Microsporidia MB is found predominantly associated with Anopheles gambiae s.s and Anopheles coluzzii in Ghana

A vertically transmitted microsporidian, Microsporidia MB, with the ability to disrupt Plasmodium development was reported in Anopheles arabiensis from Kenya, East Africa. To demonstrate its range of incidence, archived DNA samples from 7575 Anopheles mosquitoes collected from Ghana were screened. MB prevalence was observed at 1.8%. An. gambiae s.s constituted 87% of positive mosquitoes while the remaining were from An. coluzzii. Both sibling species had similar positivity rates (24% and 19%; p = 0.42) despite the significantly higher number of An. gambiae s.s analysed (An. gambiae s.s = 487; An. coluzzii = 94; p = 0.0005). The microsporidian was also more prevalent in emerged adults from field-collected larvae than field-caught adults (p < 0.0001) suggestive of an efficient vertical transmission and/or horizontal transfer among larvae. This is the first report of Microsporidia MB in Anopheles mosquitoes in West Africa. It indicates possible widespread among malaria vector species and warrants investigations into the symbiont’s diversity across sub-Saharan Africa.

) and this rate did not differ significantly between the two sibling species (χ 2 = 0.66; p = 0.42).  www.nature.com/scientificreports/ Higher MB detection rate in field-caught larvae than adults. The DNA samples used in this study comprised those extracted from adults that emerged from field-collected larvae and those that were collected from the field as adults. We were interested in establishing whether infections were more prevalent in newly emerged adults (from field-caught larvae) or in the field-caught adults. This was also to assess which of the two life stages would increase the probability of finding MB in a field study aimed at MB discovery. For this analysis, An. arabiensis, An. melas and An. funestus were excluded since they did not show positive infection. Of the remaining 7534 An. gambiae s.l mosquitoes, ~ 90% were field-caught adults. However, among the remaining 10% of samples that were initially collected from the field as larvae, 17.3% recorded positive for MB. This was significantly higher than the prevalence (0.13%) observed in field-caught adults (χ 2 = 1092; p < 0.0001).

Discussion
We make the first report of Microsporidia MB in An. gambiae s.s and An. coluzzii following identification of the symbiont in An. arabiensis. This does not only demonstrate the existence of the microsporidian in another predominant malaria vector species in Africa but also extends its incidence from East to West Africa. The prevalence of MB-positive mosquitoes was estimated to be 1.8%, which is within the rate of < 1-9% reported for An. arabiensis 2 . The present study took advantage of archived mosquito DNA samples which were either collected from the field as larvae or adults from different study sites over 5 years.
Anopheles gambiae s.s and An. coluzzii are the predominant malaria vectors in Ghana 7-9 . They are often found in sympatry with one species usually being more abundant 9,10 . Contrary to the study by Herren and colleagues 2 , a handful of An. arabiensis was analysed in the present study. Anopheles arabiensis is more commonly found in the arid north of Ghana where rainfall is observed within a few months in a year. In studies conducted in Ghana that are focused on Anopheles distribution, between 2-3% are An. arabiensis despite the collection of large numbers of mosquitoes 7,11 . However, we acknowledge that there were more collections from the south of the country, especially Greater Accra, which contributed 40% of the DNA samples used in this study and 89% of MB-positive mosquitoes. Further studies to investigate variations in mosquito species density and seasonal prevalence of Microsporidia MB will shed more light on the field dynamics of the symbiont in these mosquito populations.
DNA samples from mosquitoes initially collected as larvae from the field showed significantly higher MBpositivity than those collected as adults. Microsporidians can be transmitted both vertically and horizontally 12 . The efficiency with which MB is transmitted vertically depends on the intensity in the ovaries of the female parent. Horizontal transfer was initially speculated to occur in the larval habitat 2 which would increase the spread of MB from few infected larvae to many in the breeding site, but this transmission route has recently been shown not to occur in larvae under laboratory conditions 13 . In effect, our detection of higher prevalence among newly emerged adult mosquitoes is more likely to be attributed to highly efficient vertical transmission from female parent to offspring 2 . In the laboratory, the intensity of MB in adult mosquitoes has been observed to increase with age 2 . However, in the wild where conditions are very dynamic, the effect of larval habitat conditions and adult age on MB intensity in mosquitoes may differ. For example, larval diet has been shown to affect the microbial composition of mosquitoes in their adult life stages 14 . It is also established that microbial diversity decreases in female adults mainly because of proliferation of certain bacteria species following sugar and/or blood feeding 15 . It, therefore, makes it challenging to explain variations in bacterial diversity among field-caught adult mosquitoes since their feeding histories are unknown 16 . Several environmental and physiological statuses could potentially affect the intensity of MB observed in adults and variations in infection prevalence could be governed by factors that are yet to be investigated. Given that collecting large numbers of adult mosquitoes in the field may prove more challenging than larval sampling, our results have also shown the increasing chances of finding MB infections in a population when larvae are collected in field studies 2 .
While the data presented here shows basic information about the ecological spread of Microsporidia MB, it has nonetheless demonstrated a potential widespread occurrence of Microsporidia MB among Anopheles mosquitoes across sub-Saharan Africa. It warrants further investigation of the diversity, environmental dynamics, and interactions with other mosquito symbionts for a clearer understanding of their possible use in malaria control.

Methods
Description of Anopheles samples. We retrieved Anopheles mosquito DNA samples from various studies that have been conducted at the Noguchi Memorial Institute for Medical Research, University of Ghana between 2014-2019. Mosquitoes had been collected from different sites across the country (Fig. 1) either as larvae or adults during the rainy season period. When collected as larvae, they were reared to adults for experimental assays before DNA was extracted. Most DNA samples were from single female mosquitoes while few were extracted from pools of 25 (Supplementary Table S1). The species information on the samples were retrieved from the different projects. Where the mosquitoes were only identified as An. gambiae s.l, these were further assessed for their sibling species identification using Restriction Fragment Length Polymorphism (RFLP) 17 or SINE 18 methods when they were positive for Microsporidium MB infection.
Quality check of archived samples. Different extraction methods, including CTAB, Trizol RNA/DNA and columns had been used by the various projects to obtain DNA. It was therefore expected that the DNA integrity would differ among samples. The samples would have also gone through some freeze-thaw cycles which would compromise the quality of the DNA and result in false negatives. To address this concern, we randomly selected samples for DNA quantification with Qubit Fluorometer (Thermofisher, UK) and purity using the BioDrop (Biochrom, UK). Average DNA concentration was 34.8 ng/μL (1.02-204 ng/μ) and average purity was 1.7 (1.14-2.17). To reduce potential contaminants that would limit PCR, DNA samples were diluted 1 in 10.  Fig. S2).
Statistical analyses. Contingency analyses to compare independence observed positivity between mosquito species was performed using a two-sided Chi-squared test with Yate's correction. The distribution of An. gambiae s.s and An. coluzzii from the study sites was tested with a two-tailed unpaired Mann-Whitney nonparametric test. All test significance was accepted a p < 0.05.