Effect of Malaria Infection on Epstein-Barr Virus Persistence in Kenyan Children

Abstract

Background: The 2 cofactors in the etiology of Burkitt lymphoma (BL) are Epstein-Barr virus (EBV) and repeated Plasmodium falciparum malaria infections. This study evaluated EBV loads in mucosal and systemic compartments of children with malaria and controls. Age was analyzed as a covariate because immunity to malaria in endemic regions is age dependent.

Methods: Children (2-10 years) with clinical malaria from Western Kenya and community controls without malaria were enrolled. Saliva and blood samples were collected, EBV viral load was assessed by quantitative polymerase chain reaction, and EpiTYPER MassARRAY was used to assess methylation of 3 different EBV genes.

Results: Regardless of the compartment, we detected EBV more frequently in malaria cases compared to controls, although the difference was not significant. When EBV was detected, there were no differences in viral load between cases and controls. However, EBV methylation was significantly lower in the malaria group compared to controls in both plasma and saliva (P < .05), indicating increased EBV lytic replication. In younger children before development of immunity to malaria, there was a significant effect of malaria on EBV load in peripheral blood mononuclear cells (P = .04).

Conclusions: These data suggest that malaria can directly modulate EBV persistence in children, increasing their risk for BL.

Keywords: Burkitt lymphoma; EBV methylation; EpiTYPER; Epstein-Barr virus; malaria.

Figure 1.

Epstein-Barr virus (EBV) load in saliva, plasma and peripheral blood mononuclear cells (PBMCs) in children from both groups, clinical malaria and community controls, measured by quantitative polymerase chain reaction. (A) The mean viral load in saliva of clinical malaria group is 4.05 × 10⁸ copies/μg (standard error of the mean [SEM] = 1.34); in saliva of community controls it is 5.73 × 10⁸ copies/μg (SEM = 1.52); for plasma of clinical malaria group, the mean viral load is 1.33 × 10⁴ copies/mL (SEM = 3.60); and for community controls, it is 1.76 × 10⁴ copies/mL (SEM = 3.81). Finally, for PBMCs of clinical malaria group, the mean viral load is 1.55 × 10⁴ copies per million PBMCs (SEM = 3.70); for community controls, it is 1.03 ×10⁴ copies per million PBMCs (SEM = 3.56). (B) The EBV load in children by age. P values represent slope estimates from linear regression models.

Figure 2.

Locus-specific Epstein-Barr virus (EBV) methylation analysis. (A–C) Average EBV methylation at 4 EBV loci (A) immediate early BZLF1 promoter, (B) early BXLF1/EBV thymidine kinase promoter, and (C) BdRF1 intragenic region. (D–F) The EBV load and methylation correlation in both saliva and plasma samples analyzed. (D) Immediate early BZLF1 promoter, (E) early BXLF1/EBV thymidine kinase promoter, and (F) BdRF1 intragenic region. *, P < .05 (exact Wilcoxon-Mann-Whitney test).

Figure 3.

Dysequilibrium model for Epstein-Barr virus (EBV) persistence. Proposed model of the effect of Plasmodium falciparum infection on Epstein-Barr viral infection. The model shows how malaria infection induces both EBV lytic replication and expansion of latently infected B cells. (Figure created with BioRender.com.)