Background
Malaria morbidity and mortality remain a major public health problem globally. According to the World Health Organization (WHO) estimates, there were 229 million cases of malaria in 2019, which resulted in 409,000 deaths, primarily in children under five years of age [
1].
Despite the high death toll caused by malaria, 19 countries attained 3 consecutive years without indigenous cases of malaria, with China and El Salvador the latest to countries to achieve malaria elimination and becoming certified as malaria free by the WHO in 2019 [
1]. Unfortunately, as malaria incidence and transmission in a region decreases, it becomes increasingly difficult to identify areas of active transmission due to the rise in asymptomatic infections and lower numbers of individuals seeking treatment for malarial illness [
2‐
9]. Improved methods for identifying and monitoring foci of active malaria transmission are particularly needed in these areas of low parasite prevalence.
Entomological inoculation rates (EIR), which measure the mean number of infectious mosquito bites per individual over time, remain a widely-accepted measure of transmission, but lack precision due to heterogeneous mosquito distributions [
10]. Furthermore, this method is not without practical and ethical considerations, as it requires mosquito trapping using adult volunteers. Additionally, EIR can be difficult to extrapolate to paediatric populations [
11,
12]. Co-endemic patterns of different
Plasmodium species also pose added difficulties for malaria control efforts as vector-based interventions targeted toward
Plasmodium falciparum are less efficacious against
Plasmodium vivax [
13]. In sub-Saharan Africa,
P. falciparum is by far the most prevalent malaria parasite, but
P. vivax,
Plasmodium malariae, and
Plasmodium ovale can also be co-endemic [
14]. Human populations in many other parts of the world are also subjected to the burden of multiple
Plasmodium species, complicating epidemiological studies and elimination interventions as indicators only exist for the predominant parasite species [
8,
15,
16].
There has been an increase in the use of serological and antibody-based detection assays for measuring exposure and determining transmission intensity in regions of malaria control implementation and those undergoing elimination campaigns [
17‐
19]. Unlike diagnostic assays, which look for the presence or absence of active infection, these quantitative immunoassays offer several advantages for epidemiological studies, including more robust data generation and allowing for estimation of the individual- and population-level malaria exposure history [
8,
20‐
23]. The ability to use dried blood spot samples also makes serological methods pragmatic for field sample collection and laboratory processing [
24]. One or multiple recombinant proteins are typically used to detect anti-
Plasmodium antibodies in sera or blood samples, and common species-specific targets include circumsporozoite protein (CSP), apical membrane antigen 1 (AMA1), and merozoite surface proteins (MSP1, MSP2, and MSP3) [
17,
18,
21,
25,
26]. IgG responses to the recombinant MSP1 19 kD (MSP1
19) isoforms for the four human
Plasmodium species (
P. falciparum, P. vivax, P. malariae, and
P. ovale) have been found to be largely species-specific, giving more confidence in population seroestimates even when these four
Plasmodium spp. are co-endemic [
26].
However, unlike malaria rapid diagnostic tests, which have begun to assess pan-malaria antigen levels [
27], serological studies have yet to include pan-malaria (pan-
Plasmodium) antigens into existing antigen panels for antibody capture. The addition of pan-malaria antigens for IgG detection could improve malaria exposure estimates to aid in malaria elimination campaigns, which aim to eliminate all malaria transmission within a region, not just an individual species. Recently, the use of designed chimeric
Plasmodium proteins as serosurveillance tools has been explored, including a multi-epitope chimeric antigen that contained epitopes from eight
P. falciparum antigens [
28,
29]. While this antigen was well-recognized by
P. falciparum-infected individuals in endemic regions along the China-Myanmar border, it displayed limited cross-reactivity with
P. vivax-infected individuals [
28].
This study aims to assess the ability of a novel chimeric antigen based on
P. vivax MSP1 to serve as a pan-malaria surveillance tool that can act as a marker for exposure to any
Plasmodium species infecting humans. A chimeric
P. vivax MSP1 protein (designated PvRMC-MSP1) has been previously designed, cloned, and expressed for the primary purposes of malaria vaccine studies. This chimeric protein contains an extended version of PvMSP1
19 containing two T helper epitopes present in the PvMSP1
33 fragment of the native protein, five promiscuous T cell epitopes (capable of binding multiple MHC Class II alleles) arrayed in tandem, and a C-terminal (NANP)
6 affinity tag derived from the central repeat region of the
P. falciparum CSP [
30]. The T cell epitopes in PvRMC-MSP1 represent different regions of PvMSP1 able to bind to several MHC class II alleles, with two epitopes also functioning as B cell epitopes [
31]. A previous report showed that PvRMC-MSP1 was recognized by antibodies from individuals living in malaria-endemic areas of Brazil with no indication of genetic restriction based on HLA-DRB1 and HLA-DQB1* allele frequencies [
30]. Due to the evidence that PvRMC-MSP1 can capture IgG from naturally exposed individuals, this study evaluates the ability of PvRMC-MSP1 to induce antibodies that bind to non-vivax-specific antigenic targets and the ability of PvRMC-MSP1 to capture naturally acquired IgG antibodies from exposed individuals with known infection status from all four human
Plasmodium species. This qualitative study aims to demonstrate the potential of engineered antigens for use in serological surveys.
Discussion
Numerous new tools have been developed for mass screening of populations to detect malaria exposure. In the field, these tools are typically designed to assay for some components of
Plasmodium parasites to verify active infections. DNA and proteins produced during the parasite life cycle are attractive targets due to their high sensitivity and specificity in the detection of active infection [
36,
37]. However, especially in areas of low malaria transmission, tools for detection of infection, including rapid diagnostic tests, microscopy, and PCR, can only confirm that infection prevalence is low in the studied population but do not offer data on the exposure history of survey participants. In contrast, the history of individual and population-level exposure can be determined by assessing host-produced antibodies and the window of time for finding a “malaria positive” positive is greatly increased [
8,
25].
Serological studies offer numerous advantages over other measures of prevalence and transmission. Utilizing traditional diagnostics with passive surveillance of non-symptomatic persons in a region would underestimate the number of active cases and prevalence [
38]. Microscopy requires skilled staff and cannot detect missing low-density or submicroscopic infections, nucleic acid-based assays are costly for large scale studies and only detect active infection [
6,
8], and entomological inoculation rates are laborious to collect, have strong biases in heterogeneous environments [
10] and can be difficult to extrapolate to the paediatric population who are most at risk of complications resulting from
Plasmodium infection [
11,
12].
For malaria serological studies, tailoring the sensitivity of the assay using a highly antigenic protein has previously been suggested as a method to improve serological testing and the estimates generated by this data [
8]. A recent assessment of a multi-epitope chimeric protein for use as a serological marker in
P. falciparum elimination settings in Southeast Asia demonstrated the potential of chimeric proteins in serological studies [
28]. Another protein chimera based on the fusion of the
P. falciparum MSP1 and MSP8 antigens has shown efficacy for the induction of growth-inhibitory antibodies in animal models [
39,
40], but has been yet to be assayed against human plasma from naturally-exposed populations.
Based on data generated using these types of engineered antigens for malaria-based serological surveys, as well as our previous report on the high frequency of antibodies to PvRMC-MSP1 in individuals living in a malaria endemic area, it was sought to assess the ability of PvRMC-MSP1 to act as a pan-Plasmodium antigen target for serological surveys. Assessment of the conservation between PvRMC-MSP1 based on P. vivax Belem and the three other major Plasmodium species infecting humans revealed that for the extended 19 kD fragment of the MSP1 antigen, there is an approximately 50% conservation of the amino acid identity between the MSP1 orthologous regions and PvRMC-MSP1.
Previous publications on human antibody binding to the MSP1 19 kD antigen have shown that there is some conformational dependency and that anti-MSP1 antibodies tend to recognize epitopes that are conserved among variant sequences [
41]. However, multiple motifs have been identified within the PfMSP1 19 kD protein which are available for antibody binding [
41,
42], so sequence similarity of a single epitope would not necessarily dictate potential for cross-binding and IgG capture.
Sequence identity among the five promiscuous T cell epitopes present in PvRMC-MSP1 varied widely—from 35 to 85%, with the two epitopes that also function as B cell epitopes, PvT19 and PvT53, displaying amino acid identity ranging between 35 and 55%. This degree of conservation in the MSP1 and T cell epitopes provides a potential explanation for the binding of anti-PvRMC-MSP1 rabbit IgG to
P. falciparum blood-stage schizonts and the ability of PvRMC-MSP1 to capture IgG from travellers infected with heterologous
Plasmodium species. However, the sequence similarity or conformation of the 19 kD fragments alone would not provide a likely explanation for this finding, as human IgG among these antigens from different human
Plasmodium species is largely species-specific [
26].
After determining the level of homology between the MSP1 sequences of the major malaria parasites and PvRMC-MSP1, it was important to assess if this homology offered a functional significance that could allow anti-PvRMC-MSP1 antibodies to be used to capture different
Plasmodium proteins. It was decided to test if anti-PvRMC-MSP1 IgG would be able to bind recombinant MSP1 proteins from the major human malaria species. As expected, anti-PvRMC-MSP1 antibodies had a high degree of binding with the PvRMC-MSP1 and the recombinant
P. vivax MSP1 protein. Notably, it was observed that anti-PvRMC-MSP1 antibodies display cross-reactivity with the MSP1 proteins from
P. falciparum,
P. ovale, and
P. malariae. While the overall degree of cross-binding with non-vivax malaria appears to be very low, the binding signal remains above the detection level even at high antibody dilutions for both rabbit and mouse induced anti-PvRMC-MSP1. Furthermore, it was able to be demonstrated that this cross-reactivity extends beyond recombinant proteins, as purified IgG from rabbits immunized with PvRMC-MSP1 bound to native
P. falciparum MSP1 present in schizonts via IFA. The rabbit sera cross-binding to young parasite forms derived from culture displayed a similar binding pattern to that produced by
P. falciparum hyperimmune human sera. This antibody binding pattern is also consistent with previous observations for IFAs using PvRMC-MSP1 immunized mice and
P. vivax schizonts [
30]. Combined, these data demonstrate that PvRMC-MSP1 can induce cross-reactive antibodies that are functional for the recognition of native malaria species.
Having confirmed that anti-PvRMC-MSP1 antibodies cross-react with non-vivax MSP1 proteins, the chimeric
P. vivax MSP1 antigen was tested for its capacity to capture naturally induced IgG antibodies in plasma samples collected from US travellers with known active infection with one of the four major
Plasmodium species that infect humans. While the species of malaria parasite for the current infection was known, the full medical histories for these individuals is unknown. Therefore, a limitation of this study was not knowing if a subset of individuals would have been exposed to the same or a different
Plasmodium species previously and could display pan-species reactivity as a result of prior exposure. However, though the information was not available, as these are travellers returning to the US with malaria infection, it could be reasonably assumed that this is the first malaria exposure for many. High IgG titers are typically seen with active infection [
24], so this set of plasma from infected persons provided species-induced IgG for which to assess cross binding.
When comparing the IgG capture signal between PvRMC-MSP1 and the four recombinant
Plasmodium proteins using plasma from malaria-infected returning US travellers, irrespective of infective species, it was consistently observed a subgroup that was able to strongly recognize both PvRMC-MSP1 and the recombinant protein. However, while IgG capture between PvRMC-MSP1 and PmMSP1 and PoMSP1 was seen, due to the limited number of
P. ovale (n = 13) and
P. malariae (n = 4) samples included in this study, further assessment using a larger
P. ovale and
P. malariae exposed sample sets is required. This broad IgG binding capacity may be due to the selection of promiscuous T cell epitopes within the PvRMC-MSP1 protein, which included epitopes based on their predicted ability to bind multiple human MHC class II molecules. An assessment of synthetic peptides from
P. vivax MSP1 preceding the development of PvRMC-MSP1 showed a high degree of variability in the ability of peptides to bind HLA class II molecules and to be recognized by individuals living in endemic areas of malaria [
31].
Furthermore, data from the initial characterization of PvRMC-MSP1 demonstrated that 50.4% of individuals living in Brazil had antibodies able to recognize the PvRMC-MSP1 with no indication of genetic restriction based on HLA-DRB1 and HLA-DQB1* alleles [
30].
During the assessment of plasma from travellers based on the cause of active infection, we observed that the majority of
P. vivax-infected individuals carrying IgG antibodies had a higher degree of binding to PvRMC-MSP1 when compared to the recombinant PvMSP1 protein alone. Though these antigens were coupled to the microbeads at the same concentration, the assay signal of PvRMC-MSP1 versus PvMSP1 was increased for 89.5% (34/38) of
P. vivax-infected plasma, with some samples showing an increase of over an order of magnitude. It is possible that the reorganization of the broadly recognized epitopes from PvMSP1 into the chimeric PvRMC-MSP1 protein, and the removal of sequences of low antigenic value, made the broadly recognized epitopes more accessible to antibody binding, thereby increasing assay signal. It has previously been reported circular dichroism studies of PvRMC-MSP1 [
30], suggesting that the carboxyl-terminal region in PvRMC-MSP1 is in the same conformation as reported for native PvMSP1. Therefore, it does not appear the conformational changes to be the cause of increased binding to PvRMC-MSP1 over recombinant PvMSP1.
The increased binding observed for PvRMC-MSP1 over recombinant PvMSP1 may aid in the development of improved serological assays for use in
P. vivax endemic regions approaching elimination. Population-based studies in these regions require highly sensitive assays as a decreasing infection burden leads to increases in the frequency of asymptomatic cases and lack of treatment-seeking behaviour, making detection of cases more challenging. Many regions of South and Central America are currently within or approaching the elimination phase of malaria control, and
P. vivax is the predominant residual species in some of these areas [
43]. Having a more sensitive serological tool for detection of
P. vivax exposure would be able to assist the Americas and other global settings in mapping out regions and populations where malaria still resides. Furthermore, in countries within the Horn of Africa,
P. falciparum and
P. vivax both circulate, so serological studies based on PvRMC-MSP1 should be using in combination with recombinant PfMSP1 and PvMSP1 to obtain species specific reactivity in addition to the pan-
Plasmodium signal that could result from use of PvRMC-MSP1 alone.
Designed antigens with intentional broad antibody binding capacity could provide a valuable tool for use in serological assessment studies when compared to whole recombinant antigens alone, which are intended to be true to the genome-encoded antigen. This current study benefited from the use of the bead-based multiplex immunoassay, which allows simultaneous data collection of IgG presence and levels against multiple antigens [
20]. Including both broad-reacting, as well as
Plasmodium species-specific, antigen-coated beads in an assay panel could provide a very nuanced view of individual and population-level exposure histories as well as provide an extensive IgG profile and detailed seroestimates [
44,
45]. The population of US resident travellers returning with malaria infection would be more biased toward those with nascent exposure, but a limitation to this study is the lack of information on the number of previous episodes, or what
Plasmodium species persons would have been previously exposed to in their lives. This possibility of previous lifetime exposure provides a reasonable explanation for the finding of IgG antibodies capable of binding the MSP1 proteins from the other, non-infecting,
Plasmodium species.
Beyond known active Plasmodium infections, future studies will work to assess antibody binding to PvRMC-MSP1, and other chimeric antigens, to naturally exposed human populations in different regions of the world with different transmission intensities and co-endemic patterns. Access was not available to any Plasmodium knowlesi-infected US travellers during the sample collection period. Due to the increasing importance of this Plasmodium species in Southeast Asia, future studies should be conducted to assess the ability of this broadly recognized PvRMC-MSP1 protein to capture antibodies generated as the result of P. knowlesi exposure. Furthermore, the dynamics of the anti-Plasmodium antibody response, and particularly anti-MSP1 antibodies, is not well established. It remains to be evaluated how quickly these antibody responses form in naïve individuals and how durable these antibody responses are, with consideration given to age and transmission intensity.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.