Elevated anti-Zta IgG levels and EBV viral load are associated with site of tumor presentation in endemic Burkitt's lymphoma patients: a case control study
- 6.8k Downloads
Endemic Burkitt's lymphoma (BL) is an extranodal tumor appearing predominantly in the jaw in younger children while abdominal tumors predominate with increasing age. Previous studies have identified elevated levels of antibodies to Plasmodium falciparum schizont extracts and Epstein-Barr virus (EBV) viral capsid antigens (VCA) in endemic BL relative to malaria exposed controls. However, these studies have neither determined if there were any differences based on the site of clinical presentation of the tumor nor examined a broader panel of EBV and P. falciparum antigens.
We used a suspension bead Luminex assay to measure the IgG levels against EBV antigens, VCA, EAd, EBNA-1 and Zta as well as P. falciparum MSP-1, LSA-1, and AMA-1 antigens in children with BL (n = 32) and in population-based age-and sex-matched controls (n = 25) from a malaria endemic region in Western Kenya with high incidence of BL. EBV viral load in plasma was determined by quantitative PCR.
Relative to healthy controls, BL patients had significantly increased anti-Zta (p = 0.0017) and VCA IgG levels (p < 0.0001) and plasma EBV viral loads (p < 0.0001). In contrast, comparable IgG levels to all P. falciparum antigens tested were observed in BL patients compared to controls. Interestingly, when we grouped BL patients into those presenting with abdominal tumors or with jaw tumors, we observed significantly higher levels of anti-Zta IgG levels (p < 0.0065) and plasma EBV viral loads (p < 0.033) in patients with abdominal tumors compared to patients with jaw tumors.
Elevated antibodies to Zta and elevated plasma EBV viral load could be relevant biomarkers for BL and could also be used to confirm BL presenting in the abdominal region.
KeywordsAbdominal Tumor Viral Capsid Antigen Falciparum Antigen Kisumu City Nyanza Provincial General Hospital
Endemic Burkitt's lymphoma (BL) is the most common pediatric cancer in areas that experience stable Plasmodium falciparum transmission, and both Epstein-Barr virus infection (EBV) and holoendemic P. falciparum are thought to be etiologically linked to this B cell neoplasm [1, 2, 3, 4]. BL is an aggressive extranodal B cell lymphoma that can present in a number of different sites including jaw, abdomen, central nervous system, thyroid gland, orbital, and breast or a combination of these areas [5, 6]. However, jaw and abdominal tumors are the most common sites of presentation. Interestingly, there are different epidemiologic patterns associated with children presenting with jaw compared to abdominal tumors. For example, while the median age of onset is 6 to 7 years for BL [7, 8], jaw tumors are associated with a younger age of presentation and more frequently found in males, while abdominal tumors are common in older children, and there is a more equal distribution among males and females . Despite the fact that endemic BL has several distinct clinical pictures, most studies have looked at endemic BL as a single clinical entity.
The EBV life cycle comprises both lytic and latent phases, both of which play a role in the pathogenesis of EBV-associated malignancies, and induce different antibody responses [4, 9]. EBV nuclear antigen (EBNA)-1 is a latent antigen and is the only latent antigen consistently expressed in BL tumors . The EBV immediate early lytic transcript BZLF1 which encodes the Zta protein can also be detected in a subset of BL tumors [11, 12] but whether elevated levels of antibodies against Zta occur in children with BL is unknown. Two additional lytic antigens, viral capsid antigen (VCA) and early antigen (EAd) have been used as serologic markers of past EBV infection (e.g. VCA IgG) or viral reactivation (e.g. EAd IgG) . Seminal studies in Uganda showed that elevated titers of IgG against VCA were a prognostic risk factor for BL development . In addition, two recent case control studies in Uganda and Malawi have found that children with BL are more likely to have elevated antibody titers against VCA as compared to controls [8, 15]. These same studies also reported elevated IgG antibody titers against P. falciparum schizont extracts in BL patients compared to controls [8, 15].
Until recently, measurement of pathogen-specific antibodies was done with indirect immunofluorescence assays or ELISA assays. These assays have the disadvantage of allowing the measurement of only one antibody:antigen complex per sample, which can otherwise restrict the study of antibodies to multiple proteins when sample volumes are limited. In order to overcome this problem, multiplexed serology has recently been developed for the study of antibodies to multiple antigens within the same sample [16, 17, 18]. This assay can analyze more than 100 analytes simultaneously from a small sample volume enabling not only mass screening of several antigens, but also studies in pediatric populations with limited sample volumes [17, 18]. In addition, this assay is less laborious and has fewer reaction steps compared to the traditional ELISA  and the results are highly reproducible . We have recently developed a Luminex-based suspension bead assay, allowing the determination of levels of IgG to four different EBV antigens (EBNA, VCA, EAd and Zta) within the same sample . In this study, we used this technology to enable more detailed study of antibody responses to an increased panel of antigens in BL patients in an age-and sex-matched controls living in a region where the risk of BL is high . The panel included P. falciparum blood stage antigens (apical merozoite antigen (AMA)-1, merozoite surface protein (MSP)-1 that were specific for two strains of P. falciparum (3D7 and FV0) and a liver stage antigen (LSA)-1, in addition to the earlier mentioned EBV antigens. In addition, we also measured EBV viral load in the plasma. We report differences in both EBV serology and viral load profiles in children with BL compared to control children with the unexpected finding that the clinical presentation of the tumor correlated with levels of EBV Zta antibody and plasma viral load.
Study participant demographic characteristics
The demographic and clinical characteristics of the study participants
(n = 25)
(n = 32)
BL with abdominal
(n = 16)
BL with jaw
(n = 14)
Mean age years [range]*
n (% male)
n (% positive)
EBV viral load (Log copies/ml)[SEM]
BL patients and controls from malaria endemic region have comparable levels of IgG against P. falciparum antigens
BL patients have increased EBV-specific Zta and VCA IgG levels compared to the controls
Children with abdominal tumors have higher levels of Zta-specific IgG antibodies compared to children with jaw tumors
EBV viral load in plasma is higher in patients with abdominal tumors compared to jaw tumors
When we compared EBV viral load in plasma from BL patients with jaw tumors to BL patients with abdominal tumors, we observed significantly higher levels of EBV in plasma from BL patients presenting with abdominal tumors compared to those presenting with jaw tumors (p = 0.0337) (Fig. 4B). Elevated viral load in the plasma could be due to increased release of viral DNA from tumor cells due to apoptosis or necrosis of tumor cells, and not necessarily linked to elevated anti-Zta and VCA IgG levels. To determine if Zta and VCA IgG antibody levels correlated with EBV viral load, we did a Spearman correlation and found that Zta-specific IgG levels were positively correlated with plasma EBV viral loads in BL tumors (r = 0.3825, p = 0.0307) while there was no correlation with VCA IgG levels (r = -0.0972, p = 0.6093).
In our study, we set out to examine whether there was a difference in EBV antibody responses in BL patients with different clinical presentations, e.g. abdominal or jaw tumors. In addition, we wanted to examine whether there were elevated Zta antibody levels in BL patients relative to age-and gender-matched controls. We observed significantly higher levels of antibody against Zta IgG in all BL patients compared to malaria-exposed matched controls and significantly higher levels of Zta IgG in children with abdominal tumors compared to children with jaw tumors. These data suggest that the underlying clinical differences observed in site of BL presentation could be potentially related to differences in EBV biology.
A defining characteristic of patients with nasopharyngeal carcinoma (NPC), another EBV-associated malignancy, is elevated anti-Zta IgG and plasma DNA levels . Our data extend these observations to BL, where we observed significantly elevated anti-Zta IgG as well as anti-VCA IgG and plasma viral loads in BL patients, relative to controls. Of note is that the elevated Zta antibody levels correlated with elevated viral DNA in the plasma. That both of these antibodies to EBV lytic antigens and viral DNA were high could be linked to increased viral re-activation in the tumors or to increased viral re-activation from latently-infected cells preceding the emergence of the malignancy. This latter possibility is supported by a previous study demonstrating that elevated anti-VCA antibodies in children precedes the emergence of BL . The former is supported by evidence of EBV lytic gene expression in a subset of cells within endemic BL tumors [11, 25, 26, 27] and, thus, could lead to elevated plasma viral loads following release of virus from lytically-infected cells. What is surprising is that there is no difference in levels of EAd antibodies between controls and patients suggesting that the correlation between elevated plasma EBV DNA and EBV lytic re-activation is complex. No significant differences in EBNA-1 antibody levels between cases and controls were observed, consistent with our previous study on EBNA-1 antibody levels in BL patients using an ELISA-based method .
Interestingly, differences in humoral immune responses against EBV have also been reported under a variety of different pathological conditions. For example, in HIV patients, anti-VCA IgG correlates with EBV viral loads, while in solid-organ transplant patients, viral load correlates with both anti-EA and VCA IgG titers [29, 30]. Moreover, within a single EBV protein, i.e. Zta, different epitope responses have been described for different EBV-linked disease entities, such as infectious mononucleosis, NPC and non-Hodgkin lymphoma . Further studies are needed to understand the nature of EBV persistence and re-activation in patients with EBV-associated malignancies to determine if there are underlying alterations in the pattern of EBV persistence that might precipitate or predict the emergence of a malignant clone.
We found that BL patients with abdominal tumors had higher levels of both anti-Zta IgG and plasma DNA levels compared to patients presenting with tumors in the jaw. Although the mechanisms that lead to the elevated anti-Zta IgG and plasma DNA levels in the abdominal tumor patients is still unclear, it could result from increased tumor burden in this group, which is indicated by the higher plasma LDH levels observed in these patients. In NPC, tumor burden has also been associated with elevated anti-Zta IgG and plasma DNA levels . However, the question as to why there are also no concomitant increases in VCA or EAd IgG levels in the patients presenting with abdominal tumors, need to be addressed. In one study of BL, only mRNA from Zta could be detected in the tumors but not mRNA from another lytic transcript  suggesting that there is an abortive lytic cycle in the tumors.
Chronic exposure of children to P. falciparum malaria has been etiologically linked to Burkitt's lymphoma (reviewed in [3, 4, 14]). In support of these previous studies, two recent case control studies demonstrated that BL cases had elevated anti-malaria antibody titers compared to controls [8, 15]. However, in contrast to these studies, we found no differences in IgG levels to a broad panel of pre-erythrocytic and erythrocytic antigens from the two common circulating P. falciparum strains (3D7 and FVO) in BL patients and controls. Discrepancies between these studies and our current findings may be attributed to differences in study design. In our study, we used controls from an area that has high incidence of BL and experiences persistent P. falciparum transmission , while in the previous studies, the controls were sick children presenting at the hospital with malignancies and non-malignant conditions and were from primarily urban areas as compared to the cases [8, 15]. We chose a population-based control group because most children admitted to the Nyanza Provincial General Hospital for non-malignant illness are from Kisumu city whereas the children with BL are typically from rural regions within the Province . This was true for our cohort as most of the BL patients in the current study were from rural settings. Thus, since both our cases and controls are from rural areas where malaria burden is higher than in urban areas, it is more likely that our cases and controls have similar exposure to P. falciparum. This could have, however, resulted in over-matching of cases to controls and limited our ability to identify differences in P. falciparum antibody responses. A larger population based case-control study is clearly needed to resolve the differences between our small study and the hospital case-control studies done in Uganda and Malawi [8, 15].
Children presenting with abdominal BL have elevated plasma EBV DNA, paralleled by elevated IgG levels to Zta, compared to children presenting with jaw tumors, suggesting differences in underlying patterns of EBV replication. Whether BL presenting in the abdomen is a different clinical entity will require more detailed studies of the tumors themselves.
During the period from August 2007 and September 2008, we prospectively enrolled 34 children presenting with endemic Burkitt's lymphoma at the Nyanza Provincial General Hospital, Kisumu, Kenya. Nyanza Provincial General Hospital is located in Kisumu city. As the largest hospital in the Province, it provides out-patient and in-patient services, and maintains its own pathology and radiology laboratories. The hospital is the referral center for childhood cancer cases and is the only medical facility that maintains chemotherapeutic treatment for BL in the region. The BL patients in this study predominantly came from all Districts in Nyanza Province (e.g. Siaya, Homa Bay, Kisumu, Migori, Kisii) as well as some Districts in Western Province (e.g. Kakamega, Vihiga, Nandi, and Busia). With the exception of Kisumu District which includes Kisumu City, these Districts are predominantly rural. BL diagnosis was based on histological assessment of fine needle aspirate (FNA) stained with May-Grunewald Giemsa. Assessment was done by both the clinical cytologist and pathologist at NPGH. One patient was diagnosed with acute leukemia and was excluded from further analysis. Routine HIV testing is done on all patients admitted to NPGH. HIV exposure was determined in venipuncture blood using two rapid serological assays: Unigold (Trinity Biotech, Bray, County Wicklow, Ireland) and Determine (Abbott Laboratories, Chicago, Illinois, USA). One BL patient was HIV+ and was excluded from further analysis. All blood samples were taken from these patients prior to chemotherapy.
Controls were identified from Kanyawegi village along the shores of Lake Victoria in Kisumu West district, Nyanza province, after physical and clinical evaluation of their health status by the study clinician officers. Kanyawegi is in a rural district that has a high incidence rate of BL and experiences persistent P. falciparum transmission . After informed consent, blood samples were also collected from 25 healthy age-and gender-matched children (herein referred to as controls).
Microscopic investigation of P. falciparum parasites
Parasitemia was determined at time of blood collection by performing a thick smear and staining with 5% Giemsa. The slides were examined by two microscopists, any discrepancies in the slide reading were resolved by a third microscopist. Parasite density was expressed as the number of asexual P. falciparum per μL of blood assuming a leukocyte count of 8000 per μL.
Blood collection and processing
Ethical approval for this study was given by the Kenya Medical Research Institute (KEMRI) Ethical Review Committee, Institutional Review Board for Human studies at the Case Western Reserve University (Dr. Moormann's institutional affiliation at time of sample collection) and SUNY Upstate Medical University. After informed consent was given by the study participants' parent or guardians, finger prick blood was collected in EDTA tubes and measurements of hemoglobin levels were determined using a portable β-haemoglobin photometer (Hemocue AB Angelholm, Sweden). Additionally, 2-5 ml of venous blood was drawn from both BL patients and controls. Ficoll density gradient centrifugation of blood was done within 1 hour of blood collection; the plasma was removed for subsequent serologic or virologic analysis. The aliquots were then stored in a -80°C freezer. Biochemical analysis of lactate dehydrogenase (LDH) levels to determine tumor burden was carried out on frozen plasma samples using Selectra E auto analyzer (Vita lab, Amsterdam, The Netherlands) following standard biochemistry procedures using Fortress diagnostics kits (Fortress Diagnostics Limited, Belfast, UK).
P. falciparum and EBV antigens
P. falciparum-specific IgG was detected using five of recombinant P. falciparum antigens, MSP-1(3D7), MSP-1 (FVO), AMA-1 (3D7), AMA-1 (FVO) and LSA-1 (in collaboration with Dr. Carole Long for MSP-1, Dr. David Lanar for AMA-1, and Dr. Sheetij Dupta for LSA-1) and EBV-specific IgG was determined using four synthetic peptides representing immuno-dominant epitopes of the viral capsid antigen (VCA), EBV nuclear antigen 1 (EBNA1), diffuse early antigen complex (EAd) and immediate early protein (Zta) antigens of EBV. The definition and serological use of these EBV peptide reagents has been described before [17, 18]. In early studies we defined that the predominant response to Zta in BL patients was directed against the N-terminal domain, spanning amino acids 1-44 (Middeldorp, unpublished data). Therefore, this domain was used as synthetic Zta antigen throughout this study.
In order to detect P. falciparum and EBV-specific IgG against a panel of peptides, we used the previously described protocol . Briefly, different amounts of peptides or proteins were coupled with 1 × 106 pre-activated carboxylated microspheres (Luminex, Austin TE, USA) in 500 μl of 100 mM MES pH 6.0 buffer (peptides) or 50 mM MES pH 5.0 buffer (proteins). The beads were then washed and stored in PBS, 0.1% BSA, 0.002% Tween-20, 0.05% Sodium Azide, pH 7.4 at 4°C until use. The amounts of peptides used per 500 μL coupling reaction were 20 μg for VCA, EBNA1 and Zta, and 50 μg was used for EAd. The amounts of recombinant malaria proteins per 500 μL reaction were 10 μg for the AMA-1 and MSP-1 antigens, and 20 μg for LSA-1. During the development of the assay, several sample dilutions were tested (ranging from 1:35 to 1:51,200) to determine the best dilution for testing IgG levels to different antigens in the test panel. It appeared that the Luminex assay was more sensitive than ELISA assays performed in parallel, and that in order for a majority of results to fall within the linear part of a standard curve (e.g. dilutions of a positive control sample versus fluorescent units) a dilution of 1:6400 was most appropriate. Therefore, we tested all samples at the typical 1:100 dilution as well as 1:6400. During the analysis, it appeared that indeed the 1:6400 dilution allowed for better distinction between high and low antibody levels than the 1:100 dilution. Thus, antigen specific IgG was measured by incubating 1000 beads of each antigen per well with plasma diluted 1:100 and 1:6400 in a final volume of 100 μl, but the results of the 1:6400 dilution were used in the final analysis. After washing, a 1:200 dilution of PE-conjugated Goat F(ab)2 anti Human IgG (Biosource, Camarillo, CA) was added. At least 75 beads of each antigen were then acquired on a Bioplex reader (Bio Rad, Hercules, CA). Sera from North Americans who had no prior exposure to P. falciparum were used as negative controls while pooled plasma from individuals who were constantly positive for P. falciparum antibodies were used as positive controls for P. falciparum antigens. In addition, EBV-seronegative and -seropositive controls were used in each plate. The results of the assay were expressed in Mean Fluorescent Intensity (MFI) of at least 75 beads for each P. falciparum and EBV antigen tested.
Quantitative PCR to quantify EBV DNA
DNA was extracted from 200 μl plasma using Qiagen DNAeasy kit (Qiagen, Valencia, CA) according to the manufacturers' protocol. DNA was eluted off the column in an equivalent volume of H20 and stored at -20°C. Previously designed primers and probes that detect a 70 bp region of the EBV BALF5 gene were used . The quantitative (Q)-PCR cycle was as follows: 2 min at 50°C, 10 min at 95°C, 42 cycles of 15 sec at 95°C and 1 min at 60°C using thermal cycler model I Cycler™ optical module (BioRad Laboratories, Hercules, CA). IQ SuperMix (BioRad Laboratories, Hercules, CA) was used for all reactions. To generate a standard curve, we used an EBV positive plasmid generated from the PCR product. The viral loads were log-transformed and then calculated based on EBV genome copies/ml.
GraphPad prism version 5 (GraphPad Software, Inc, La Jolla, CA) was used for all the data analysis. Differences in the levels of P. falciparum and EBV-specific IgG levels between controls and BL patients were compared using Mann-Whitney U test. The same test was used to compare both antigen-specific IgG levels and viral loads between children presenting with jaw and abdominal tumors. Across group comparisons of antigen-specific IgG levels and viral loads in controls, BL patients with jaw and abdominal tumors, was carried out using Kruskal-Wallis test. The association between categorical variables for healthy and BL patients was assessed using Fisher's exact test. Two-way ANOVA was used to evaluate the association between antigen-specific IgG levels with diseases status. Correlations between antigen specific IgG levels and viral load were done using Spearman's Rho test. p ≤ 0.05 was considered statistically significant.
This work was supported by NIH R01 CA102667 (RR) and NIH R01 CA134051 (AMM). We thank the parents and the guardians of both the BL patients and controls in allowing us to enroll their children in this study. We are also grateful to the medical staff at the Nyanza Provincial General Hospital in assisting with both recruitment of the study participant and collection of blood samples and Dr. Chandy John of the University of Minnesota/KEMRI project for allowing us to use his Bioplex Luminex machine. We also appreciate Dorine Omenah's efforts in co-ordinating the Burkitt's lymphoma study. This work is published with the permission of the Director of the Kenya Medical Research Institute.
- 5.Mwanda OW, Rochford R, Rainey J, Wilson ML: Challenges in the epidemiological and clinical aspects of Burkitt's lymphoma in Kenya: linking evidence and experience. East Afr Med J. 2004, S111-116.Google Scholar
- 11.Xue SA, Labrecque LG, Lu QL, Ong SK, Lampert IA, Kazembe P, Molyneux E, Broadhead RL, Borgstein E, Griffin BE: Promiscuous expression of Epstein-Barr virus genes in Burkitt's lymphoma from the central African Country Malawi. Int J Cancer. 2002, 99: 635-643. 10.1002/ijc.10372.PubMedCrossRefGoogle Scholar
- 13.Rickinson A, Kieff E: Epstein-Barr Virus. Fields Virology. Edited by: Knipe DM, Howley PM. 2007, Philadelphia: Lippincott Williams and Wilkins, 2655-2700. 5Google Scholar
- 14.de-The G, Geser A, Day NE, Tukei PM, Williams EH, Beri DP, Smith PG, Dean AG, Bronkamm GW, Feorino P, Henle W: Epidemiological evidence for causal relationship between Epstein-Barr virus and Burkitt's lymphoma from Ugandan prospective study. Nature. 1978, 274: 756-761. 10.1038/274756a0.PubMedCrossRefGoogle Scholar
- 15.Mutalima N, Molyneux E, Jaffe H, Kamiza S, Borgstein E, Mkandawire N, Liomba G, Batumba M, Lagos D, Gratrix F, et al.: Associations between Burkitt lymphoma among children in Malawi and infection with HIV, EBV and malaria: results from a case-control study. PLoS One. 2008, 3: e2505-10.1371/journal.pone.0002505.PubMedPubMedCentralCrossRefGoogle Scholar
- 16.Piriou E, Kimmel R, Chelimo K, Middeldorp JM, Odada PS, Ploutz-Snyder R, Moormann AM, Rochford R: Serological evidence for long-term Epstein-Barr virus reactivation in children living in a holoendemic malaria region of Kenya. J Med Virol. 2009, 81: 1088-1093. 10.1002/jmv.21485.PubMedPubMedCentralCrossRefGoogle Scholar
- 17.Gu AD, Mo HY, Xie YB, Peng RJ, Bei JX, Peng J, Li MY, Chen LZ, Feng QS, Jia WH, Zeng YX: Evaluation of a multianalyte profiling assay and an enzyme-linked immunosorbent assay for serological examination of Epstein-Barr virus-specific antibody responses in diagnosis of nasopharyngeal carcinoma. Clin Vaccine Immunol. 2008, 15: 1684-1688. 10.1128/CVI.00135-08.PubMedPubMedCentralCrossRefGoogle Scholar
- 21.Meij P, Vervoort MB, Aarbiou J, van Dissel P, Brink A, Bloemena E, Meijer CJ, Middeldorp JM: Restricted low-level human antibody responses against Epstein-Barr virus (EBV)-encoded latent membrane protein 1 in a subgroup of patients with EBV-associated diseases. J Infect Dis. 1999, 179: 1108-1115. 10.1086/314704.PubMedCrossRefGoogle Scholar
- 22.Drouet E, Brousset P, Fares F, Icart J, Verniol C, Meggetto F, Schlaifer D, Desmorat-Coat H, Rigal-Huguet F, Niveleau A, Delsol G: High Epstein-Barr virus serum load and elevated titers of anti-ZEBRA antibodies in patients with EBV-harboring tumor cells of Hodgkin's disease. J Med Virol. 1999, 57: 383-389. 10.1002/(SICI)1096-9071(199904)57:4<383::AID-JMV10>3.0.CO;2-3.PubMedCrossRefGoogle Scholar
- 24.Dardari R, Menezes J, Drouet E, Joab I, Benider A, Bakkali H, Kanouni L, Jouhadi H, Benjaafar N, El Gueddari B, et al.: Analyses of the prognostic significance of the Epstein-Barr virus transactivator ZEBRA protein and diagnostic value of its two synthetic peptides in nasopharyngeal carcinoma. J Clin Virol. 2008, 41: 96-103. 10.1016/j.jcv.2007.09.010.PubMedCrossRefGoogle Scholar
- 25.Labrecque LG, Xue SA, Kazembe P, Phillips J, Lampert I, Wedderburn N, Griffin BE: Expression of Epstein-Barr virus lytically related genes in African Burkitt's lymphoma: correlation with patient response to therapy. Int J Cancer. 1999, 81: 6-11. 10.1002/(SICI)1097-0215(19990331)81:1<6::AID-IJC2>3.0.CO;2-2.PubMedCrossRefGoogle Scholar
- 30.Stevens SJ, Smits PH, Verkuijlen SA, Rockx DA, van Gorp EC, Mulder JW, Middeldorp JM: Aberrant Epstein-Barr virus persistence in HIV carriers is characterized by anti-Epstein-Barr virus IgA and high cellular viral loads with restricted transcription. Aids. 2007, 21: 2141-2149. 10.1097/QAD.0b013e3282eeeba0.PubMedCrossRefGoogle Scholar
- 31.Tedeschi R, Foong YT, Cheng HM, dePaoli P, Lehtinen T, Elfborg T, Dillner J: The disease associations of the antibody response against the Epstein-Barr virus transactivator protein ZEBRA can be separated into different epitopes. J Gen Virol. 1995, 76 (Pt 6): 1393-1400. 10.1099/0022-1317-76-6-1393.PubMedCrossRefGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.