Malaria caused by Plasmodium falciparum is a major health care challenge in sub-Saharan Africa (SSA). The burden of malaria is high with an estimated 94% of all malaria deaths occurring in SSA (WHO, 2020). Over the last two decades, local transmission of malaria in some countries in Africa is declining, and child mortality has also dropped as a result of the control measures adopted by governments (Campbell and Steketee, 2011).

In Ghana, about 30% of out-patients in health facilities have malaria (NMCP, 2016). Malaria transmission in Ghana is seasonal, varying from high in the rainy season to low in the dry season. Malaria transmission also differs according to the three ecological zones of Ghana, being low in the coastal scrub, medium in the semi-deciduous and transitional forest, and high in the guinea savanna zone (Afari et al., 1992).

The conventional tools used in estimating malaria transmission intensity are entomological inoculation rates (EIR) based on human landing catches and parasite prevalence (Hay et al., 2000; Corran et al., 2007). Alternative methods such as serology (antibody responses to vector-specific or parasite-specific antigens) have been proposed and tested to be robust and sensitive (Corran et al., 2007; Badu et al., 2012; Kusi et al., 2014; Badu et al., 2015). Serology can thus reveal micro-differences in a larger transmission zone, especially those with medium to low transmission. Local parasite transmission is dependent on environmental factors, land use, human settlements and behaviour, which may influence vector abundance and thus parasite transmission patterns (Yankson et al., 2019).

Plasmodium falciparum merozoite surface protein 2 (PfMSP2) is a glycoprotein which is encoded by the MSP2 gene on chromosome 2. It is made up of five blocks with the central block being the most polymorphic. MSP2 is one of the numerous malaria vaccine candidates under consideration for a vaccine (Howard and Pasloske, 1993). However, it exhibits extensive polymorphism and the genes of MSP2 in the block 3 regions of MSP2 have alleles grouped into two families, the IC3D7 and FC27 (Smythe et al., 1991). These have been used to determine the multiplicity of P. falciparum infection (Felger et al., 1994) which is an indication of the diversity of the circulation parasites in an area. Plasmodium falciparum circumsporozoite protein (CSP) is a protein expressed extensively on sporozoite surfaces (Ménard et al., 1997), and exposed to the immune system for a brief time after an infectious bite. Meaning anti-CSP antibodies may have a relatively shorter half-life (Kerkhof et al., 2016) than most blood-stage antigens, and thus positive responses may indicate a recent infection. With the recent decline in malaria transmission, it is becoming increasingly difficult to use the conventional entomological inoculation rate (EIR) or parasite prevalence to estimate transmission intensity due to the very low prevalence of infected mosquitoes, thus proposing serology as an alternative (Corran et al., 2007; Rosas-Aguirre et al., 2013). For example, anti-CSP antibodies have been found to be ideal for monitoring malaria transmission intensity and pattern (Kusi et al., 2014). We recently reported on the suitability of multiple malaria antigens to estimate transmission intensity and pattern in three ecological zones in the Volta Region of Ghana (Kyei-Baafour et al., 2021). However, the local differences in transmission intensity and pattern within each ecological zone were not highlighted. Here, we report on the micro-differences in P. falciparum transmission within a medium malaria transmission zone in the Hohoe Municipality.

With the prevalence and transmission of malaria declining, sensitive tools are needed to effectively monitor transmission intensity and pattern at the micro-level. Serology is one key tool used to estimate transmission intensity and pattern because it can detect heterogeneity in transmission intensity at the micro level using antibody responses to malaria antigens.

We set out to determine the association between parasite prevalence and antibody responses to two P. falciparum merozoite antigens (MSP2-3D7, MSP2-FC27) and a sporozoite antigen (CSP) within different settlements of the same transmission zone. Our data indicate that the parasite prevalence was higher in rural settlements compared to urban settlements. Also, antibody responses to all antigens were higher in the rural settlements, and antibody seroprevalence to MSP2-3D7, and CSP were also higher in the rural settlements. In a multivariate analysis, we also found responses to PfMSP2-3D7 were associated with high transmission and thus rural settlement. This is consistent with studies from Gabon (Imboumy-Limoukou et al., 2016) where higher antibody responses were found in rural children compared to those in the urban areas.

Though the larger area of Hohoe municipality and its environs is designated as having medium but seasonal malaria transmission (Afari et al., 1992; Kweku et al., 2015; Kyei-Baafour et al., 2021), the micro-ecological and transmission differences and practices between the rural and urban settlements in the municipality may result in different vector exposure rates. The overall parasite prevalence was 7.3% by microscopy and 13.5% by PCR. Our data indicate that P. falciparum was more prevalent in the rural settlements compared to the urban settlement confirming previous reports (Robert et al., 2003; Lourenço, 2012). The high parasite prevalence in the rural settlements is indicative of higher transmission at those sites compared to the urban centre. The population in the rural communities mostly engage in agrarian activities with rice and vegetables being the main crops farmed. The water used in the rice farms could potentially serve as breeding sites for malaria vectors, thereby increasing the parasite exposure of people living in rural areas. Also, practices in rural settlements such as storing water in open receptacles, coupled with the bushes around could be good breeding grounds for the vectors (Azabre et al., 2013). Other practices in many rural settlements in Ghana likely to increase the risk of malaria are the habit of sleeping outside, especially in the dry season, and early evening and morning activities (Monroe et al., 2015). On the other hand, the urban centres have less favourable breeding grounds for vectors due to improved infrastructure and easy access to health care (Lindsay et al., 1990; Alirol et al., 2011). That notwithstanding, urbanization without the proper infrastructure may increase vector breeding sites and increase transmission as is recently reported (Mzilahowa et al., 2016).

We found higher total IgG responses to all the three antigens in the rural settlements compared to the urban settlements. This was confirmed by the higher seroprevalence (the proportion of children that responded to each antigen) in the rural settlements. This higher response may be due to the high parasite prevalence in rural communities. Also, parasitaemic children in the rural communities had higher total IgG responses to MSP2-3D7 and CSP, while no differences were observed in the urban settlements between parasitaemic and non-parasitaemic children. Parasite specific antibody responses can be boosted by Plasmodium infection by an estimated 20% increase in level (Kaddumukasa et al., 2015) and thus the high responses observed in the rural settlements are likely to reflect the higher parasite prevalence there. The low responses and low seroprevalence in the urban settlement may be due to the awareness of the population about malaria-related illness and protecting children from infection risk. Also, some of the practices in the villages, such as storing water in open receptacles, are hardly observed in the urban communities, which are served by piped water. A study comparing antibody responses between older Gambians and younger Ghanaians suggest that the younger children are unable to produce long-lived IgG antibody secreting cells compared to their older counterparts (White et al., 2014). Thus, children with intermitnent infection such as those in the rural community, are more likely to bost and sustain antibody production comapared to those less exposed (Fowkes et al., 2012). This may explain the higher responses in the rural children compared to those in the urban area. There are reports of antibody half-lives to different antigens (Ondigo et al., 2014) where some responses were found to long-lived. This long-lived responses may be a limitation to the use of such antigens to monitor malaria transmission intensity.

In an age-related analysis, high antibody responses, and seroprevalence in rural area in older children supports previous report of age-related antibody acquisition (Stanisic et al., 2015). Though there was no difference in the ages of the children from both settlements, higher seroprevalence associated with age in the rural settlement may be an indication of higher vector exposure. Interestingly, seroprevalence to MSP2-FC27 in the younger children was high in the urban area and the reason for this is not readily understood. Although Hohoe is an urban settlement, it is also a commercial town, and women are mostly involved in trading. This exposes the younger children who are carried at the back by their mothers to early morning bites when the mothers go out to the market at dawn. Though we do not have data on the specific strains of circulating parasites in our study area, which is a limitation, the specificity of the seroprevalence to MSP2-FC27 in the younger children may be an indication this parasite clone may be the predominant clone in the urban settlement.

In a logistic regression model adjusting for confounders, we found higher odds of responses to MSP2-3D7 to be associated with parasite carriage and thus higher transmission in the rural settlement which may indicate its circulation in the rural settlement. Serology, which was the tool employed in this study has confirmed that in an area of low to medium malaria transmission, micro-differences in transmission can be detected using anti-malaria antibody responses (Kusi et al., 2014; Kusi et al., 2016). However, the use of some of these anti-malaria antibody responses such as CSP to monitor transmission intensity and pattern in the future may be challenging with the introduction of RTS,S vaccine. This is because the RTS,S vaccine has CSP as the main component (Cohen et al., 2010; Laurens, 2020), and responses to the vaccine induced CSP may dilute naturally acquired responses. Thus, the need to increase the repertoire of antigens.

In summary, the study highlights heterogeneity in malaria transmission at the micro level and also confirms serology as a robust tool for transmission monitoring, especially in low to medium transmission settings. This has implications for malaria control and the administration of vaccines in such communities.

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

The studies involving human participants were reviewed and approved by The Research Ethics Committee of the University of Health and Allied Sciences (UHAS-REC A.1 (5) 18–19). Written informed consent to participate in this study was provided by the parent/legal guardian.

Conceived and designed the experiments: EK-B, LH, MK and MFO. Performed the experiments: EK-B, MO, AF, BAc, EAO. Analyzed the data: EK-B, KK, and MFO. Contributed reagents/materials/analysis tools: KK, BA, and MT. Wrote the manuscript: EK-B, KK, BA, LH and MFO. All authors contributed to the article and approved the submitted version.