First report of natural infection of Anopheles gambiae s.s. and Anopheles coluzzii by Wolbachia and Microsporidia in Benin: a cross-sectional study

Malaria is an infectious disease caused by a parasite of the Plasmodium genus with half of the global population at risk of this disease. In 2020, there were globally 247 million cases, and 619,000 deaths due to malaria [1]. Sub-Saharan Africa, where Plasmodium falciparum remains the most prevalent malaria parasite, bears the greatest global burden of disease [2]. The cornerstones of malaria vector control have been long-lasting insecticidal nets (LLINs) and indoor residual spraying (IRS) and have averted 1.5 billion malaria cases and 7.6 million malaria deaths, with these interventions accounting for 68% and 10% of these achievements, respectively [3]. With the scale-up of these interventions, the disease burden in Africa is expected to be significantly reduced by 2030. However, widespread insecticide resistance [4] and changes in vector behavior [5] may sabotage elimination in the upcoming decades. For that, biological control tools such as exploitation of Wolbachia, Spiroplasma, and Microsporidia endosymbionts that can be used alone or in combination with insecticide based-tools, have been developed to improve the control of vector-borne diseases, including malaria [6‐9]. These bacteria have a large array of interactions including mutualism, commensalism, and parasitism within their hosts [10]. Wolbachia can colonize certain mosquito populations, and impact pathogen development, thereby reducing their infection and transmission potential [7, 8, 11]. Laboratory experiments have shown an absence of dengue virus infection in populations of Aedes aegypti artificially infected with Wolbachia [7, 12]. In addition, other laboratory trials showed that some Wolbachia strains impede infection of Anopheles vectors with Plasmodium species [13‐16], making it an alternative option for malaria control. However, evidence of an impact of Wolbachia infection on malaria transmission at the community level is still scarce [13, 14]. It has long been assumed that Wolbachia is absent from natural populations of Anopheles [17]. It is only recently that studies have reported that Anopheles gambiae sensu stricto (s.s.), Anopheles coluzzii and Anopheles arabiensis can be found naturally infected by Wolbachia in Burkina Faso and Mali [18‐20] and Anopheles moucheti and Anopheles demeilloni have been reported infected by Wolbachia in Cameroon, Kenya and the Democratic Republic of the Congo, with evidence of the capacity to induce cytoplasmic incompatibility [15]. Negative correlations between the presence of Wolbachia and development of Plasmodium has been demonstrated in An. gambiae in Mali and An. coluzzii in Burkina Faso [20, 21]. This supports the need for developing new vector control tools based on Wolbachia-Anopheles interactions.

The first report of Microsporidia in An. arabiensis was in Kenya, where Microsporidia infected mosquitoes were unable to be infected with P. falciparum [22]. The presence of this endosymbiont in wild vector populations, warrants screening for it in other endemic regions in Africa.

To progress the development of endosymbiont-based malaria control tools, it is important to continue identifying, and characterizing the native range of endosymbiont-infected Anopheles vector populations. The present study conducted in Southern Benin aims to identify the natural presence of Wolbachia and Microsporidia in Anopheles gambiae s.l., the main malaria vector in this region.

Given that the efficacy of insecticide-based control tools are under threat because of the emergence of resistance, there is a growing interest in the use of alternative, effective biological vector control strategies. For that, the search for natural endosymbiont-Anopheles systems capable of reducing vector competence has become essential. The present study is the first that reports the presence of Wolbachia and Microsporidia in both An. gambiae s.s. and An. coluzzii in Benin.

A trial recently conducted in Kenya showed that Microsporidia, a vertically transmitted bacteria was capable of disrupting Plasmodium development in An. arabiensis [22]. Moreover, it has been demonstrated that some mosquitoes can have their longevity reduced by Wolbachia, which prevents the completion of the life cycle of some infectious pathogens, thereby interrupting transmission [28]. Findings of the present study shows the natural presence of Microsporidia and Wolbachia in the microbiota of An. gambiae s.l. in Benin. These results confirm those of Gomes et al. [21] and Dada et al. [29] that demonstrated the ability of An. gambiae s.l. to host Wolbachia and Microsporidia.

In the present study, the infection rates (5.1% in An. gambiae s.s. and 1.3% in An. coluzzii) to Wolbachia was overall lower compared to those observed in Burkina-Faso (46% in An. coluzzii and 33% in An. arabiensis). The same trend was observed at the complex level (An. gambiae s.l.), with infection rates ranging between 46 and 78%, depending on the study site in Mali [21]. The general low infection prevalence of Wolbachia in the study area could be due to low density levels that were difficult to detect by PCR or reflect the insensitivity of the end-point PCR technique used. In a previous study in Mali, nested PCR failed to identify 21.7% of infected An. gambiae s.l. samples infected with Wolbachia wAnga-Mali with poor concordance between technical replicates, suggesting that Wolbachia levels were close to the limit of detection of these assays [21]. qPCR methodologies, recently developed for Wolbachia Anga, may have improved detection levels [21]; however, were not feasible with the limited laboratory resources. Thus, the null infection rate to Wolbachia observed in some study communes should not necessarily be interpreted as an absence of this endosymbiont. Taken together these results suggest that natural infection of An. gambiae s.l. to Wolbachia is highly variable across sites in Africa. A similar result was observed in China where the prevalence of Wolbachia natural infection was highly variable in field-collected mosquitoes (Aedes albopictus, Anopheles sinensis, Armigeres subalbatus, Cx pipiens, and Culex tritaeniorhynchus) collected across 25 surveyed provinces [30]. Moreover, Wolbachia natural infection could also be highly variable in various Anopheles species as previously reported in Gabon, Central Africa [31]. Of note, there is a huge diversity of Wolbachia strains with different effects in nature [19].

The deployment of a Wolbachia-based control tool for controlling mosquito borne diseases through the production of sterile insects or pathogen blocking, requires the inducement of cytoplasmic incompatibility to drive the bacterium into natural arthropod populations [32]. While Wolbachia has been shown to impact P. falciparum development, previous works revealed that Wolbachia detected in the present study do not confer cytoplasmic incompatibility [20] and, therefore, would not be feasible to use for control purposes.

The findings show a strong presence of Microsporidia in both An. gambiae s.s. and An. coluzzii, with a mean infection rate of 53.4%. This corroborates previous findings from Akorli et al. [33] who demonstrated that Microsporidia was highly associated with An. gambiae s.s. and An. coluzzii in Ghana. Also, a higher respective, albeit non-significant infection rate to Microsporidia was observed in An. coluzzii than in An. gambiae both in the present trial (57% vs 41%) and the one (80.7% vs 76.0%) of Akorli et al. [34]. Thus, one aspect worth investigating in future trials would be whether An. coluzzii is more susceptible to infection with Microsporidia, compared to An. gambiae s.s.

Though the present trial is a cross-sectional one, it is worth mentioning that investigating the dynamics or variations in bacterial diversity in field-collected adult populations of An. gambiae s.l. is challenging, as bacterial diversity is strongly influenced by several factors such as seasonality, locality-dependent acquisition of environmental microbes [34], diet at larval stage [35], sugar/blood feeding, mating [36], and other factors likely not yet studied.

Overall, in both HLCs and PSCs, the most frequent mosquito species collected were Culex spp, and Mansonia spp, followed by Anopheles spp, and Aedes spp. The same trend was previously observed in Cove, Ouinhi and Zangnanando communes located 156 km away from Cotonou, the economic capital of Benin [37]. Molecular species identification revealed the presence of a mixture of An. coluzzii and An. gambiae s.s. which is consistent with findings from several other previous trials conducted in Southern Benin [21, 35, 36]. Overall, the SR was similar in An. gambiae s.s. and An. coluzzii, which corroborates previous findings from Akogbeto et al. [38] in Northern Benin. However, given the P. falciparum infection rate was assessed at the mosquito level, while the infection rate to each endosymbiont was evaluated at the pool level, it was not possible to assess the influence of the presence of each endosymbiont on the Plasmodium sporozoite infection, which is a limitation for the study. Failure to carry out phylogenetic analyses in order to identify relationships between Microsporidia and Wolbachia detected in An. gambiae s.l. from Benin and those observed in other regions in Africa also constitutes another drawback of this study.

The present study is the first to report the natural presence of both Wolbachia and Microsporidia in natural populations of An. gambiae in Benin. Sustained efforts should be made to widen the spectrum of bacteria identified in mosquitoes, with the potential to develop endosymbiont-based control tools; such interventions could be the game-changer in the control of malaria and arboviral disease transmission.