Deficient EBV-specific B- and T-cell response in patients with chronic fatigue syndrome

Abstract

Epstein-Barr virus (EBV) has long been discussed as a possible cause or trigger of Chronic Fatigue Syndrome (CFS). In a subset of patients the disease starts with infectious mononucleosis and both enhanced and diminished EBV-specific antibody titers have been reported. In this study, we comprehensively analyzed the EBV-specific memory B- and T-cell response in patients with CFS. While we observed no difference in viral capsid antigen (VCA)-IgG antibodies, EBV nuclear antigen (EBNA)-IgG titers were low or absent in 10% of CFS patients. Remarkably, when analyzing the EBV-specific memory B-cell reservoir in vitro a diminished or absent number of EBNA-1- and VCA-antibody secreting cells was found in up to 76% of patients. Moreover, the ex vivo EBV-induced secretion of TNF-α and IFN-γ was significantly lower in patients. Multicolor flow cytometry revealed that the frequencies of EBNA-1-specific triple TNF-α/IFN-γ/IL-2 producing CD4(+) and CD8(+) T-cell subsets were significantly diminished whereas no difference could be detected for HCMV-specific T-cell responses. When comparing EBV load in blood immune cells, we found more frequently EBER-DNA but not BZLF-1 RNA in CFS patients compared to healthy controls suggesting more frequent latent replication. Taken together, our findings give evidence for a deficient EBV-specific B- and T-cell memory response in CFS patients and suggest an impaired ability to control early steps of EBV reactivation. In addition the diminished EBV response might be suitable to develop diagnostic marker in CFS.

Figure 1. EBNA antibody response is reduced in CFS patients.

(A) Serum IgG titers were assessed for healthy controls and CFS patients by ELISA for EBV VCA-IgG (control n = 57, CFS n = 63), (B) EBNA-IgG (control n = 57, CFS n = 63), (C) EBV VCA-IgM (control n = 57, CFS n = 63), (D) CMV-IgG (control n = 32, CFS n = 41) and (E) CMV-IgM (control n = 32, CFS n = 41). Statistical analysis was performed using the two-tailed Mann-Whitney-U test and for EBNA-IgG and EBV VCA-IgM Fisher's exact one-tailed test for association analysis with * p<0.05.

Figure 2. EBNA-1-IgG is reduced in a subset of patients but total IgG and B-cell subpopulations are not different in EBNA-1-IgG positive and -negative CFS patients.

(A) Serum IgG titers were assessed in CFS patients for EBV-IgG and EBV-EBNA-1-IgG (n = 387), (B–E) EBNA-1 negative (neg, n = 7) and positive (pos, n = 8) CFS patients were compared for (B) total IgG, (C) the absolute numbers of CD19⁺ B cells/nl blood, (D) frequencies of IgD⁺IgM⁺CD27⁺ marginal zone B cells, and (E) frequencies of IgD⁻CD27⁺ class switched memory B cells. Statistical analysis was performed using the two-tailed Mann-Whitney-U test and for IgG Fisher's exact one-tailed test for association analysis with *** p<0.0001.

Figure 3. EBV-specific antibody secreting cells are reduced in CFS patients.

(A–D) Frequencies of ASCs in healthy controls and CFS patients 7 days after polyclonal stimulation of total PBMCs. Secreted total and specific IgG was assessed with the ELISpot assay. IgG-secreting B cells are shown as frequencies from 1×10⁶ seeded cells for (A) total IgG (control n = 12, CFS n = 17), (B) EBV-lysate-specific IgG (control n = 12, CFS n = 17), VCA-specific IgG (control n = 12, CFS n = 17), IgG against EBNA-1 peptides (control n = 12, CFS n = 16) and (C) HSV- (control n = 8, CFS n = 11) and (D) CMV-lysate-specific IgG (control n = 6, CFS n = 6). (E) Comparison of frequencies of ASCs in polyclonal stimulation of total PBMCs (T-cell dependent) and stimulation of isolated B cells with CD40L (T-cell independent) in CFS patients for total IgG, EBV-lysate, (n = 8), VCA or EBNA-1 peptides (n = 6). Statistical analysis was performed using the two-tailed Mann-Whitney-U test with * p<0.05.

Figure 4. CFS patients show diminished cytokine response against EBV.

Whole blood of healthy controls and CFS patients was analyzed by Multiplex-Immunoassay for (A) IFN-γ production after stimulation with either EBV-lysate (control n = 29, CFS n = 22), EBNA-1 peptide (control n = 24, CFS n = 11) or SEB (control n = 21, CFS n = 11) and (B) after EBV-lysate stimulation for TNF-α (control n = 29, CFS n = 22), IL-2 (control n = 29, CFS n = 22) and IL-10 (control n = 25, CFS n = 13). Statistical analysis was performed using the two-tailed Mann-Whitney-U test with * p<0.05 and *** p<0.001.

Figure 5. CFS patients show reduced EBV-specific memory T-cell response.

(A) Comparison of cytokine producing CD4⁺ (upper panels) and CD8⁺ T cells (lower panels) of CFS patients and healthy controls after 10 days of stimulation with EBNA-1 (left panel, Control n = 17, CFS n = 23). Boolean gating strategy was applied to analyze IFN-γ/TNF-α/IL-2 triple, IFN-γ/TNF-α double, and IFN-γ and TNF-α single cytokine producing T cells after intracellular staining of isolated PBMCs incubated with Brefeldin A for 16 h. Stimulation with CMV pp65 (right panel, Control n = 7, CFS n = 5) is shown for IFN-γ/TNF-α/IL-2 triple, and IFN-γ single cytokine producing T cells. (B) Frequencies of PD-1 expression were analyzed for IFN-γ/TNF-α double producing CD4⁺ and CD8⁺ T cells after 10 days of stimulation with EBNA-1 or pp65 (n = 8). Statistical analysis was performed using the two-tailed Mann-Whitney-U test with ** p<0.01.

Figure 6. Latent EBV can be detected more frequently in CFS patients.

(A) EBV DNA was analyzed via nested real-time PCR in total PBMCs of 20 healthy donors and CFS patients for EBER-1. EBER-copies were calculated in accordance to Namalwa standard. (B) BZLF-1 RNA was analyzed via nested real-time PCR in total PBMCs of 20 healthy donors and CFS patients but no BZLF-1 cDNA was detected. cDNA of EBV cell line 293T/B95-8 was used as positive control. Statistical analysis was performed using the one-tailed Mann-Whitney-U test with ** p<0.01.