Silence, activate, poise and switch! Mechanisms of antigenic variation in Plasmodium falciparum

Abstract

Phenotypic variation in genetically identical malaria parasites is an emerging topic. Although antigenic variation is only part of a more global parasite strategy to create adaptation through epigenetically controlled transcriptional variability, it is the central mechanism enabling immune evasion and promoting pathogenesis. The var gene family is the best-studied example in a wide range of clonally variant gene families in Plasmodium falciparum. It is unique in its strict selection of a single member for activation, a process termed monoallelic expression. The conceptual advances that have emerged from studying var genes show striking common epigenetic features with many other clonally variant gene families or even single-copy genes that show a variegated expression in parasite populations. However, major mechanistic questions, such as the existence of a potential expression site and the identity of transcription factors or genetic elements driving singular gene choice, are still unanswered. In this review we discuss the recent findings in the molecular processes essential for clonal variation, namely silencing, activation, poising and switching. Integrating findings about all clonally variant gene families and other mutually exclusive expression systems will hopefully drive mechanistic understanding of antigenic variation.

Fig. 1

Var gene activation and silencing throughout asexual blood stage development of P. falciparum. A. The single var gene is transcribed at the beginning of the asexual blood stage cycle (ring) soon after merozoites invasion of a red blood cells. Shortly before parasite DNA replication starts, var gene transcription ceases but it remains in a poised state (trophozoite and schizont) ready to be activated at the next blood stage cycle. B. All var genes, central as well as subtelomeric ones, are tethered to the nuclear periphery (blue) and silent var genes (red) cluster in repressive centers (grey) (Lopez-Rubio et al., 2009). A single active var gene (green) segregates away from repressive centres and forms a perinuclear expression site containing required transcription factors (orange). The transition from the active to the poised state (yellow) is still poorly defined and it is unknown whether positional memory establishes throughout late stages. However, recently a putative methyltransferase PfSet10 has been specifically associated with the poised var gene (Volz et al., 2012). Variable var gene switching rates have been measured in vitro (Horrocks et al., 2004), but it is unclear at which point in the cell cycle switching occurs.

Fig. 2

Epigenetic switching in different clonally variant gene families of P. falciparum. Several gene families can be differentially expressed in genetically identical parasites (Rovira-Graells et al., 2012). The number of active genes (green) varies but silent genes are widely associated with epigenetic silencing mark H3K9me3 (red) (Lopez-Rubio et al., 2009). Combinatorial switching of the different gene families can result in a tremendous range of phenotypic variation within genetically identical populations. Gene families involved in immune evasion have different numbers, switching rates and strictness in gene counting. While var genes undergo strict monoallelic expression, multiple stevor genes can be transcribed in the same cell (Kaviratne et al., 2002). Gene families not involved in immune evasion can also undergo clonal variation. Here three examples with different ‘switching modes’ are shown. From the 12 members of the acs gene family, encoding for acyl-CoA synthetases 4 members have been shown to be clonally variant in blood stages (Rovira-Graells et al., 2012). The clag3 family, containing two members important for red blood cell permeation pathways, is the only other gene family besides var for which mutually exclusive expression has been shown (Cortes et al., 2007). A single member of the putative AP2 transcription family (PF3D7_1222600) is associated with H3K9me3. It is thought to be involved in inducing gametocytogenesis (Kafsack et al., MPM meeting 2012 Woods Hole). In this simple form epigenetically regulated activation at low frequency could constitute a developmental switch in the life cycle.