Evaluating New Biomarkers for Epstein Barr Virus Infection in Egypt

1. INTRODUCTION:

 Epstein-Barr virus (EBV) is a ubiquitous gamma herpes virus with tropism for B cells. It is the most common persistent virus infection in humans with approximately 95% of the world’s population showing an asymptomatic life-long carrier status. This sustained life-long latent infection is the result of the unique interaction of EBV with B cells, specifically memory B cells, which are the EBV reservoir in healthy individuals. The disruption of this finely regulated balance between virus and host immune system can result in EBV-associated lymphoproliferations (LPD) of B, T and NK cell derivation [Dojcinov et al., 2018].

Primary infection of EBV is usually asymptomatic and occurs early in life. When symptomatic, it is usually a self-limited disease occurring in adolescents or young adults manifested as acute infectious mononucleosis (IM) [Rostgaard et al., 2019]. The liver is frequently affected by EBV infection; in more than 90% of cases of EBV-related mononucleosis, liver involvement is present [Schechter and Lamps, 2018]. Moreover, EBV infection underlies the development of hematologic (i.e., Burkitt and Hodgkin lymphoma) and epithelial (i.e., nasopharyngeal carcinoma) tumors [Mawson and Suvankar, 2017]. Additionally, it was being as well associated to some autoimmune disorders (e.g., Multiple Sclerosis) [Guan et al., 2019].

Conventional diagnosis of EBV depends on EBV-VCA for IgM and IgG, but recent studies had shown high false positive rates for these antibodies [Elansary et al., 2016]. Cluster of differentiation (CD) molecules are surface molecules expressed on cells of the immune system that play key roles in immune cell-cell communication and sensing the microenvironment. These molecules are essential markers for the identification and isolation of leukocytes and lymphocyte subsets [Kalina et al., 2019].

Some CDs were reported to be tightly reflecting B cells condition, those are CD19⁺, CD23⁺ and CD24⁺ B cells [Delage et al., 2019]. CD19⁺B cell level was estimated in patients with anti-EBVCA IgM positivity [Luo et al., 1991]. CD23- known as Fc epsilon RII, or FcεRII- is a B cell activation marker, proved to be closely associated with EBV infection [Selb et al., 2017]. CD23 expression was subsequently shown to be required for EBV induced immortalization [Kang and Kieff, 2015]. In addition, CD24⁺- known as signal transducer CD24 or heat stable antigen CD24 (HSA) is a cell adhesion molecule, was found to be expressed at the surface of most B lymphocytes and differentiating neuroblasts [Daniel et al., 2016].

Going with this line, the present work aimed to investigate other modalities as CD19⁺-, CD23⁺- and CD24⁺- B lymphocytes in early EBV mono-infected Egyptian patients, that can improve early diagnosis of EBV.

 2. SUBJECTS & METHODS:

 2.1 Study design:

This case-control study was applied on 50 subjects, divided into 2 groups (control and EBV mono-infected patients). All patients were selected from outpatients of the "Internal Medicine" clinic, Alexandria University Hospital, Alexandria city- Egypt, through January 2015 to January 2016. Apparently healthy individuals were driven from blood donors and acted as control group. All of subjects were from the same governorate, but had different occupations. The ethical approval of the faculty was performed on the study protocol, and enrolment of the individuals to the study was conditioned by an informed consent.

The control group included 25 apparently healthy subjects, while the EBV group included 25 newly diagnosed patients. Both groups were matched in age and gender. All enrolled subjects were subjected to full medical history, and complete clinical examination. Diagnosis of EBV was based on clinical examination, laboratory tests, as well as ultrasonography in some cases.

 2.2  Exclusion Criteria:

 Patients were selected to have no history of cardiovascular disease or diabetes mellitus, recent drug intake (of any kind) at least 14 days before sampling, and also negative for: serum HCV-RNA, HBs-Ag, cytomegalovirus (CMV), autoimmune hepatitis, and sclerosing cholangitis. Women undergoing pregnancy and lactation, and patients with underlying diseases requiring long-time corticosteroids or immunosuppressive treatments were also excluded from this study.

 2.3 Laboratory Analyses:

Serum and whole EDTA-blood samples were collected via venipuncture the median cubital vein in closed sterile tubes. Any hemolyzed, icteric, or turbid sample was avoided. All samples involved in this study were subjected to duplicate analyses. 

 2.3.1 Hematological and biochemical analyses:

total leucocytic count and hemoglobin were measured using (Siemens- Advia 2120i, Germany), liver enzymes: alanine transaminase (ALT) and aspartate transaminase (AST) were measured by commercial kits using UV-visible spectrophotometer (Thermo Spectronic Unicam Helios Alpha, UK).

2.3.2 Serological analysis for EBV infection:

The diagnosis of primary EBV-hepatitis was made based on both: positive immunoglobulin M antibody to EBV viral capsid antigen (anti-EBV-VCA) IgM and Epstein-Barr virus EA IgG. It was accomplished by using (Robonik Read Well Stirp ELIZA Analyzer) apparatus, and kits had been used were of catalog numbers “ESR1361G” and “ESR1363G” from SERION ELISA classic, Germany. The results of EBV IgM and IgG measurements were expressed as optical density (450nm). All the selected patients were reactive to EBV-VCA IgM, indicating acute infection.     

 2.3.3 Immunophenotyping CD19⁺, CD23⁺, and CD24⁺ B-lymphocytes:

For quantification of the percentage of B-cell for CD 19⁺, 23⁺ and 24⁺, “Ficoll-Diatrizoate” density gradient solution for isolation of lymphocytes (#824012: Bio-Rad, Hercules, CA, USA) had been used for isolation of lymphocytes from diluted blood samples onto the density gradient. Then lymphocytes from the blood samples were prepared and fixed in a 1.0% (w/v) solution of paraformaldehyde was prepared by dissolving paraformaldehyde in phosphate buffered saline (PBS) at 70°C and sterile filtration through a 0.45-pm filter. The pH of the solution was confirmed to be 7.4, then samples were preserved in 4°C until analysis [Lal et al., 1988].

Identification of B-cell subpopulation's percentage in whole blood was carried out using flow cytometry technique. A region (gate) was defined around total lymphocytes in peripheral blood, depending upon cell size (forward scatter) and complexity (side scatter). The sample is incubated with fluorochrome labeled monoclonal antibodies that are specific for cell surface antigens: CD19⁺, CD23⁺, and CD24⁺ (Recombinant Anti-CD19 antibody [BU12] (ab254170), APC Anti-CD23 antibody [2G8] (ab25457), and Biotin Anti-CD24 antibody [30-F1] (ab25215): abcam-USA). Then, analyzed on a flow cytometer; cells that have bound the labeled antibodies can be quantified on the bases of their fluorescence emission. Light scatter gating for the expected lymphocyte region is carried out using linear forward angle (FSC) and side light scatter (SSC). Manufacturer’s instructions were strictly followed up and the results were expressed as percentages.

 2.4  Statistical Analysis:

 Statistical package for the social sciences (SPSS) computer program- version 22- was used (IBM SPSS Statistics for Windows, Armonk, NY) to analyze the results. Data were illustrated as mean± standard error (M±SE) and independent sample t-test was performed for comparing quantitative variables between the two-tested groups. Bivariate correlation between the estimated CD markers and the other parameters was performed too. Receiver operating characteristic (ROC) curve analysis was applied to detect the potentiality of each CD measured in this work as a diagnostic tool for EBV at best cutoff value for each. P- value

 3.RESULTS:

 3.1      Subjects’ characteristics & laboratory analyses studies:

 Collectively, 50 subjects involved in this work, consisting of 22 men (44%) and 28 women (56%), aged between 31-52 years (M±SE= 41.60±0.75). Demographic, haematological and biochemical results are shown in table (1).

 3.2      Flow cytometry analyses:

 Resulted flow cytometry analyses for CD19⁺, CD23⁺, and CD24⁺- B lymphocytes frequencies- in both groups- is illustrated in the following figure (1). While the precise measured frequencies for the estimated CDs are illustrated in underneath table (2).

    3.3  Correlation between tested CD markers and other parameters:

 Pearson’s correlation was performed to get the correlations between the parametric variants (frequencies of CD19⁺, CD23⁺, and CD24⁺ B-lymphocytes) and the other investigated laboratory analyzed parameters among the whole tested cohort, as shown in table (3).  

   3.4      Diagnostic performance of investigated CDs in EBV:

       ROC curve analysis was performed to examine the potentiality of each of CD19⁺, CD23⁺, and CD24⁺ B-lymphocytes in differentiating EBV patients from apparently healthy controls as illustrated in figure (2). Comparison with each other concerning: AUC, sensitivity, specificity, 95% confidence interval (95% CI), positive and negative predictive values, in addition to significance of the test are all shown in the underneath table (4).

  Detailed frequency-estimation of the three-tested CD surface markers were performed by subclassifying their results into weak-, moderate- and high- positivity in each group, as illustrated in the following table (5):

4.DISCUSSION:

 Immunophenotyping is used in the diagnosis of EBV infection- as  symptoms of EBV infection are like other illnesses- and expected to open a new era in the field of EBV disease research and EBV disease diagnostics [Arai, 2019]. Thus, this paper attempts to investigate the variant frequency of CD19⁺, CD23⁺ and CD24⁺ markers in EBV patients, and to examine their potentiality in diagnosing early EBV infection.

Table (1) shows that age and gender were non-significantly changed between groups. TLC was significantly elevated (P<0.001) in EBV group compared to the control group (M±SE: 11.12±0.77 vs 6.66±0.30), respectively. This result was in harmony with other studies that noticed elevation in TLC in about 40-70% of patients with EBV [Yang et al., 2014; Rochford, 2016]. It was attributed to a complex interaction between virally encoded proteins and B-cell specific cellular proteins which constitute the proliferation-inducing program [Shrivastava et al.,2016]. In contrast, Hgb level did not show significant change between the two groups in our study; and Zhang and coauthors [2019]were in agree with this result.

ALT and AST (table: 1) were significantly increased (P<0.001) in EBV group compared to the control group (M±SE: 32.20±1.34 vs 15.48±0.64 for ALT), and (M±SE: 28.24±1.14 vs 14.72±0.72 for AST), respectively. EBV should considered as a possible causative agent for this elevation if there was no other reason for acute hepatitis [Yang et al., 2014; Rochford, 2016]. In spite of being within normal range, this condition of slight elevation of liver transaminases was explained by a study made by Fugl and Andersen [2019] who found that liver enzymes peaked within 14 days and returned to normal after the first presentation of EBV infection such as fever, sore throat, or lymph node swelling.

Concerning CD- surface markers (table: 2), quantification of CD19⁺ frequency revealed a significant (P<0.001) decrease in the EBV group compared to the control group (27.64±0.75 vs 48.15±1.05), respectively. This was in concordance with other studies performed by Shoman et al., [2014] and Luo et al., [2018]. Recently, it was also indicated that more than half (56%) of the EBV patients had decreased number of CD19⁺ B-cells compared with healthy individuals in United States, which augmented our resultstoo [Chatterjee et al., 2019]. In contrary, Luo and coworkers in 1991 were against these findings; they demonstrate elevation in frequency of CD19⁺ B cells in the EBV patients. CD24⁺ was also decreased significantly (P<0.001) in the EBV group vs the control group (20.94±0.52 vs 27.59±0.33), respectively. Other studies had the same results as ours,  and attributed it to the cross-linking of CD24⁺ with apoptosis in CD24⁺ B- lymphocyte cells [Kondo et al., 2011; Tsai  et al.2017].

On the other hand, CD23⁺- B lymphocyte cells were the only tested-CD member that significantly (P<0.001) increased in EBV patient’s group compared with the control group (49.43±0.58 vs 21.18±1.10), respectively as shown in table (2). This result was in harmony with other studies [Luo et al., 1991; Pan et al., 2014; Selb et al., 2017]; which referred this result to the stimulatory role of CD23⁺ in EBV-transformation of B-cells. Investigations had shown that only cells co-expressing Epstein-Barr nuclear antigens and CD23⁺ undergo immortalization [Mühe and Wang, 2017; Yap et al., 2019]. Additional study applied by Mrozek-Gorska and coauthors [2019] supported our result too, as CD23⁺ was one of certain activation markers (CD23⁺, CD27⁺, CD38⁺) that were induced by viral infected B cell.

Correlation between each of tested CD- surface marker frequency and the investigated laboratory parameters (table: 3) was done. Pearson’s correlation revealed that CD19⁺ had a strong negative correlation with ALT (r=-.859, P=0.01), AST (r=-.830, P=0.01), and fair negative correlation with TLC (r=-.539, P=0.01) at the 2-tailed significance level. CD23⁺ was also strongly positively correlated with ALT (r=.838, P=0.01), with AST (r=.822, P=0.01), and fair positive correlation with TLC (r=517, P=0.01) at the 2-tailed significance level. No correlations had been obtained for CD24⁺ B-lymphocytes frequency with any of the estimated laboratory parameters.

 Early in this year [2019], Yap et al. mentioned thatatypical cell expressed- CD23⁺ was positive for confirming the EBV diagnosis. So, diagnostic efficacy for each tested CD marker was examined by performing ROC curve analysis (figure: 2). It showed a promising role for CD23⁺ far greater than CD24⁺, while CD19⁺ did not show such role (table: 4). It has a sensitivity and specificity (96%, 100%), respectively, at cut off value of 44.4%. It covers almost all the AUC (0.998), possesses ultimate NPV (100%) and 96.2% PPV, with a high significance value (P<0.001).

The unexpected results for CD19⁺ and CD24⁺ in ROC curve analysis recruited us to subclassify the frequencies of both into three grades after Subramanian and coauthors[2005] (as shown in table: 5). It obviates that: for CD19⁺, all the control members were in the moderate- and high- positive grade, while all the EBV group members laid in the weak-positive grade. This may explain the zero response of CD19⁺ as an EBV- diagnostic tool. For CD24⁺-very weak response for ROC curve study (AUC= 0.026, with sensitivity= 4%, and specificity= 36%, as shown in table: 4), it may be attributed to that all tested subjects had the same weak-positive grade (control and EBV patients).

 5.CONCLUSION:

 Overall, these findings may nominate cellular CD23⁺ to play a role as a complementary biomarker in early diagnosis of EBV, together with the conventionally used EBV–VCA antibodies.

 6.RECOMMENDATIONS:

 Correlation between these CD-surface markers and more clinicopathologic features in EBV- associated liver disease is recommended.

Table (2): Frequencies of the investigated CD- surface markers.

             CD: cluster of differentiation; P: significance.

Correlation is significant at the 0.01 level (2-tailed); r: correlation; P: significance.

    Table (4): Diagnostic efficacy of the estimated CD- surface markers for EBV.

CD makers are represented in % of circulating B-lymphocytic cells; PPV: positive predictive value; NPV: negative predictive value.

            Table (5): Grades of each tested-CD in both groups.

  Fig. (1): Performance of flow cytometry on peripheral blood of participated subjects to estimate           CD19⁺-, CD23⁺-, and CD24⁺- B lymphocytes percentages; M2: positive population.

Fig. (2): Diagnostic performance of the estimated CDs.