The use of a chimeric antigen for Plasmodium falciparum and P. vivax seroprevalence estimates from community surveys in Ethiopia and Costa Rica.

BACKGROUND: In low-transmission settings, accurate estimates of malaria transmission are needed to inform elimination targets. Detection of antimalarial antibodies provides exposure history, but previous studies have mainly relied on species-specific antigens. The use of chimeric antigens that include epitopes from multiple species of malaria parasites in population-based serological surveys could provide data for exposure to multiple Plasmodium species circulating in an area. Here, the utility of P. vivax/P. falciparum chimeric antigen for assessing serological responses was evaluated in Ethiopia, an endemic country for all four human malarias, and Costa Rica, where P. falciparum has been eliminated with reports of sporadic P. vivax cases.
METHODS: A multiplex bead-based assay was used to determine the seroprevalence of IgG antibodies against a chimeric malaria antigen (PvRMC-MSP1) from blood samples collected from household surveys in Ethiopia in 2015 (n = 7,077) and Costa Rica in 2015 (n = 851). Targets specific for P. falciparum (PfMSP1) and P. vivax (PvMSP1) were also included in the serological panel. Seroprevalence in the population and seroconversion rates were compared among the three IgG targets.
RESULTS: Seroprevalence in Costa Rica was 3.6% for PfMSP1, 41.5% for PvMSP1 and 46.7% for PvRMC-MSP1. In Ethiopia, seroprevalence was 27.6% for PfMSP1, 21.4% for PvMSP1, and 32.6% for PvRMC-MSP1. IgG levels in seropositive individuals were consistently higher for PvRMC-MSP1 when compared to PvMSP1 in both studies. Seroconversion rates were 0.023 for PvMSP1 and 0.03 for PvRMC-MSP1 in Costa Rica. In Ethiopia, seroconversion rates were 0.050 for PfMSP1, 0.044 for PvMSP1 and 0.106 for PvRMC-MSP1.
CONCLUSIONS: Our data indicate that chimeric antigen PvRMC-MSP1 is able to capture antibodies to multiple epitopes from both prior P. falciparum and P. vivax infections, and suitable chimeric antigens can be considered for use in serosurveys with appropriate validation.

Introduction

In 2021, the World Health Organization reviewed the progress made towards malaria elimination since 2000 and placed increased emphasis on examining the global trends in the burden of malaria to identify any variations in malaria burden by age as malaria transmission declines [1]. Reliable measures of transmission are particularly critical in the setting of malaria elimination as decreasing malaria incidence is associated with an increase in the proportion of asymptomatic infections as well as a decrease in treatment-seeking behavior and identified cases for epidemiological estimates [2-10]. Therefore, conventional diagnostic methods which provide data on active infection, such as microscopy or malaria rapid diagnostic tests, are not as well suited for estimating the true level of recent or ongoing malaria exposure in a population.

In recent years, serological surveillance methods have repeatedly demonstrated utility in measuring transmission in low malaria-endemic regions [2-10]. Serological methods offer multiple advantages well suited for use in surveys, primarily their ability to generate estimates of malaria exposure history in a population by detecting antibodies against parasite antigens [11,12]. Antibody seroprevalence and the age of participants are used to estimate the rate of antimalarial antibody acquisition in a population [9,11]. These estimates can be used to monitor differences among populations in an area, or changes in transmission over time, or in response to an intervention [7].

While serological methods have seen increased use in the assessment of malaria transmission, these studies have mainly relied on recombinant antigens based on a single antigen or on epitopes that share sequence identity with naturally occurring Plasmodium antigens [13]. Multiplex assays allow for the simultaneous assessment of antibody responses to multiple antigens from a single sample [12,14]. It remains to be determined if chimeric antigenic constructs from different malaria parasite species can be used to collect serology data for multiple species in serosurveys.

A recent study from our laboratory described the pattern of serological response in U.S. travelers [15] using a chimeric protein based on Plasmodium vivax MSP1 and P. falciparum circumsporozoite protein (CSP) repeat epitopes [16], known as PvRMC-MSP1. It was evident from this study that this chimeric antigen captured IgG from a majority of returning U.S. travelers with PCR confirmed malaria infection regardless of the Plasmodium species responsible for infection [15]. Furthermore, an increased assay signal was observed for PvRMC-MSP1 compared to recombinant PvMSP1 in 34 out of 38 active P. vivax infections [15].

A better understanding of the utility of chimeric antigens for capturing antibody responses in endemic populations with different levels of malaria exposure will help determine future applications of such tools for malaria serological studies. Therefore, this study evaluated the ability of PvRMC-MSP1 to capture IgG from dried blood spot samples (DBS) collected from Ethiopia and Costa Rica. Ethiopia was selected for this study because the country is co-endemic for both P. falciparum and P. vivax, making up 69% and 27% of malaria cases, respectively [17]. Costa Rica was selected as a representative low, mono-species Plasmodium endemic region, and as of 2017, 98% of the country was considered malaria-free, with only three remaining foci of active transmission for P. vivax [18]. A multiplex bead-based assay was used to evaluate the population-level IgG binding to PvRMC-MSP1 in comparison to recombinant P. falciparum MSP1 and recombinant P. vivax MSP1 via modeled seroprevalence curves and LOESS regression curves.

Materials and methods

Study sites, design, enrollment, and ethics statements

Ethiopian samples were collected as part of the 2015 Ethiopia Malaria Indicator Survey (MIS) [19], a national household survey that occurred between September 30 and December 10, 2015, which coincided with the high malaria transmission season and covered 13,875 households in malarious areas between 2,000m and 2,500m above sea level. Two-stage cluster randomized sampling was used to select 555 areas and 25 households within each area to ensure that the survey participants were representative of malaria endemic regions throughout the country. In each household, blood samples were obtained from all children under five years of age upon obtaining informed consent from their parent/guardian. In Ethiopia, the informed consent forms and questionnaires were translated into Amharic, Oromiffa, and Tigrigna languages and read aloud in their entirety by the survey team in the participant’s language of choice to ensure understanding before verbal consent was obtained from the adult participant, the parent/guardian of any participant below the age of 18, and verbal assent for children below the age of 18. Verbal consent/assent was documented along with responses to the questionnaire. In every fourth household, persons of all ages were enrolled in the study upon obtaining informed consent. Blood was spotted onto filter paper and allowed to dry (creating dried blood spots, DBS) before each sample was individually packaged. The MIS-2015 protocol received ethical clearance from Ethiopia’s National Research Ethics Review Committee. The survey protocol underwent human subjects review at the US Centers for Disease Control and Prevention (CDC) and received non-research determination. Additional ethical clearance for the present serology study was obtained from the Institutional Review Board of the College of Health Sciences of the Addis Ababa, University (AAUMF 03–008). From the 2015 Ethiopia MIS, 7,077 DBS were available for serological data collection.

In the Costa Rican canton of Matina in 2015, a total of 851 individuals were enrolled in their households and provided blood samples for DBS creation. The canton of Matina in the province of Limón was selected for this study, given that it was one of the last locations in the country where malaria cases were present. In Costa Rica, informed consent forms were read aloud in Spanish by the survey team before signatures were obtained from participating adults older than 18, parental/guardian consent for any participant below the age of 18, and assent from children 12–17. At enrollment, a questionnaire was administered and whole blood samples were collected on filter paper for molecular and serological analysis. The study was approved by the Costa Rica Ministry of Health (DVS-721-2015), and the study protocol underwent human subjects review at CDC and received non-research determination.

For both Ethiopian and Costa Rican studies, the laboratory staff at the CDC and Emory University had no contact with the participants of this study and no access to any personal identifiers.

Recombinant and chimeric Plasmodium antigens used for multiplex assay

The structure and production of the chimeric P. vivax merozoite surface protein 1 has been previously described [15,16]. Briefly, the P. vivax recombinant modular chimera based on MSP1 (PvRMC-MSP1) was based on the P. vivax Belem sequence (GenBank: XP_001614842.1) and contains five promiscuous T cell epitopes, with two also functioning as B cell epitopes, as well as an extended version of the P. vivax MSP1 19 kDa fragment containing two T helper epitopes from the MSP1 33 kDa fragment, and the (NANP)6 repeat peptide from P. falciparum circumsporozoite protein (CSP) [16]. Each T cell epitope included in PvRMC-MSP1 is 19 amino acids in length. In addition to PvRMC-MSP1, recombinant PfMSP1 and PvMSP1 antigens and the NANP repeat peptide from PfCSP were also used; the production and use of these antigens has been reported previously [8,20].

Multiplex bead-based assay (MBA)

Magnetic MagPlex microspheres (Luminex Corp.) were covalently linked to the chimeric PvRMC-MSP1 protein, the recombinant PfMSP1 and PvMSP1 antigens, and the PfCSP peptide as previously described [12,15], with all three recombinant proteins conjugated at a concentration of 20 μg/mL and the peptide at a concentration of 30 μg/mL. To obtain whole blood from the filter paper, a 6 mm diameter hole punch from individual DBS samples were eluted overnight at 4°C in buffer containing: PBS, 0.05% Tween 20, 0.5% polyvinylpyrrolidone, 0.5% poly(vinyl) alcohol, 0.1% casein, 0.5% BSA, 0.02% NaN3, and 3 μg/mL E. coli extract to prevent nonspecific binding. Samples were subsequently diluted to 1:100 serum concentration for the immunoassay.

For the multiplex immunoassay, all data were collected with an overnight incubation assay as described previously [14]. Briefly, in 5 mL of reagent buffer (PBS, 0.05% Tween 20, 0.5% BSA, 0.02% NaN3), a bead master mix was prepared with all conjugated beads included, and 50 μL of the bead mix was pipetted into each well of a BioPlex Pro plate (Bio-Rad, Hercules, CA). Beads were washed twice on a handheld magnet (Luminex Corp, Austin, TX) with 100 μL wash buffer (PBS, 0.05% Tween 20). After washing, 50 μL of reagent mix (in 5 mL of reagent buffer: 1:500 anti-human IgG, 1:625 anti-human IgG4, 1:200 streptavidin-PE) was added to all wells, then 50 μL samples (or controls) were added to the reagent mix in the appropriate wells. Anti-IgG4 was added separately to the anti-human IgG mixture as we have previously found that anti-IgG4 responses are difficult to determine using only the standard anti-IgG cocktail [21]. All anti-human antibodies were obtained from Southern Biotech (Birmingham, AL) and streptavidin-PE from Invitrogen (Waltham, MA). Plates were incubated overnight with gentle shaking at room temperature and protected from light. The following day (after ~ 16 h total incubation time), plates were washed three times, and beads were resuspended with 100 μL PBS and read on a MAGPIX machine (Luminex Corp). MFI signal was generated for a target of 50 beads/region. A background MFI from wells including only elution buffer was subtracted from each sample to give a final value of MFI-bg.

Statistical analyses

A seropositivity threshold was generated for each antigen by assaying a panel of 92 U.S. resident blood donors without recent travel outside the country and determining the lognormal mean plus three standard deviations for the MFI-bg signal. A reversible catalytic model was fit to the seropositivity by age data for each antigen, and the estimates for the serological conversion rate (SCR) and serological reversion rate (SRR) per year were directly calculated from the likelihood model [22] utilizing R version 3.3.2 (R Foundation for Statistical Computing, Vienna, Austria). Local regression (LOESS) curves for IgG levels by age were generated in SASv9.2 (Cary, NC) by the sgplot procedure with cubic interpolation and a smoothing degree of 1.

Results

Ethiopia and Costa Rica study populations

From the 2015 Ethiopia MIS, 7,077 DBS were available for serological data collection, and 851 DBS were available from the 2015 Costa Rica survey. In the Ethiopia survey, 53.1% of participants with serological data were male, and nearly half (47.2%) of persons were five years of age or younger (Table 1). From the Costa Rica survey, 32.5% of participants with serological data were male, and the youngest participant was 12 years old.

Table 1

Demographic characteristics of the Costa Rica and Ethiopia study populations.

Study, Variable	Number (%)
Ethiopia
Male	3757 (53.1%)
Female	3320 (46.9%)
Age 0-5y Age 6-10y Age 11-20y Age 21-30y Age >30y	3342 (47.2%)774 (10.9%)933 (13.2%)749 (10.6%)1279 (18.1%)
Costa Rica
Male	277 (32.5%)
Female	574 (67.5%)
Age 0-5y Age 6-10y Age 11-20y Age 21-30y Age >30y	0 (0.0%)0 (0.0%)119 (14.0%)139 (16.3%)593 (69.7%)

Population-level response to antigens in Ethiopia and Costa Rica

The percentage of participants from each survey who were IgG positive (seropositive) for each of the markers varied substantially by antigen and study population. Overall seroprevalence in Costa Rica was 3.6% for PfMSP1, 41.5% for PvMSP1 and 46.7% for PvRMC-MSP1. In Ethiopia, overall seroprevalence was 27.6% for PfMSP1, 21.4% for PvMSP1 and 32.6% for PvRMC-MSP1. For both surveys, increases in both seropositivity and IgG levels were noted with an increase in age (Figs 1 and 2), though these increases were minor for the PfMSP1 antigen in the Costa Rican population. For Ethiopia specifically, plotting the IgG assay signal against age revealed that antibodies against the PfMSP-1 antigen were highest within the Ethiopian populace, followed by PvRMC-MSP1, and lastly, PvMSP-1 (Fig 1B). A different pattern was noted in the Costa Rican populace, with the PvRMC-MSP1 antigen providing the highest IgG signal among all ages, followed by IgG against PvMSP1 and very low IgG levels against PfMSP1 for all ages (Fig 2B).

Fig 1

Population-level responses to PvRMC-MSP1, PfMSP1, and PvMSP1 from Ethiopian persons.

A) Modelled seroprevalence curves for the Ethiopian study population for the three antigens, with seroconversion estimates shown in Table 2. B) LOESS regression curves for IgG assay signal of the study population to each antigen by age. A minimal anonymized data set for the responses to PvRMC-MSP1 in Ethiopia is included with the Supporting Information for this publication.

Fig 2

Population-level responses to PvRMC-MSP1, PfMSP1, and PvMSP1 from Costa Rican persons.

A) Modelled seroprevalence curves for the Costa Rican study population for the three antigens, with seroconversion estimates shown in Table 2. B) LOESS regression curves for IgG assay signal of the study population to each antigen by age. The x-axis for the LOESS regression curves starts at age 12, corresponding to the youngest participant from the Costa Rican survey. A minimal anonymized data set for the responses to PvRMC-MSP1 in Costa Rica is included with the Supporting Information for this publication.

Table 2

Serological conversion rate for plasmodium antigens in the Ethiopian and Costa Rican study populations.

Country	Antigen	Point Estimate	Lower 95% bound	Upper 95% bound
Ethiopia
	PfMSP1	0.05	0.045	0.055
	PvMSP1	0.044	0.039	0.049
	PvRMC-MSP1	0.106	0.095	0.116
Costa Rica
	PfMSP1	0.00086	-0.001	0.003
	PvMSP1	0.023	0.02	0.03
	PvRMC-MSP1	0.03	0.02	0.04

Modeling for seroconversion rates

For the Ethiopian data, the highest seroconversion rate was observed for PvRMC-MSP1 at 0.106 (95% CI: 0.095–0.116), compared to PfMSP1 at 0.050 (95% CI: 0.045–0.055) and PvMSP1 at 0.044 (95% CI: 0.039–0.049) (Table 2). Analysis of the Costa Rican data revealed the seroconversion rate of PvRMC-MSP1 to be the highest of the three antigens assayed, at 0.030 (95% CI: 0.02–0.04), followed by PvMSP1 at 0.023 (95% CI: 0.02–0.03), and then PfMSP1 at 0.00086 (95% CI: -0.001–0.003).

Assay signal between PvRMC-MSP1 and PfCSP

Although PvRMC-MSP1 is based primarily on the P. vivax Belem sequence, six copies of the NANP repeat sequence from the P. falciparum circumsporozoite protein (PfCSP) are present at the C-terminus of PvRMC-MSP1 and were included during the design of PvRMC-MSP1 to serve as an additional purification tag. To determine if antibody recognition and binding to the PfCSP region is responsible in part for the assay signals observed for PvRMC-MSP1, the MFI-bg signals were compared between the chimera and a peptide containing five copies of the PfCSP NANP peptide (NANPx5, S1 Fig). For the Ethiopian study population, a majority of samples produced higher MFI-bg signals in response to antigen capture by PvRMC-MSP1 than for PfCSP NANPx5 peptide, with a comparatively smaller subset of samples tracking along the x = y reference line or showing increased signal for PfCSP than PvRMC-MSP1 (S1A Fig). The MFI-bg signal obtained for the PvRMC-MSP1 and PfCSP antigens from individuals from Costa Rica was also compared, revealing primary binding to PvRMC-MSP1 as compared to PfCSP, with relatively few samples following the x = y reference line (S1B Fig).

Discussion

Malaria serological data offers the advantage of expanding the window of time to assess prior malaria infection in the human population because exposed individuals maintain anti-malaria IgG antibodies for months to years after exposure to malaria parasites [22], and this methodology can use targets that are not affected by short-term variations in malaria transmission [5]. Additionally, these methods can detect antibodies from convenient samples, such as DBS, which are easier to collect and store than blood fractionation techniques and cold storage [23]. Assessing the presence of antibodies for multiple antigen targets through bead-based multiplex assay makes serological studies high throughput and cost-effective for application in seroepidemiologic studies aimed at understanding the burden of malaria transmission within a community [9,11]. As a result, these characteristics make malaria serological studies an excellent tool for evaluating prevalence in malaria elimination settings.

Although there are still few reports of malaria seroprevalence studies using designed antigens, the potential of a chimeric antigen approach has been recently demonstrated in a study assessing malaria transmission intensity along the China-Myanmar border [13]. This study used a designed antigen that included 11 epitopes from eight P. falciparum antigens, including single epitopes from CSP and MSP2, and two epitopes each from AMA1 and MSP1, which was able to differentiate individuals living in endemic and non-endemic regions within the study area [13]. However, unlike the PvRMC-MSP1 antigen reported here, the design of the antigen used in the China-Myanmar study aimed to avoid cross-reactivity with P. vivax, highlighting the potential to customize engineered seroprevalence antigens for a seroprevalence study to either include all Plasmodium species or target only one. However, the chimeric antigen PvRMC-MSP1 used here was originally designed as an experimental vaccine construct that included promiscuous T cell epitopes from P. vivax Belem strain, P. vivax MSP-1 19kDa antigen and six repeats of NANP sequence from the CSP antigen of P. falciparum [15,16,24]. Therefore, the use of the PvRMC-MSP1 antigen allowed for the investigation of how a hybrid recombinant antigen with epitopes from two different malaria parasites could capture antibody reactivity patterns from endemic populations with varying history of transmission pattern and exposure history.

Two different malaria-endemic settings were utilized for this study. Ethiopia is a country in the horn of Africa that is known to be endemic for all four human malarias [5] and is unique in the African context in that P. vivax accounts for approximately one-third of reported clinical malaria cases [17]. Ethiopia has seen reductions in its malaria burden and aims for malaria elimination by 2030. Costa Rica is a country in Central America that has drastically suppressed malaria transmission to the point where nearly all identified cases are identified as being imported infections [25]. Additionally, though P. falciparum was noted in Costa Rica many decades ago [26], most recent cases are caused by P. vivax [18]. The malaria transmission settings for these two countries were reflected in both the seroprevalence to the antigen targets used in the multiplex assay as well as the estimated seroconversion rates.

For the species-specific antigens in the Ethiopia study, more PfMSP1 seropositive individuals were identified when compared with PvMSP1, with 28% and 21% of study participants seropositive for these two targets, respectively. The chimeric PvRMC-MSP1 antigen provided even more seropositive calls, with 33% of all blood samples having IgG against this target. This increase in the seroprevalence estimate is likely due to the cumulative capture of antibodies to multiple epitopes such as universal T cell epitopes of PvMSP1 and PfCSP epitopes, in addition to PvMSP1 19kDa antigen (S1A and S1B Fig) [15,16,24]. Though the PvRMC-MSP1 antigen provided the highest seroprevalence estimate in the Ethiopian population, the assessment of the IgG levels found that by far the PfMSP1 antigen had the highest titers in the population, which may be due to the increased transmission as well as multiple exposures of the Ethiopian study population to P. falciparum.

In the Costa Rican setting, seropositivity to PfMSP1 was almost nonexistent, with only the oldest individuals showing any notable levels (approximately 10% seropositivity in persons over 70 years of age, versus 2.9% in persons younger than 70). The minimal response to PfMSP1 was not a surprising finding because P. falciparum has rarely been reported in Costa Rica over the past fifty years [25,26]. Seroprevalence was 42% for PvMSP1 and 47% for PvPMC-MSP1, consistent with our hypothesis that this reactivity pattern can be attributed to cumulative response due to the presence of multiple epitopes of P. vivax MSP1 besides the 19kDa domain in the chimeric antigen.

Estimates for seroconversion rates for the three antigens in the Ethiopian and Costa Rican studies provided further information for the transmission dynamics of malaria parasites in these populations over time and displayed the utility of the chimeric antigen for malaria serological estimates. In Ethiopia, the increased seroconversion rate for PvRMC-MSP1 over the PfMSP1 and PvMSP1 antigens (0.11 versus 0.05 and 0.04 respectively) showed the increased rate at which antibodies that recognize this antigen are acquired in children compared to the other MSP1 antigens. Indeed, the 95% confidence interval for the regression curve fitting for the chimeric antigen was completely separated from the confidence intervals for the other two antigens in the 0-15y age group (Fig 1A), and the overall estimated seroconversion rate was statistically higher for the chimera (Table 2). The population considered seropositive plateaued for all three antigens around age 30, though average IgG levels continued to increase as persons aged in the population (Fig 1A versus 1B). The IgG signal for PvRMC-MSP1 is primarily due to anti-P. vivax antibodies, but since the antigen can also capture anti-PfCSP antibodies, these seroprevalence and seroconversion data should be interpreted as exposure to either of these two malaria parasites.

Assessment of seroconversion rates in Costa Rica provided further evidence of the increased sensitivity of PvRMC-MSP1 over PvMSP1 for detecting P. vivax exposure in this low-endemic setting. The seroconversion rate for PvRMC-MSP1 for the study population from the canton of Matina in Costa Rica was again the highest of the three antigens tested. However, all three of these rates were more than an order of magnitude lower in Costa Rica when compared to the Ethiopian estimates (Table 2), alluding to the much lower exposure of the Costa Rican population to malaria parasites over the past decades when compared with the Ethiopian populace. Two limitations to the Costa Rica study included enrollment of persons only aged 12 years and older and the relatively fewer numbers (n = 851) of persons enrolled. For these reasons and the very low overall incidence of malaria transmission, modeling for seroconversion rates showed wider and overlapping confidence intervals with both PvRMC-MSP1 and PvMSP1. Not surprisingly, seropositivity to PfMSP1 was almost nonexistent except in the oldest persons who may have been exposed decades ago [18,26,27].

Some limitations of this study include a) The chimeric construct used here was designed to elicit broad T cell reactivity as a vaccine candidate and contained only selected antigenic domains. It will be useful to compare different chimeric constructs with broader inclusion of B cell epitopes from multiple antigens in future studies; b) The study population was limited to two different endemic populations with very different transmission patterns. Future studies that include specimens from multiple other endemic settings (including diverse age groups) will provide additional insights for assessing the value of chimeric antigens. Despite these limitations, the increased sensitivity of PvRMC-MSP1 to capture antimalarial IgG even in younger populations and in a low endemic setting highlights the potential utility of well-designed chimeric antigens for use in seroepidemiological studies. Chimeric antigens can be constructed using recombinant technology as applied in the development PvRMC-MSP1 construct or synthesized using longer multiple antigen peptides with defined epitopes from different species or target antigens [28]. Exploring such chimeric antigen constructs can broaden the toolbox for seroepidemiological studies and may help to improve their utility in populations with limited ongoing exposure.

Conclusion

The data from this study demonstrate the potential use of chimeric antigens for population-level serosurveillance as these antigens can be tailored to capture antibodies to multiple epitopes in different endemic settings/populations. Engineered antigens based on highly recognized epitopes may also allow for increased sensitivity due to the reorganization of the antigen to expose these highly recognized epitopes while simultaneously removing sequences of low antigenic value. Taken together, the use of engineered antigens and the bead-based multiplex technique can allow for robust data collection for seroepidemiological studies, especially for malaria elimination campaigns.

Cross-binding of anti-PfCSP IgG with PvRMC-MSP1.

A) Scatterplot of PfCSP signal compared to PvRMC-MSP1 for the Ethiopian study population. B) Scatterplot of PfCSP signal compared to PvRMC-MSP1 for Costa Rican study population. The hashed reference line shown is y = x.

(TIF)

Click here for additional data file.

(XLS)

Click here for additional data file.

(XLS)

Click here for additional data file.

2 Mar 2022

11 Mar 2022

Response to Reviewer Comments

Manuscript Title: The use of a chimeric antigen for Plasmodium falciparum and P. vivax seroprevalence estimates from community surveys in Ethiopia and Costa Rica

PLOS ONE Submission ID: PONE-D-22-01720

Reviewer 1 - Review Comments to the Author:

1. Throughout the article the acronym “PvRMC-MSP1” was used a few times for the chimeric malaria antigen however, it would be better to use the term (PvRMC-MSP1/PfCSP) on the whole text, as the use of the first term could lead to a misinterpretation, that the antigen used had only specific targets for vivax.

We thank the reviewer for this suggestion, but this has previously been the name designated to this antigen by our multiple previous publications, and it should remain the same for consistency. Throughout the manuscript, we state multiple times that PvRMC-MSP1 also contains the epitopes for PfCSP, so this is made clear throughout the whole text for the reader’s interpretation.

2. In the introduction, it would be better to use the most current reference from the World Health Organization – Malaria Report 2021 (References 1, 17, 18).

The reviewer is correct that newer references should be used where possible. The World Health Organization Malaria Report citation has been updated to the 2021 version for references 1. The first sentence has also been updated to “In 2021, the World Health Organization reviewed the progress made towards malaria elimination since 2000 and placed increased emphasis on examining the global trends in the burden of malaria to identify any variations in malaria burden by age as malaria transmission declines (1).” to reflect the focus of the most recent WHO Malaria report.

The 2018 Malaria Country Profiles for Ethiopia and Costa Rica are the most recent versions of these reports available for these countries from the WHO.

3. In the “Materials and Methods - Study sites, design, enrollment, and ethics statements” in relation to Ethiopia, briefly explain how the areas of study were selected. Insert in this topic the total number of participants from Ethiopia.

The household survey samples obtained from the multiple sites in Ethiopia correspond to the sites included in the 2015 Multiple Indicator Survey. The study was conducted from September 30, to December 10, 2015, covering a sample of 13,875 households in malarious areas between 2,000m and 2,500m above sea level. The goal was to ensure that the survey participants were representative of malaria endemic regions throughout the country. After completion of the main goals of the Multiple Indicator Survey, the remaining 7,077 DBS were made available for serological data collection for this study.

Lines 88-93 have been updated to: “Ethiopian samples were collected as part of the 2015 Ethiopia Malaria Indicator Survey (MIS) (19, 20), a national household survey that occurred between September 30 and December 10, 2015, which coincided with the high malaria transmission season and covered 13,875 households in malarious areas between 2,000m and 2,500m above sea level. Two-stage cluster randomized sampling was used to select 555 areas and 25 households within each area to ensure that the survey participants were representative of malaria endemic regions throughout the country.”

4. Explain - From the Costa Rica the youngest participant was 12 years old however, in Figure 2b, there are data for children under 12 years old.

The authors thank the reviewer for suggesting additional clarification in figure 2. In panel A of Figure 2, seropositivity to each antigen is modeled down to age 0 using data collected from Costa Rican participants ages 12 and up. In panel B showing the LOESS regression curves, the x-axis starts at 12 years and the curve stops at 90 years. We have revised the legend for panel B to indicate the x-axis start to avoid confusion: “The x-axis for the LOESS regression curves starts at age 12, corresponding to the youngest participant from the Costa Rican survey.”

5. In the discussion, in relation to Ethiopia, it would be recommended comparing the data from the different selected areas of the country relating to seroconversion rate and the prevalence of the different species of plasmodium. Use tables and figures to relate this data.

The authors thank the reviewer for this interesting question. Based on the Ethiopian Public Health Institute’s National Strategic Plan 2021-2025, the most complete and recent aggregated regional data, the P. falciparum/P. vivax ratio has maintained at 70/30% for these parasite species for most regions of the country. Even with changes over time, P. falciparum has remained the dominant malaria species in Ethiopia. Specifically with serological data, this study examined population exposure over long periods of time due to the longevity of IgG responses. Based on these factors, it would be difficult to correlate temporal differences throughout the country with serological data.

Submitted filename: McCaffery_etal__PvRMCMSP1_PLOS_One_Response to Reviewers.docx

Click here for additional data file.

18 Apr 2022

19 Apr 2022

Response to Reviewer Comments

Manuscript Title: The use of a chimeric antigen for Plasmodium falciparum and P. vivax seroprevalence estimates from community surveys in Ethiopia and Costa Rica

PLOS ONE Submission ID: PONE-D-22-01720

Review Comments to the Author:

Reviewer #1: The authors carefully answered the questions raised above.

The modifications contributed to a better understanding of the study.

The research described in this article is unprecedented and highly relevant.

The authors thank the reviewer for their positive feedback of our manuscript.

Reviewer #2:

1. Introduction:

a. Line 69: Sorry if I misunderstood, but reference 15 (PMID: 27708348) doesn’t show results about U.S travelers, only 16 (PMID: 33579292).

The authors apologize for the confusion. Original citation 15 is the original production of the chimeric protein as a vaccine candidate and includes the relevant details regarding design. Original citation 16 is the results of the serology studies from malaria infected US travelers.

We have revised lines 68-72 to include these citations in their appropriate locations.

b. Additionally, you should mention the results of a seroepidemiological study involving individuals naturally exposed to P. vivax and cross-reactivity in individuals with active infection by P. falciparum from 16 (Figure 6).

The authors thank the reviewer for this suggestion. Lines 70-74 of the introduction currently state “It was evident from this study that this chimeric antigen captured IgG from a majority of returning U.S. travelers with PCR confirmed malaria infection regardless of the Plasmodium species responsible for infection (15). Furthermore, an increased assay signal was observed for PvRMC-MSP1 compared to recombinant PvMSP1 in 34 out of 38 active P. vivax infections (15).”

c. Moreover, based on which results did the authors decide to use a P. vivax chimeric protein to evaluate P. falciparum antibodies? It is important to justify. I felt confused reading the introduction.

Thank you for pointing out that the rationale of this study is not sufficiently clear. Currently the rationale of this study is summarized on lines 65-67: “It remains to be determined if chimeric antigenic constructs from different malaria parasite species can be used to collect serology data for multiple species in serosurveys,” and lines 75-82: “A better understanding of the utility of chimeric antigens for capturing antibody responses in endemic populations with different levels of malaria exposure will help determine future applications of such tools for malaria serological studies. Therefore, this study evaluated the ability of PvRMC-MSP1 to capture IgG from dried blood spot samples (DBS) collected from Ethiopia and Costa Rica. Ethiopia was selected for this study because the country is co-endemic for both P. falciparum and P. vivax, making up 69% and 27% of malaria cases, respectively (17). Costa Rica was selected as a representative low, mono-species Plasmodium endemic region, and as of 2017, 98% of the country was considered malaria-free, with only three remaining foci of active transmission for P. vivax (18).”

It is important to note that we are not suggesting that all anti-P. falciparum antibodies are captured by this chimeric antigen since it is based on P. vivax MSP1 and the NANP-repeat region of P. falciparum CSP. Due to the CSP region of the chimeric protein, as well as the promiscuous B cell epitopes from P. vivax MSP1, we have the capacity to capture some amount of anti-P. falciparum IgG. We want to emphasize in this publication is that there currently isn’t an ideal antigen that can capture all IgG responses induced following exposure to the different malaria species. To address this gap, we present a proof of concept that an engineered antigen can be used to capture antibody responses from exposure to different malaria species, and that this is a useful approach to consider when developing future engineered antigens. We believe that this idea is sufficiently covered in the discussion.

Regarding the contribution of the P. vivax MSP1 portion of the chimera versus the P. falciparum CSP NANP repeat region to contribute to the capture of IgG from P. falciparum-exposed persons, as part of our previous publication where we investigated the ability of US travelers with active malaria infection to recognize PvRMC-MSP1 as compared to recombinant MSP1 proteins from the four main malaria species, we included in figure 7 a comparison of the PfCSP signal to the PvRMC-MSP1 signal for the 181 individuals with P. falciparum infection.

Overall, we found “many samples were double-positive, responding to both of these antigens (Fig. 7). Additionally, some plasma samples showed a correlation of assay signals between the two antigens, tracking on a y = x reference line. However, some of these assay signals from P. falciparum infections were non-existent for PfCSP yet showed very high PvRMC-MSP1 IgG binding. No samples were IgG positive to PfCSP alone.”

We interpreted the double positive signals as normal for a population with P. falciparum exposure and did not conclude that PfCSP was responsible for the signals observed from P. falciparum patients as “some of these assay signals from P. falciparum infections were non-existent for PfCSP yet showed very high PvRMC-MSP1 IgG binding. No samples were IgG positive to PfCSP alone.”

2. Material and methods

a. 86-97: The authors should include the number of individuals enrolled in the study Costa Rica = 7,077 ?

In lines 165-166 of the results, we state “From the 2015 Ethiopia MIS, 7,077 DBS were available for serological data collection, and 851 DBS were available from the 2015 Costa Rica survey.”.

On line 107 of the methods, we state “In the Costa Rican canton of Matina in 2015, a total of 851 individuals were enrolled in their households and provided blood samples for DBS creation.”

For improved clarity, we have also added this statement to line 106-106: “From the 2015 Ethiopia MIS, 7,077 DBS were available for serological data collection.”

b. 127-134: To clarify. The protocol was performed overnight. The mix reagent containing conjugated antibodies (anti-Human IgG) and samples were added to the well. Did the reaction happen with all components (primary, secondary, and Streptavidin-PE) add at the same time? Why did the authors include IgG4 in to mix reaction?

The reviewer is correct, the binding of beads to hIgG, hIgG to biotinylated detection Ab, and binding of biotinylated Ab to streptavidin-PE all happens simultaneously when these reagents are mixed together.

The supplementary anti-IgG4-BIOT is added as the anti-hIgG included in the formulation from Southern Biotech does not pick up hIgG4 very well. This was noted from previous experiments and published in 2020 in the AJTMH by Jeffrey W. Priest et al. To explain this better in the manuscript text, we have updated lines 145-147 to include “. Anti-IgG4 was added separately to the anti-human IgG mixture as we have previously found that anti-IgG4 responses are difficult to determine using only the standard anti-IgG cocktail,” and cited this publication.

c. 138: MFI signal was generated for a target of 50 beads/region = 50 beads per antigen?

The reviewer is correct that a minimum of 50 beads per region (fluorescent signal) are required by the Luminex MagPix to produce signal, as stated in line 149 “MFI signal was generated for a target of 50 beads/region.”. Region in this case is the pre-gated bead region, or fluorescence signal from each bead type used in the assay. We separately conjugate each fluorescently tagged bead obtained from the manufacturer to one antigen of interest to allow us to obtain unique signals for PvRMC-MSP1 vs recombinant PvMSP1 vs recombinant PfMSP1. The multiplex platform allows us to obtain the level of antibody recognizing each antibody simultaneously in one well.

3. Results

a. 153-156: The authors should describe data from Table1 according to table data. First, Costa Rica and, than Ethiopia.

As suggested, we have revised table 1 to include the demographic data from Ethiopia first, above that of Costa Rica, to keep discussion of Ethiopia and then Costa Rica consistent throughout the manuscript.

b. 162-164: I suggest the authors a table with seroprevalence data according to the region to make easy the reading of data.

We thank the reviewer for this interesting question. It is possible to plot and compare the seroprevalence of antibodies to each of the three MSP1 antigens tested by region within Ethiopia, but we believe that this is outside of the scope of this study. Instead, we want to emphasize the seroprevalence to each antigen in the population as a whole, especially since Costa Rica is approaching elimination and our samples were obtained from only one region of Costa Rica.

c. 194-205: NANP6 or NANPx5 peptide?

The reviewer is correct that this passage is somewhat confusing. We have revised accordingly.

The PvRMC-MSP1 protein contains six copies of the NANP repeat peptide from the P. falciparum circumsporozoite protein, abbreviated as (NANP)6 on lines 124-125: “the (NANP)6 repeat peptide from P. falciparum circumsporozoite protein (CSP)”. This is also indicated on lines 211-214 “Although PvRMC-MSP1 is based primarily on the P. vivax Belem sequence, six copies of the NANP repeat sequence from the P. falciparum circumsporozoite protein (PfCSP) are present at the C-terminus of PvRMC-MSP1 and were included during the design of PvRMC-MSP1 to serve as an additional purification tag.”

Additionally, we have revised lines 214-216 to state “To determine if antibody recognition and binding to the PfCSP region is responsible in part for the assay signals observed for PvRMC-MSP1, the MFI-bg signals were compared between the chimera and a peptide containing five copies of the PfCSP NANP peptide (NANPx5, S1 Figure).”

4. Discussion

a. 241-247: To clarify, the NANP sequence was used as a tag to improve the purification o protein (NANP-6, was included at the C-terminus for biochemical characterization of antigenic integrity and to provide an optimal affinity purification tag, ref 15).

In previously published works (refs 15 and 16), a cross-reactivity with P. falciparum was described. However, it was demonstrated that this reactivity is not due to the presence of the PfCSP NANP repeat region alone. How do the authors justify the importance of this sequence on the reactivity of antibodies in this work? I felt confused When I read previous data.

Thank you for taking the time to look into our relevant previous publications as part of your review of this current manuscript.

The reviewer is correct that the NANP sequence was initially included in the PvRMC-MSP1 protein as a purification tag during its design as a vaccine candidate. In the publication where we investigated the ability of US travelers with active malaria infection to recognize PvRMC-MSP1 as compared to recombinant MSP1 proteins from the four main malaria species, we included in figure 7 a comparison of the PfCSP signal to the PvRMC-MSP1 signal for the 181 individuals with P. falciparum infection.

Overall, we found “many samples were double-positive, responding to both of these antigens (Fig. 7). Additionally, some plasma samples showed a correlation of assay signals between the two antigens, tracking on a y = x reference line. However, some of these assay signals from P. falciparum infections were non-existent for PfCSP yet showed very high PvRMC-MSP1 IgG binding. No samples were IgG positive alone to PfCSP.”

We interpreted the double positive signals as normal for a population with P. falciparum exposure and did not conclude that PfCSP was responsible for the signals observed from P. falciparum patients as “some of these assay signals from P. falciparum infections were non-existent for PfCSP yet showed very high PvRMC-MSP1 IgG binding. No samples were IgG positive alone to PfCSP.”

Submitted filename: McCaffery_etal_Response to Reviewers_Revision 2.docx

Click here for additional data file.

12 May 2022

The use of a chimeric antigen for Plasmodium falciparum and P. vivax seroprevalence estimates from community surveys in Ethiopia and Costa Rica

PONE-D-22-01720R2

Dear Dr. McCaffery ,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Luzia Helena Carvalho, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors corrected the suggested changes. And all the parts that had generated doubts were clarified.

Reviewer #2: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

16 May 2022

PONE-D-22-01720R2

The use of a chimeric antigen for Plasmodium falciparum and P. vivax seroprevalence estimates from community surveys in Ethiopia and Costa Rica

Dear Dr. McCaffery:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Luzia Helena Carvalho

Academic Editor

PLOS ONE