Prevalence of Epstein-Barr virus infection and characteristics of lymphocyte subsets in newly onset juvenile dermatomyositis

Qi Zheng ·Kun Zhu ·Cai-Na Gao ·Yi-Ping Xu ·Mei-Ping Lu

Abstract Background The underlying etiology of juvenile dermatomyositis (JDM) is unknown.T cell deficiency as well as Epstein-Barr virus (EBV) infection had been suspected to be involved in the pathogenesis,but it has been poorly evaluated in JDM patients.Methods This study described the traits of T and B lymphocyte subsets in newly onset JDM patients and the incidence of EBV infection in JDM patients compared with match controls.Newly developed JDM patients from 2014 to 2018 were included in the study.Lymphocytes with different markers (CD3 +,CD3 + CD4 +,CD3 + CD8 +,CD3 - CD19 + and CD3 - CD16 + CD56 +)were tested with flow cytometry in the first admission or after 6 months of treatment.Statistical analysis was conducted to compare the EBV infection in the group of JDM patients and controls.Results We observed that JDM patients had higher positive rate of Epstein-Barr nuclear antigen-immunoglobulin G (IgG)(P ＜ 0.0001) as well as EBV capsid antigen-IgG (P ＜ 0.05) than normal controls.CD3 - CD16 + CD56 + lymphocyte was found to be extremely low in early stage of JDM patients,but increased after 6 months of treatment (P =0.0091).Conclusions The level of CD3 - CD16 + CD56 + cells may associate with the clinical course of JDM.EBV may act as anenvironmental factor predisposing patients to the development of JDM.

Keywords Epstein-barr virus·Juvenile dermatomyositis·Lymphocytes

Introduction

Juvenile dermatomyositis (JDM) is a multi-system disease which results in nonsuppurative inflammation of striated muscle and skin.It is one of the most common inflammatory myositis in children.The incidence of JDM is about two to three per million around the world [1].It is a heterogenous group of disorders manifest as symmetric proximal muscle weakness,cutaneous skin changes (heliotrope,Gottron’s papules) and elevation of muscle enzyme [2].The disease course can be monocyclic,polycyclic or chronic [3].

The pathogenesis of JDM remains unclear.The major histocompatibility complex is the strongest genetic risk factor for JDM [4,5].Except genetic factor,environmental factors such as early exposure to viral infection and air pollutants were considered as potential pathogenic causes of JDM [6].Serologic evidences displayed that Epstein-Barr virus (EBV),coxsackie B and influenza were associated with the incidence of JDM,but these results were not supported by case-control studies and viral genome polymerase chain reaction (PCR) evidence [7-9].T-cell lymphoma mimicking dermatomyositis was associated with chronic active EBV infection,which in turn suggested that viral infection may be associated with the clinical phenotype of dermatomyositis [10].T-cell deficiency is a feature of many chronic autoimmune diseases.It is supposed that impaired CD8+T cell results in decreased control of EBV infection and clonal expansion of EBV-infected autoreactive B cells in the target organ [11].These autoreactive B cells can produce pathogenic autoantibodies and provide costimulatory survival signals to autoreactive T cells which would otherwise die in the target organ by activation-induced apoptosis [12].While others found that subsets of peripheral blood lymphocytes such as CD3-CD16-CD56+(natural killer,NK) cells and human leukocyte antigen (HLA)-DR-CD86+myeloid dendritic cells correlate with disease activity in JDM [13].Recently,Piper et al.[14]identified that immature transitional B cells were mass expanded in active JDM and type I interferon as well as toll like receptor-7 pathway may be involved in this process.Based on these studies,we compared lymphocyte subsets of newly onset JDM with the convalescent stage of JDM.We also detected seroprevalence of EBV infection in JDM patients compared with healthy individuals,to further prove whether EBV infection is associated with the onset of JDM.

Methods

Patients and samples

The study population consisted of 27 patients who fulfilled the Bohan and Peter classification criteria for JDM [15].Those patients were hospitalized in Children’s Hospital of Zhejiang University School of Medicine,during the year of 2014-2018.General information including age,gender,chief complaint,physical examination,medical history,date of disease onset,date of first admission and laboratory findings were recorded by group members.The study was approved by the pediatric review board committee,and informed consent from patients and healthy individuals was acquired before the study.Blood samples from 30 healthy individuals were collected to detect EBV antibodies.

Peripheral blood mononuclear cell

Peripheral blood mononuclear cell from whole blood of JDM patients (27 samples from first admission and 15 samples from 6 months’ post-treatment) were isolated by standard density centrifugation (Ficoll Separation Solution:GIBCO,Germany).After centrifugation,the mononuclear cells were collected and washed in phosphate-buffered saline(pH 7.2) two times.The cells were re-suspended in MEM medium (Gibco,Grand Island,USA) with 10% heatedactivated fetal calf serum (Gibco,South America) for flow cytometric analysis.

DNA extraction and quantitative PCR (qPCR)

Qiagen DNeasy blood mini kit (Qiagen,Germany) was used to extract DNA from plasma.EBV loads were quantified by expression of the BamH1W repeats in the viral genome using ABI PRISM 7900 sequence detector (Applied Biosystems,United States).Human β2 microglobulin sequence was detected as an internal control.

Cell staining and flow cytometric analysis

According to the manufacturer’s instructions,lymphocytes were stained with LIVE/DEAD fixable Aqua Stain(Life Technologies,USA),then fixed and permeabilized with FACS Lyse and FACS Perm II (BD Pharmingen,USA).Fluorescent antibodies were used to stain lymphocytes which included CD3,CD4,CD8,CD19 and NK cells.CD3-CD19+represents B lymphocyte,CD3+CD4+represents helper-T lymphocyte,CD3+CD8+represents cytotoxic T-lymphocyte and CD3-CD16+CD56+represents NK cells.All functional values were background subtracted by the negative control.Analysis and compensation were performed using FlowJo flow cytometric analysis software (BD company,USA).

Analysis of anti-EBV IgG antibodies

Blood samples from 27 JDM patients and 30 blood donors were collected and tested for anti-EBV viral capsid antigen (VCA;IgG),EBV capsid antigen (EBVCA;IgM),Epstein-Barr nuclear antigen (EBNA;IgG) and early antigen (EA;IgG) antibodies.Anti-EBV antibodies were tested with commercial enzyme-linked immunosorbent assay kits (Euroimmun Medical Diagnostics,Germany).The absorbance was measured at a wavelength of 450 nm and a reference wavelength of 630 nm.The signal-tocutoffratio ≥ 1.1 was considered as positive.The normal reference of quantitative EBVCA-IgG is 0-22 U/mL.Procedures were conducted strictly in accordance with the instructions.

Statistics

Data were analyzed using GraphPad Prism software (version 7).Mann-WhitneyUtest was used for comparing medians in skewed continuous variables.Two-sample Student’sttest was used to test for significant differences in the means of continuous variables.Proportions were compared using Fisher’s exact test appropriately.Two-tailedPvalues of below 0.05 were considered significant.

Results

Age at onset,sex ratio and seasonal distribution

Data suggest a median age at onset of 7.0 years,with 56%of children ＞ 8 years of age and a female-to-male ratio of 1:1.7.Onset was especially common from the 8th to the 12th year.The highest number of JDM cases occurred in winter especially in December,but in other months there were only sporadic cases.Autumn had the least number of cases (Fig.1).

Fig.1 Seasonal distribution of JDM.The highest number of JDM cases occurred in winter especially in the month of December(13/27).Autumn had the least number of cases (2/27).JDM juvenile dermatomyositis

Clinical and laboratory features

Rash,muscle weakness and fever are the most common complaints of early stage JDM (within 3 months of onset).The insidious disease progression may predate diagnosis by 2-3 months or even longer.In this study,78% (21/27)patients had muscle weakness,70% (19/27) patients had elevated creatine kinase and 89% (24/27) patients had increased lactate dehydrogenase.Magnetic resonance examination showed 89% (24/27) cases with high signal intensity in short tau inversion recovery as well as edema in the muscles.For electromyogram,92% (22/24) patients had specific myogenic damage.Muscle biopsy (hematoxylin and eosin stain) demonstrated minor atrophic muscle fibers with varying degrees of inflammatory cell infiltration.T lymphocyte infiltration was observed in the epimysium,perimysium and endomysium.

JDM patients had higher prevalence of EBV infection than controls

Since most cases in this study had symptoms above 1 month before admission,we detected serum IgG antibody for EBVCA,EBNA and EA as well as IgM antibody for EBVCA on first admission.The results showed that JDM patients had higher positive rate of EBNA-IgG (P＜ 0.0001) as well as EBVCA-IgG (P＜ 0.05) than matched controls.The frequency of EBNA-IgG positivity for each serologic test in JDM patients and controls was 74% (20/27) and 60.0%(18/30),respectively.EBVCA-IgG positivity for each serologic test in JDM patients and controls was 88.9% (24/27)and 66.7% (20/30).There was no statistical difference in EAIgG (P=0.13) and EBVCA-IgM (P=0.19) between JDM patients and controls (Fig.2).None of the JDM patients or controls showed positive EBV-DNA in plasma.

Lymphocyte subsets in JDM patients

The percentage of lymphocyte subsets including CD3+,CD4+,CD8+and CD19+in JDM patients was in normal range,except NK cells which was found to be extremely low in the early stage of JDM.In this study,77.8% (21/27)newly developed JDM patients had reduced percentage of NK cells.NK cell was re-tested in 15 patients who received treatment for 6 months (Fig.3).The percentage of NK cell was elevated in convalescent stage compared with earlyonset JDM (P=0.0091,n=15).

Discussion

The possibility that JDM might be related to EBV infection has been a topic of interest for some time [7,16].In the present study,significant higher titers of EBNA and EBVCA IgG antibody were found in JDM patients than in healthy controls.There was no statistical difference of EBEA-IgG between JDM patients and controls.EA-IgG appeared in the late stage of acute EBV infection and gradually disappeared within 3 months.Therefore,EA-IgG is regarded as a marker of virus proliferation or recent EBV infection.Active EBV infection in children occurs more frequently in the spring.In our study,the highest number of JDM cases occurred in winter,especially in December.We compared the incidence of EBV infection with the occurrence of JDM in different months.The regularity shows that active infection of EBV does not coexist with the incidence of JDM,which shows that active EBV infection may not be the cause of JDM (data not shown).EBV replication cycle has two phases which include lytic or latent viral gene expression in host cells[17].EBV latent gene products consist of the six EBNAs and the two latent membrane proteins [18].Latent antigenspecific responses are mostly made to epitopes derived from the EBNA family which was found to be cross-reactive with“self-proteins”[19].In dermatomyositis,molecular mimicry between EBV antigens and JDM auto-antigens may induce the production of pathogenic autoantibodies and provide costimulatory survival signals to autoreactive T cells [20,21].Based on our study,we deduced that latent EBV infection rather than lytic viral infection may act as a pathogenic factor predisposing patients to the development of JDM.James et al.tested 36 childhood myositis patient’s sera for anti-EBVCA antibodies and found a negative relationship between EBV infection and childhood myositis [22].The control group (mean age 15.40 ± 2.51) they used was antigen,EBVCAEBV capsid antigen,Igimmunoglobulin,NSnot significant.*P＜ 0.05,‡P＜ 0.0001 suitable for systemic lupus erythematosus (SLE),but not JDM.The truth is that positive rate of anti-EBV antibodies will increase with age,while the mean onset age of JDM(7 years) is younger than that of SLE (12 years).

Fig.2 Comparison of anti-EBV antibodies in JDM patients and controls.EBV Epstein-Barr virus,JDM juvenile dermatomyositis,EBNA Epstein-Barr nuclear antigen,EBEA Epstein-Barr early

Fig.3 Comparison of natural killer cells (CD3 - CD16 + CD56 + T cells) before and after treatment.† P ＜ 0.01

Previous studies indicated that subsets of peripheral blood lymphocytes such as NK cells and HLA-DR-CD86+myeloid dendritic cells seems to correlate with disease activity in JDM [13].In our study,we found that the percentage of NK cell was extremely low in early-onset JDM patients.Over three-fourths of newly developed JDM patients had reduced percentage of NK cells according to the population-based normal values and increased to normal after 6 months of treatment.In line with our finding,patients with active dermatomyositis were found to have low numbers of NK cells with declined cell function that reverted to normal upon disease inactivation [23].However,not all clinical institutions are able to perform NK cell function test.Percentage analysis of NK cells by flow cytometry may serve as a useful tool for disease activity monitoring.

In summary,the level of NK cells may associate with the clinical course of JDM.Higher incidence of latent EBV infection was observed in JDM patients than control,which indicates EBV may act as an environmental factor predisposing patients to the development of JDM.But findings need to be verified in a larger,independent cohort.More attention should be paid to the role of NK cell function in the pathogenesis of dermatomyositis.

AcknowledgementsWe are very grateful to the Leukemia Lab of Children’s Hospital of Zhejiang University School of Medicine for flow cytometry.We also want to thank all participating patients and volunteers for their support.

Author contributionsQZ contributed to manuscript writing.KZ contributed to histopathological examination.CNG contributed to statistical analysis.YPX contributed to sample collection and patient informed consent.MPL contributed to paper editing.All authors approved the final version of the manuscript.

FundingThis work was supported by Zhejiang Medical and Health Science and Technology Project (No.2017KY440).

Compliance with ethical standards

Ethical approvalThe study was approved by the pediatric review board committee (approval number:2019-IRB-099),and informed consent from patients and healthy individuals was acquired before the study.

Conflict of interestNo financial or nonfinancial benefits have been received or will be received from any party related directly or indirectly to the subject of this article.