Background
Malaria remains a major global public health problem affecting hundreds of millions of people annually, mainly in sub-Saharan Africa. Each year approximately half a million people die, mostly children younger than 5 years [
1]. Gametocytes, the sexual stage of the parasites, are essential for transmission of the parasite from man to mosquito. Malaria transmission can be interrupted by drug treatment affecting gametocytes [
2,
3], causal chemoprophylaxis, vector control as well as the acquisition of immunity to sexual stage parasites by the human host [
4,
5].
Transmission blocking interventions that target gametocyte development and gamete fertilization are considered an essential part of malaria control, especially if containment or eradication of the disease is the aim. Transmission blocking vaccines (TBVs) would have a great public benefit in malaria-endemic countries by breaking the life cycle and decreasing the number of new infections. In addition, it is assumed that TBVs could help in containing the spread of parasites resistant to drugs or malaria vaccine components directed against asexual blood stage or pre-erythrocytic stages [
6,
7]. To better understand immunity against the sexual stage of the plasmodial life cycle and for the design and development of TBVs, profiling the response to mature gametocytes is of relevance.
Antibodies (Abs) are important mediators of sexual stage immunity against
Plasmodium and other apicomplexan parasites [
8‐
12]. Such Abs can affect malaria transmission either by inhibiting gametocyte development [
5] or by directly affecting viability of mature sexual stages [
13‐
15]. The latter might happen within the body or once they are ingested by mosquitoes [
5,
16‐
18], e.g. through opsonization of gametes followed by phagocytosis [
12]. In malaria-endemic areas, the age-dependent decline of the duration of gametocyte carriage [
19,
20] is most likely due to an increase in gametocyte exposure and development of sexual stage specific immune responses, in parallel to the asexual immunity acquired with age [
21].
Indirectly, immune responses to asexual stage antigens may decrease transmission by limiting the number of asexual parasites that develop to gametocytes [
21], similar to the decrease of gametocytogenesis that results from the elimination of asexual infections by drugs [
22]. However, development of sexual-stage immunity is different from the immune response directed to asexual stage antigens [
13,
15]. Gametocytes have distinct gene expression patterns [
23] and proteomic profiles [
24] compared to asexual stages. Similarly early and late stage gametocytes differ; for example, the latter have a comparatively low representation of active export machinery proteins. However, some overlaps are expected in the proteomic profiles and exported proteins between the different stages of the parasite’s life cycle [
24].
Naturally acquired sexual-stage antibodies are produced against gametocyte-infected erythrocyte surface antigens or gamete-specific antigens in the circulation and also against mosquito-stage parasites that act following ingestion of the parasite [
25]. There are only few studies on natural immune responses to gametocyte-infected erythrocyte surface antigens. Saeed et al. [
15] showed that 34% of Gambian children had plasma antibodies recognizing stage V gametocyte-infected erythrocytes in vitro, with no recognition of stages I–IV. In the same study Abs to gametocyte surface antigens were associated with lower gametocyte densities indicating the importance of Abs in reducing gametocyte carriage. Most other studies on immune responses to sexual stage antigens have focused on few specific antigens, mainly the TBV candidates Pfs230 [
18,
26‐
31] and Pfs48/45 [
18,
27‐
32]. The association of Ab response to these single antigens and transmission reducing activity is not consistent. After testing antibody response to both antigens, some authors reported a correlation of transmission reduction with both antigens [
31], while others found associations only with Pfs230 [
18,
28] or only with Pfs48/45 [
29,
30]. Even though correlation might be confounded by exposure history to earlier malaria infections, these results suggest that Ab responses to other gametocyte-specific antigens may play an additional role in controlling transmission [
5]. Here, Ab responses to gametocyte-infected erythrocyte surface antigens were measured in individuals from a malaria-endemic country (Gabon).
In the present study, the concentration of anti-gametocyte Abs against in vitro differentiated mature gametocytes of one
Plasmodium falciparum clinical isolate and one laboratory strain (NF54) was measured by flow cytometric immunofluorescence assay (IFA) in sera from malaria-exposed children and semi-immune adults. Since exposure to asexual blood stage antigens and co-infection with other highly prevalent parasites may modulate immune responses [
33,
34], here the anti-gametocyte responses were related to infection status with intestinal helminths. Assuming a reduced anti-gametocyte antibody response after vaccination with a malarial vaccine, additionally the anti-gametocyte antibody response to antibodies induced by vaccination was related with the asexual blood stage vaccine candidate GMZ2, a recombinant fusion protein of
P. falciparum glutamate-rich protein (GLURP) and merozoite surface protein 3 (MSP3) [
35]. Therefore, the collected serum samples during the GMZ2 trials were used to investigate this in depth.
Discussion
Understanding the development of Ab responses to sexual stages of
P. falciparum in malaria-exposed populations is important for understanding transmission patterns and the design and development of TBVs. In this study in a highly malaria endemic area, adults showed higher Ab responses to sexual stage antigens than children. The effect of age on anti-gametocyte Ab production and transmission reducing activity has been addressed in several studies and resulted in contradictory results. Some have shown an increase in antibody response to fixed whole parasites [
51] and to Pfs48/45 and Pfs230 with age [
46], others a decline in transmission reducing activity of sera with age [
27]. The difference may be due to the varying age ranges, different epidemiological settings and inconsistent assays for anti-gametocyte Ab measurements between the studies. Since effective humoral immunity to intraerythrocytic gametocytes requires prolonged exposure to the parasite [
52], the epidemiological setting shall play an important role and the interpretation of divergent results needs to account for that.
The anti-gametocyte seropositivity rate was 77% in adults and 57% in children. Similar seropositivity rates have been recorded earlier, more than 50% in Ghanaian children with asymptomatic infection [
53] and 34% in Gambian children with uncomplicated malaria and gametocytaemia [
15]. In another cohort of Gambian children, who were gametocytaemic, an Ab response was detected in all participants [
51]. This response might have been so high, as they fixed gametocytes with acetone that simultaneously permeabilizes the cell [
54]. The same study reported a lower (42%) sero-prevalence when live gametes were used. Unfortunately, gametocyte carriage rates were not recorded in our studies. Due to different methodologies, direct comparison to the previously reported results is not possible. Of note, gametocyte prevalence is not always associated to antibody response to gametocytes [
55] and transmission capacity [
56,
57].
The effect of anti-GMZ2 Ab and Ab responses to whole asexual stage antigens and the effect of GMZ2 vaccination on the development of sexual-stage immunity was evaluated. The result showed a significant but weak positive correlation between Ab response to gametocyte antigens and whole asexual stage antigens, but not to GMZ2 antigen in adults, showing that exposure is correlated to Ab response to gametocytes in the investigated population. The level of sexual Ab response was significantly reduced after vaccination in GMZ2 vaccinated adults, but this difference was not seen in the rabies control group. An effective asexual malaria vaccine should lead to a reduced asexual parasite load and, therefore, reduced gametocyte production. The recently reported results of the GMZ2 vaccine shows 11–14% efficacy in children [
38], therefore, the expected effect on gametocytes would be rather low. There is no evidence for the expression of either of the GMZ2 component proteins (MSP3 and GLURP) on the surface of gametocyte infected erythrocytes even though it was previously reported that GLURP is expressed in different stages of the parasite life cycle including the pre-erythrocytic stage [
58]. The results show no boosting of the anti-gametocyte Ab response, supporting the data that none of the components is present at the sexual stage.
In children, there was no correlation between the Ab response to gametocyte antigens and GMZ2 (and other asexual antigens). The difference of the correlation results between adults and children might be due to the very low level of anti-gametocyte immune response in children complicating to see differences between groups. Additionally, Ab response to matured gametocytes did not change in children following vaccination with GMZ2 or the rabies control vaccine. Unfortunately our analysis has been underpowered to detect subtle differences in the study populations. Our result shall be confirmed with a bigger sample size.
As expected, antigen recognition by immune sera was significantly enhanced after fixation and permeabilization of gametocyte-infected erythrocytes as reported earlier [
51]. This effect may be partly due to intracellular targets of transmission-blocking Abs [
52] that are not exposed on the infected erythrocyte surface. However, the response measured after fixation and permeabilization of cells might not be gametocyte specific, but represent a response to the cocktail of internal asexual and sexual stage proteins [
59]. Therefore, non-fixed and non-permeabilized gametocyte-infected erythrocytes were used to analyse the effect of age, vaccination, coinfection, and parasite strain variation on Ab responses to gametocytes. The natural Ab response to gametocyte antigens might inhibit gametocyte development and thereby interrupt the transmission of the parasite as shown in vitro by co-cultivation of early gametocytes with plasma from malaria patients [
5]. In addition, Abs may also act following exflagellation in the mosquito midgut [
60]. To get a deeper insight, characterization of naturally acquired transmission blocking Abs might improve the portfolio of TBVs.
Recognition of
P. falciparum gametocytes of the laboratory strain NF54 and a clinical isolate by serum antibodies of semi-immune adults was assessed. Both lab strain and clinical isolate have been detected by the serum antibodies and no significant difference in the level of Ab recognition of the two strains could be seen. This was unexpected as the clinical isolate was collected from the area where the study participants were recruited. However, similar to the finding presented here, it has been shown earlier by Dinko et al. that the plasma antibodies from Ghanian participants recognizing the laboratory strain 3D7 and a clinical isolate from Kenia [
55].
The immune response to malaria is modulated by co-infection with other infectious agents [
33]. There were lower Ab responses in
Trichuris trichiura infected children when the assay was done with permeabilized gametocytes. However this might not represent the suppression of the immune response to gametocytes but rather the response to the asexual or both stages of the parasite. This corroborates earlier findings in the same population, which showed that
Trichuris trichiura infection is associated with 3.4-fold reduced Ab levels to the blood-stage asexual GMZ2 vaccine antigen while the response was increased in
Ascaris lumbricoides-infected participants [
34]. However, the effect of
Trichuris trichiura infection could not be seen in the assay using non-permeabilized gametocyte-infected erythrocytes. In line with this, a cross-sectional study has reported a significant reduction of gametocyte specific (Pfs48/45) antibody titers in individuals infected with
Schistosoma haematobium, though this effect is not seen when comparing another gametocyte specific antigen (Pfs230) [
61].
The response to any gamete-specific antigens was not measured and therefore activation of non-fixed gametocytes cannot be ruled out. Mature gametocytes should be handled with care and a drop in temperature could provoke exflagellation. Therefore, the small fraction of recognized erythrocytes with gametocytes could also be a small fraction that was accidentally activated. However, we tried to ensure that non-fixed and non-permeabilized erythrocytes in the assay were intact and controlled for exflagellation by microscopy. Therefore, the measured response most probably represents largely the antibody response to antigens displayed on the surface of gametocyte-infected erythrocytes. Another explanation for the low response could be that only a fraction of gametocytes transport proteins to the outside and express surface proteins on the surface of erythrocytes. A reduced transport of proteins to the erythrocyte surface is known for asexual parasites of laboratory strains in long-term culture. To check this, the prevalence of positive cells was tested also for one clinical isolate (JH013), but did not find a difference in the percentage of recognized cells when compared to NF54.
Infection with
Ascaris lumbricoides resulted in an increased anti-gametocyte immune response compared to the uninfected participants. This was surprising but similar effects on the anti-malarial immune response have been observed earlier [
34]. However, due to the relatively small sample size and exploratory nature of the experiment, confirmatory studies will be required. Therefore, monitoring of anti-gametocyte responses should be done in further larger studies. In addition, it would be interesting to validate the immuno-modulatory effect of co-infections on the development of asexual and sexual stage immune responses and the transmissibility of malaria in other co-endemic areas.
Authors’ contributions
The study was conceived and experiments designed by TG AA BM JH. TG performed the experiments and TG AA BM JH analysed the data. TG AA MT ME UAN SI AAA PGK BM JH contributed reagents/materials/analysis tools and wrote the paper. All authors read and approved the final manuscript.