To the Editor:
In their interesting report, Prang et al1 showed the presence of peripheral blood mononuclear lymphocytes (PBMC) positive for Epstein-Barr virus (EBV) lytic proteins. This is in agreement with the updated theory of EBV persistence.2 However, this is not the first paper reporting lymphocytes expressing EBV lytic proteins in vivo.
It used to be believed that epithelial cells are responsible for the primary infection and the persistence of EBV. EBV lytic infection in epithelial cells is differentiation-associated and virus released there infects lymphocytes later on.3 The first report proposing a different view for the role of EBV-infected lymphocytes in the persistence of the virus was published in 1988.4 Later in 1990, Rickinson5 proposed a new model which shows how lymphocytes can contribute to the persistence of the virus through expansion at the germinal center, and also to the transmission of the virus to distant epithelial cells by spontaneous lytic infection. Rickinson's model still emphasizes the role of epithelium in the primary infection of EBV. In 1994, Niedobitek and Young2 proposed that both primary EBV infection and the persistence of the virus are mediated by lymphocytes. How can mucosal lymphocytes become infected by EBV without the help of epithelial cells if the epithelium is not damaged? The presence of intraepithelial lymphocytes may provide an explanation. I have previously shown the presence of EBV-infected intra-epithelial lymphocytes in the nasopharyngeal mucosa of EBV seropositive people who had no EBV-associated disease or malignancy6 and the detection of rare lymphocytes positive for EBV lytic proteins in nasopharyngeal mucosa.7 EBV lytic infection can also be shown by in situ hybridization for EBV lytic transcripts (eg, BHLF1) or EBV DNA. BHLF1+ stromal or intraepithelial lymphocytes have already been detected in tonsillar specimens of infectious mononucleosis (IM) patients.8
In addition, another point should be clarified in their discussion: as yet there is no report of the usage of the real lytic F promoter (Fp)9 for EBNA1 in PBMCs. Although one paper reported the detection of “Fp-initiated transcripts” in PBMCs,10 the transcripts detected likely originated from the Q promoter (Qp) because the “Q” primer (coordinate 62440-62457 of B95-8 genome) was used to detect the QUK splicing of EBNA1 transcript.11-13 Before the characterization of the real Qp in 1995, it was assumed that these transcripts originated from Fp. Thus, there has been no unequivocal detection of Fp activity in PBMCs yet.
In the course of the primary infection (infectious mononucleosis, IM) with the Epstein-Barr virus (EBV), high titers of antibodies specific for viral proteins early (EA) and late (VCA) in the lytic replication are detected. The VCA-specific antibodies persist lifelong, probably due to restimulation by periodic shedding of EBV into the saliva and readsorption to mucosal lymphocytes. Antibodies to EA, however, disappear and indicate efficient immunologic control of the permissive cells in the blood at the very early phases of replication. Therefore, there is a difference between lytic replication of EBV during IM from that in healthy seropositive persons after convalescence.
A previous report by Anagnostopoulos et al1-1 demonstrated transcription of lytic cycle genes in virus-infected B cells in the tonsils during IM. Tao et al1-2 reported the expression of lytic cycle proteins in the nasopharyngeal mucosa in healthy donors. Both reports showed lytic gene expression in B cells in vivo. Because the investigators did not succeed in demonstrating viral lytic replication in epithelial cells, they proposed that the B cells are the only source for EBV to spread in vivo. However, there are other reports demonstrating lytic replication in differentiated epithelial cells in the parotis or in the tonsills during IM.1-3-1-5 Furthermore, lytic replication in epithelial cells in the case of OHL6,7 by AIDS patients is similar to the immunologic situation during IM, where the EBV-specific immune response is not yet established. Therefore, due to the particular localization of these B cells in the nasopharyngeal epithelium, one cannot determine whether these cells were the source of the infectious EBV or whether they have recently become infected by the virus.
The recent report by Prang et al1-8 showed that during IM replication of EBV also takes place in the B cells of the peripheral blood. Replication in these cells was characteristic for IM, cases of reactivation of EBV, chronic active EBV (CAEBV), and correlated with elevated levels of EA-specific Igs. The experimental data of neither Tao et al nor of Anagnostopoulos et al showed the activity of EBV in the peripheral B lymphocytes. The group of Tierney1-9 also did not succeed (with one exception) to demonstrate the expression of lytic cycle genes in the peripheral B cells during IM. However, their analysis of the activity of the different promoters for EBNA-1 did not contradict with our own observations and therefore there was no reason for us to reexamine F and Q promoter usage.
In conclusion, our results reinforced the often proposed model of transmission of EBV from the reservoir of small resting cells to the sites of virus production within lytically infected B cells. In the report of Prang et al the tendency for lytic replication in B lymphocytes was demonstrated also in healthy seropositive donors, but in contrast to the situation present during IM, viral lytic replication is controlled by an intact immune system. Therefore, an active production of the virus can only proceed in immunologically unpatrolled sites of the body. Whether these sites are B cells, infiltrating the epithelium of the parotis, or whether differentiating epithelial tissues give rise to infectious virus which further reinfect B cells remains unclear. Both possibilities have been published and the truth probably lies somewhere in between.