The samples used in this study were collected as part of a case-control study whose objective was to evaluate the effect of LLIN use on malaria morbidity in clusters of low and high vector resistance to pyrethroids [29] . Dried blood drops (DBD) on Whatman^® 903 paper were collected for passive case detection, conducted between December 2011 and July 2012 in four hospitals in four communes (Pobè, Kétou, Sakété, and Ifangni) of the Plateau department [29] .

This is a cross-sectional case-control study with one control for one case that started in November 2011 and ended in July 2012 in the above-mentioned communes. The epidemiological results were published in 2022 by Tokponnon et al. [29] .

In these communes, clusters with low and high insecticide resistance of malaria vectors where the coverage of LLINs distributed to the population is greater than or equal to 80% were selected. According to the level of resistance, we compared the data obtained in the two groups of children (cases and controls) and also evaluated the influence of resistance in the two R+ and R+++ areas (Sovi et al., 2014). This prior study provided information on the resistance status of malaria vectors in the villages of origin of the children to be included in the study. A total of 32 clusters, 16 of which were high and 16 low resistance, were selected for recruitment of children.

The overall sample size was estimated at 240 children under 5 years of age, with 60 children per health center.

The sample consisted of all children under five years of age residing in the study localities for at least six months, who were seen in medical consultations or in vaccination services and whose parents had given their informed consent. We recruited the first 120 children under 5 years of age with fever and the first 120 children under 5 years of age without fever who were seen in medical consultation. Four reference health facilities were identified in the four communes to house our work. These were the Health Centers of Kétou and Ifangni, and the Hospitals of Sakété and Pobè. Recruitment in each health facility involved the first 30 children under 5 years of age with fever and the first 30 children under 5 years of age without fever.

A questionnaire is asked to the children’s companions on the use or not of LLINs and a thick drop sample is taken to assess malaria infestation.

A filter paper was made per patient for the determination of anti-saliva antibodies.

Technique for preparation of confetti

The filter paper with 5 circles of 12 mm diameter was impregnated with 5 drops of blood taken from each spot. The confetti was air-dried for 24 hours, stored in individual plastic bags and then in airtight straps (containing a desiccant, silica gel) and a humidity control at room temperature until the genotypic and ELISA tests were performed and transported to the CREC/IRD laboratories for storage and handling.

It should be noted that each DBS is labeled with the date of preparation, the initials of the name and surname of the individual and his identification number, according to the following Figure 1.

Blood samples on filter paper (FP) are dried completely to prevent fungal growth. The technician must ensure that each filter paper has the same information as the thick drip slide. The moisture indicator should be checked periodically (once a week) in case it changes color and is replaced routinely.

This was done according to the technique of Fontenille described by Beier et al. (1998) and SOP N˚13 [30] . We prepare the “Blood Meal” Buffer (to be kept at +4˚C, 1 to 2 weeks) for 1/2 liter, take 2.5 g of casein in 50 ml of 0.1 N NaOH.

For each selected child (N = 240, including 120 cases with temperature above 37.5˚C and 120 controls without fever), a standardized DBS (1 cm in diameter) was eluted by incubation in 350 μL of phosphate-buffered saline (PBS-Tween 0.1%) at +4˚C for 48 hours. Enzyme-linked immunosorbent assay (ELISA) was performed on the eluates to assess the level of IgG responses to the gSG6-P1 peptide antigen, as described previously [13] . Briefly, the wells, of Maxisorp plate (Nunc, Roskilde, Denmark), were coated with gSG6-P1 (20 μg/mL) in PBS. After washing (distilled water + 0.1% Tween), the DBS eluate from each child was incubated in an incubator at 37˚C. This sensitized plate was then incubated (in duplicate) at +4˚C overnight at a 1/40 dilution (PBS-Tween 1%). This optimal dilution was determined by preliminary experiments. A biotinylated mouse antibody to human IgG (BD Pharmingen, San Diego, CA) was incubated at a 1/1000 dilution in PBS with 1% Tween (1.5 hours at 37˚C) and streptavidin (1/1,000; 1 hour at 37˚C) conjugated to peroxidase (Amersham, Les Ulis, France). Colorimetric development was performed using ABTS (2,2’-azino-bis (3-ethylbenzothiazoline6-sulfonic acid) diammonium; Sigma, St Louis, MO) in 50 mM citrate buffer (pH = 4)

containing 0.003% H₂O₂. Optical density (OD) was measured at 405 nm after 2 hours at room temperature in the dark. In parallel, each tested sample was evaluated in a blank well containing no gSG6-P1 antigen (ODn) to measure nonspecific reactions. Known positive controls were included on each ELISA plate to control for plate variation and assay reproducibility. Specific anti-gSG6-P1 antibodies were also assessed in individuals not exposed to Anopheles (N = 12; negative control from France) to quantify the non-specific background of the ELISA. Individual results were expressed as the ΔOD value: ΔOD =ODx−ODn, where ODx represents the average of the individual ODs in the two antigen wells OD the two antigen wells [13] [31] [32] .

The study was submitted to the National Ethics Committee for Health Research of Benin, which gave its approval (Agreement N˚ 007 of May 25, 2010). Informed consent was obtained in writing from the parents of the children included in the study.

Data confidentiality and anonymity were respected. Health check-ups and treatment of uncomplicated malaria in the targeted children were free of charge. An information note presenting the objectives, benefits, risks, and inconveniences of the study was read to the study participants.

Data were entered independently in duplicate into a database using Epi data software. Data analyses were done with SPSS 16.0 software on the one hand and GraphPad Prism5 statistical software^® (GraphPad, San Diego, CA, USA) for the evaluation of the median of anti-saliva antibody levels. Differences in proportions were analyzed by the test of equality of two proportions. The distribution of ADOs is not normally (Gaussian) distributed. Thus, nonparametric tests were used to compare antibody levels. The Mann-Whitney Test was used for comparison of antibody levels between two independent groups, and the nonparametric Kruskal-Wallis test was used to compare ADOs between more than two groups. All differences were considered significant at p < 0.05.

The socio-demographic characteristics of the children included in this study are similar to those described in previously published epidemiological studies [29] . Thus, 240 children were recruited from the 4 health facilities participating in the study. A total of 185 children tested negative among the children under five years of age included in the study. The 120 cases and 120 controls were well identified. The median age of the children was 21 months (minimum 1 and maximum 59 months). The sex ratio (M/F) was 1.01. The 120 children referred to as cases (i.e., with an axillary temperature greater than 37˚5) suffering from malaria and the 120 others referred to as controls were selected at random according to their origin and progressively according to their order of arrival at the health center on the list of cases and controls. Their mothers or guardians were asked to fill out a questionnaire. Of the parents or caregivers of children at the health center, 169 (70.4%) reported that their children had slept under LLINs the day before they arrived at the health center. This rate represents 70.8% (85 children) among cases and 70.0% (84 children) among controls with no significant difference (p = 0.8). When considering the status of the villages, the rate of use is 68.0% in the high resistance zone and 74.2% (69 children) in the low resistance zone with no significant difference (p = 0.3). The use of impregnated mosquito nets the day before the survey, the frequency of use of impregnated mosquito nets during the two weeks preceding the day of the survey, the availability of impregnated mosquito nets at the household level, and the use of other means of protection against mosquito bites were the same in both groups of children (cases and controls) (p > 0.05) in both low and high vector resistance localities [29] .

We assessed levels of anti-gSG6-P1 antibodies to malaria in children to determine their level of exposure to mosquito bites. We analyzed and presented the results of anti-gSG6-P1 IgG assays, by ELISA, of 240 individuals (120 cases and 120 controls constituting the population of this study). This biomarker is thought to better explain variations in exposure or human-vector contact. These analyses were done taking into account the use of LLINs, the clinical status of the children, the presence or absence of the parasite, the status of the clusters found, and the communes in which the health centers used for data collection were located.

For the “resistance” effect, we compared the profile of individual IgG responses of subjects in low-resistance clusters (n = 93; 38.75%) with those in high-resistance clusters (n = 130; 61.25%), across all ages of children. The evolution of median-specific IgG levels between children was similar (Figure 2(a)). This suggested that vector resistance at the cluster level does not affect the anti-saliva IgG response and that specific IgG levels were similar in both groups (p = 0.4146). Figure 2(b) shows that the median antibody level was similar in children whether or not they used LLINs the day before the survey (p = 0.7724).

In this study, gender and especially age, major factors of variation in the adaptive immune response, do not have a significant effect on the development of the Ac (IgG) response.

We tested whether individuals infected with P. falciparum (the only plasmodial species diagnosed in this study) had different profiles or levels of specific IgG response to An. gambiae total saliva. For this purpose, we took all the samples and compared the specific IgG levels between uninfected individuals (n= 185) and P. falciparum infected individuals (n= 55). The results of this comparison are shown in Figure 3(b).

Figure 3(a) shows that the median level of An. gambiae total saliva specific IgG was not significantly lower in uninfected individuals compared to those infected with P. falciparum (p = 0.1152; non-parametric Mann-Whitney test).

Large variabilities were observed at study site (Figure 2(a)). The medians of the graph on the left clearly indicate that the exposure levels are different and position Pobè as the commune with the highest level according to the Kruskal Wallis test (p = 0.03).

In addition, the individual specific IgG levels of the infected individuals were

mostly above 0.250 (Reaction Threshold); this was not the case for the non-infected individuals in whom several OD (individual) < 0.250 were observed.

The results of individual gSG6-P1-specific IgG responses in cases and controls are shown in Figure 4. No differences in exposure level were observed in cases and controls according to the level of insecticide resistance of the vectors. The Kruskal Wallis test (p = 0.7724) noted no difference in exposure levels even with LLIN use.