Background
Malaria parasites have a complex developmental cycle in both the vertebrate and vector mosquito hosts. The sporozoites are the most versatile invasive stage of the parasite and are unique in having the ability to invade two types of cells, their primary targets in their vertebrate hosts and those of the mosquito salivary glands [
1]. Furthermore, target-cell specificity supports the hypothesis that the localization and penetration of mosquito salivary glands by sporozoites is mediated by receptors [
2‐
5]. Two mosquito proteins, circumsporozoite binding protein (CSPBP) and saglin, were identified as receptors for sporozoites and these interact with the parasite-expressed circumsporozoite protein (CSP) and thrombospondin-related anonymous protein (TRAP), respectively [
4,
5]. Merozoite apical erythrocyte-binding ligand (MAEBL) is another proposed sporozoite ligand involved in attachment and invasion of mosquito salivary glands [
6], however, no receptor has been characterized yet.
CSP is the predominant sporozoite surface antigen and important for mediating recognition and invasion of the salivary glands (as reviewed [
7,
8]). It is expressed by a single copy gene [
9], and has a large central immuno-dominant domain comprising tandem repeats of small peptides. The amino-(NH
2) terminus comprises a domain of hydrophobic amino acid that makes up the secretory signal peptide, and a pentapeptide (KLKQP) designated Region I, that contains a proteolytic cleavage site. Cleavage is required for vertebrate hepatocyte cell invasion, but is not necessary for salivary gland infection in the mosquito host [
10,
11]. Furthermore, the NH
2-terminus together with the repeat domain is responsible for sporozoite development and
Plasmodium berghei mutants lacking these regions do not produce free sporozoites [
12]. The carboxyl-(C) terminus, which has sequence similarity to the thrombospondin type-1 repeat (TSR) superfamily, has an 18-amino acid sequence (EWSXCXVTCGXG(V/I)XXRX(K/R) designated Region II [
13‐
15], and a putative glycosylphosphatidylinositol (GPI) anchor attachment site. Both Regions I and II are conserved highly among
Plasmodium species [
16].
Plasmodium falciparum CSP binds with greater avidity to the medial lobe and distal portions of the lateral lobes of
Anopheles stephensi salivary glands than any other mosquito organs that are in contact with the haemolymph, and these lobes are invaded preferentially by sporozoites [
17]. A synthetic peptide encompassing Region I, plus the additional five amino acids immediately adjacent to the amino-end of the endogenous sequence, inhibits CSP binding to salivary glands [
17]. In addition, synthetic peptides encompassing
P. falciparum Region I inhibit by 80 % sporozoite invasion of salivary glands [
18]. Mutant sporozoites expressing a CSP in which Region I is deleted invade salivary glands with a 10–15 % lower efficiency compared with controls, and mutants with a complete N-terminal deletion have ten-fold fewer salivary gland sporozoites compared with controls [
11]. In contrast,
P. berghei mutant sporozoites expressing a
P. falciparum CSP lacking Region I showed no impairment of motility or infectivity in the vector host, however, disruption of Region II interfered with motility and impaired sporozoite invasion of salivary gland [
19].
Transient (recombinant Sindbis virus) and stable (piggyBac-mediated mosquito transformation) expression systems were used to study competition between different peptides from Plasmodium gallinaceum and P. falciparum based on NH2-terminus or TSR domains of CSP and conspecific sporozoites for receptors in the salivary gland of Aedes aegypti. The data presented here shows that only NH2-terminal peptides interfere with sporozoite penetration of mosquito salivary glands. Furthermore, a heterologous P. falciparum CSP partial NH2-terminus peptide also can block P. gallinaceum sporozoite invasion of the Ae. aegypti salivary glands. These results support the conclusion that the partial NH2-terminus peptides from both P. gallinaceum and P. falciparum may act as ligands in the invasion of mosquito salivary glands, and the possibility of using the peptides to prevent parasite transmission.
Discussion
CSP represents 5–15 % of the total protein expressed in sporozoites and is highly immunogenic [
28,
29]. These characteristics make CSP an excellent candidate for malaria vaccine development. RTS,S AS01, a recombinant protein candidate malaria vaccine that targets the
P. falciparum CSP is the first to reach phase three clinical testing and is partially effective against clinical disease in young African children up to 4 years after vaccination, according to final trial data [
30]. Other important properties of CSP include its ligand-like interaction with mosquito salivary gland receptors and this makes it a target for disrupting parasite invasion in this tissue and thus blocking transmission. The results presented here are the first to show that endogenous production of peptides encompassing NH
2-terminus of CSP protein using either a dsSindbis virus system or transgenic approach, inhibit sporozoite penetration into salivary gland using
Ae. aegypti–
P. gallinaceum model. Moreover this is the first report of a heterologous effect of the
P. falciparum NH
2-terminus of CSP-peptide in the
P. gallinaceum model.
Controversy accompanies the published conclusions of the roles of CSP Regions I and II in receptor-mediated salivary gland recognition [
11,
17‐
19]. The results presented here support the conclusion that the NH
2-terminus (containing Region I) has a role in salivary gland recognition but not TSR (with Region II) in
P. gallinaceum infection in a model system. Previous reports of a role for Region II were based on experiments performed with
P. berghei [
19,
31], however, the genetically-ablated parasites used were disabled in their gliding ability, and this makes it difficult to distinguish between disruption of a receptor-ligand interaction or the lack of motility as a cause of the reduced salivary gland infections. It also is possible that a mutation in Region II inhibits parasites egression from oocysts even though they produce the same numbers of oocyst sporozoites [
31]. Since the previous study did not measure the number of sporozoite in the haemocoel, it is possible that this is the cause of the impairment phenotype observed in both studies. Furthermore, once the NH
2-terminus is removed from mutated parasites, the TSR domain is exposed, switching parasites from a migratory state to an adhesive state reducing the numbers of sporozoite capable of reaching and invading the salivary glands [
11]. Thus, it is unlikely that the TSR + Region II is acting as a ligand for sporozoites for salivary gland invasion.
An evolutionarily-conserved role for the partial NH
2-terminus portion of CSP in salivary gland invasion is supported here, since it was observed that not only peptides from
P. gallinaceum but also from
P. falciparum are able to disrupt
P. gallinaceum sporozoites penetration of
Ae. aegypti salivary glands. Also,
P. gallinaceum and
P. falciparum are thought to be more closely-related than
P. falciparum with other human malaria parasites [
32‐
35]. Thus, information obtained with
Ae. aegypti–
P. gallinaceum model studies may be relevant to
P. falciparum.
Transient expression of Sindbis-PgpNT resulted in a lower mRNA production than the other dsSindbis constructions and this most likely resulted in reduced peptide production, which in turn would not be enough to interfere with sporozoite penetration of the salivary glands. Also, the non-specific amplification product was most likely due to virus instability in cell culture passages that could select viruses with different sequences or for those lacking the transgene [
36].
The choice of the
VgI promoter in the transgenesis construct was based on its high level of expression and peak induction time (~24 h PBM). Previous work has shown its effectiveness in expressing heterologous proteins in
Ae. aegypti transgenic lines [
24,
27]. The signal peptide from the
Ae. aegypti maltase-
like I gene also was added to direct recombinant peptides into mosquito haemolymph [
22].
Although all the elements chosen to generate the transgenic line were tested in previous reports, transgenic lines P#2 and F#8 did not express the transgene and this could be related to position effects, a phenomenon frequently seen in mosquito transgenesis [
37]. Furthermore, F#8 had an abnormal fragment when amplified with specific primers, and a recombination event might have occurred disrupting the integrity of the transgene, so subsequent experiments were performed with line P#1 that did not present these problems.
The initial analysis of parasite challenges of the pBac-PgpNT line P#1 did not reveal a statistically-significant difference for
P. gallinaceum sporozoite in salivary glands between them and non-transgenic controls. Although this result first suggested the incapability of transgenic mosquitoes to block parasite entrance in the salivary gland, a more profound analysis of that initial outcome was performed. Since ligand/parasite competition should be interfered by parasite excess binding to the salivary gland, it was defined a threshold (<2086 sporozoites/salivary gland) to determine mosquitoes with low infection levels and a significant reduction in parasites recovered from salivary glands was observed between pBac-PgpNT line P#1 and control mosquitoes below the threshold, but not above (Additional file
3). Therefore, if the number of parasites is not in excess (saturation), it is possible to block the receptors thus inhibiting sporozoite salivary gland invasion, on the other hand, if the number of produced sporozoites is high, they will be able to overcome the blockade and penetrate the salivary gland. Moreover, the mosquitoes expressing TSR-based peptides Sin-PgRTSR, Sin-PfRTSR, Sin-PfpTSR were unable to generate significant difference compared to control even if a threshold is established (Additional file
4) reinforcing the results presented by the second analysis of pBac-PgpNT line P#1.
The reduction (~30 %) of sporozoites invading salivary glands in the presence of peptides based on NH
2-terminus of CSP from either
P. gallinaceum or
P. falciparum in
Ae. aegypti is lower than that observed (60–80 %) when
A. stephensi were injected with synthetic peptides from
P. falciparum and infected by
P. berghei [
18]. It is likely that the amounts of recombinant peptides produced by both dsSindbis virus and the transgene were lower than the amount injected in that study, and this is reflected in the blocking efficiency like stated previously. It was reasoned that because codon optimization is crucial for heterologous expression of CSP in transgenes in
Anopheles gambiae [
38], it might also be crucial in
Ae. aegypti. Also, CSP is not the only protein described that could promote salivary gland invasion [
5,
6] and, therefore, complete inhibition may not be possible with a single class of peptides.
Recent technological advances in cellular and molecular biology, genetics and bioinformatics, have enabled the development of a transgenic
A. stephensi transgenic lines that block the progression of the
P. falciparum cycle rendering salivary glands free of sporozoites [
39]. This transgenic line is a groundbreaking tool towards an innovative, population modification method to control vector-borne diseases. However, human malaria parasites can be transmitted by a ~465 formally recognized
Anopheles species and ~41 are considered to be dominant vector species/species complexes, capable of transmitting malaria at a level of major concern to public health [
40]. Thus, further studies on the biology of different vectors and host-pathogen interactions are needed to provide a broad spectrum of control possibilities that include all mosquitoes-parasites combinations.
Authors’ contributions
BBK conceived and designed the experiments. MLC, BBK, CM and DAH performed the experiments. BBK, DOC, ALCdS, OM, and MLC analysed the data. MCB, AAJ and MLC contributed reagents/materials/analysis tools. BBK, ALCdS, DOC, MLC, and AAJ wrote the paper. All authors read and approved the final manuscript.