Genetic polymorphism of circumsporozoite protein of Plasmodium falciparum isolates in children in Brazzaville, Republic of Congo

Abstract

Background

The first WHO-approved malaria vaccines (RTS, S/AS01, and R21/Matrix M) target part of the Plasmodium falciparum circumsporozoite protein (PfCSP), displays a degree of polymorphism that may raise concerns about vaccine efficacy. As a prelude to vaccine implementation, the study here reports investigation on Pfcsp gene polymorphisms in isolates from Congolese individuals in the Republic of Congo.

Methods

Blood samples were collected from 202 children infected with P. falciparum during a cross-sectional study from March to October 2021. The full-length Pfcsp gene was amplified by nested PCR and sequenced using the Oxford Nanopore platform.

Results

Overall, 30 haplotypes were identified in the N-terminal region of the protein. The A98G mutation was the most frequent (25.6%), while KLKQP was the most conserved motif. In the central repeat region, 50 haplotypes were found, with a predominance of the NANP motif, although some haplotypes contained the NVDP, NEDP and NVNP variants. C-terminal region was highly polymorphic, with 76 haplotypes identified among the 174 sequenced samples. In the C-terminal Th2R region, the major mutations identified included K317E (87.4%) and N321K (83.9%), with a nucleotide diversity of π = 0.16. In the Th3R region, the E357Q (75.9%) mutation was predominant, with a nucleotide diversity of π = 0.08. Neutrality tests revealed contrasting patterns of evolution between the Th2R and Th3R regions.

Conclusion

This study provides baseline data on Pfcsp genetic diversity in a defined area of the Republic of Congo. The results highlight the degree of variation in natural parasite populations at gene loci relevant to vaccine-targeted epitopes of (PfCSP) antigen. Further research incorporating immunological data will be needed to conduct in-depth assessments of vaccine efficacy.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12936-025-05502-7.

Background

Malaria remains a major public health issue, particularly in sub-Saharan Africa. In 2024, an estimated 263 million cases and 597,000 deaths were reported worldwide, with 94% of cases and 95% of deaths occurring in Africa [1]. In the Republic of Congo, malaria is the leading cause of medical consultations (59%), hospitalizations (61%) and mortality (22%). Although its incidence has declined in recent years, children remain the most vulnerable population group [2].

The persistence of malaria is exacerbated by multiple factors, including parasite resistance to antimalarial drugs, mosquito resistance to insecticides and most especially, the absence of fully effective vaccines [3]. As part of global malaria control efforts, the World Health Organization (WHO) has launched an initiative to promote the development and implementation of malaria vaccines [4]. These vaccines, when combined with other control measures, are expected to contribute synergistically to malaria elimination [5]. One of the primary vaccine targets is the Plasmodium falciparum circumsporozoite protein (PfCSP), a key antigen expressed on the surface of sporozoites that is essential for hepatocyte invasion. PfCSP is partitioned into N-terminal, central repeat and C-terminal (Th2R/Th3R) regions. The RTS, S/AS01 and R21/Matrix-M vaccines, based on the PfCSP protein, are recommended by WHO for children in malaria-endemic regions [1, 6]. This recommendation is supported by pilot studies conducted in Ghana, Malawi and Kenya which showed RTS, S/AS01 reduced clinical and severe malaria cases by 46% and 36%, respectively, after 18 months [7]. R21/Matrix-M demonstrated 78% efficacy over two years with a booster dose [8]. Although both vaccines target the PfCSP antigen, their structure differs. RTS, S/AS01 combines a PfCSP-HBsAg fusion with unfused HBsAg, forming virus-like particles to enhance antigenic presentation [9]. R21/Matrix-M also includes a PfCSP-HBsAg fusion, but is devoid of unfused HBsAg, maintaining a PfCSP/HBsAg ratio of 11:1 due to a saponin-based adjuvant [10].

Reduced vaccine efficacy against some P. falciparum strains exhibiting variations in the PfCSP protein region targeted by the RTS, S/AS01 vaccine have been reported [11–13]. Thus, the genetic diversity of Pfcsp could have a significant impact on vaccine effectiveness [11]. Moreover, Pfcsp polymorphism varies depending on P. falciparum strains and transmission levels in endemic areas [14, 15]. WHO has planned to distribute both RTS, S/AS01 and R21/Matrix-M vaccines in the majority of African countries in the next years [1]. To date, these vaccines have been integrated into the Expanded Programme on Immunization in Ghana, Kenya and Malawi, where the pilot study was conducted, as well as in other countries including Cameroon, Burkina Faso, Liberia, Benin, Niger, Sierra Leone, the Central African Republic, and Côte d’Ivoire [16]. Additional African countries are also preparing to implement this new malaria control strategy in children. As the introduction of RTS, S/AS01 and R21/Matrix-M vaccines is being considered in the Republic of Congo, it is important to establish baseline data on local pfcsp strains in circulation. This would help anticipate potential challenges for a successful deployment and optimize vaccination strategies. This study, therefore, aims to characterize the polymorphism of the pfcsp gene in P. falciparum isolates from children in Southern Brazzaville, Republic of Congo.

Methods

Study sites

This study is part of a previous survey [17] conducted in two regions of the Republic of Congo: the rural district of Goma Tsé-Tsé and the urban health district of Madibou. In the rural area, samples were collected in Djoumouna and Ntoula, where numerous watercourses favour the proliferation of malaria vectors [18]. The local economy is based on agriculture and fishing. In urban areas, samples were collected both in the community and hospital from the Madibou district in the southern part of Brazzaville [19]. All these regions have a humid climate, with an average temperature of 25 °C [20] and an annual rainfall of approximately 1,100 mm, distributed over two rainy seasons of nine months. Malaria transmission is perennial in these areas, with P. falciparum as the predominant species [21]. The primary vector is Anopheles gambiae, with a global entomological inoculation rate of 47.5 infective bites per person per year (62.0 ib/p/y in rural and 33.6 ib/p/y in urban areas) [22]. The P. falciparum multiplicity of infection in the area is estimated at 2.6 (2.9 in urban hospital and rural community, and 2.4 in the urban community) [18, 19].

Ethical statement

The samples analysed in this study were collected as part of malaria field surveys approved by the Institutional Ethics Committee of the Congolese Foundation for Medical Research (Approval No. 013/CIE/FCRM/2018 and Decision No. 031/CIE/FCRM/2020). Since the study target participants were children, written informed consent was obtained from parents or legal guardians before their enrolment.

Sample collection and study population

During rainy and dry seasons, intravenous blood samples (2–3 mL) were collected in EDTA-containing tubes from participants under 15 years of age diagnosed with P.falciparum malaria who had resided in the study areas for at least two weeks before sampling. Participants were enrolled at the urban hospital, urban and rural communities. Parasitaemia and haemoglobin level were determined for each participant and a standardized questionnaire was used to record demographic and clinical data, including age, gender, weight, season of collection, and symptoms (fever ≥ 37.5 °C, headache, vomiting, nausea).

Malaria diagnosis

Screening for malaria infection was carried out using a rapid diagnostic test (Abbott SD BIOLINE Malaria Ag Pf), confirmed by microscopy and PCR. Thick blood smears were prepared for each RDT-positive sample and examined by two independent microscopists to confirm the presence of Plasmodium spp. and determine the parasitaemia, in accordance with WHO guidelines [23].

Molecular analysis

DNA extraction

Genomic DNA extraction from whole blood was performed using the QIAamp DNA Mini Kit (QIAGEN, Ref: 51306), following the manufacturer’s instructions. Extracted DNA was stored at -20 °C until further use for PCR amplification.

Molecular identification of Plasmodium species

A nested PCR targeting the 18S rRNA gene was performed to identify different Plasmodium species, following published method [24]. To ensure the quality of the analysis, each amplification run included a negative control (nuclease-free water) without DNA template. Additionally, genomic DNA from reference Plasmodium laboratory strains served as positive controls for the targeted species.

Amplification of the pfcsp gene

After species identification, only P. falciparum-positive samples were selected for pfcsp gene typing. A nested PCR amplifying the full-length pfcsp gene was performed according to a published method [25]. The first round PCR was performed using the primers 5′-TTAGCTATTTTATCTGTTTCTTC-3′ and 5′-TAAGGAACAAGAAGAGATAATACC-3′, to generate a 1,177 bp fragment. The reaction mixture (20 µL) contained 1X PCR buffer, 2.5 mM MgCl₂, 0.1 mM dNTPs, 0.3 µM of each primer, 0.625 U of DNA polymerase, and 2 µL of DNA template. The thermal cycling conditions were as follows: initial denaturation at 94 °C for 10 min, followed by 45 cycles of denaturation at 94 °C for 1 min, annealing at 55 °C for 1 min, extension at 72 °C for 1 min 30 s, and a final extension at 72 °C for 10 min. The second-round PCR used the same reaction mixture condition (20 µL) with 2 µL of amplicon as template, amplifying a 1,027 bp fragment with followed primers: 5′-GAAATGAATTTATATGGGAAACAG-3′/5′-GAAGGGATAACCACTTATTAATCC-3′. The cycling conditions included 35 cycles of denaturation at 94 °C for 1 min, annealing at 41 °C for 1 min, extension at 72 °C for 1 min 20 s, and a final extension at 72 °C for 10 min. Negative controls (without DNA template) and positive controls (P. falciparum strains 3D7, Dd2, HB3, and D10) were included in each PCR round for quality control. The amplification products were analysed by 1.5% agarose gel electrophoresis (120 V, 400 mA, 60 min) to verify amplicon size and specificity (Figure S1). A 1 kb DNA Ladder (New England Biolabs) was included to confirm expected fragment lengths.

Sequencing of the pfcsp gene

PCR products were purified using Ampure XP beads (Oxford Nanopore Technology, Oxford, UK) and quantified using Qubit (Thermo Fischer Scientific, Ref Q33226, Qubit 4 fluorometer, Singapore). Amplicons normalized to 100 ng were then barcoded using the SQK-RBK110.96 kit (Oxford Nanopore Technology, Oxford, United Kingdom), pooled and purified once more. The purified product was prepared as a library and loaded on the R9 flowcell for sequencing using Oxford Nanopore platform (GridION GRD BX5003).

Bioinformatics analysis

To ensure high-quality Nanopore reads, raw data were first assessed using NanoPlot (with the command: NanoPlot --fastq inputfile -o outputfile -t 5 --X) for read quality visualization. Poor quality reads were filtered using NanoFilt (command: NanoFilt input-file -l 40 --maxlength 50 or 1200 -q 5> output file) to retain only sequences with Q scores >5. Filtered high-quality reads were then processed via the Galaxy Europe platform. The FASTQ Groomer tool was used to standardize read format for compatibility with downstream tools while Bowtie2 was used for alignment against the P. falciparum 3D7 reference (XM_001351086.1) in Ubuntu (v20.04). Ivar Consensus was applied for consensus sequences calling.

For polymorphism analysis, only sequences with overall base quality > 80% were selected and submitted in GenBank with the accession numbers (PV994480–PV994615) and (PV953072-PV953108). Alignment and manual curation were performed using MAFFT (for multiple sequence alignment in Ubuntu). The fastA nucleotide and amino acid sequences of the pfcsp gene were visualized using BioEdit (v7.7.1), MEGA (V11.01), and DnaSP 6 for complementary analysis. The number of segregating sites (S), the average number of nucleotide differences per pair (K), haplotype diversity (Hd), and nucleotide diversity (π) were calculated using DnaSP [26]. Tajima’s D test (α = 0.05) [27], along with Tajima’s D and Fu and Li’s D and F statistics (α = 0.05) [28], were conducted using MEGA 11 [26]. Positive values for Tajima’s D and Fu and Li’s D and F (D > 0, F > 0) indicated positive selection, whereas negative values (D < 0, F < 0) suggested purifying selection. To assess the neutral theory of natural selection, the recombination parameter (R), which includes effective population size and the probability of recombination between adjacent nucleotides per generation, as well as the minimum number of recombination events (Rm), were analyzed [26]. The dominant allele was retained in the genotype file at loci with mixed reads (reference/non reference).

Statistical analysis

The data were entered and structured using Microsoft Excel, then analysed using GraphPad Prism 8. Non-parametric variables were compared using the Mann–Whitney or Kruskal–Wallis tests, while parametric variables were analysed using Student’s t-test or ANOVA. Mutations proportions in N-terminal region, Th2R and Th3R epitopes of pfcsp gene were compared between isolates from urban hospital and those from urban and rural communities using Chi-square test. The significance threshold was set at 0.05. Haplotype (haploid genotype) is defined as a set of variants along a single chromosome that is inherited together.

Results

General characteristics of the study population

Plasmodium falciparum mono-infection was the most common among the study population, 155/198 (78.3%). However, Plasmodium malariae/P. falciparum, Plasmodium ovale/P. falciparum, and P. ovale /P. falciparum/P. malariae co-infections were reported for 27/198 (13.6%); 9/198 (4.5%); and 7/198 (3.5%) of the analysed samples respectively. The present study focused on P. falciparum. A total of 198 children with a thick blood smear positive for P. falciparum were enrolled in urban hospital (N = 65), urban community (N = 28), and rural community (N = 105) settings. The median body temperature of the participants was 37 °C (interquartile range of 36.5–37.5 °C), and 27.8% (55/198) of participants had fever at the time of enrolment (T ≥ 37.5 °C). Additionally, the median haemoglobin level was 11.1 g/dL (interquartile range of 9.9–11.9 g/dL), with an overall anaemia prevalence of 77.3% (153/198). The participants’ age ranged from 1 to 14 years, with a median age of 9 years (interquartile range of 5–12 years). Age group distribution showed that children aged 5–10 years were the most affected group representing 44.4% of this P. falciparum-infected study population. The median parasitaemia was 1,227 parasites/μL with an interquartile range of 203–11,311 parasites/μL, and 48% (95/198) of the participants had a parasite density < 1,000 trophozoites/µL. All the participants from urban hospital (N = 65) were symptomatic while only 45.7% (48/105) and 46.4% (13/28) of the participants were symptomatic in rural and urban communities, respectively.

Polymorphism of the N-terminal region of the pfcsp gene

Of the 202 microscopy-positive samples included in the study, 198 were successfully amplified by nested PCR. Of these, 176 yielded high-quality full-length PfCSP sequences suitable for analysis and were submitted in GenBank with the accession numbers (PV994480-PV994615) and (PV953072-PV953108). Amplification or sequencing failure was more frequently observed in samples with very low parasite density, suggesting a potential bias linked to parasitaemia levels. Overall, sequence analysis revealed low polymorphism in the N-terminal region. A total of 30 distinct haplotypes were identified. The predominant haplotype (similar to the reference sequence) accounted for 57.4% (101/176) of the isolates, while the less frequent haplotype was observed in a single isolate (1/176). Furthermore, the KLKQP motif, which is involved in sporozoite invasion of hepatocytes, was conserved across all isolates (Fig. 1).

Fig. 1.

N-terminal polymorphism analysis. H urban hospital, R rural community, U urban community, T total isolates. H1-H30 haplotypes. Dots represent amino acids identical to the reference sequence; the presence of letters indicates a difference in amino acid compared with the reference sequence at the position where the letter is found; Ref = reference sequence; 3D7, Dd2, D10 and HB3 were used as laboratory strains controls

Mutation analysis of the N-terminal region of the pfcsp gene, performed on 176 sequences, identified 28 mutations. A98G was the most frequent mutation, observed in 25.6% (45/176) of haplotypes, followed by N75K and G73E, detected in (22/176, 12.5%) and (14/176, 8.0%) of haplotypes, respectively. However, mutations such as D76E, S69R and Y55C were detected in a single haplotype (1/176). No significant difference was observed between isolates from urban hospital, urban and rural communities, concerning these investigated mutations (Table.S1).

Polymorphism of the central region of the pfcsp gene

Analysis of the central repeat region of the pfcsp gene revealed significant polymorphism, with 50 distinct haplotypes identified among 176 isolates (Fig. 2). The predominant haplotype (H01), similar to the 3D7 reference sequence, accounted for 51.1% (90/176) of all sequences and was the most frequent in all study sites while haplotypes H17 to H50 were observed in a single isolate (1/176). Distinct motifs emerged in the central region, with NVNP and NADP motifs being specific to some haplotypes. Although not detected in the 3D7 reference sequence (strain used in the RTS,S vaccine), NVNP and NADP motifs were detected in 7 and 12 isolates, respectively, from this study where they represented less than 5% of repeats (Fig. 3). Other motifs, such as NEDP, NENP, NEEP and others, were found in a subset of isolates, highlighting the genetic diversity of pfcsp. The analysis revealed a predominant of the NANP motif in most of the haplotypes. Some haplotypes had alternative motifs such as NVDP, NEDP and NVNP, while others, including NAHP, NADP and IVDP, were less frequent. Haplotypes H01 to H20 were dominated by NANP motif, while haplotypes H21 to H50 showed a greater diversity of motifs compared to haplotypes (H25 to H35) (Fig. 3).

Fig. 2.

Polymorphism of the central repeat region. H Urban hospital, U urban community, R rural community and T total isolates, H1-H50 haplotypes. The symbol in the sequences correspond to the different motifs found in the central repeat region: 1 = NANP, 2 = NVDP, 3 = NVNP, 4 = NEDP, 5 = NADP, 6 = NENP, 7 = NAKP, 8 = NEEP, 9 = NVDT, @ = NAHP, 0 = IVDP and * = NAIP. Ref = reference sequence; 3D7, Dd2, D10 and HB3 were used as laboratory strains controls

Fig. 3.

Composition and frequencies of repeated motifs per haplotype in the central region with repeated motifs. H01 is similar to 3D7 Ref sequence used in RTS,S vaccine

Furthermore, the results revealed that sequences containing 42 motif repeats were the most prevalent, accounting for 70% (123/176) of cases. Sequences with 44 and 45 motif repeats followed, with similar proportions (approximately 10.2% (18/176) and 9.1% (16/176), respectively). The proportions of 43, 46, 48, 53, and 54 motif repeats ranged from 1.1% (2/176) to approximately 4% (7/176).

Polymorphism of the C-terminal region of the Pfcsp gene

Analysis of the C-terminal region of the Pfcsp gene revealed a high degree of polymorphism, with 76 distinct haplotypes identified among 176 sequences, compared to the 3D7 reference strain. The predominant haplotype was observed in 13.6% (24/176) of isolates, while minor haplotypes were detected in a single isolate (1/176). None of the analysed P. falciparum sequences were similar to the reference sequence (Fig. 4).

Fig. 4.

C-terminal polymorphism analysis. H urban hospital, R rural community, U urban community, T total isolates. H1-H76 haplotypes. Dots represent amino acids identical to the reference sequence; the presence of letters indicates a difference in amino acid compared with the reference sequence at the position where the letter is found; Ref = reference sequence; 3D7, Dd2, D10 and HB3 were used as laboratory strains controls

Prevalence of mutations in the Th2R epitope of the C-terminal region of the pfcsp gene

The results identified 15 distinct mutations within the Th2R epitope. Some of these mutations were particularly prevalent across all study sites, especially K317E (152/176, 86.4%) and N321K (146/176, 83.0%), which were highly prevalent in the three sites. A high prevalence was also noted for the K322T mutation (79/176, 48.9%). Conversely, some mutations, such as K314Q, were less prevalent across study sites (ranging from 29.8% (17/57) to 33.3% (32/96)). Scarce mutations, including N321Q (2/176, 1.1%) and Q324R (2/176, 1.1%), were also detected among the analysed isolates. Although K317T was not frequent, its distribution differed significantly among the three sites (p = 0.0132) (Table S2).

Prevalence of mutations in the Th3R epitope of the C-terminal region of the pfcsp gene

Analysis of the C-terminal region of the pfcsp gene identified nine mutations within the Th3R epitope among isolates collected across the three study sites. The overall prevalence of the E357Q mutation was 75.6% (133/176), with site-specific frequencies of 78.9% (45/57) in hospital settings, 73.9% (17/23) in urban community, and 74.0% (74/96) in rural community. The A361E mutation was also common, with an overall prevalence of 60.8% (107/176), corresponding to 57.9% (33/57) in hospital settings; 52.2% (12/23) in urban community, and 64.6% (62/96) in rural community. The frequencies of the D359N, N352G and D356N mutations were 8.5% (15/176), 7.9% (14/176) and 5.7% (10/176), respectively, whereas the N352D, P354S, A361I, and L358Y mutations were detected at 4.0% (7/176), 2.8% (5/176), 1.1% (2/176), and 0.6% (1/176), respectively (Table S3).

Nucleotide polymorphism of C-terminal region, natural selection and genetic recombination

Analysis of nucleotide diversity (π) in the C-terminal region of the pfcsp gene across study sites revealed distinct patterns of variation within the Th2R and Th3R epitopes. Overall, nucleotide diversity exhibited peaks at specific positions, with similar trends observed across the three study sites. In the Th2R epitope, nucleotide diversity was particularly elevated between positions 939 and 981, reaching a peak of approximately 0.16. In contrast, the Th3R epitope displayed lower but noticeable peaks of nucleotide diversity, particularly between positions 1053 and 1092, where π reached approximately 0.08. Despite these variations, no significant differences were observed among the three study areas (P > 0.05) (Fig. 5).

Fig. 5.

Analysis of the nucleotide diversity of Th2R and Th3R motifs in the C-terminal region by study sites. Hospital = Urban hospital; Urban_Com = Urban community; Rural_Com = Rural community

Nucleotide diversity (π) across the entire C-terminal region ranged from 0.029 to 0.032, with urban hospital isolates exhibiting higher diversity (π = 0.032) than rural community isolates (π = 0.029). The number of segregation sites (S) varied between 14 and 22, with high haplotypes diversity (Hd = 0.95–0.97). Seasonal analyses showed stable π values from one season to the next, and diversity remained constant between symptomatic (π = 0.031) and asymptomatic (π = 0.029) groups, although the number of haplotypes was higher in symptomatic cases (49 versus 33). Nucleotide diversity increased with the parasitaemia (π = 0.029–0.0322) and was highest in children under five years of age (π = 0.033). It was also higher in the Th2R region (π = 0.077) than in Th3R (π = 0.045). Th2R had more segregation (S = 11) and recombination (Rm = 4) sites compared to Th3R (Rm = 2). Tajima's D, Fu and Li's D, and Fu and Li's F showed no significant deviation from neutral equilibrium, except for seasonal and motif-related variations (Table 1).

Table 1.

Nucleotide diversity of C-terminal region, natural selection and gene recombination

Variables	N	π	K	S	H	Hd ± SD	No.of mutation	Tajima's D	Fu and li's D	Fu and li's F	Rm
By study site
Hospital	56	0.032	5.20	22	31	0.95 ± 0.02	26	− 0.26430a	− 0.6987a	− 0.6465a	6
Urban community	24	0.031	5.0	14	18	0.97 ± 0.02	18	0.12915a	0.3228a	0.3083a	4
Rural community	97	0.029	4.68	18	45	0.95 ± 0.01	21	0.42916a	1.33416a	1.18782a	6
By season
Dry	111	0.030	4.8	23	47	0.94 ± 0.012	27	− 0.1554a	0.0582a	− 0.0301a	7
Rainy	64	0.031	5.0	17	35	0.97 ± 0.01	20	0.5444a	1.7092b	1.5381a	6
Symptomology
Symptomatic	119	0.031	5.0	23	49	0.94 ± 0.012	27	− 0.0381a	0.0328a	0.0059a	6
Asymptomatic	56	0.029	4.7	17	33	0.97 ± 0.011	21	0.0850a	− 0.1274a	− 0.0624a	7
Parasite density group
< 1000 P/µl	80	0.029	4.7	19	38	0.95 ± 0,013	23	0.0500a	0.2677a	0.2230a	7
1000–10000 P/µl	48	0.0302	4.9	18	31	0.96 ± 0,015	21	0.1085a	0.2929a	0.2719a	5
> 10,000 P/µl	48	0.0322	5.2	21	24	0.94 ± 0,017	25	− 0.2419a	− 0.6884a	− 0.6324a	5
Age group
< 5 years	32	0.033	5.4	17	22	0.97 ± 0.018	20	0.3462a	0.3595a	0.4182a	5
5–10 years	76	0.028	4.5	19	42	0.96 ± 0,014	23	− 0.1123a	− 1.2194a	− 0.9673a	7
< 15 years	67	0.031	5.0	21	31	0.94 ± 0,016	25	− 0.1287a	− 0.5734a	− 0.4897a	6
Epitopes
Th2R	176	0.077	3.2	11	44	0.938 ± 0.009	14	0.77562a	1.5319c	1.49514d	4
Th3R	176	0.045	1.6	10	14	0.764 ± 0.021	11	− 0.3697a	− 0.0537a	− 0.20209a	2

N = number of sequences matching the respective sequence in the alignment, across the different sampling sites; π = nucleotide diversity; S = segregation sites, H = number of haplotypes; Hd = haplotype diversity; Rm = minimum number of recombination; K = the average number of nucleotide differences per pair; a = P > 0.10; b = P < 0.02; c = P < 0.05; d = P > 0.05

N = nombre des isolats; π = diversité nucléotidique; K = S = sites de ségrégation, H = nombre d’haplotypes; Hd = diversité des haplotypes; dS = taux de mutations synonymes; dN = Taux de mutation non synonymes; Rm = nombre minimum de recombinaison; a = P > 0.10; b = P < 0.02; c = P < 0.05; d = P > 0.05

Discussion

This is the first report on polymorphism of Pfcsp protein from isolates collected in the Republic of Congo. The findings indicate that the polymorphism in N-terminal region was higher than that reported in Cameroon in 2024 and in Sudan in 2021 [14, 29]. The high degree of polymorphism observed in Congolese isolates may be attributed to the selective pressure induced by host immune responses, particularly in populations exposed to high and sustained malaria transmission. This immune-driven selection is further supported by recent clinical evidence showing that monoclonal antibodies targeting circumsporozoite protein can provide protective efficacy against some variants [30]. In such contexts, genetic variations may confer a selective advantage to parasites able to evade acquired immunity. Indeed, previous studies have reported that the N-terminal region of pfcsp can evolve rapidly in response to host population-specific immunity, thereby enhancing parasite survival [31].

In the N-terminal region, a total of 28 mutations were identified, among which the A98G mutation was the most prevalent (25.6%). This finding is consistent with previous reports in other African countries [32, 33]. However, the functional impact of this mutation remains unexplored. Isolates from rural areas exhibited greater genetic variability compared to those from urban settings, probably reflecting the different level of malaria transmission. The N-terminal region plays a crucial role in hepatocyte invasion by P. falciparum sporozoites [34, 35]. In the present study, the KLKQP motif, which mediates hepatocyte binding, was found to be conserved across all isolates. This stability underscores its essential functional role in the parasite’s life cycle, potentially by facilitating hepatocyte invasion or contributing to immune evasion. Consequently, this region represents a promising target for malaria vaccine development [36].

The analysis of the central repeat region revealed a high degree of polymorphism, consistent with previous studies conducted in China, Sudan and India [14, 25, 33]. However, these findings contrast with those of a study carried out in Equatorial Guinea that identified only 14 haplotypes [37]. This discrepancy suggests that polymorphism within the central repeat region of pfcsp may be influenced by variations in local selective pressures, and epidemiological contexts. The central region contains B-cell epitopes composed of repeat motifs and plays a crucial role in the maintaining of the pfcsp gene stability, which is affected by the number of repeat units [38]. In this study, isolates from participants exhibited a range of 42 to 54 repeats, with approximately 70% of sequences containing 42 repeats. This level is lower than in other African studies, where the average number of repeats was reported as 43 [33, 39]. However, an increased number of repeats has not been associated with a reduction in vaccine efficacy [11]. Additionally, beyond the well-documented NANP and NVDP motifs, four new motifs (NEDP, NENP, NEEP, NVDT) were identified in this study. Thus, the functional or immunological implications of these new motifs need to be investigated.

The C-terminal region exhibited a high degree of polymorphism, with 76 distinct haplotypes, confirming observations from other countries [15, 25, 33]. None of the isolates was similar to the reference strain. This underscores the potential limitations of current PfCSP-based vaccines in this setting as genetic diversity, particularly mutations at gene loci relevant to vaccine-targeted epitopes of PfCSP, could be associated with reduced efficacy of the vaccine.

A total of 24 mutations were identified between the Th2R and Th3R epitopes with some mutations being particularly prevalent, especially K317E (87.4%) and N321K (83.9%). Previous studies have shown that vaccine efficacy declined as the number of amino acid differences increased, suggesting that the allele specific immunity is important in eliciting protection. The Th2R epitope exhibited greater diversity than Th3R, which is consistent with others studies’ findings from Africa [33]. However, some mutation identified within the Th3R region (P454S, D356N, L358Y, D359N, A361I) could potentially impact vaccine efficacy [11], by altering CD8 + T-cell activity [38]. In fact, residues with larger side chains such as tyrosine or glutamine in the CSP peptide have been shown to block the HLA groove thereby reducing binding interaction [40].

Analysis of the Th2R and Th3R epitopes revealed two peaks of genetic variation, indicating regions of high diversity. Tajima’s D and Fu’s D values for the Th2R epitope were positive, suggesting positive selection, whereas those for the Th3R epitope were negative, indicating either neutral evolution or purifying selection. These findings highlight the complex evolutionary pressures on these regions of the protein. Finally, genetic diversity was highest in children under five years old (π = 0.033) and decreased with age; suggesting progressive natural selection [41]. The number of haplotypes peaked in children aged 5 to 10 years which was also the most affected group (44.4% of the study population), highlighting a phase of intense allele acquisition and recombination [42].

Many African countries (Benin, Burkina Faso, Burundi, Cameroon, Central African Republic, Côte d’Ivoire, Chad, Democratic Republic of the Congo, Ghana, Kenya, Liberia, Malawi, Mozambique, Niger, Nigeria, Sierra Leone, South Sudan, Sudan, Uganda) were offering malaria vaccines (RTS, S or R21) as part of their childhood immunization programmes [16], and according to their national malaria control plans. In addition, at least 30 countries in Africa plan to introduce the malaria vaccine into their childhood immunization programmes and as part of their national malaria control strategies. Thus, the variation in polymorphism across different geographical regions in sub-Saharan Africa raises the potential need for region-specific vaccine updates if polymorphism affects efficacy.

The present study has few limitations. First, the cross-sectional design cannot address temporal dynamics, and single-time-point sampling may miss seasonal shifts. Second, only samples from southern Brazzaville were considered in this study and this may limit generalizability of findings to the rest of the Republic of Congo. Findings from this study underscore the potential limitations of current PfCSP-based vaccines in this setting. However, future studies integrating sero-epidemiology or functional assay (e.g., sporozoite neutralization) should be conducted to link pfcsp sequence variation to immunogenicity.

Conclusion

This study provides the first baseline data on pfcsp genetic diversity in the Republic of Congo prior to the introduction of malaria vaccines. It highlights differences between the haplotypes incorporated in the RTS, S/AS01 and R21/Matrix-M vaccines and those in circulation in Congolese P. falciparum isolates. In the study area, pfcsp was found to be highly polymorphic including 30, 50 and 76 distinct haplotypes for N-terminal, the Central and C terminal region and greater nucleotide diversity in the Th2R epitope compared to Th3R.

Abbreviations

18S rRNA

18S ribosomal RNA

3D7

A laboratory-adapted strain of Plasmodium falciparum

Dd2

Chloroquine-resistant Plasmodium falciparum strain

D10

Plasmodium falciparum Strain known for its pyrimethamine sensitivity

DNA

Deoxyribonucleic Acid

EDTA

Ethylenediaminetetraacetic acid

HB3

Plasmodium falciparum Strain derived from Honduras

pfcsp

Plasmodium falciparum Circumsporozoite protein gene

PfCSP

Plasmodium falciparum Circumsporozoite protein

PCR

Polymerase Chain Reaction

R21/Matrix-M

Recombinant malaria vaccine candidate formulated with Matrix-M adjuvant

RDT

Rapid Diagnostic Test

RTS, S/AS01

Recombinant circumsporozoite protein-based malaria vaccine formulated with AS01 adjuvant

Th2R

T-helper 2 repeat region

Th3R

T-helper 3 repeat region

WHO

World Health Organization

Author contributions

Conceptualization, MTB, JCD and FN; Methodology, MTB, JDMN, AGKE, and CCMM; Formal analysis and bioinformatics, MTB and CCMM; Writing original draft, MTB and JCD; Validation JCD and FN; Investigation, MTB, JCD, JDMN, AL, AGKE, CCMM and VAE; Writing review and editing, MTB, JCD, JDMN, AL, AGKE, VAE, CCMM, AM and FN; Supervision, JCD and FN. All authors have read and agreed to the published version of the manuscript.

Funding

This project is part of the EDCTP2 programme (CANTAM (Central Africa Clinical Research network), grant number EDCTPRegNet2015-1045) supported by the European Union and the Alexander von Humboldt foundation (project HR-COCA). JCD and CCMM were supported by the Fondation Mérieux and the Alexander von Humboldt foundation (project HR-COCA), respectively.

Data availability

Sequence data that support the findings of this study have been deposited in GenBank with the accession numbers (PV994480—PV994615) and (PV953072-PV953108).

Declarations

Competing interests

The authors declare no competing interests.