Natural Pathogenic Fungi Infection Attributes of Malarial Vectors Anopheles Maculipennis S. L. and Anopheles Superpictus S. L. in Central Iran

Background: Due to the effect of synthetic and commercial insecticides on non-target organisms and the resistance of mosquitoes, in recent years non-chemical and environmentally friendly methods have become prevalent. In this research, we aimed to isolate entomopathogenic fungi with toxic effects on mosquitoes in natural larval habitats. Methods: Larval mosquitos were collected from Central, Qamsar, Niasar and Barzok Districts in Kashan County, Central Iran by standard dipping method, from April to late December 2019. Dead larvae and larvae with signs of infection to fungal mycelium detectable on the outer surface of its body were isolated from the rest of the larvae and were sterilized with 10% sodium hypochlorite for two minutes, then washed twice with distilled water and transferred to PDA (potato-dextrose-agar) and WA (water-agar) media and incubated at a temperature of 25 ± 2° C for 3 - 4 days. Larvae and fungi were identied morphologically based on identication keys. Results: A total of 9789 larvae were collected from urban and rural areas in Kashan County. The genera i.e. Anopheles (7.89%), Culiseta (17.42%), and Culex (74.69%) including 13 species were identied. A total of 105 larvae including Anopheles superpictus s. l., An. maculipennis s. l., Culex deserticola, Cx. perexiguus, and Culiseta longiareolata were found to be infected by Nattrassia mangiferae, Aspergillus niger, A. fumigatus, Trichoderma spp. and Penicillium spp. Fungi. Penicillium spp. was the most abundant fungus isolated and identied from the larval habitats while An. superpictus s.l. was the most infected mosquito species. Conclusions: Based on the observations and results obtained from our study, isolated fungi had the potential for pathogenicity on mosquito larvae. Therefore, it is suggested that their effects on mosquito larvae be investigated in the laboratory. The most important point, however, is the proper way of exploiting these biocontrol


Background
Transmission of malaria, lariasis, Japanese encephalitis, dengue fever, and other arbovirus diseases by mosquitoes has turned mosquitoes into the most important group of arthropods in medicine and health [1]. In Iran, mosquitoes are vectors of two protozoal, two bacterial, four larial, and seven arboviral diseases [2,3]. There are 70 species and 8 (or 12) genera of Iranian mosquitoes depending on the classi cation of the tribe Aedini [4]. Anopheles mosquito species are responsible for the transmission of malaria, but the majority of mosquito species from the genera of Culex and Aedes are responsible for the transmission of arboviruses to humans [5]. Aedes aegypti, a highly anthropophilic species, Ae. albopictus the highly invasive mosquito and Culex species are important targets for the prevention of arboviruses including dengue virus, chikungunya virus, yellow fever virus, Zika virus, Japanese encephalitis virus and West Nile virus [5,6].
Anopheles species is responsible for the transmission of malaria. So far, seven malaria vectors have been recognized and reported in Iran, including: Anopheles stephensi, An. culicifacies, An. dthali, An. uviatilis, An. superpictus, An. maculipennis and An. sacharovi [14]. The rst ve of these vectors can be found in the southeast of the country, together with the majority of malaria cases. Also, An. pulcherrimus has been considered as a potential malaria vector in this area based on immunological parasite detection (two-site immunoradiometric assay (IRMA)) [15].
Anopheles maculipennis s.l. distributed in Eurasia and North America and comprises nine Palearctic members [16,17]. Some research indicated the occurrence of this malaria vector in central, the Caspian coast in the north, and North West of Iran [18][19][20].
Anopheles superpictus s.l. is distributed in Europe, Asia, and North Africa [21][22][23][24][25]. This species is one of the seven species of malaria vectors and reported in the Iranian Plateau, the slopes of the Alborz Mountains and southern Zagros, as well as the coastal plains of the Caspian Sea and the Persian Gulf in both malaria-endemic and non-endemic areas [8,22]. Oshaghi et al in 2008 reported three genotypes X, Y, and Z in Iran. Interestingly, while the sympatric Y and Z genotypes appear to be exclusive to the populations from the southeastern part of the country, genotype X is geographically separated, and present in the North, the West, the South and the Central territories [23].
A malaria eradication programme was initiated in Iran in 1951 and changed to malaria control in 1985 due to the challenges and restrictions [14]. Iran has been in an elimination stage since 2010. In 2017, the total number of recorded cases was 89, and incidence cases decreased from 0.01/1000 cases in 2017 to zero in 2019 [14,26].
The goal of all control methods is to reduce the size of vector populations. There is a risk of insecticide resistance and off-target effects on other arthropod species in chemical control [27]. Biological control is biodegradable and ecologically friendly [28]. Entomopathogenic fungi were rst used on An. gambiae with a fungus from the genus of Coelomomyces [29]. Azari-Hamidian and Abaei reported Coelomomyces sp. from the larvae of An. culicifacies s.l. in Sistan and Baluchestan Province, Southeast Iran, where 5.8% of larvae were infected with the fungus [30]. The use of pathogenic insect fungi against mosquito larvae has been reported in many studies and fungi is proven to be an effective way of killing mosquito larvae.
Unlike other biological control agents, pathogenic insect fungi can infect mosquitoes directly by penetrating the cuticle [31]. Some insect pathogenic fungi have been used effectively in recent years to control vector mosquitoes and have a wide range of species diversity. This group of pathogens is found among all phyla of fungi. The Ascomycota is the largest group of fungi. This group is extremely ecologically diverse, just like the pathogenesis pathogen of plants, animals, and humans. Pathogenic insect ascomycetes include a large group of fungi that attack a wide range of insects and are the most common insect pathogens [32]. The entomopathogenic ascomycete fungi including Metarhizium anisopliae, and Beauveria bassiana have been reported as insecticides [33]. The use of B. bassiana for control of Ae. aegypti [34] and Lagenidium giganteum in California targeted to control Cx. tarsalis [35] reduced the survival rate, blood-feeding, fecundity, and disease transmission power of targeted mosquitoes. Two species, Metarhizium anisopliae, and M. brunneum, are pathogenic to a wide range of mosquitoes in the genera of Aedes, and Culex [36,37]. In many studies, spores and secondary metabolites of insect pathogenic fungi have been reported as biocontrol agents against mosquitoes [38][39][40][41]. The fungal hyphae produces endotoxins and penetrates through the larval. These toxins cause larval damage and toxicity in the hemocoel and larval mosquito guts [42,43]. Metabolites of B. bassiana caused changes in the body and tissues of treated Cx. pipiens larvae, especially in the cuticle and midgut [44].
This study aimed to isolate and identify entomopathogenic fungi associated with mosquito larvae in Kashan County, Central Iran, and their infection, effects on mosquito larvae, and the investigation of new ways of biological control for disease vector mosquitoes.
A total of 23 larval habitats were selected in Central, Qamsar, Niasar, and Barzok Districts. These larval habitats were natural or arti cial, permanent or temporary, with or without vegetation, sunlight or shaded and clear or stagnant water (Fig. 1).

Larval sampling
Using a standard 350 ml capacity mosquito dipper, larvae and pupae of the mosquitoes were collected from April to late December 2019. Twenty dips were taken in each larval habitat in the morning (08:00-12:00 h) or afternoon (15:00-18:00 h). For sampling larvae from small water bodies, we used an eyedropper. Larvae were observed under a stereomicroscope. Dead larvae showing signs of infection and larvae and pupae with a white coating of fungal mycelium on the outer surface of their bodies were isolated from the rest of the larvae. Larvae of mosquitoes were identi ed based on a Culicidae identi cation key of Iran at the species level [45].
Isolation and diagnosis of fungi associated with mosquito larvae A white coating of fungal mycelium is observed on the surface of some of the larvae and pupae (Fig. 2).
These larvae and pupae were removed from the water of the larval habitat at the laboratory and were sterilized with 10% sodium hypochlorite for two minutes (to remove surface contaminants that con ict with the main pathogen), then washed twice with distilled water and the remaining water was removed and passed through the lter paper sterilizer [46]. They were then transferred to PDA (potato-dextroseagar) and WA (water-agar) media. Parts of the body of some other larvae were degraded or broken, or the outer epithelial layer of the larva were cut and collapsed. Consequently, the outer surface of the larvae was wrinkled. Obviously, these larvae were sick (Fig. 3). Therefore, to determine the possibility of fungal infection, they were also transferred to PDA and WA media after surface disinfection, then incubated the Petri dishes in the incubator at a temperature of 25 ± 2° C for 3-4 days. Fungi were identi ed based on phenotypic characteristics and characteristics of the culture medium, such as shape and color of the fungus colony, lament growth pattern as well as microscopic properties such as shape, size, and color of spores, mycelium, and conidiophore structure [47].

Resalts
Larval sampling results A total of 9789 larvae were collected from urban and rural areas of Central, Qamsar, Niasar, and Barzak Districts in Kashan County and were identi ed based on a valid diagnostic key and at the species level.
Three genera, i.e. Anopheles (7.89%), Culiseta (17.42%), and Culex (74.69%), consist of 13 species were identi ed (Table 1). Some mosquito specimens were deposited in the museum of Medical Entomology, Tehran University of Medical Sciences (TUMS). Fungi associated with mosquito larvae Five species of fungi were isolated from mosquito larvae ( Table 2). These fungi were isolated from larvae and pupae obtained from natural larval habitats in Qamsar and Barzak Districts. Five out of 13 mosquito species were found to be infected by fungi. A total of 105 larvae were infected with morphological or behavioral changes and fungal mycelium were observed in all infected larvae. Nattrassia mangiferae was only isolated from An. superpictus s. l. larvae or pupae in Qamsar District in August. The white hyphae of A. niger, A. fumigatus and Trichoderma spp. had grown on the surface of the larvae, and penetrated the body. Penicillium spp. was identi ed from larvae whose parts of their bodies were wrinkled, degenerated, or broken (Figs. 4, 5). In this study, Penicillium spp. was isolated from 57 mosquito larvae (54.29% of infected larvae) and it was the most abundant fungus isolated, and identi ed from larval mosquito habitats in Kashan County (Fig. 6). This fungus was identi ed from larvae collected from a natural larval habitat with vegetation in Barzok. Anopheles superpictus s. l. had the highest number of larvae infected with the fungi in all larval habitats, and from 105 infected larvae collected, 59 larvae were related to this species (Table 2).

Discussion
Insect pathogenic fungi can grow in liquid and solid environments, and their spores can attack and kill mosquito larvae [48]. In the present study, ve fungi species were identi ed from mosquito pupae or larvae. All of these fungi were isolated from larvae and pupae collected from natural larval habitats and this is the rst report of natural infection of mosquito species with these fungi from Iran.

Effects of fungi mycelia and secondary metabolites on mosquito larvae
Nattrassia mangiferae is common in the tropics and is best known as a plant pathogen (the cause of dieback and trunk cankers in trees). It can also cause fungal infections in human nails [49]. This fungus had not previously been isolated from insects and we are the rst to report this from mosquito larvae and pupae. Aspergillus has more than 180 species, some of them are pathogenic or allergenic to humans and animals. In different studies, several species of this fungus have been reported in mosquito larvae. Secondary metabolites of A. niger were also effective against the larvae of these three mosquito species [31]. In our study, also, it was found that two species of Aspergillus, including A. niger, and A. fumigatus, can grow in an aquatic environment on mosquito larvae and infect them.
In the present study, Trichoderma spp. were also isolated and identi ed from mosquito Larvae. This fungus is capable of attacking other organisms and microorganisms by producing antibiotics and other extracellular enzymes. Because of this ability, Trichoderma spp. have been known as biocontrol agents of plant pathogens for about 70 years [52]. This fungus is widely used in agriculture to control plant diseases as well as to increase crop yield. Podder and Ghosh investigated the effect of T. asperellum against anophelinae larvae and their study was the rst report on the use of T. asperellum as mosquito larvicides. They observed that the internal tissues of the larvae were destroyed after larval death [53]. It has been con rmed that some fungal toxins can cause tissue damage and dehydration of the host tissues [54]. In a study in 2016, researchers reported M. brunneum blastospores kill Aedes larvae much faster than conidia of this fungus in natural habitat, in freshwater. Blastopores easily penetrated the larval cuticle and results in rapid larval death. Conidia cause stress-induced mortality, which takes a slightly longer time [48].
Effects of fungi on morphological or behavioral changes of mosquito larvae In the present study, Penicillium sp. was isolated from wrinkled and degenerated larvae and the larvae whose parts of their cuticle were destroyed. Ragavendran et al. [55] studied the effect of larvicidal of seven fungal isolates and their metabolites on Ae. aegypti and Cx. quinquefasciatus in vitro and reported that Penicillium sp. had the best larvicidal effect compared to other fungi. The mycelia extract of this fungus had toxic effects on many parts of the larval body, including thorax, abdomen, anal gills, such as the loss of external hair, crumbled epithelial layer of the outer cuticle and shrinkage of the larvae. After 30 minutes of exposure of the larvae to the fungal metabolites, the behavioral symptoms of the treated larvae were observed including upward, downward, horizontal and vertical movements of the larvae and damage at the bottom of the larval body. Damage to the cuticle layers was also one of the morphological changes in the treated larvae. In our study also Penicillium sp. was isolated from larvae that had morphological changes in cuticle layers. Lethargy and inactivity were among behavioral changes observed in infected larvae.
Fungi infection in this study was present in An. superpictus s. l. larvae. Omrani et al. reported the rst case of a microsporidium infection (a microsporidium species from the genus Parathelohania) in An. superpictus s. l. from Iran [56]. Parathelohania legeri was reported in An. maculipennis s. l. about 110 years ago [57].
In addition to parasitic effects and their potential for mosquito control, mosquito associated fungi also have nonpathogenic interactions with mosquitoes such as impact on breeding site selection and impact on larval and adult feeding behavior. It has been demonstrated that secondary metabolites produced by Trichoderma viride have effects on attract gravid Cx. quinquefasciatus females and nd oviposition sites. Studies on nonpathogenic fungi of mosquitoes are very scarce and have not been done in Iran. Therefore study of impact of pathogenic and nonpathogenic fungi on behavior of mosquitoes can help to develop new vector control strategies [58,59].

Conclusion
Based on the observations and results obtained from our study and investigations of other researchers, entomopathogenic fungi have the potential for mosquito control and can e ciently kill mosquito larvae in laboratory and eld conditions. We did not study the lethal effects of these fungi on larvae in the laboratory, and only reported natural fungi infection in mosquito larvae in their natural habitats.
Therefore, it is suggested that their effects on mosquito larvae be investigated in the laboratory. The most important point, however, is the proper way of exploiting these biocontrol agents to maximize their effect on reducing the population of vector mosquito larvae without any negative effect on non-target organisms. Declarations