Cytomegalovirus, Epstein‐Barr virus, and human herpesvirus‐6 infections in patients with myalgic еncephalomyelitis/chronic fatigue syndrome

Abstract Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) is a disabling multisystem chronic disease. The etiology and pathogenesis of ME/CFS are unknown. Infections of cytomegalovirus (CMV), Epstein‐Barr virus (EBV), and human herpesvirus‐6 (HHV‐6) are suspected as etiological agents for ME/CFS. This study aims to estimate prevalence and type (active/latent) of EBV, CMV, and HHV‐6 infections in Bulgarian ME/CFS patients. In the study were included 58 patients with ME/CFS and 50 healthy controls. Virus‐specific antibodies were detected by enzyme‐linked immunosorbent assay and viral genomic sequences in peripheral blood mononuclear cell (PBMCs) and plasma samples by nested polymerase chain reaction (PCR). We did not observe any significant differences in virus‐specific immunoglobulin G and immunoglobulin M positivity rates between patients with ME/CFS and control group. In ME/CFS plasma samples, EBV DNA was found in 24.1%, CMV DNA in 3.4%, and HHV‐6 DNA in 1.7% of samples. EBV DNA was detected in 4%, and CMV and HHV‐6 DNA were not found in plasma samples of controls. The frequency of viral genome detection in PBMCs of patients and controls was 74% vs 78% for CMV, 81% vs 84% for EBV, and 82.8% vs 82% for HHV‐6. The difference in frequency of EBV active infection in ME/CFS and control group was statistically significant (P = .0027). No ME/CFS and control individuals with active CMV and HHV‐6 infection were observed. In conclusion, this study using both serological and PCR‐based techniques for distinguishing between active and latent infection showed high rate of active EBV infection among patients with ME/CFS indicating that at least in a subset of cases, EBV is important factor for the development of disease.

Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) is a disabling multisystem chronic disease. The main clinical sign is debilitating persisting chronic fatigue, not relieved by rest. In addition to fatigue, patients with ME/CFS also suffer from a variety of other symptoms including postexertional malaise, cognitive impairment, musculoskeletal pain, sleep dysfunction, sore throats, lymphadenopathy, orthostatic intolerance, and gastrointestinal symptoms.
The disease is poorly understood and no diagnostic biomarkers are currently available. Therefore, the diagnosis of ME/CFS is difficult and requires exclusion of other medical conditions. It is based on several different sets of diagnostic criteria/case definitions, of which the most widely used are Fukuda case definition, Canadian consensus criteria, and International Consensus Criteria. [1][2][3] The etiology and pathogenesis of ME/CFS are still unknown. Dysregulation of immune system, autonomic nervous system, and metabolic disturbances are the most popular explanatory models for ME/CFS. [4][5][6][7][8] The hypotheses for etiology include genetic predisposition, immune dysfunction, infectious agents, metabolic disturbances, brain dysfunction, toxins, stress, trauma, circulatory abnormalities, or a combination of any of these factors. As in many patients with ME/CFS, the disease starts suddenly with a "flu-like'' illness, it was suggested that an infectious agent can trigger the syndrome.
Numerous viruses have been associated with the development of ME/CFS including enteroviruses, herpesviruses, retroviruses, parvovirus B19, hepatitis C virus, Ross River virus (RRV). [8][9][10][11][12][13][14][15] It was shown that the severity of acute Epstein-Barr virus (EBV) and RRV infection and the host response may determine the course of postinfectious fatigue and ME/CFS and was suggested that inflammatory cytokines influence the CNS, resulting in neurocognitive disturbances. 16,17 In addition, according to Duvignaud et al, 18 CFS-like illness may develop in 26% of patients with chronic fatigue as a result of postchikungunya chronic disease, induced by chikungunya virus. However, although the correlation between viral infections and ME/CFS has been studied for a long time, the role of viruses in the etiology of ME/CFS is still uncertain.
Herpesviruses have frequently been associated with ME/CFS. Infections with EBV, human herpesvirus-6 (HHV-6), and cytomegalovirus (CMV) are considered as triggering factors for ME/CFS. 9 After an acute infection, these viruses persist life-long in various cells of the body and may reactivate. There are hypotheses that the reactivation of a latent virus could damage the immune system and contribute to the morbidity of ME/CFS. Another possibility is that patients with ME/CFS are susceptible to acute viral infections as a consequence of immune dysfunction. At the same time, these viruses are ubiquitous in the general population and, therefore, it is difficult to prove their causative roles.
Despite multiple studies on the association of EBV, CMV, and HHV-6 with ME/CFS, the data are not consistent. This study aims to estimate the prevalence and type of EBV, CMV, and HHV-6 infections in Bulgarian patients with ME/CFS using both serological and polymerase chain reaction (PCR)-based techniques.

| Study participants
A total of 108 subjects were recruited for this study-58 patients with ME/CFS and 50 healthy persons as a control group. The patients were diagnosed with ME/CFS according to Fukuda criteria. 1 They were aged between 19 and 60 years (average 39 years) and women were more prevalent (72%) than men (28%).
The control group included 34 females and 16 males with average age of 42 years.

| Ethical issues
This study has been approved by the Ethical Committee of the National Center of Infectious and Parasitic Diseases, Sofia, Bulgaria. All participants provided informed written consent before their enrollment.

| Sample collection and processing
Blood samples were collected in ethylenediaminetetraacetic acid

(for peripheral blood mononuclear cells [PBMCs] and plasma) and
Gel/Clot Activator (for serum) vacutainers. PBMCs were prepared from whole blood by Histopaque-1077 (Sigma-Aldrich) density gradient separation. Serum samples were used for enzyme-linked immunosorbent assay (ELISA) testing. DNA was isolated from PBMCs and blood plasma samples.

| Enzyme-linked immunosorbent assay
ELISA testing was used to detect immunoglobulin G (IgG) and immunoglobulin M (IgM) class antibodies specific to CMV, EBV, and HHV-6. All serum samples were tested by using commercial ELISA   instructions. The quality of the PBMCs DNA and the absence of contamination of plasma DNA by cellular DNA were evaluated by β-globin gene amplification as previously described. 19 Nested polymerase chain reaction (nPCR) was used to amplify specific CMV, EBV, and HHV-6 DNA sequences in PBMCs and cell-free blood plasma from patients with ME/CFS and healthy controls.
Detection of CMV, EBV, and HHV-6 DNA by nPCR was mainly based on previous studies of Allen et al, 20 Cinque et al, 21 (Table 3).

| Statistical analysis
There was no statistically significant difference between ME/CFS patients and the control group concerning CMV (P = .6597), EBV (P = .8018), and HHV-6 (P = 1.00) DNA detection in PBMCs samples.   We assessed the number of EBV DNA copies in plasma samples of ME/CFS patients with active EBV infection by quantitative PCR.
EBV DNA loads were relatively low, in the range between 790 and 1540 copies/mL of plasma samples.

| DISCUSSION
Association of herpesviruses CMV, EBV, and HHV-6 with ME/CFS has been investigated for long time, however the results are inconsistent. [9][10][11]13,15,[23][24][25][26] To contribute to the understanding of ME/CFS, in the present study, we continued these investigations and determined prevalence and type of CMV, EBV, and HHV-6 infections among Bulgarian patients with ME/CFS. We have simultaneously tested ME/CFS patients and healthy controls for the presence of CMV, EBV, and HHV-6 DNAs in cell-free plasma and PBMCs samples as well for the virus-specific antibodies.
The key finding of this study is the higher prevalence of EBV several EBNA-6 peptides. 25 On the contrary, according to a recent study, no increased EBV-CA IgG reactivity over EBNA-1 in ME/CFS cohorts is found and EBNA-6 peptide IgG reactivity is not significantly different between the ME/CFS and healthy control samples. 27 In addition, some earlier studies also reported no differences in IgG titers against EBV-CA, EBNA-1 and EA. 10 All these conflicting results may be attributed to the methodological differences, not well-characterized patients with ME/CFS, pathogenesis-related EBV genetic variants, and heterogeneity of studied ME/CFS populations. 33 Thus, heterogeneity among patients with ME/CFS is well recognized and subtypes of ME/CFS may reflect particular etiological factors. 34 Zhang et al found evidence of subtypespecific relationships for EBV among patients with ME/CFS analyzing EBV antibody markers in patients with ME/CFS which had been grouped into eight subtypes based on clustering of real-time PCR expression data for 88 CFS/ME-associated genes, 12 of them associated with EBV infection. It is assumed that heterogeneous host response to EBV reactivation could explain the heterogeneous occurrence of many of the immune and neurological abnormalities reported in patients with CFS/ME. 15,35 Our results based on serological as well as on PCR-based techniques distinguishing between active and latent infection have also shown that there are no significant differences in the frequency of CMV and HHV-6 active infection in patients with ME/CFS compared to the control group. We also detected high but quite similar frequency rates of CMV and HHV-6 latent infection among both, ME/ CFS and control groups. These results confirm some previous studies indicating no correlation between CMV and HHV-6 infection and ME/CFS. 10,14,22,26,36 At the same time, other studies indicate such a link. [37][38][39] Moreover, an association between active HHV-6 infection and ME/CFS has been demonstrated in studies distinguishing between active and latent infection using immunofluorescence assays directed against HHV-6A antigens or early antibody assays. 40,41 Active HHV-6 infection was detected more often in patients with ME/CFS than in controls, and this infection correlated with the occurrence of the clinical symptoms. 11 In a recent meta-analysis, however, we were not able to find a statistically significant difference between reported studies that have found no correlation between HHV-6 and ME/CFS and publications that noted a correlation. 13 The present study had some limitations including the small size of studied ME/CFS population, a potential limitation of methodology approach for estimating elevated levels of virus-specific antibodies, the results are indicative for characterization of herpesvirus infection at a particular moment and not for the entire course of the disease.
In conclusion, this study using both serological and PCR-based techniques for distinguishing between active and latent infection showed a high rate of active EBV infection among patients with ME/ CFS indicating that at least in a subset of cases, EBV is an important factor for the development of the disease.

ACKNOWLEDGMENTS
This study was supported by the Bulgarian National Science Fund

CONFLICT OF INTERESTS
The authors declare that there are no conflict of interests.

AUTHOR CONTRIBUTIONS
ES and MM conceived the idea; ES wrote the first draft of the manuscript; MM contributed towards writing and editing of the manuscript; VR collected clinical data; and AK, SR, DA, and NC carried out laboratory analyses. All authors read and approved the final manuscript.