The epigenetic control of antigenic variation in Plasmodium falciparum

Much of what is known about antigenic variation in the human malaria parasite Plasmodium falciparum has been established by the study of phenotypic changes at the surface of parasitized red blood cells. Although this has contributed to our fundamental understanding of immune escape, nothing conclusive has been elucidated about the molecular mechanisms that determine activation and silencing of members of the antigenic variation var gene family. Recent findings indicate that reversible chromatin modifications and perinuclear gene movement are epigenetic factors that define the silent and active states of telomere-adjacent var genes.

Introduction

The most severe form of malaria is caused by the parasite Plasmodium falciparum, which undergoes antigenic variation. Antigenic variation sometimes indicates the accelerated generation of diversity in an antigenic gene family, or can refer to the ability of individual populations to sequentially express members of such a family while preventing expression of other family members (see [1] for review). Although both strategies are employed by the parasite P. falciparum, this review focuses on the latter phenomenon.

The best-characterized family of variant antigens in P. falciparum is the PfEMP1 family, which is encoded by the var family of genes [2, 3, 4]. Each genome contains 60–70 members of this family, which are found in groups at subtelomeric regions and in several groups at chromosome-central loci [5]. All var genes share a two-exon structure, with a long 5′ exon followed by a short 3′ exon. The 5′ exon encodes the extracellular portion of the PfEMP1 protein, which mediates cytoadhesion of infected erythrocytes to various host substrates. The 3′ exon encodes the cytoplasmic tail, which is conserved between var genes. Comparison of the var repertoire between different strains indicates that although there is some overlap, there is otherwise very high sequence variability between strains. Telomeres in P. falciparum form physical clusters of 4–7 chromosome ends, a phenomenon that probably enhances ectopic recombination in subtelomeres [6]. Such recombination leads to an accelerated generation of diversity at subtelomeric var loci, with shuffling within and between domains creating new PfEMP1 molecules that have potentially novel antigenic and cytoadhesive properties. Freitas-Junior et al. [6] showed that meiosis in P. falciparum (which only occurs in the mosquito midgut) indeed involves enhanced generation of new subtelomeric var forms. Ectopic mitotic recombination might also contribute to var diversity, although this has yet to be demonstrated.

Most work on var gene regulation has been conducted with in vitro cultures. Although natural infections generate parasite populations that express one or very few var genes simultaneously, the absence of immune selection in in vitro systems eventually results in antigenically heterogeneous populations. To obtain homogenous populations, parasite-infected erythrocytes can be selected for that have binding phenotypes that correspond to specific underlying PfEMP1 family members [7, 8, 9]. Such studies have shown that parasites can upregulate one var gene while silencing most or all other var genes. The best-studied system of antigenic variation in eukaryotes is Trypanosoma brucei, which consecutively expresses different members of the variant surface glycoprotein (VSG) family. In T. brucei, only a single VSG expression site is active in the genome, and DNA rearrangements move other VSGs into the active site (for a recent review see [10]). It is now clear that DNA rearrangement is not required for var gene activation or silencing [11]. Additionally, promoters from active or silenced var genes constitutively drive expression of episome-based reporter genes, which suggests that all var promoters are competent for transcription but that the majority of them are silenced by epigenetic factors [11, 12]. The transcription profile of var genes in individuals and in populations is controversial. Some studies suggest that relaxed transcription occurs in early (ring-stage) parasites and that specific transcription occurs in later (trophozoite stage) parasites [8, 9]. Others suggest that multiple var transcripts are detectable [13, 14], but that only one type of PfEMP1 protein is expressed [14]. The sensitivity of reverse transcription-PCR for detecting var gene transcripts might confound these studies by elevating the apparent importance of transcripts that might in fact be present only at negligible levels [15, 16•]. Another additional confusing element for these studies is the persistent transcription of one var gene, var1csa, into trophozoite stages, the regulation of which now appears to be an exception to the rule [16^•]. Most groups in the field now report that homogeneous populations express only one or very few dominant var genes at substantial quantities [15, 16•, 17, 18•, 19].

Several questions have arisen concerning the regulation of var gene transcription. For example, how is allelic exclusion maintained – what are the cis and trans determinants of silencing? How is the active var gene imprinted as the privileged family member? How does switching take place? In this review, we will focus on the molecular mechanisms that establish the silent and active states of var genes. The recent investigation of the nuclear biology of P. falciparum has revealed several epigenetic factors that determine or are connected with the repression of var gene activity.