Background
Plasmodium falciparum is a major cause of mortality and disease in sub-Saharan Africa. Immunity to malaria in areas with intense transmission is acquired during childhood as a broad repertoire of specific protective antibodies to parasite-derived polymorphic variant antigens present on the infected erythrocyte surface, develops [
1‐
3].
Plasmodium falciparum erythrocyte membrane protein 1 (PfEMP1) is the best characterized variant surface antigen [
4‐
6]. A single parasite express only one or a few variants at a time [
7‐
11] as the parasite develops from ring-shaped early forms into trophozoite stages, simultaneously with the onset of adhesion and antigenicity of the infected erythrocyte [
12]. Members of the PfEMP1 family mediate the cyto-adherence of infected erythrocytes to host receptors, allowing parasites to avoid splenic clearance and an immense sequence variation within the protein family has evolved to escape counteracting host antibody responses [
13,
14]. The PfEMP1 family is encoded by approximately 60
var genes per parasite genome [
6,
15], which can be divided in two four groups A-E based on their 5´upstream region [
16,
17]. Three
var genes,
var1 (UPSA),
var2csa (USPE)
, and var3 (aka Type 3) (UPSA) are conserved in their full length in the global parasite population [
18‐
22]. Apart from these, most parasite genomes contain a set of
vars encoding more diverse domain cassettes
not spanning the full length genes [
23]. PfEMP1s are composed of multiple Duffy binding-like (DBL) and cysteine-rich inter-domain region (CIDR) domains. Among the PfEMP1s the VAR1, VAR2CSA and VAR3 variants have unique domain compositions void of CIDR domains. While VAR1 is particularly, long consisting of seven DBL domains and one CIDR, VAR3 and VAR2CSA distinguishes themselves by not containing any CIDR domains. In addition VAR3 is the smallest of the PfEMP1 proteins [
15] and is the only PfEMP1 consisting of only two domains; DBLα/ζ and DBLε [
23].
Var3 genes have arisen from a recombination between a DBLζ-DBLε encoding sequence only found in the 3´end of
var exon1 and an N-terminal DBLα1 sequence. Sequence analysis has shown that only the DBLζ-DBLε part of VAR3, which is 99% identical between VAR3 sequences, is unique to the protein sub-family [
23].
VAR2CSA binds chondroitin sulphate A (CSA) in the placenta and facilitates the parasite sequestration causing pregnancy-associated malaria. Similarly, organ specific PfEMP1 mediated sequestration has been linked to severe malaria in children [
24], and evidence indicates that a restricted and antigenically conserved subset of variant surface antigens causes the most severe malaria syndromes in children [
25‐
32]. Thus, it is hypothesized that defined PfEMP1 subtypes confer specific adhesion phenotypes for the parasites [
23]. Several studies have aimed to identify the
var types expressed in severe malaria [
29,
30,
33‐
36]. Most of these have relied on defining the most prevalent
var mRNA species by PCR amplification and sequencing of a short 350 bp DBLα tag present in all
var genes except
var2csa and
var3. As this approach does not capture
var3 sequences, quantitative PCR has been deployed to investigate
var3 transcript levels in patient samples [
29,
37]. The overall conclusion from these studies is that members of the UPSA
var genes are associated with severe malaria in children.
Because VAR3 is conserved, belongs to group A PfEMP1 and has an unusual domain structure, this protein could play a particular important role for the parasite and in development of malaria disease. Malaria-endemic populations has been shown to have acquired antibodies that react with recombinant VAR3 protein [
38], suggesting that the protein is expressed and immunogenic during natural infections. However, quantitative PCR studies of
var3 transcript levels in malaria patient samples have shown that
var3 transcripts was not associated to any particular syndrome of severe malaria [
37]. A formal demonstration of VAR3 on the surface of
P. falciparum-infected erythrocytes has never been achieved and evidence presented by Epp
et al.[
39] suggests that
var3 transcription may be regulated differently from the rest of the
var genes and expressed in a non-mutually exclusive manner due to an atypical intron activity
.
In an effort to unravel some of the mystery surrounding this PfEMP1, parasites of different genetic background were successfully manipulated to express VAR3 on the surface of the infected erythrocytes. Analysis of these parasites suggests that both var3 mRNA and VAR3 protein is expressed similar to other PfEMP1s. Furthermore, in this study, evidence for in vivo expression of VAR3 in malaria-sick children is presented.
Discussion
The PfEMP1 mediated parasite sequestration causes harmful inflammation and severe pathology of malaria disease and evidence suggests that particular PfEMP1 subtypes are associated with different syndromes of malaria (reviewed in [
55]). In pregnant women, malaria is caused by VAR2CSA mediated parasite sequestration in the placenta [
53] and the subgroup of UPSA flanked
var genes which includes the conserved
var3 variant has been associated with severe malaria in children [
28‐
32]. The particular involvement of VAR2CSA in pregnancy malaria and the ubiquitous and conserved nature of VAR2CSA make it an apparent vaccine target. The same might apply for VAR3, but no particular function has been ascribed to VAR3 and analysis of
var3 transcripts in patient isolates [
29,
37] does not predict a particular association with severe malarial disease in children. Nevertheless, the data presented in this paper indicate that VAR3 expressing parasites are viable
in vivo and can bestow significant proportions of parasite populations in children sick from malaria. Specifically, IgG from children down to the age of one year reacted with VAR3 PfEMP1-expressing parasites where the level of recognition seemed to increase with age and find a plateau within the first three years of age. The VAR3-expressing parasites were better recognized by the children´s IgG than a parasite line expressing a Group B
var gene. Although VAR3 is less polymorphic than Group B PfEMP1s, these data may indicate an earlier exposure to VAR3-expressing parasites than to Group B PfEMP1-expressing parasites, suggesting a particular involvement in early childhood malaria as predicted for the UPSA type PfEMP1 in general [
17,
28‐
30]. Of the 130 acute-convalescent samples from children under five years of age, only two children were found to acquire IgG antibodies to VAR3 as a result of an acute malaria episode requiring hospitalization. Patient A suffered from hyperparasitaemia but neither of the two patients was anaemic nor showed signs of cerebral malaria. Parasites isolated at hospitalization were found to transcribe
var3-like genes at high levels, which together with the acquired antibodies indicate that VAR3-expressing parasites have contributed to the development of malaria symptoms in these patients. However, other
var type transcripts were also detected in these patients and thus the VAR3-expressing parasites did not appear to be the only cause of disease in these two patients.
The pattern of Tanzanian children´s acquisition of antibodies to live VAR3-expressing parasites compared to UPSA and UPSB type PfEMP1-expressing parasites indicate a virulence of VAR3 similar to UPSA type PfEMP1. Yet, in spite of the conserved nature of
var3 and the presence in almost every parasite genome, it is striking that only a fifth of 1,342 individuals from malaria-endemic areas in Tanzania showed acquisition of antibodies to this PfEMP1 [
38]. This may reflect either that VAR3 is not very immunogenic and/or parasites expressing VAR3 are not very virulent as also suggested by QPCR studies [
29,
37] and the acute/convalescent plasma study described here.
These interpretations of
var3 transcribing parasites and VAR3 antibodies in malaria patients can be challenged if
var3 genes are regulated and expressed abnormally as appears to be the case for
var1. Studies of laboratory parasite lines show that
var1 is constitutively transcribed throughout the blood stages cycle, while the transcription of other
var genes is silenced when parasites develop into schizonts [
56]. In spite of this and the fact that malaria endemic populations develop anti-VAR1 antibodies [
38], VAR1 erythrocyte surface expression has not been demonstrated. Analysis of available sequenced parasite genome shows that
var1 often lacks or has a truncated exon2 which may cause a corrupted transcription regulation resulting in protein synthesis and host antibody development upon parasite destruction. It remains unclear if VAR1 serves a particular function for the parasite and why
var1 is maintained in the population. Indeed, Epp and colleagues [
39] found that the atypical
var intron of the
PFF0020c var3 gene did not show any transcriptional activity as opposed to other non-VAR3
var introns. Non-coding RNA is transcribed from
var introns and is physically associated with chromatin and believed to be involved in the transcriptional regulation of
var genes [
39]. Thus, it was speculated that
var3 genes in general could be regulated differently from other
var genes. From the present data it does not seem that
var3 genes are transcribed and expressed differently from other
var genes. In both 3D7VAR3 and FCR3VAR3
var genes other than
var3 were silenced and the
var3 transcript profile during the intra-erythrocytic life cycle was similar to that of other
var genes in other parasite lines [
43,
52] and predicts as expected the surface expression of VAR3 in early trophozoite stages.
Also the VAR3 expression on the erythrocyte surface appeared similar to that of other PfEMP1. Antibodies raised towards a recombinant 3D7 VAR3 variant successfully selected VAR3 erythrocyte surface-expressing parasites from both 3D7 and IT/FCR3 parasite populations. The surface expression of VAR3 was confirmed by antibody staining in both flow cytometry and confocal microscopy with expression levels and a punctuate pattern characteristic for PfEMP1 expression in knobs. While the VAR3 in the IT/FCR3 line most likely is the only PfEMP1 on the infected erythrocyte surface, it is possible that both PFI1820w and PFF0020c VAR3 PfEMP1s are expressed by the same parasites, as previously seen in 3D7 parasites [
11].
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
CWW carried out molecular biology studies, analysed data and wrote the paper. TL and TGT participated in the design, coordination and analysis of the study and helped to draft the manuscript. LT designed and carried out molecular biology studies, analysed data and helped to draft the manuscript. DCB, SBS, AMM and MA carried out molecular biology studies and revised the manuscript. PAM and JPL participated in the coordination of the study and revised the manuscript. All authors read and approved the final manuscript.