for NK cell degranulation as previously reported [9]. PBMCs in the presence of mAbs for CD107a-PE and CD107b-FITC (BD Biosciences, San Diego, USA) were stimulated with either K562 cells (E:T of 25:1) or PMA (50 ng/ml) plus ionomycin (0.5 μg/ml) for 1 h at 37 °C with 5 % CO 2 . Monensin (BD Biosciences, San Diego, USA) was added to the PBMCs and the cells were then incubated for an additional 3 h. An unstimulated control sample included PBMCs incubated in only RPMI 1640 media. Post 4 h incubation, the cells were washed and incubated with mAbs against CD56-APC, CD16-BV711 and CD3-BV510 (BD Biosciences, San Diego, USA) for 25 min which was followed by flow cytometric analysis.

NK cell lytic proteins and maturation marker
Intracellular staining was used to measure the lytic proteins perforin, granzyme A and granzyme B contained within the secretory granules of NK cells [27,42]. Surface expression of CD57 was measured as a marker for NK cell maturation [43]. The PBMCs were incubated with mAbs for CD56-PE-Cy7, CD16-BV711, CD3-BV510 and CD57-PE-cyanin-based fluorescent dye (CF)594 for 25 min. The PBMCs were then permeabilised with BD fixation/permeabilisation solution for 20 min, washed in BD perm/wash buffer and then incubated with mAbs including perforin-APC (Miltenyi Biotec, Cologne, BG), granzyme A-FITC and granzyme B-V450 (BD Biosciences, San Diego, USA) for 30 min which was followed by flow cytometric analysis.

Multiparametric flow cytometry analysis
Data were collected on a 14-parameter LSR-Fortessa X20 flow cytometer (BD Biosciences, San Diego, USA). Cell signalling technology beads (BD Biosciences, San Diego, USA) were run on a daily basis to ensure optimal flow cytometry performance and application settings were employed to standardise target values for the duration of the experiments. A total of 2500-5000 CD56 positive events were acquired. Data generated for NK cell cytokines, degranulation, lytic proteins and cell maturation was analysed on FlowJo (version 10.0.8) and phosphorylation data were analysed on Cytobank (version 5.0) [45]. NK cell analysis was performed on cells which fell within the lymphocyte population according to forward and side scatter properties. CD56 + CD3 − NK cells were gated to determine total NK cells which was extrapolated to a plot of CD56 and CD16 to identify CD56 bright CD-16 dim/− and CD56 dim CD16 + NK cells for the analysis of each marker for cytokines, degranulation, phosphorylation, lytic proteins and cell maturation. A combination of appropriate fluorescence minus one controls, isotype controls matched to antibody concentrations and unstimulated samples were used to determine NK cell gating for each analysis.

Statistical analysis
Statistical analysis of the data was performed on the Statistical Package for the Social Sciences (version 22) and GraphPad Prism (version 6). All data sets were tested for normality using the Shapiro-Wilk test. The independent Mann-Whitney U test was used to identify any significant differences in the NK cell parameters between the CFS/ME and NFC groups. A Kruskal-Wallis multiple comparisons test was used to identify significant differences in NK cell parameters before and after stimulation within the CFS/ME and NFC cohorts. Significance was set at p < 0.05 and the data is presented as median ± interquartile range unless otherwise stated.

Participant inclusion, blood parameters and NK cell phenotypes
14 CFS/ME patients meeting the 1994 Fukuda definition (mean age [years] ± standard error of the mean (SEM) = 53.5 ± 2.17) and 11 NFC (mean age [years] ± SEM = 48.82 ± 3.46) were included in this study. Comparison of the group ages and blood parameters including erythrocyte sedimentation rate, high sensitivity C-reactive protein and full blood counts of white and red blood cells between CFS/ME and the NFC revealed no significant differences (Table 1). Total NK cells were identified as two phenotype populations-CD56 dim CD16 + and CD56 bright CD16 dim/− NK cells-and compared between CFS/ME patients and NFC. No significant differences were observed for NK cell phenotypes (Additional file 1: Figure S1).

ERK1/2 significantly reduced in CD56 dim CD16 + NK cells from CFS/ME patients
After incubation with K562 cells at an E:T ratio of 25:1, ERK1/2 was significantly reduced in CD56 dim CD16 + NK cells from CFS/ME patients when compared to NFC (Fig. 1). PMA/I induced a significant increase in ERK1/2 phosphorylation in CD56 dim CD16 + NK cells compared to the US and K562 stimulated cells from CFS/ME and NFC participants. Comparison of ERK1/2 in CD56 bright CD16 dim/− NK cells revealed no significant differences between CFS/ME and NFCs (Additional file 2: Figure S2).

MEK1/2 and p38 significantly increased CD56 bright CD16 dim/− NK cells from CFS/ME patients
In CFS/ME patients, phosphorylation of MEK1/2 and p38 was significantly increased in CD56 bright CD16 dim/− NK cells following incubation with K562 cells at an E:T ratio of 25:1 compared to the NFC (Fig. 2). Stimulation with PMA/I induced a significant increase in MEK1/2 and p38 compared to US and K562 stimulated cells in both CFS/ME and NFC cohorts. Comparison of MEK1/2 and p38 in CD56 dim CD16 + NK cells from CFS/ME and NFC revealed no significant differences (Additional file 2: Figures S3, S4). Measurement of additional MAPK proteins including Stat-3, NF-κβ, Iκβ, protein kinase c-α and JNK revealed no significant differences between CFS/ME and the NFC cohorts (Additional file 2: Figures S5-S9).

NK cell cytotoxic activity reduced in CFS/ME
In both CFS/ME patients and NFC, NK cell cytotoxic activity at 25:1 was significantly increased compared to 12.5:1 and 6.25:1 ratios. Compared to NFC, CFS/ME was reduced at 25:1 and 12.5 ratios, although this was not statistically significant (Additional file 3: Figure S10).

CD107a and CD107b increased on CD56 dim CD16 + NK cells after stimulation
Surface expression of CD107a and CD107b on CD56 dim CD16 + and CD56 bright CD16 dim/− NK cells was significantly increased following stimulation with PMA/I and K562 cells in both CFS/ME and NFC (Additional file 3: Figures S11, S12). Comparison of CD107a and CD107b expression between CFS/ME and the NFC under each stimulatory condition revealed no significant differences. CD56 dim CD16 + NK cells from CFS/ ME patients displayed increased CD107a following K562 stimulation, although this increase was not significant.

No significant differences in NK cell lytic proteins from CFS/ ME patients
NK cell lytic proteins perforin, granyzme A and granzyme B were measured in CD56 dim CD16 + and CD56 bright CD-16 dim/− NK cells from CFS/ME patients and NFC. Comparison between the two groups revealed no significant differences. Surface expression of CD57 was measured as a marker for NK cell maturation on CD56 dim CD16 + and CD56 bright CD16 dim/− NK cells and no significant differences were observed between the CFS/ME patients and the NFC (Additional file 3: Figures S13, S14).
CD56 dim CD16 + and CD56 bright CD16 dim/− NK cell cytokine production increased after PMA/I stimulation CD56 dim CD16 + and CD56 bright CD16 dim/− NK cell cytokine production was measured under two stimulatory conditions with PMA/I or K562 cells. INF-γ, TNF-α and GM-CSF production in CD56 dim CD16 + and CD56 bright CD16 dim/− NK cells increased following stimulation with PMA/I in both the NFC and CFS/ME patients (Additional file 4: Figures S15-S17). Comparisons of CD56 dim CD16 + and CD56 bright CD16 dim/− NK cell cytokine production between CFS/ME patients and the NFC under the different stimulatory conditions revealed no significant differences between groups.

Discussion
This is the first study to investigate ERK1/2 and MEK1/2 MAPK intracellular signalling in CD56 dim CD16 + and   [46,47]. Downstream activation of ERK1/2 is the result of intracellular signalling networks propagating activating signals through phosphorylation cascades [48,49]. Sequential phosphorylation of MAPK kinase kinase (MAPKKK) and MAPK kinase (MAPKK/MEK1/2) activates ERK1/2 through dual phosphorylation of threonine and tyrosine residues [48,49]. Phosphorylation of ERK1/2 induces a significant conformational change which is required for NK cell cytotoxic activity as it increases substrate accessibility to phosphorylate paxillin [50,51]. Paxillin is an adaptor protein which provides a docking site for regulatory proteins such as ERK1/2 and structural proteins including microtubules and actin cytoskeleton [50,51]. Colocalisation of phosphorylated ERK2 and paxillin to the microtubules and the microtubule organising centre (MTOC) facilitates polarisation of the secretory granules towards the immune synapse [14,15,46,51,52]. In CFS/ME patients, abnormal signalling through ERK1/2 may interfere with and delay release of the lytic proteins to induce cytotoxic lysis of target cells.
NK cell cytotoxic activity was reduced in the CFS/ME cohort compared to the NFC. The significant reduction of ERK1/2 in CD56 dim CD16 + NK cells may disrupt For a representative individual, CD56 bright CD16 dim/− NK cells were identified using CD56 and CD16 (a). CD56 bright CD16 dim/− NK cell MEK1/2 was determined on a histogram (b) and comparisons were drawn between CFS/ME and the NFC under different stimulatory conditions (c). In comparison to the NFC, MEK1/2 was significantly increased in CFS/ME patients after K562 cell incubation (*p < 0.05). p38 MFI in CD56 bright CD16 dim/− NK cells was also measured on histograms (d) and comparisons were drawn against the CFS/ME patients and the NFC (e) before and after stimulation. In CFS/ME patients, p38 was significantly increased (*p < 0.05) after K562 incubation compared to the NFC. Data are presented as MFI with interquartile range intracellular signalling required for secretory granule polarisation through the MAPK pathway. As the MAPK cascade integrates signals received from the cell surface, the pathway is subject to complex regulatory and feedback mechanisms which may contribute to the reduction observed in ERK1/2 from CFS/ME patients [46,53]. ERK1/2 is under constant regulation which also functions to determine specificity of ERK1/2 to target the secretory granules in cytotoxic NK cells [14,15,46,53]. Regulatory mechanisms of ERK1/2 include phosphatases MKP3 and MKPX which dephosphorylate protein tyrosine kinases to inhibit activation [46,53]. Receptor desensitisation and dissociation of the receptor-ligand interaction changes the strength and duration of activation signals [46,53]. Scaffold proteins and subcellular localisation of the cascade regulate phosphorylation by directing ERK1/2 to target substrates in the cytoplasm or nucleus [46,53]. The integration and crosstalk of ERK1/2 with other signalling pathways also acts as a feedback mechanism to regulate phosphorylation levels [46,53]. As ERK1/2 is subject to a number of distinct mechanisms of regulation, further investigations in CD56 dim CD16 + NK cells from CFS/ME patients are required to determine if these regulatory mechanisms contribute to reduce ERK1/2 phosphorylation.
Degranulation of cytotoxic NK cells was measured to investigate if potential impairments in intracellular signalling through ERK1/2 contribute to reduced cytotoxic activity in CFS/ME patients. Whilst no significant differences were observed in NK cell surface expression of CD107a and CD107b, CD56 dim CD16 + NK cells from CFS/ME patients displayed increased CD107a following K562 stimulation. In support of this current finding, we have previously reported a significant increase in CD107a on NK cells following K562 stimulation in a larger cohort of CFS/ME patients [27]. This finding suggests that the reduction in ERK1/2 may delay movement of the secretory granule and MTOC towards the immune synapse but does not prevent degranulation [14,15]. Increased degranulation of CD56 dim CD16 + NK cells from CFS/ME patients suggests the cells may be under a continuum of activation due to an inability to induce cytotoxic lysis and subsequent removal of the target cells [27].
Continual activation of NK cells in CFS/ME patients may be the result of prolonged contact with target cells. Kinetic priming facilitated by sustained NK cell contact with target cells retains convergence of the secretory granules and the MTOC at the plasma membrane [54]. This mechanism is known as 'serial killing' as subsequent lysis of target cells is more rapid due to pre-docking of the secretory granules, bypassing the need for ERK1/2 to initiate polarisation of the secretory granules towards the immune synapse for degranulation [55]. Further investigations are required to determine if the secretory granule completely fuses with the NK cell membrane to release the entire lytic protein content or if deficiencies in the lytic proteins may contribute to reduced target cell lysis in CFS/ME patients [27,36,56]. Reduced perforin and granzyme B has been reported in NK cells from CFS/ ME patients which may be a consequence of 'serial killing' [27,36]. Whilst it has been identified that NK cells from CFS/ME patients are degranulating, the inability of NK cells to eliminate target cells by cytotoxic activity suggests that the NK cells may be highly activated through a potential mechanism of inefficient 'serial killing' .
NK cell production of cytokines including IFN-γ and TNF-α has been identified as an integral part of NK cell cytotoxic activity and increased production of IFN-γ has previously been reported in CFS/ME [27,29,57]. NK cells differentiate and mature from CD56 bright CD16 dim/− to CD56 dim CD16 + NK cells with predominant cytokine or cytotoxic effector function [6,[58][59][60]. This differentiation process suggests that together CD56 bright CD16 dim/− and CD56 dim CD16 + NK cells function to optimise an efficient NK cell response which may be impaired in CFS/ME patients [6,[58][59][60]. NK cell production of IFN-γ has been reported to augment cytotoxic activity by upregulating expression of the adhesion molecule ICAM-1 on tumour target cells through the NF-κβ pathway which improves conjugate formation and adherence with cytotoxic NK cells [60]. Conversely, it has also been reported that IFN-γ treatment of tumour cells with high basal levels of ICAM-1, such as K562 cells, up-regulates major histocompatibility class I which acts as a ligand for inhibitory receptors on NK cells and reduces NK cytotoxic activity [60][61][62]. In CFS/ME patients, further investigations are required to determine if increased IFN-γ may contribute to the proposed inefficient mechanism of 'serial killing' resulting in increased degranulation or if IFN-γ desensitises K562 cells to NK cell mediated cytotoxic activity.
Cytokine synthesis by MEK1/2 and p38 is tightly controlled and each tier of the MAPK signalling cascade is subject to regulation which may be impaired in CFS/ ME patients [46,53]. Phosphatase MKP1 is located in the nucleus and downregulates MEK1/2 and p38 activity by dephosphorylating threonine and tyrosine residues, attenuating cytokine production [46,53]. Further investigations into the regulation of MEK1/2 and p38 in CD56 bright CD16 dim/− NK cells from CFS/ME patients are required to determine if a regulatory mechanism such as MPK1 may contribute to increased MEK1/2 and p38 activity and IFN-γ cytokine production.
Investigations into the MAPK intracellular signalling pathway in NK cells from CFS/ME patients has revealed novel findings which may explain previous reports of reduced NK cell cytotoxic activity and increased cytokine production. To our knowledge, this is the first study to report significant differences in CD56 dim CD16 + NK cell ERK1/2 from CFS/ME patients. CD56 dim CD16 + NK cell cytotoxic activity is dependent on synergistic action of CD56 bright CD16 dim/− NK cell cytokine production. Consequently, increased MEK1/2 and p38 may increase IFN-γ production which in turn may desensitise K562 cells against NK cell cytotoxic activity in CFS/ME patients. The novel, preliminary findings of this study provide a rationale for further investigations into a larger cohort and particular clinical subgroups of CFS/ME-including severity-to elucidate the cause of reduced NK cell cytotoxic activity.

Conclusions
The results from this study highlight the importance of intracellular signalling through the MAPK pathway for synergistic function of CD56 dim CD16 + and CD56 bright-CD16 dim/− NK cells to ensure efficient clearance of target cells in CFS/ME patients. Further investigations are required to determine if regulatory mechanisms contribute to the aberrations in MAPK intracellular signalling in CD56 dim CD16 + and CD56 bright CD16 dim/− NK cells in a larger cohort of CFS/ME patients.