Assessment of exposure to malaria vectors is important to our understanding of spatial and temporal variations in disease transmission and facilitates the targeting and evaluation of control efforts. Recently, an immunogenic Anopheles gambiae salivary protein (gSG6) was identified and proposed as the basis of an immuno-assay determining exposure to Afrotropical malaria vectors. In the present study, IgG responses to gSG6 and 6 malaria antigens (CSP, AMA-1, MSP-1, MSP-3, GLURP R1, and GLURP R2) were compared to Anopheles exposure and malaria incidence in a cohort of children from Korogwe district, Tanzania, an area of moderate and heterogeneous malaria transmission. Anti-gSG6 responses above the threshold for seropositivity were detected in 15% (96/636) of the children, and were positively associated with geographical variations in Anopheles exposure (OR 1.25, CI 1.01–1.54, p = 0.04). Additionally, IgG responses to gSG6 in individual children showed a strong positive association with household level mosquito exposure. IgG levels for all antigens except AMA-1 were associated with the frequency of malaria episodes following sampling. gSG6 seropositivity was strongly positively associated with subsequent malaria incidence (test for trend p = 0.004), comparable to malaria antigens MSP-1 and GLURP R2. Our results show that the gSG6 assay is sensitive to micro-epidemiological variations in exposure to Anopheles mosquitoes, and provides a correlate of malaria risk that is unrelated to immune protection. While the technique requires further evaluation in a range of malaria endemic settings, our findings suggest that the gSG6 assay may have a role in the evaluation and planning of targeted and preventative anti-malaria interventions.
Citation: Stone W, Bousema T, Jones S, Gesase S, Hashim R, Gosling R, et al. (2012) IgG Responses to Anopheles gambiae Salivary Antigen gSG6 Detect Variation in Exposure to Malaria Vectors and Disease Risk. PLoS ONE 7(6): e40170. https://doi.org/10.1371/journal.pone.0040170
Editor: Erika Martins Braga, Universidade Federal de Minas Gerais, Brazil
Received: March 31, 2012; Accepted: June 5, 2012; Published: June 29, 2012
Copyright: © 2012 Stone et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by the Intermittent Preventive Treatment in Infants (IPTi) Consortium and the Gates Malaria Partnership, both of which are supported by the Bill and Melinda Gates Foundation. T.B. is supported by a Gates Grand Challenges Exploration Grant (grant 51991); C.D. is supported by a grant from the Wellcome Trust (grant 078925). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Heterogeneity in malaria exposure is present at all levels of endemicity  but is most readily observed in areas of low transmission and following periods of extensive control –. Recent evidence of decreasing malaria incidence , , has fuelled calls for malaria elimination from the world’s public health, political and philanthropic authorities , . As a result the interest in malaria heterogeneity and its potential effect on malaria control has increased , , . Hotspots of higher malaria transmission are likely to hamper malaria elimination efforts, as residual foci of persistent malaria infection may seed transmission to the wider community –.
Although not all factors that affect malaria heterogeneity are fully understood, variation in the exposure to malaria vectors is likely to be of key importance , –. In sub-Saharan Africa, the transmission of Plasmodium falciparum is maintained by three key mosquito species; Anopheles gambiae, An. arabiensis and An. funestus . Mosquito exposure is typically assessed as a component of the entomological inoculation rate (EIR), which is defined as the number of infectious Anopheles bites per person per unit time (ib/p/yr) , . Despite its value in malaria research, a direct assessment of EIR to determine small-scale variation in malaria exposure is operationally unattractive at low levels of transmission (EIR<10 ib/p/yr) –. The development of accurate and sensitive tools for identifying micro-epidemiological variations in vector exposure and malaria risk is important in assessing the efficiency of control efforts and focusing interventions to those areas or populations that are most affected by malaria. Serological assessments of malaria exposure are receiving increasing interest in this respect and have been used for quantifying malaria transmission intensity  and its temporal  and spatial variation , , . Recently, serological markers of malaria exposure were also used to quantify heterogeneity in the efficacy of malaria interventions . Recombinant malaria blood stage antigens have been most widely used for these purposes , while responses to the infective sporozoite specific circum-sporozoite protein (CSP) are currently viewed as the best available serological tool to detect exposure to infectious mosquito bites , –. A similar tool to identify spatial patterns of cumulative exposure to Anopheles biting could be integral to the detection of malaria hotspots and play a role in forecasting the risk of malaria epidemics or the dynamics of malaria resurgence in areas where parasite carriage in human populations has decreased but exposure to malaria vectors persists .
Our understanding of the human immune response to mosquito saliva has until recently been largely restricted to culicine mosquitoes and the clinical consequences of allergy –. Humoral responses to the saliva of various disease vectors have been exploited epidemiologically, revealing significant correlation with disease seropositivity and vector exposure. Such assays have now been described for Ixodes ticks , , triatomine bugs , Glossina tsetse flies  and Lutzomyia and Phlebotomus sand flies , . Recently, transcriptome analysis of the salivary glands of An. gambiae females identified over 70 putative secreted salivary proteins –. A small (∼10 kb) immunogenic protein, gambiae salivary gland protein 6 (gSG6), that is well conserved in the three major Afrotropical malaria vectors (An. gambiae, An. arabiensis and An. funestus) and restricted to anopheline mosquitoes , has been identified as a suitable candidate for a bioassay of Anopheles exposure , . Antibody responses to a gSG6 peptide (gSG6-P1) described Anopheles exposure in areas of low vector density  and in response to vector control programs  with some success, and were recently shown to reflect Anopheles heterogeneity at the district level in Dakar, Senegal . Recombinant full length gSG6 has also shown strong immunogenicity among rural populations in Burkina Faso, which appears to be sufficiently short lived to correlate with seasonal changes in Anopheles abundance , .The relationship between malaria case incidence and anti-gSG6 response has not been studied, despite early indications that humoral responses to Anopheles whole saliva were positively associated with malaria infection .
Using a subset of samples collected during a large study of intermittent presumptive treatment among infants (IPTi) , along with entomological data from an intensive survey in the same area , we present the first evaluation of IgG antibody responses to the recombinant gSG6 salivary antigen for describing spatial heterogeneity in vector exposure between and within geographically defined subvillages in an area of moderate and heterogeneous malaria exposure in northern Tanzania. At the individual level, we determine the association of gSG6 reactivity with household Anopheles exposure and subsequent malaria incidence. In addition, we determined reactivity against a selection of malaria antigens that have been more commonly used in epidemiological studies, namely CSP and four blood stage proteins, AMA-1, MSP-1, MSP-3, and glutamate-rich protein (GLURP).
Witnessed written consent was provided by the caregivers of all children involved in serological sampling, and by heads of households for participation in the entomological survey. Ethical approval was granted by the review board of the National Institute for Medical Research of Tanzania, and the London School of Hygiene and Tropical Medicine ethics committee.
Study Area and Subjects
Plasma samples were collected from children recruited over 18 months as part of a longer-term study (2004–2008) carried out in the district of Korogwe, Northern Tanzania, an area of moderate malaria endemicity. Korogwe district is situated ∼600 m above sea level, and has a seasonal pattern of rainfall (800–1400 mm/year) . Malaria transmission in the Korogwe region has declined in recent years , such that an EIR of 1–14 ib/p/yr was estimated in 2007 . The original study investigated the relative impacts of different drug regimens for intermittent presumptive treatment (IPTi) among a total of 1280 infants .
Entomological Data Collection
In the final year of the IPTi study a randomly selected subset of 600 children were enrolled in a detailed entomological survey, aiming to describe spatial patterns of malaria incidence in relation to mosquito exposure . In the room of each selected child, mosquitoes were sampled with miniature CDC light traps (Model 512; John W. Hock Company, Gainesville, Florida) for one night at the end of the wet season (May), again at the beginning (July) and finally the end (September) of the dry season in 2008. Mosquito exposure at the household level was highly correlated between all surveys (correlation coefficient: May/July = 0.462, May/September = 0.497, July/September = 0.444; p<0.0001). Mosquito data from first of the three sampling points, during the peak transmission season when Anopheles abundance was highest, was therefore deemed adequate in displaying variation in exposure. Of the total Anopheles females caught during sampling, An. gambiae s.l. made up 80.3%, An. funestus 18.6% and other anophelines 1%.
Clinical Data and Plasma Samples
Malaria incidence was assessed by passive monitoring for signs of illness throughout the 22 months following recruitment, during which time free access to clinical treatment was provided . The average age at recruitment was 9.4 weeks (range 8–17 weeks) and infants were recruited at different times of the year, i.e. at different time-points in the transmission season. Plasma samples used in the current study were taken at 9 months of age when infants were presented at clinics as part of the Expanded Program on Immunisation (EPI). Blood samples were collected by finger prick and after plasma separation samples were stored at −20°C until processing. In our analyses, we included malaria incidence in the period between serum collection at 9 months of age and the end of follow-up. This gave an effective follow up period of approximately 15 months and ensured that the follow-up period included one or more peak malaria transmission seasons for each child. The current analyses are an ancillary study and many of the blood samples had been used previously for other IPTi specific investigations. As a result of this non-systematic exhaustion of samples, sera were available for a subset of 636/1280 children for gSG6 ELISA; 247/636 children from this subset were involved in the household level entomological survey.
ELISA was performed as previously described with few modifications , . Briefly, Maxisorp 96-well plates (Nunc M9410) were coated with gSG6 at 5 ug/ml. Test and negative control serum were analysed in duplicate at 1∶100 in phosphate buffered saline with 0.05% Tween 20 (PBST)/1% skimmed milk powder (Marvel, UK). On every plate blank wells (PBST/Marvel) were included to correct sample ODs for background antibody reactivity, and positive control sera (1∶40 in PBST/Marvel) were analysed to allow standardisation of OD values for day-to-day and inter-plate variation. Positive control sera was provided, with consent, by an employee of the London School of Hygiene and Tropical medicine who was exposed weekly to the bites of approximately 50–100 laboratory bred An. gambiae s.s (Kisumu strain) during colony feeding.
Sera from 39 Europeans with no recent history of travel to malaria endemic countries were used as negative controls for calculation of IgG seroprevalence. Cut off for seropositivity among samples was determined as the mean OD of the unexposed sera plus 3 standard deviations.
P. falciparum ELISA and Luminex Assays
For this analysis, IgG antibody responses were chosen in preference to IgM for their high antigen specificity. IgG antibody responses against CSP (Gennova, 0.009 µg/ml), AMA-1 (BPRC, 0.3 µg/ml) and MSP-119 (CTK Biotech, 0.2 µg/ml) were detected as previously described , . Test sera were analysed in duplicate at 1∶200 (CSP), 1∶1000 (MSP-119) or 1∶2000 (AMA-1) in PBST/Marvel. Blank wells, positive control sera from a hyper-endemic region in the Gambia , and a serial dilution of pooled hyper-immune sera were included in duplicate on each plate to correct for non-specific reactivity and allow standardisation of inter-plate variation. Seroprevalence of IgG antibodies to these non-salivary antigens was calculated using a mixture model as described previously , .
Recombinant proteins corresponding to the R1, R2 (Central repeat and C-terminal repeat regions of GLURP), and the C-terminal region of MSP-3  were covalently coupled to carboxylated luminex microspheres according to the manufacturer’s protocol and tested as previously described . Cut-off for positivity was calculated as the mean reactivity in malaria non-exposed European individuals plus 2 standard deviations.
To examine the relationship between patterns of gSG6 reactivity and small scale spatial variation in Anopheles exposure, antibody responses were described at the level of subvillages, which are defined by their geographical location (Figure 1) . The arithmetic mean mosquito exposure for each village was used for ranking villages from low to high mosquito exposure; this rank was related to antibody prevalence and mean log 10 adjusted antibody level per subvillage. This enabled analyses relating to geographic variations in Anopheles abundance for all individuals, irrespective of their involvement in the entomological survey (Figure 2).
Sampling in Korogwe district was conducted in 5 areas, which are marked on the map: Korogwe, Majengo, Magasin, Mnyuzi, and Mandera. Within these areas, our study population were resident in 15 subvillages. Korogwe consisted of the following subvillages: Kwasemangube (KS), Lwengera (LW) Msambazi (MS) and Masuguru (MU). Majengo consisted of the following subvillages: Kilole (KI), Majengo (MJ) and Manundu (MA). Magasin consisted of the following subvillages: Kwagunda (KW) and Maguga (MG). Mnyuzi consisted of the following subvillages: Gereza (GE), Lusanga (LU), Mkwakwani (MK), Mnyuzi (MY) and Shambakapori (SH). Mandera (MD) was an isolated subvillage.
Numbers of households sampled for each subvillage, in order of Anopheles exposure, were as follows: MY = 45, MS = 23, MA = 26, MU = 21, MJ = 29, KS = 24, LW = 65, LU = 61, MD = 45, MK = 14, KW = 99, SH = 13, GE = 47, KI = 30, MG = 91.
For infants for whom both household mosquito data and plasma samples were available, it was possible to investigate associations between Anopheles exposure and antibody reactivity against salivary and malaria antigens at an individual level. For this purpose, households were analysed in quintiles of Anopheles exposure (Table 1).
Statistical analysis was conducted in STATA (Version 10, STATA statistical software StataCorp) and GraphPad Prism (Version 5.0, GraphPad Software Inc., La Jolla, CA) software packages. IgG responses to salivary or malaria antigens between two independent groups were compared by Wilcoxon rank-sum tests (Mann-Whitney U test), with Bonferroni correction for multiple comparisons between subgroups. Comparisons of multiple groups were carried out by Kruskal-Wallis test. Seroprevalence comparisons were made using Chi-square test, with a test for trend in proportions. Correlations between IgG and malaria or entomological measures were made using Spearman correlation or with linear regression analysis after log10-transformation of OD data. IPTi treatment arm was included in our analyses as potential confounder. As a small number of sample ODs were lower than their ELISA plate blank value, some normalised ODs had negative values and an arbitrary positive value (+1) was therefore added to all ODs before transformation.
Small Scale Spatial Variation in Anopheles Exposure and anti-gSG6 Responses
The recombinant gSG6 protein elicited significant anti-gSG6 IgG responses in children from Korogwe district (mean OD 0.109, maximum OD 2.014). European sera were used as negative controls for exposure to Anopheles mosquitoes, the responses of which were pooled to determine a cut-off for seroprevalence at OD 0.167 (Table 2). Mean OD among antibody negative children from Korogwe was 0.052, and ranged from 0.001–0.166 (standard deviation 0.040). IPTi treatment arm was not associated with gSG6 antibody prevalence (p = 0.23) or density (p = 0.38) and did not show any evident association with any of the other antigens tested, nor was it found to be a confounder in any of the associations presented below (data not shown).
When mean mosquito exposure was plotted against log 10 adjusted anti-gSG6 IgG level for each of the 15 subvillages, a significant positive association was observed between mean mosquito exposure and antibody reactivity (Figure 3). Similarly, despite significant variability in gSG6 response between subvillages, there was a significant positive association between mean mosquito exposure per subvillage and anti-gSG6 IgG seropositivity, wherein an average increased exposure of 10 mosquitoes was associated with a 25% increase in antibody positivity (odds ratio [OR] 1.25, CI 1.01–1.54, p = 0.04).
Anti-gSG6 IgG levels are given as the log-10 adjusted mean anti-gSG6 OD per subvillage. Mosquito exposure is given as the ascending and sequential mean Anopheles female count for each of 15 subvillages (x-axis), as in Figure 2. The trend-line from the linear regression is shown as a dashed line (r2 = 0.436, p = 0.007).
Household Level Mosquito Exposure and anti-gSG6 Response
Information on household-level mosquito exposure was available for the households of 247 children. At the level of individual households, exposure to Anopheles females showed a significant positive correlation with anti-gSG6 IgG level (correlation coefficient 0.188, p = 0.003) but not with levels of anti-CSP IgG (correlation coefficient 0.036, p = 0.59). When households were grouped into quintiles according to their relative exposure to Anopheles (Table 1), there was a statistically significant positive association between Anopheles exposure in quintiles and anti-gSG6 IgG levels (p = 0.001) and prevalence (test for trend in proportions, p = 0.001) (Figure 4). There was no evident association between individual Anopheles exposure in quintiles and individual CSP antibody level (p = 0.544) or prevalence (test for trend in proportions p = 0.422). Similarly, no significant associations were observed between Anopheles exposure in quintiles and individual responses to any blood stage antigen, save MSP-3 for which there was a significant positive association with antibody level (p = 0.017).
A. Box plots showing anti-gSG6 IgG level between groups sorted according to Anopheles exposure in quintiles. Boxes show the median and 25th/75th percentiles, whiskers show the 5th/95th percentiles, and outliers are represented by dots. Where outliers were excluded from the graph but not analysis they are marked with a + and included in parentheses. P values for pairwise comparisons were determined by Mann-Whitney test with Bonferroni correction (*), and for all groups by Kruskal-Wallis test (**). B. Seroprevalence of anti-gSG6 and anti-P. falciparum IgG antibodies plotted against Anopheles exposure in quintiles. Error bars indicate 95% confidence intervals (CI). P values were determined by a test for trend in proportions (***).
Malaria Incidence and anti-gSG6 and Anti-malaria Responses
Antibody levels were positively associated with the frequency of malaria episodes recorded after serum collection for all antigens except AMA-1 (gSG6 correlation coefficient 0.240, p<0.0001 (Figure 5A); CSP correlation coefficient 0.183, p = 0.004; MSP -1 correlation coefficient 0.256, p<0.0001; MSP-3 correlation coefficient 0.141, p = 0.0008; GLURP R1 correlation coefficient 0.126, p = 0.003; GLURP R2 correlation coefficient 0.101, p = 0.017 [data not shown]). The prevalence of IgG responses varied significantly with grouped malaria incidence for gSG6 (p<0.0001), AMA-1 (p = 0.004), MSP-1 (p<0.0001) and GLURP R2 (p = <0.001). No significant variation in seroprevalence of antibodies to CSP, MSP-3 and GLURP R1 was present between groups of malaria incidence (Figure 5B). A strong positive association was observed between grouped malaria incidence and the prevalence of antibody responses against gSG6, MSP-1 and GLURP R2, while this relationship was present but only marginally significant for MSP-3 (Figure 5B).
Malaria incidence is grouped into 0 episodes, 1–2 episodes or >3 episodes. A. Box plots showing anti-gSG6 IgG level between groups sorted according to malaria incidence subsequent to serological sampling. Boxes, whiskers and P values are as in Figure 4. n = 269. B. Seroprevalence of anti-gSG6 and anti-P. falciparum IgG antibodies plotted against grouped malaria incidence. Sample sizes vary by antigen according to the available serological methodology; CSP n = 246, AMA-1 n = 227, MSP-1 n = 227, MSP-3 n = 566, GLURP R1 n = 566, GLURP R2 n = 566. P values were determined by a test for trend in proportions (***). Error bars denote 95% CI.
In the present study we show that the antibody responses of young children to the recombinant An. gambiae salivary protein, gSG6, reflect small scale spatial variation in malaria transmission, and are strongly associated with malaria risk in an area of moderate transmission intensity in northern Tanzania where An. gambiae and An. funestus are the main malaria vectors.
Reactivity to both the peptide and recombinant forms of the anopheline gSG6 protein has previously been associated with seasonal or regional patterns in mosquito exposure –, . The current study is the first to describe antibody responses to the recombinant gSG6 protein in relation to village of residence, and individual level mosquito exposure and malaria incidence. For this, we utilised a detailed entomological dataset from Korogwe district, Tanzania, that revealed significant heterogeneity in Anopheles abundance between and within villages . Despite generally low reactivity among our infant study population, anti-gSG6 IgG level and prevalence effectively described varying levels of exposure to Anopheles between subvillages, corroborating recent findings from Senegal where gSG6-P1 responses reflected spatial variation in Anopheles exposure between districts in urban Dakar . The first studies to assess IgG responses to recombinant gSG6 were carried out in two rural villages in Burkina Faso, and revealed >50% seroprevalence in children during the peak transmission season . The lower responses observed in this study confirm the lower transmission intensity in the current study area.
At the level of subvillages, anti-gSG6 antibody responses closely followed patterns in malaria incidence and community-level antibody responses to malaria-specific antigens AMA-1 and MSP-119 . This broad agreement in estimates of malaria incidence and Anopheles and malaria-specific antibody responses at subvillage level is unsurprising , , , . Patterns may diverge when assessed at an individual level, as Anopheles abundance and biting behaviour may be unevenly distributed between households , ,  and intense mosquito exposure may not necessarily mean a high malaria exposure if anophelines are not infected. This commonly happens at the start of the wet season when mosquitoes have just emerged and are unlikely to have completed a sporogonic cycle , but mosquito sporozoite rates may also show spatial variation . Associations between mosquito exposure, malaria incidence and immune responses are further complicated by the fact that individuals with the highest malaria exposure will acquire protective immunity most rapidly and may experience lower malaria incidence in some settings , . In general, it is complex to disentangle markers of exposure from markers of protection when analysing malaria blood stage antigens. Recent studies highlight the importance of considering malaria heterogeneity when determining the protective effect of antibody responses on clinical malaria. Initially, counterintuitive observations that higher blood stage immune responses were associated with increased malaria incidence , , were explained by adjusting for heterogeneity in malaria exposure and excluding non-parasitaemic individuals. This revealed a protective effect among immune responders, reflecting either true or surrogate humoral immune mediation . This methodological challenge, first described by Bejon and colleagues , , has highlighted the need for markers that capture heterogeneity in malaria exposure but are not associated with clinical protection , , . Markers of mosquito exposure, as described in this manuscript, may play this role by identifying those individuals most at risk of malaria.
No clear associations were apparent between Anopheles exposure at an individual level and antibody responses to any of the malaria-specific antigens (CSP, AMA-1, MSP-1, MSP-3, GLURP R1, GLURP R2). Anti-CSP reactivity might be expected to correlate with exposure to infected mosquito bites and therefore perhaps also with overall mosquito biting, but in our analysis did not. This may be a consequence of the relatively small sample size, and low EIR , ; in moderate to low endemic areas the proportion of infected vectors is frequently lower than 1% , , . Contrary to this, individual-level anti-gSG6 responses were strongly associated with household Anopheles exposure. Interestingly, mosquito exposure was assessed towards the end of the study, starting approximately 15 months after the serum sample that was used for serology was collected. This suggests that heterogeneity in mosquito exposure is consistent over time in our study area, supporting the hypothesis of stable hotspots of malaria transmission , .
We previously showed that antibody responses to blood stage malaria antigens determined in clinic attendees reliably predicted spatial patterns in malaria incidence in a cohort of children living in the same area . We here extended these analyses and showed that an individual’s antibody responses to MSP-1, MSP-3 and GLURP-R2 are all positively associated with subsequent malaria incidence. The selection of malaria antigens we used in this study was not intended to be exhaustive, nor did we aim to identify the malaria antigen with the highest discriminative power to detect variation in malaria exposure. We chose 4 malaria antigens to put our findings with gSG6 in an epidemiological context. Our findings are consistent with previous reports from areas of heterogeneous exposure where malaria specific antibody responses as markers of past exposure predict future exposure , . Strikingly, in our analyses anti-gSG6 responses also provided a strong association with malaria incidence, indicating that malaria heterogeneity is associated with heterogeneous biting behaviour . Unlike responses to transmission and blood stage malaria antigens ,  responses to gSG6 confer no protection to malaria, thus avoiding any confounding associations with immunity and malaria incidence. In such a way, the gSG6 assay may provide a useful marker for exposure to malaria for use in clinical studies .
Though the sampling framework of the current study was not designed to evaluate the temporal dynamics of the anti-gSG6 response, there are indications that, as with responses to the salivary proteins of other haematophagous arthropods, it elicits short lived antibody responses, reflecting only recent Anopheles exposure , , . As blood-feeding is transitory, and saliva is only released into the skin during probing with the majority likely to be re-ingested with the blood meal, this limits the development of a humoral immune response to mosquito saliva –. This short exposure to antigen explains the low anti-gSG6 responses observed among children from Korogwe. These low level responses highlight inherent problems in assessing exposure using an arbitrarily defined cut off for seropositive individuals. Identifying individuals never exposed to malaria is relatively straightforward but the same cannot be said for individuals never exposed to Anopheles, a genus which has a very wide geographical distribution. The nature of mosquito feeding, with the strength of the correlations observed in our analyses between spatial and individual level mosquito exposure and antibody OD, supports the use of antibody level rather than seroprevalence as a finer tool for assessment of Anopheles exposure intensity.
This is the first report that antibody responses to the recombinant An. gambiae salivary protein gSG6 in children can reflect small-scale spatial variation in exposure to anophelines at village and household level. Importantly, our analysis also provides the first evidence for a reliable association between malaria incidence and anti-gSG6 response; a relationship only previously observed using whole An. gambiae saliva . Caution is required in extrapolating findings from this study to other age groups because our analyses were restricted to plasma samples from children aged 9 months and a role of maternal transfer of IgG during breastfeeding can therefore not be excluded. This limitation of the current study does not alter our conclusions that these antibody responses are suitable markers of micro-epidemiological differences in Anopheles exposure. Potential uses for this assay include establishing Anopheles biting exposure to include indoor and outdoor biting, controlling for exposure in highly heterogeneous settings, and as a measure of receptivity to inform programs that are moving toward elimination where there is a high risk for re-introduction. However, its utility in low endemic and pre-elimination settings first needs to be assessed . To this end, it will be important to establish the assays suitability for use with scalable antibody sources such as dried filter paper blood-spots. The identification and analysis of other salivary proteins may also help increase the sensitivity of the approach in such settings .
Conceived and designed the experiments: WS TB RG DC CD BA. Performed the experiments: WS SJ TT SG. Analyzed the data: WS TB IC DC. Contributed reagents/materials/analysis tools: SG RG RR DM BA CD RH. Wrote the paper: WS TB RG TT BA CD.
- 1. Kreuels B, Kobbe R, Adjei S, Kreuzberg C, von Reden C, et al. (2008) Spatial Variation of Malaria Incidence in Young Children from a Geographically Homogeneous Area with High Endemicity. Journal of Infectious Diseases 197: 85–93.
- 2. Bhattarai A, Ali AS, Kachur SP, Mårtensson A, Abbas AK, et al. (2007) Impact of Artemisinin-Based Combination Therapy and Insecticide-Treated Nets on Malaria Burden in Zanzibar. PLoS Med 4: e309.
- 3. Clark TD, Greenhouse B, Njama-Meya D, Nzarubara B, Maiteki-Sebuguzi C, et al. (2008) Factors Determining the Heterogeneity of Malaria Incidence in Children in Kampala, Uganda. Journal of Infectious Diseases 198: 393–400.
- 4. Okiro E, Hay S, Gikandi P, Sharif S, Noor A, et al. (2007) The decline in paediatric malaria admissions on the coast of Kenya. Malaria Journal 6: 151.
- 5. Hommel M (2008) Towards a research agenda for global malaria elimination. Malaria Journal 7: S1.
- 6. Das P, Horton R (2010) Malaria elimination: worthy, challenging, and just possible. The Lancet 376: 1515–1517.
- 7. O’Meara WP, Mangeni JN, Steketee R, Greenwood B (2010) Changes in the burden of malaria in sub-Saharan Africa. The Lancet Infectious Diseases 10: 545–555.
- 8. Moonen B, Cohen JM, Snow RW, Slutsker L, Drakeley C, et al. (2010) Operational strategies to achieve and maintain malaria elimination. The Lancet 376: 1592–1603.
- 9. World Health Organisation (2007) Malaria elimination: a field manual for low and moderate endemic countries.
- 10. Bousema T, Griffin JT, Sauerwein RW, Smith DL, Churcher TS, et al. (2012) Hitting Hotspots: Spatial Targeting of Malaria for Control and Elimination. PLoS Med 9: e1001165.
- 11. Bousema T, Drakeley C, Gesase S, Hashim R, Magesa S, et al. (2010) Identification of Hot Spots of Malaria Transmission for Targeted Malaria Control. Journal of Infectious Diseases 201: 1764–1774.
- 12. Smith DL, Drakeley CJ, Chiyaka C, Hay SI (2010) A quantitative analysis of transmission efficiency versus intensity for malaria. Nat Commun 1: 108.
- 13. Oesterholt M, Bousema JT, Mwerinde OK, Harris C, Lushino P, et al. (2006) Spatial and temporal variation in malaria transmission in a low endemicity area in northern Tanzania. Malaria Journal 5: 1–7.
- 14. World Health Organisation (2010) World Malaria Report.
- 15. Drakeley C, Schellenberg D, Kihonda J, Sousa CA, Arez AP, et al. (2003) An estimation of the entomological inoculation rate for Ifakara: a semi-urban area in a region of intense malaria transmission in Tanzania. Tropical Medicine & International Health 8: 767–774.
- 16. Hay SI, Rogers DJ, Toomer JF, Snow RW (2000) Annual Plasmodium falciparum entomological inoculation rates (EIR) across Africa: literature survey, internet access and review. Transactions of the Royal Society of Tropical Medicine and Hygiene 94: 113–127.
- 17. Hii JLK, Smith T, Mai A, Ibam E, Alpers MP (2000) Comparison between anopheline mosquitoes (Diptera: Culicidae) caught using different methods in a malaria endemic area of Papua New Guinea. Bulletin of Entomological Research 90: 211–219.
- 18. Corran P, Coleman P, Riley E, Drakeley C (2007) Serology: a robust indicator of malaria transmission intensity? Trends in Parasitology 23: 575–582.
- 19. Mbogo CN, Glass GE, Forster D, Kabiru EW, Githure JI, et al. (1993) Evaluation of light traps for sampling anopheline mosquitoes in Kilifi, Kenya. Journal Of The American Mosquito Control Association 9: 260–263.
- 20. Drakeley CJ, Corran PH, Coleman PG, Tongren JE, McDonald SLR, et al. (2005) Estimating medium- and long-term trends in malaria transmission by using serological markers of malaria exposure. Proceedings of the National Academy of Sciences of the United States of America 102: 5108–5113.
- 21. Stewart L, Gosling R, Griffin J, Gesase S, Campo J, et al. (2009) Rapid Assessment of Malaria Transmission Using Age-Specific Sero-Conversion Rates. PLoS ONE 4: e6083.
- 22. Bejon P, Williams TN, Liljander A, Noor AM, Wambua J, et al. (2010) Stable and Unstable Malaria Hotspots in Longitudinal Cohort Studies in Kenya. PLoS Med 7: e1000304.
- 23. Bousema T, Youssef RM, Cook J, Cox J, Alegana VA, et al. (2010) Serologic markers for detecting malaria in areas of low endemicity, Somalia, 2008. Emerging infectious diseases 16: 392–399.
- 24. Cook J, Kleinschmidt I, Schwabe C, Nseng G, Bousema T, et al. (2011) Serological Markers Suggest Heterogeneity of Effectiveness of Malaria Control Interventions on Bioko Island, Equatorial Guinea. PLoS ONE 6: e25137.
- 25. Drakeley C, Cook J (2009) Chapter 5 Potential Contribution of Sero-Epidemiological Analysis for Monitoring Malaria Control and Elimination: Historical and Current Perspectives. In: Rollinson D, Hay SI, editors. pp. 299–352. Advances in Parasitology: Academic Press.
- 26. Mendis C, Del Giudice G, Gamage-Mendis AC, Tougne C, Pessi A, et al. (1992) Anti-circumsporozoite protein antibodies measure age related exposure to malaria in Kataragama, Sri Lanka. Parasite Immunology 14: 75–86.
- 27. Proietti C, Pettinato DD, Kanoi BN, Ntege E, Crisanti A, et al. (2011) Continuing Intense Malaria Transmission in Northern Uganda. The American Journal of Tropical Medicine and Hygiene 84: 830–837.
- 28. Satoguina J, Walther B, Drakeley C, Nwakanma D, Oriero E, et al. (2009) Comparison of surveillance methods applied to a situation of low malaria prevalence at rural sites in The Gambia and Guinea Bissau. Malaria Journal 8: 274.
- 29. Smith DL, McKenzie FE, Snow RW, Hay SI (2007) Revisiting the Basic Reproductive Number for Malaria and Its Implications for Malaria Control. PLoS Biol 5: e42.
- 30. Brummer-Korvenkontio H, Lappalainen P, Reunala T, Palosuo T (1994) Detection of mosquito saliva-specific IgE and IgG4 antibodies by immunoblotting. Journal of Allergy and Clinical Immunology 93: 551–555.
- 31. Palosuo K, Brummer-Korvenkontio H, Mikkola J, Sahi T, Reunala T (1997) Seasonal Increase in Human IgE and lgG4 Antisaliva Antibodies to Aedes Mosquito Bites. International Archives of Allergy and Immunology 114: 367–372.
- 32. Peng Z, Simons FER (2007) Advances in mosquito allergy. Current Opinion in Allergy and Clinical Immunology 7: 350–354.
- 33. Schwartz BS, Ribeiro JMC, Goldstein MD (1990) Anti-Tick Antibodies: An Epidemiologic Tool in Lyme Disease Research. American Journal of Epidemiology 132: 58–66.
- 34. Schwartz BS, Ford DP, Childs JE, Rothman N, Thomas RJ (1991) Anti-tick Saliva Antibody: A Biologic Marker of Tick Exposure That Is a Risk Factor for Lime Disease Seropositivity. American Journal of Epidemiology 134: 86–95.
- 35. Schwarz A, Sternberg JM, Johnston V, Medrano-Mercado N, Anderson JM, et al. (2009) Antibody responses of domestic animals to salivary antigens of Triatoma infestans as biomarkers for low-level infestation of triatomines. International Journal for Parasitology 39: 1021–1029.
- 36. Poinsignon A, Remoue F, Rossignol M, Cornelie S, Courtin D, et al. (2008) Human IgG Antibody Response to Glossina Saliva: An Epidemiologic Marker of Exposure to Glossina Bites. The American Journal of Tropical Medicine and Hygiene 78: 750–753.
- 37. Barral A, Honda E, Caldas A, Costa J, Vinhas V, et al. (2000) Human immune response to sand fly salivary gland antigens: a useful epidemiological marker? The American Journal of Tropical Medicine and Hygiene 62: 740–745.
- 38. Marzouki S, Ahmed MB, Boussoffara T, Abdeladhim M, Aleya-Bouafif NB, et al. (2011) Characterization of the Antibody Response to the Saliva of Phlebotomus papatasi in People Living in Endemic Areas of Cutaneous Leishmaniasis. The American Journal of Tropical Medicine and Hygiene 84: 653–661.
- 39. Cornelie S, Remoue F, Doucoure S, NDiaye T, Sauvage F-X, et al. (2007) An insight into immunogenic salivary proteins of Anopheles gambiae in African children. Malaria Journal 6: 75.
- 40. Arcà B, Lombardo F, Valenzuela JG, Francischetti IMB, Marinotti O, et al. (2005) An updated catalogue of salivary gland transcripts in the adult female mosquito, Anopheles gambiae. Journal of Experimental Biology 208: 3971–3986.
- 41. Lanfrancotti A, Lombardo F, Santolamazza F, Veneri M, Castrignanò T, et al. (2002) Novel cDNAs encoding salivary proteins from the malaria vector Anopheles gambiae. FEBS Letters 517: 67–71.
- 42. Ribeiro JMC, Mans BJ, Arcà B (2010) An insight into the sialome of blood-feeding Nematocera. Insect Biochemistry and Molecular Biology 40: 767–784.
- 43. Rizzo C, Ronca R, Fiorentino G, Mangano V, Sirima S, et al. (2011) Wide cross-reactivity between Anopheles gambiae and Anopheles funestus SG6 salivary proteins supports exploitation of gSG6 as a marker of human exposure to major malaria vectors in tropical Africa. Malaria Journal 10: 206.
- 44. Orlandi-Pradines E, Almeras L, Denis de Senneville L, Barbe S, Remoué F, et al. (2007) Antibody response against saliva antigens of Anopheles gambiae and Aedes aegypti in travellers in tropical Africa. Microbes and Infection 9: 1454–1462.
- 45. Poinsignon A, Cornelie S, Ba F, Boulanger D, Sow C, et al. (2009) Human IgG response to a salivary peptide, gSG6-P1, as a new immuno-epidemiological tool for evaluating low-level exposure to Anopheles bites. Malaria Journal 8: 198.
- 46. Drame PM, Poinsignon A, Besnard P, Cornelie S, Le Mire J, et al. (2010) Human Antibody Responses to the Anopheles Salivary gSG6-P1 Peptide: A Novel Tool for Evaluating the Efficacy of ITNs in Malaria Vector Control. PLoS ONE 5: e15596.
- 47. Drame P, Machault V, Diallo A, Cornelie S, Poinsignon A, et al. (2012) IgG responses to the gSG6-P1 salivary peptide for evaluating human exposure to Anopheles bites in urban areas of Dakar region, Senegal. Malaria Journal 11: 72.
- 48. Rizzo C, Ronca R, Fiorentino G, Verra F, Mangano V, et al. (2011) Humoral Response to the Anopheles gambiae Salivary Protein gSG6: A Serological Indicator of Exposure to Afrotropical Malaria Vectors. PLoS ONE 6: e17980.
- 49. Remoue F, Cisse B, Ba F, Sokhna C, Herve J-P, et al. (2006) Evaluation of the antibody response to Anopheles salivary antigens as a potential marker of risk of malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 100: 363–370.
- 50. Gosling RD, Gesase S, Mosha JF, Carneiro I, Hashim R, et al. (2009) Protective efficacy and safety of three antimalarial regimens for intermittent preventive treatment for malaria in infants: a randomised, double-blind, placebo-controlled trial. The Lancet 374: 1521–1532.
- 51. Mmbando BP, Vestergaard LS, Kitua AY, Lemnge MM, Theander TG, et al. (2010) A progressive declining in the burden of malaria in north-eastern Tanzania. Malaria Journal 9: 216.
- 52. Corran PH, Cook J, Lynch C, Leendertse H, Manjurano A, et al. (2008) Dried blood spots as a source of anti-malarial antibodies for epidemiological studies. Malaria Journal 7: 195.
- 53. Theisen M, Soe S, Brunstedt K, Follmann F, Bredmose L, et al. (2004) A Plasmodium falciparum GLURP–MSP3 chimeric protein; expression in Lactococcus lactis, immunogenicity and induction of biologically active antibodies. Vaccine 22: 1188–1198.
- 54. Cham GK, Kurtis J, Lusingu J, Theander TG, Jensen AT, et al. (2008) A semi-automated multiplex high-throughput assay for measuring IgG antibodies against Plasmodium falciparum erythrocyte membrane protein 1 (PfEMP1) domains in small volumes of plasma. Malaria Journal 7: 108.
- 55. Poinsignon A, Cornelie S, Mestres-Simon M, Lanfrancotti A, Rossignol M, et al. (2008) Novel Peptide Marker Corresponding to Salivary Protein gSG6 Potentially Identifies Exposure to Anopheles Bites. PLoS ONE 3: e2472.
- 56. Woolhouse MEJ, Dye C, Etard J-F, Smith T, Charlwood JD, et al. (1997) Heterogeneities in the transmission of infectious agents: Implications for the design of control programs. Proceedings of the National Academy of Sciences 94: 338–342.
- 57. Smith DL, Dushoff J, McKenzie FE (2004) The Risk of a Mosquito-Borne Infectionin a Heterogeneous Environment. PLoS Biol 2: e368.
- 58. Bousema T, Kreuels B, Gosling R (2011) Adjusting for Heterogeneity of Malaria Transmission in Longitudinal Studies. Journal of Infectious Diseases 204: 1–3.
- 59. Clarke SE, Bøgh C, Brown RC, Walraven GEL, Thomas CJ, et al. (2002) Risk of malaria attacks in Gambian children is greater away from malaria vector breeding sites. Transactions of the Royal Society of Tropical Medicine and Hygiene 96: 499–506.
- 60. Greenhouse B, Ho B, Hubbard A, Njama-Meya D, Narum DL, et al. (2011) Antibodies to Plasmodium falciparum Antigens Predict a Higher Risk of Malaria But Protection From Symptoms Once Parasitemic. Journal of Infectious Diseases 204: 19–26.
- 61. Bejon P, Cook J, Bergmann-Leitner E, Olotu A, Lusingu J, et al. (2011) Effect of the pre-erythrocytic candidate malaria vaccine RTS,S/AS01E on blood stage immunity in young children. The Journal of infectious diseases 204: 9–18.
- 62. Bejon P, Warimwe G, Mackintosh CL, Mackinnon MJ, Kinyanjui SM, et al. (2009) Analysis of Immunity to Febrile Malaria in Children That Distinguishes Immunity from Lack of Exposure. Infect Immun 77: 1917–1923.
- 63. Kinyanjui SM, Bejon P, Osier FH, Bull PC, Marsh K (2009) What you see is not what you get: implications of the brevity of antibody responses to malaria antigens and transmission heterogeneity in longitudinal studies of malaria immunity. Malaria Journal 8: 242.
- 64. Beier JC, Killeen GF, Githure JI (1999) Short report: entomologic inoculation rates and Plasmodium falciparum malaria prevalence in Africa. The American Journal of Tropical Medicine and Hygiene 61: 109–113.
- 65. Greenwood BM (1990) Immune responses to sporozoite antigens and their relationship to naturally acquired immunity to malaria. Bulletin of the World Health Organization 68:
- 66. Fowkes FJI, Richards JS, Simpson JA, Beeson JG (2010) The Relationship between Anti-merozoite Antibodies and Incidence of <italic>Plasmodium falciparum</italic> Malaria: A Systematic Review and Meta-analysis. PLoS Med 7: e1000218.
- 67. Matsuoka H, Yoshida S, Hirai M, Ishii A (2002) A rodent malaria, Plasmodium berghei, is experimentally transmitted to mice by merely probing of infective mosquito, Anopheles stephensi. Parasitology International 51: 17–23.
- 68. Rosenberg R, Wirtz RA, Schneider I, Burge R (1990) An estimation of the number of malaria sporozoites ejected by a feeding mosquito. Transactions of the Royal Society of Tropical Medicine and Hygiene 84: 209–212.
- 69. Medica DL, Sinnis P (2005) Quantitative Dynamics of Plasmodium yoelii Sporozoite Transmission by Infected Anopheline Mosquitoes. Infect Immun 73: 4363–4369.
- 70. King JG, Vernick KD, Hillyer JF (2011) Members of the Salivary Gland Surface Protein (SGS) Family Are Major Immunogenic Components of Mosquito Saliva. Journal of Biological Chemistry 286: 40824–40834.