In this study, a multiplex assay to simultaneously measure responses to 13 peptides derived from pre-erythrocytic P. falciparum Ags and 2 peptides derived from one A. gambiae salivary Ag has been developed.
Though multiplex immunoassays have already proven to be useful in serological [39–42] and malaria research [10, 43], the present multiplex assay is the first to include 15 peptides simultaneously and to combine P. falciparum and A. gambiae Ags.
No interference between bead sets was observed in the present study (Figure 2a) or in work from others . Furthermore, previous reports have shown that multiplex results are consistent with those obtained by ELISAs and that both methodologies are equally sensitive [14, 44]. The multiplex assay is a flexible technique, allowing new targets to be included when required. The simultaneous evaluation of multiple Abs leads to a reduction in both the time needed for the measurement and in its cost. The small volume of serum required is an important criterion, as it is fitted to the screening against several Ags of large populations of children, from whom minimal amounts of blood can be obtained. Therefore, multiplex assays can reliably test Abs to multiple Ags at one time in a fast and affordable manner. The high price of the Luminex machine prevents however that it is as widespread as the ELISA machines.
Thirteen pre-erythrocytic peptides from P. falciparum were included in the assay; such peptides are useful for monitoring the exposure to the malaria parasite. Using peptides in place of intact proteins in such immunological essay could lead to a loss of sensitivity (i.e. a restriction on the number of epitopes) and may on occasion give unexpected cross-reactivities. However, the peptides chosen for the present study had been previously compared to recombinant proteins in ELISA essays and were shown to give consistent results. A high prevalence of Ab response to pre-erythrocytic Ags has been found in individuals living in malaria endemic areas [23, 26, 27], and the level of exposure to P. falciparum correlates with prevalence rates for those Ags . In agreement with previous data, the present results showed that for some of these peptides, such as Lsa1-41, Lsa1J, Lsa3NR2, Glurp, Salsa2, StarpR, CSP and SR11.1, the proportion of seropositive individuals clearly increases with malaria exposure levels (Figure 3), and higher Ab response rates were seen with the Lsa1-41, Glurp, Salsa2, and CSP peptides. This data suggests that such Ags can be considered as possible biomarkers of exposure, and further studies in such direction would be valuable. Moreover, for all peptides shown in Figure 3, the proportion was higher in exposed individuals living in malaria endemic countries, compared to transiently exposed travellers. This trend was similar for median fluorescence intensities of Lsa1-41, Glurp, Salsa1, Salsa2 and CSP SR11.1 peptides (Figure 4). The very low responses recorded towards LSA3-RE are in sharp contrast with the high prevalence rates and titers obtained previously when using the same peptide in ELISA assays (, and unpublished observation in the same study populations investigated here). This may reflect alterations of antigenicity related to the coating to BSA or to multiplex beads, and raise doubts about the results obtained with that particular peptide in the multiplex assay.
When evaluating the distribution of the mean number of peptides against which individuals were seropositive, the value was found related to the level of malaria endemicity to which they were exposed. Figure 5 shows that such measurements can clearly stratify individuals based on their exposure levels, with a clearly higher number of peptides recognized by individuals exposed to high malaria transmission.
In addition to pre-erythrocytic P. falciparum Ags, the assay also included two peptides derived from one A. gambiae salivary Ag. As previously shown, the evaluation of immune responses to mosquito salivary Ags can indicate exposure to vector bites. Studies have demonstrated that children living in malaria endemic regions develop Abs against the vector's salivary proteins . Thus, Abs against A. gambiae salivary Ags represent an immunological marker of exposure to malaria vector . One of these Ag, the gSG6 salivary protein, was first described in 1999 . The gSG6 Ag is highly conserved among Anopheles species. It was reported to be potentially antigenic in travellers briefly exposed to Anopheles bites and was more recently confirmed as antigenic in Senegalese children .
The two peptides derived from gSG6 protein included in the assay (Saliv1 and Saliv2) were antigenic, but the intensity of their IgG responses was peptide-dependent, where Saliv1 showed the higher response . In the present study, Saliv1 was the target of an Ab response only in regularly exposed individuals (Figures 3 and 4, and Table 2). The results of the present study indicated that the immune response to Saliv1 peptide could be used to differentiate individuals, based on their exposure to Anopheles bites, short-term exposure in travellers and long-term exposure in people living in endemic countries. Furthermore, immune responses against such peptides are difficult to detect in briefly exposed travellers, though a response was observed when the Abs were tested directly against mosquito saliva .
It has been reported recently that the gSG6-P1 peptide (corresponding to Saliv1) can be used to evaluate low-level exposures to Anopheles bites, notably in Senegalese children ≥2 years of age . Unfortunately, in the present study population, the number of young children was too low to confirm these findings. Furthermore, the low-level exposure group (travellers) was exposed to Anopheles bites at a lower rate compared to Senegalese children due shorter exposure times and systematic use of anti-vectorial devices (impregnated bed nets, long-sleeved battledresses and repellents). This factor, together with the different assay used - multiplex at a 1:100 dilution of sera in our case and ELISA at a 1:20 dilution of sera in the study performed by Poinsignon et al  - could explain why Abs against Saliv1 were not detected in the travellers group. Further, the epitopes used could be less representative of the native ones present in humans and, therefore, less antigenic.