Evaluation of Pathogenicity of Entomopathogenic Oomycetes Lagenidium giganteum and L. ajelloi against Anopheles Mosquito Larvae

Malaria is a mosquito-borne life-threatening parasitic disease of humans and the leading cause of mortality and morbidity in sub-Saharan Africa. Despite the major eforts made towards malaria control, it is facing challenges of development of parasite resistance towards antimalarial drugs coupled with Anopheles vector resistance towards insecticides being used in control. Tere is, therefore, a need to develop complementary control strategies that are economical and environmentally friendly. Biological control using entomopathogenic fungi against the immature malaria mosquito vectors presents an untapped opportunity. Tis study sought to isolate and characterize entomopathogenic oomycetes Lagenidium giganteum and L. ajelloi from wild Anopheles larvae from Ahero rice felds in western Kenya and test their pathogenicity against laboratory-reared Anopheles gambiae larvae. Laboratory-reared A. gambiae larvae (3 rd and 4 th instar) were exposed to fve diferent concentrations of L. giganteum and L. ajelloi zoospores; 1000, 2000, 3000, 4000, and 5000 zoospores/mL, respectively. Te larval mortality was recorded after 24, 48, 72, and 96 hours post-exposure, until all larvae were dead. Te results obtained showed that L. giganteum was not pathogenic to A. gambiae larvae after 24 and 48hours post-exposure to all concentrations. Larval mortality was recorded at 72 and 96 hours. Tere were no signifcant diferences observed in the mortalities ( p > 0 . 05 ) from all treatments. No mortalities were observed in deionized water (negative control) whereas 100% mortality was recorded in larvae exposed to Bti (positive control). Probit analysis showed that LC 50 after 72hours and 96hours was 2.32 × 10 4 and 3.51 × 10 3 zoospores/ml, respectively. L. ajelloi caused larval mortalities at all the 5 test concentrations after 24, 48, 72-and 96-hours post-exposure with LC 50 values of 1.18 × 10 5 , 1.43 × 10 4 , and 6.05 × 10 2 , and 27.08 zoospores/ml, respectively. Tis study isolated and tested two species of Lagenidium from feld collected larvae . Lagenidium ajelloi recorded greater pathogenicity than that of L. giganteum against A. gambiae larvae, making them potential candidates for use in the development of bio-larvicide for the control of Anopheles larvae.


Introduction
Malaria remains the leading cause of mortality and morbidity in sub-Saharan Africa, where Plasmodium falciparum is the main malaria parasite. Te World Malaria Report indicates that approximately, 241 million cases of malaria were reported in 2020 in 85 malaria endemic countries. Te WHO African region accounts for 95% (228 million) of all reported malaria cases. An increase in malaria cases was recorded compared with 227 million cases reported in 2019, which is attributed to the disruption of services due to the COVID-19 pandemic [1]. In addition, there is increasing evidence of the emergence of partial resistance towards artemisinin-based combination therapies (ACTs), the main chemotherapeutic intervention for malaria in the African Region [2]. Consequently, WHO encourages an integrated approach towards vector-borne disease control, calling on countries to prioritize research on the exploration and development of novel mosquito control strategies [3]. Te current study was aimed at exploring alternative biological interventions for malaria vector control using fungi.
Sub-Saharan Africa continues to record high malaria infections in the world mainly due to the presence of efcient vectors of the Anopheles gambiae and Anopheles funestus complexes, favorable weather (hot and humid) that allows the transmission to occur all year round, scarce resources and social-economic instability that have hindered efcient malaria control [4]. In western Kenya, malaria is hyperendemic, with the perennial transmission of parasites by Anopheles mosquitoes. Both Anopheles gambiae and A. funestus complexes are present at Ahero, with A. arabiensis being the predominant species [5,6]. Although considerable gains have been made towards malaria elimination in Africa, the development of the recent transformative technologies such as vaccines and gene-drive mosquitoes may soon quicken malaria control [7]; however, total eradication of the disease faces challenges [4]. For a long time, the control of malaria in most parts of the world and indeed in Kenya has been based on case management and vector control using long-lasting insecticidal nets (LLINs) and indoor residual spraying (IRS) [7,8]. However, the use of chemical pesticides has led to the increasing development of insecticide resistance of the malaria vectors to pyrethroid-based insecticides, hence compromising the efectiveness of insecticide-based vector control programs [9][10][11]. Besides, the persistent use of chemical pesticides contaminates the environment and also poses a health risk to nontarget organisms [12], and the need for ecologically safe alternatives is clear [13]. To counteract these challenges, integrated approach to malaria vector control by use of both chemical and nonchemical methods including biological control and environmental management has been proposed [14]. Integrated approach strategies work by combining diferent control methods that target either adult mosquitoes or/and the immature aquatic stages of mosquitoes.
Immature stages of mosquito control mainly follow the WHO recommendation on larval source management, which includes habitat modifcation, habitat manipulation, larviciding, and biological control [15]. Controlling mosquitoes by targeting their immature stages is convenient since they are confned in their aquatic habitat, where they have a relatively limited mobility range, and therefore controlling them is far easier than controlling the adult stage [16]. Biological control agents that can be used to control mosquitoes and other insect pests include plants, fsh, nematodes, viral, bacterial, protozoan, and fungal pathogens [16,17]. Of these, entomopathogenic fungi are the most well-suited for development as biopesticides against vector mosquitoes [18,19]. Fungal biopesticides are easy to deliver, and they provide fexibility in improving their formulations and hence being able to target and control a wide range of insects with diferent feeding habits [20]. Entomopathogenic fungi are pathogens of insects or they cause some level of harm within insects [21,22]. Eighty-fve genera and over 750 species of entomopathogenic fungi are known. Amongst these, two genera, Beauveria and Metarhizium, which belong to the class Hyphomycetes in Deuteromycota are being used as an ingredient for "myco-insecticides" or "mycoacaricides" [23]. Control of mosquito larvae supplements adult mosquito control strategies, and it is advantageous since larvae are confned in their habitats, unlike adult mosquitoes that fy, making their control difcult [24,25].
Entomopathogenic fungi have shown natural variations in their prevalence and virulence against mosquitoes. It is, therefore, advisable to carry out pathogenicity tests of different isolates of the same fungus from diferent geographical locations to ascertain their susceptibility to mosquito larvae obtained from the habitat of interest.
Te genus Lagenidium is a group of aquatic fungi that belong to the class Oomycota and whose members are either saprophytes or parasites [26]. Only one species of this group, Lagenidium giganteum has been shown to be parasitic to mosquito larvae [27]. It was frst described by Couch [28] from a combined collection of copepods and mosquito larvae (Culex and Anopheles) in North Carolina, USA. Te infective stage is the motile spore (zoospores) which attacks and destroys mosquito larvae in water. Depending upon temperature and zoospore density, the larvae die of starvation within 1-4 days [29]. Lagenidium ajelloi is a water mold, and although it has previously been isolated in mammals in the past, little is known about its entomopathogenic potential against mosquito larvae and its interaction with other organisms, except for its morphology [30]. Te current study therefore aimed to isolate L. giganteum and L. ajelloi from wild Anopheles larvae from Ahero rice felds in western Kenya and to test their potency against laboratory-reared Anopheles gambiae larvae.

Study Area.
Te study was carried out at the Ahero rice irrigation scheme (34.9°E, −0.17°S), located in Kisumu County along the shores of Lake Victoria in western Kenya. Te region receives long and short rains between the months of March to May and September to December, respectively, with annual variability. River Nyando passes through Ahero and provides water to the local population for irrigation, with the main cash crops being rice and sugar cane [10]. Due to the presence of water, Ahero provides favorable breeding grounds for mosquitoes throughout the year. Te main malaria vectors in Ahero belong to A. gambiae and A. funestus complexes. Both A. arabiensis and A. gambiae sensu stricto occur sympatrically in the Ahero region [31].

Mosquito Larvae Sampling. A sampling of wild
Anopheles mosquito larvae was carried out in July, 2018, from 10 randomly selected irrigation canals within a radius of 1 km. Anopheles larvae were distinguished from Culex by observing their resting positions on the water surface. Anopheles larvae were sampled and put in plastic containers and then transported to the laboratory at the Technical University of Kenya in Nairobi, where fungal isolation and characterization were carried out.

Fungal Isolation.
In the laboratory, the dead larvae were surface sterilized by dipping them in a 2% sodium hypochlorite solution for two minutes to kill any external microbial contamination. Te larvae were then introduced into potato dextrose agar (PDA) media for fungal isolation. Gentamycin was added as a bacteriostatic agent in the media to prevent bacterial contamination. Te larvae were inoculated into the media with a sterile pair of forceps. To ensure complete fungal isolation, twenty replicates were carried out with each plate containing 10 larvae. A negative control was set with 3 plates in which no larvae were inoculated. Te plates were placed in the incubator set at 24 ± 2°C for fungal cultivation for 14 days, with observations carried out, starting from the 3 rd day for fungal growth and colony characteristics. On the eighth day, a fungal colony survey was carried out and subculturing carried out on Sabouraud dextrose agar (SDA) plates supplemented with gentamycin to obtain pure cultures.

Fungus Identifcation.
Fungal species were then identifed using both macroscopic characteristics of colonies and microscopic observation of morphological features of mycelia, hyphae, and spores. Microscopic identifcation was performed under objectives X40 and X100 using an OPTIKA B-380 compound microscope manufactured by OPTIKA Srl in Italy at Ponteranica, and images were taken using an OPTIKA 4083.B2 digital camera attached to the B-380 microscope. Microscopic slide preparations were performed using a modifed Riddell's slide culture technique to enable viewing of the specimen in its live state with minimal distortion to morphological features [32]. Existing published work and taxonomic keys [28,30,33] were used in the identifcation of the fungi. Only Lagenidium species were selected for further tests after carrying out molecular analysis to ascertain their identities.

DNA Extraction and Sequencing.
Sabouraud dextrose broth was used to inoculate the 2 fungal strains in a 250 mL fask. Incubation was then performed for 72 h at 37°C on a rotating shaker at 150 rpm. Merthiolate (0.02%, wt/vol) was used to kill the cultures after incubation before fltration to obtain the hyphal cell mass for DNA extraction. Te hyphal cell mass was grounded using a pestle and mortar in the presence of the CTAB bufer. Te mixture was subjected to sodium dodecyl and proteinase K treatment and incubated for 1 h at 60°C before adding CTAB/NaCl (heated at 65°C) and incubating at 10 min. Te total DNA was extracted from the solution with phenol: isoamyl alcohol (25 : 24 : 1). ITS gene was amplifed by hot start PCR protocol using primers for Lagenidium strains as described by Vilela et al. [34]. ITS is recognized as a fungal barcode because it is the most sequenced region of fungi and is routinely used for systematics, phylogenetics, and identifcation [35]. Te PCR products were analyzed in 2% agarose gel and purifed using Exosap-it ® (Afymetrix, Santa Clara, CA), as per the manufacturer's instructions. Sequencing was performed using BigDye ® Terminator v3.1 Cycle Sequencing Kit reaction mix (Applied Biosystems, USA) using a 3500xL sequencer.

Phylogenetic Analysis.
Te raw sequence data in the form of chromatograms from both forward and reverse primers were edited manually to remove poor quality reads. Te edited reads were then exported to BioEdit software version 7.0.5.3 [36], aligned using MAFFT alignment software [37] and assembled into contigs after resolving potential ambiguities by eye. Multiple sequence alignment is important in the identifcation of gaps, matches, and mismatched nucleotides in the three genes, which is a prerequisite for the construction of phylogenetic trees. Te two ITS genes from L. giganteum and L. ajelloi were analyzed independently against twenty-three lagenidium species' sequences obtained from the National Centre for Biotechnology Information (NCBI), of which four sequences were from L. giganteum and the other four were from L. ajelloi. Tis led to the construction of a phylogenetic tree showing the evolutionary relationship of both L. giganteum and L. ajelloi with other Lagenidium species obtained from NCBI. Phylogeny construction was carried out in PhyloSuite [38]; the models assumed the rates and patterns of substitution were uniform among the four nucleotide sites. Te evolutionary history was inferred by using the maximum likelihood method and Kimura 2-parameter model [39]. Te test for phylogeny was bootstrap resampling [40], where the number of bootstrap replications was set at 5000 for the ML tree constructed.

Anopheles gambiae Larvae Colony. Anopheles gambiae
colony was established at the University of Nairobi insectary with eggs obtained from the International Centre of Insect Physiology and Ecology (ICIPE), Mbita point, western Kenya. As per the standard operating procedures (SOPS) of the University of Nairobi insectary, environmental conditions in the insectary were set at temperatures of 32°C and relative humidity of 70-80%. Te adults were kept separately at room temperatures of 26°C-28°C. From this colony, 3 rd and 4 th instars of A. gambiae larvae were picked and used for the bioassay according to WHO recommendations [15]. Because both A. gambiae and A. arabiensis occur in the study area, A. gambiae sensu stricto was selected for laboratory bioassays as it was readily available.

Pathogenicity Tests on Mosquito
Larvae. Pure cultures of the two Lagenidium species were removed from the Petri dishes in storage at 4°C and each blended together with 100 mL of deionized water, then kept for 12 hours to initiate zoospore production. For each fungus, the number of zoospores in an aliquot of 10 microliters of this stock solution was counted by using a hemocytometer, and the number of spores per mL was determined. Ten, concentrations of 1000, 2000, 3000, 4000, and 5000 zoospores/mL were calculated using the formula C 1 V 1 � C 2 V 2 . Tese were then used as the concentrations upon which the late instars of Anopheles mosquitos were tested. Tis was informed by Shathele [41], who found that a concentration of 2000 L. giganteum zoospores/ml caused a 100% kill of 1 st instar of Culex quinquefasciatus. Each sample was added to a 500 mL plastic container flled with 100 mL deionized Psyche: A Journal of Entomology 3 water in which 25 late instar A. gambiae larvae were introduced, with each concentration having 4 replicates [15]. For each concentration, deionized water was used as a negative control where no fungus was added. Bacillus thuringiensis var israelensis (Bti) at a concentration of 0.21 mg/L was used as a positive control [42]. Te larval mortality was recorded after 24, 48, and 72 hours, and any dead larvae were recorded [43].
2.9. Data Analysis. Data were subjected to Finney Probit analysis [44] using Biostat 2009 to determine the concentration of the fungal zoospores required to kill 50% of the larvae (LC 50 ) and the time taken by a specifc concentration of zoospores to kill 50% of larvae (LT 50 ). Te data were also subjected to ANOVA to determine any signifcant diferences between the treatments and the positive control. Comparisons of mortalities caused by the two fungal species were performed by ANOVA using the Prism software package.

Results
Based on the observation of macroscopic and microscopic morphological features obtained by DNA analysis, 2 species of Lagenidium (L. ajelloi and L. giganteum) were identifed.

Morphological Characteristics of Lagenidium giganteum.
L. giganteum colonies were circular, velvety, and orangewhite in colour on an agar plate and were about 1.5 cm in diameter on the 3 rd day of growth (Figure 1(a)). Te hyphae were aseptate at frst but later became septate, hyaline, and constricted at the septum and had almost parallel branching to form mycelium with a whitish gleam evident in the protoplasm (Figure 1(b)) at ×40 magnifcation. Zoosporangiophores appeared and formed zoospores that were subglobose, colorless, thick-walled, and formed inside sporangia (Figure 1(c)) as observed under a microscope at ×100 magnifcation. Zoospores were pale black, laterally biciliate, and swum sluggishly. Homothallic sexual reproduction was observed (Figure 1(d)) to form a spherical zygote with a subcentral sphere of fat surrounded by cytoplasm (Figure 1(e)). Both structures were observed under a compound microscope at ×100 magnifcation. Finally, the hyphae cellulose changed to purple on staining with chloro-zinc iodide (Figure 1(f )). Tis observation was in agreement with the fndings reported by Couch [28].

Morphological Characteristics of Lagenidium ajelloi.
Macroscopically, the colony was irregular, 4.5 cm in diameter on the 7 th day, and yellow in colour (Figure 2(a)). Microscopically, the thallus was flamentous with strong restrictions at the septa forming long detached segments. Hyphae were septate and hyaline, with protoplasm with pale whitish gleam and with conspicuous shiny globules (Figure 2(b)). Exit tubes were present and were short and extended through hyphal wall superfcially. Zoosporangiophores were club-shaped with sporangia at the tip, and a conspicuous shiny globule was present at the foot, while some were constricted at the center with pale black coaxial fabric running in the middle. Zoosporangiophores contained two sporangia in most cases, which were lobed and irregular and sometimes ovoid in shape. Zoospores were pale black to golden yellow, laterally bifagellate, swum aggressively in circular paths and rolled in gyrates, and after sometime, formed cysts that were pointed at the anterior and rounded at the posterior end (Figure 2(c)). Homothallic sexual reproduction was characterized by antheridia moving horizontally along the hyphae to meet archegonia and hence formed pyriform zygote with subcentral male and female gametes. Tese observations were in agreement with the fndings reported by Mendoza et al. [30].

Phylogenetic Analysis.
Molecular phylogeny was performed to confrm the identity of the two species already identifed from morphological characteristics. Initial trees for the ML heuristic search were attained through the application of Neighbor-Join and BioNeighbor-Join algorithms to pairwise distances matrix generated from Lagenidium sequences. Te fnal tree selected was the tree topology with a superior log likelihood value. Bootstrap resampling was selected as a test of phylogeny, and it measured the reliability of the tree generated [40]. In addition, NCBI Blast search was also conducted using L. ajelloi as query sequence (13a) and yielded 100% identity of organisms with GenBank accession numbers KJ506135, KJ506135, and KJ506135 that correspond to L. ajelloi. Tis is in line with the ML phylogeny obtained, where all the fve sequences from L. ajelloi species are grouped together, forming a clade that is strongly supported by a bootstrap value of 100, implying the query sequences (13a) was indeed L. ajelloi (Figure 3). Similarly, 100% sequence identity was observed in KT257472, AY151183, KY965927, and KT257343 which correspond to L. giganteum when blast search was carried out using L. giganteum (11) as the query sequence. Identity of L. giganteum has also confrmed from the ML phylogeny obtained where all the fve L. giganteum species have been grouped together in a clade which is supported by a strong bootstrap support value of 97 ( Figure 3). Hence, the ITS genes from the two species strongly indicated that the species were indeed L. ajelloi and L. giganteum as previously identifed from the morphological characteristics.

Pathogenicity Tests.
It was observed that L. ajelloi was more toxic against Anopheles gambiae larvae than L. giganteum. No larval mortalities were recorded due to L. giganteum zoospores at 24 and 48 hours post-exposure to all 5 zoospores concentrations tested. However, larval mortalities were recorded for L. ajelloi after 24, 48, 72, and 96 hours post-exposure in all the 5 test concentrations. For both fungi, it was observed that the larval mortalities increased with exposure time (Figure 4). Te LC 50 of L. giganteum zoospores against A. gambiae larvae after 72 and 96 hours was 2.32 × 10 4 and 3.51 × 10 3 zoospores/mL, respectively, while that of L. ajelloi zoospores after 24, 48, 72, and 96 hours was 1.18 × 10 5 , 1.43 × 10 4 , 6.05 × 10 2 , and 27.08 zoospores/ml, respectively (Table 1). No mortalities were recorded in deionized water (negative control), while 100% mortality was recorded in larvae exposed to Bti (positive control) within 24 hours. It was observed that the longest LT 50 was in larvae exposed to 1000 zoospores/mL concentration for both oomycetes, which was 105 hours for L. giganteum and 62.58 hours for L. ajelloi. On the other hand, the shortest LT 50 was observed under 5000 zoospores/mL, which was 89 hours for L. giganteum and 58.19 hours for L. ajelloi, respectively (Table 2). Te results indicate that the increase in zoospore concentrations was directly proportional to larval mortality for both fungi, with L. ajelloi causing signifcantly more larval mortalities than L. giganteum ( Figure 4). However, the larvicidal activity of both L. giganteum and L. ajelloi was signifcantly slower than the positive control (Bti) in all test zoospore concentrations ( Figure 5).

Discussion and Conclusion
Te results of this study indicate that Anopheles mosquito larvae from the Ahero rice irrigation scheme in Western Kenya are naturally parasitized by Lagenidium giganteum and L. ajelloi. It was evident that both species are infective to the larvae of A. gambiae and the pathogenicity was observed to increase proportionately with an increase in the concentration of zoospores. L. ajelloi was found to be more pathogenic than L. giganteum to A. gambiae larvae under all concentrations of zoospores tested. Pathogenicity due to L. ajelloi and that of Bti were found to be similar, whereas signifcant diferences were recorded between pathogenicity of L. ajelloi and that of L. giganteum. As a facultative parasite, L. giganteum can grow vegetatively both as a pathogen on mosquito larvae or as a saprophyte in aquatic environments [45][46][47]. Isolation of L. giganteum from Anopheles larvae has previously been reported by studies performed in the USA [26,28,33]. Psyche: A Journal of Entomology 5 Couch [28] was the frst scientist to discover the fungus in copepods (Daphnia sp.) and a mixture of Culex and Anopheles larvae in North Carolina, USA. Tereafter, Umphlett [48] and Kerwin et al. [49], both in the USA, isolated and tested the pathogenicity of the fungus on several species of Culex mosquitoes. Umphlett found L. giganteum to be toxic against larvae of Culex restuans and Culex quadrimaculatus [48], while Kerwin et al. found L. giganteum toxic against larvae of Culex quinquefasciatus [49]. Te highest concentration (5000 zoospores/mL) of L. giganteum zoospores tested against A. gambiae larvae killed 68% of the exposed larvae in 96 hours. However, these results difered slightly from those of Golker et al. [50], who found that 56% of A. gambiae larvae exposed to L. giganteum zoospores were protected from death by the larval immune defense. Both fndings support the fact that the pathogenicity of L. giganteum to A. gambiae increases with zoospore concentration and that diferent strains of the fungus may produce diferent virulence and pathogenicity [51]. Te weak pathogenicity of L. giganteum zoospores observed in the study can also be attributed to their inability to sometimes recognize late instars of otherwise susceptible mosquito larvae [52]. However, the fndings of this study may not authentically suggest that L. giganteum zoospores have weak pathogenicity against A. gambiae larvae since laboratory fndings may not necessarily always refect the ideal feld situations due to a wide range of diferences in environmental factors [53]. Although little previous entomopathogenic research work has been carried out on L. ajelloi, it was found to be more pathogenic with the highest tested concentration (5000 zoospores/mL) killing 94% larvae in 96 hours. It was observed that L. ajelloi zoospores swim faster than those of L. giganteum, which agreed with the fndings by Mendoza et al. [30]. Zoospores that swim faster could imply many come into contact with larvae, consequently resulting in more infections and pathogenicity as compared to slow swimming zoospores.
Previous works report that Anopheline mosquito larvae have been found to produce immunity characterized by melanization response, resulting in the encapsulation of foreign particles and parasites with a dark layer of melanin    Psyche: A Journal of Entomology [54]. Te larval immune melanization starts at the cuticle before moving on to the hemocoel; hence, some larvae that mount intense cuticle defense escape infection. Further studies by Golker et al. [50] revealed that some A. gambiae larvae that do not mount enough cuticle defense get infected but may also escape death. Tis is because after the fungi enter the hemocoel, there is complete melanin encapsulation of the fungal hyphae rendering them unable to exchange both gases and metabolites leading to death of the invading oomycete due to asphyxia or starvation. However, the present study has shown L. ajelloi to be more potent against the larvae of A. gambiae causing signifcant larval mortalities   that are closely comparable to Bti with the only diference in exposure time. Tese results may suggest that melanization by L. ajelloi was more intense as compared to that caused by L. giganteum, indicating a diference in the number of infecting zoospores and hence pathogenicity. Faster swimming L. ajelloi zoospores could result in more infections than the sluggish swimming L. giganteum zoospores. Infection by many zoospores would then lead to overwhelming of the immune defense regardless of melanin development.
Te delayed larval mortalities due to L. giganteum and L. ajelloi infections, as compared to mortalities due to Bti could be attributed to the mode of infection, and action between the two treatments. Infection is faster with Bti since the larvae directly ingest the bacteria in water as opposed to Lagenidium infection where the zoospores take time to recognize (by chemotaxis) and penetrate the larval cuticle [26]. Secondly, the larvae die within hours after ingestion of Bti due to protein toxins produced by the bacteria inside the gut of the larvae [25]. On the other hand, larval death due to Lagenidium infection is slower since the fungi have to penetrate and proliferate within the larvae [26], which takes longer than pathogenicity due to Bti infection.
Unlike viruses and bacteria parasites that need to be ingested by the mosquito larvae in order to cause pathogenicity [55], infection due to Lagenidium is advantageous since after killing the mosquito larvae; more zoospores are produced in the cadavers resulting in the recycling of the fungus in the water and hence being able to infect current and subsequent generations of mosquito larvae [46]. Besides, due to the specifcity of Lagenidium zoospores to mosquito larvae, it is expected to have minimal to no risk to natural enemies of mosquito larvae [28].
In general, the fndings of this study are in tandem with earlier fndings that L. giganteum is a facultative entomopathogenic oomycete of mosquito larvae [28,43,49]. Te fndings of this study support the fndings of earlier studies that susceptibility of mosquito larvae to L. giganteum is variable and depends upon the mosquito species [50] and also upon diferent isolates of the fungus from diferent geographical locations [56]. Te successful discovery and isolation of L. ajelloi as a mosquito larval pathogen is encouraging and brings on board a new candidate for further evaluation. Tere is a need for further research on molecular characterization to ascertain the possibility of having several strains of L. giganteum and L. ajelloi in mosquito-infested water from diferent parts of the country. Perhaps, diferent strains of the oomycete would present diferent toxicity levels against A. gambiae larvae [51]. Despite the promising opportunities and advantages associated with entomopathogenic fungi, there has been little work on the use of fungal biopesticides against malaria vectors. Tere is also a need to explore possibilities of producing fungus-based biocontrol agents that will complement existing adult-based strategies and also slow mosquito resistance development [18,19].

Data Availability
Te data used to support the fndings of this study are available from the corresponding author upon request.

Conflicts of Interest
Te authors declare that they have no conficts of interest.

Authors' Contributions
SSI and VWW conceived the study. SSI provided expertise with feld work and sample collection. MMS did sample collection and laboratory experiments and prepared the frst manuscript draft. VWW provided expertise on laboratory trials. FMM did fungal characterization and statistical analysis. SSI, VWW, and FMM reviewed and contributed to the fnal manuscript. All authors have read and approved the fnal manuscript.