Control of Aedes mosquito populations using recombinant microalgae expressing short hairpin RNAs and their effect on plankton

New biocontrol strategies are urgently needed to combat vector-borne infectious diseases. This study presents a low-cost method to produce a potential mosquito insecticide that utilizes the microalgae released into suburban water sources to control mosquito populations. Chlorella microalgae are ubiquitous in local waters, which were chosen as the host for genetic transfection. This species facilitated the recombinant algae to adapt to the prevailing environmental conditions with rapid growth and high relative abundance. The procedure involved microalgae RNAi-based insecticides developed using short hairpin RNAs targeting the Aedes aegypti chitin synthase A (chsa) gene in Chlorella. These insecticides effectively silenced the chsa gene, inhibiting Aedes metamorphosis in the laboratory and simulated-field trials. This study explored the impact of recombinant microalgae on the phytoplankton and zooplankton in suburban waters. High-throughput sequencing revealed that rapid reproduction of recombinant Chlorella indirectly caused the disappearance of some phytoplankton and reduced the protozoan species. This study demonstrated that a recombinant microalgae-based insecticide could effectively reduce the population of Aedes mosquitoes in the laboratory and simulated field trials. However, the impact of this technology on the environment and ecology requires further investigation.

Introduction Aedes spp. is important for transmitting several mosquito-borne diseases. Aedes aegypti and Aedes albopictus are the primary carriers of pathogens, such as dengue, Zika, yellow fever, and chikungunya viruses, thus impacting public health [1,2]. Approximately half of the world's population is infected with these mosquito-borne diseases annually, with 96 million manifesting as clinical diseases with varying severity levels [3]. In China, the incidence of dengue fever was low between 2000 and 2011 and has increased rapidly since then. A 2014 outbreak of dengue fever involved 46,864 reported cases. Between 2015 to 2018, the incidence varied with a clear upward trend in 2019 (22,599 reported cases) [4]. Another disease, the Zika virus (a member of the Flaviviridae family), is associated with neonatal microcephaly and Guillain-Barré syndrome. Chikungunya virus, a member of the Togaviridae family, causes a febrile illness characterized by severe chronic arthralgia [5,6].
Many Aedes mosquitoes live in or near human dwellings, and their life cycle involves four stages: egg, larva, pupa, and adult. When the egg is exposed to water, it hatches larvae within a few hours, develops into fourth instars within 4-7 days, and enters the pupal stage. The pupa undergoes metamorphosis, molting into the winged adult form. At this time, the females are reproductively mature and are capable of transmitting pathogens while feeding on human hosts. Unfortunately, mosquitoes are developing resistance to insecticides very fast and produce numerous offspring, generating more populations with different genetic characteristics and worsening insect resistance to insecticides [7,8]. Alternative vector control strategies, such as predatory fish and fungi, have been developed to control vector populations [9]. Bacillus thuringiensis toxin and Wolbachia infection can effectively control mosquitoes [10][11][12][13]. However, their effectiveness requires an in-depth assessment for long-term use.
In this study, the mosquito chitin synthase A (chsa) gene was the RNA interference (RNAi) target for controlling mosquito multiplication. Chitin, an N-acetylglucosamine homopolymer composed of N-acetyl-D-glucosamine-[GlcNAc] linked by beta-1,4-glycosidic bonds, is a major structural component of the exoskeleton cuticle and an important component of the peritrophic matrix (PM) in the mosquito intestine and eggshell [14][15][16]. Chitin synthase (CHS), a member of the glycosyl transferase family, completes the final step of chitin biosynthetic. Insects usually have two chitin synthase genes: chsa and chsb. Chsa mainly synthesizes chitin in the exoskeleton cuticle and eggs, while chsb synthesizes chitin in the intestinal PM [17][18][19]. As chitin is crucial for mosquito life stages, it is an attractive target for controlling mosquito development and life cycle. Chitin, an important structural component of invertebrates, is absent in vertebrates, making it an attractive insect-specific target with low vertebrate toxicity.
In this study, RNAi technology was widely used to examine gene function in numerous insect species and has been explored as an alternative for insect population control [20][21][22][23]. In target cells, double standard RNA (dsRNA) is cleaved by RNase III-dicer into small interfering RNA (siRNA) of 20-25 nucleotides in length. Subsequently, the argonaute protein assembles with the siRNA to form an RNA-induced silencing complex (RISC), which binds to and degrades complementary endogenous messenger RNA (mRNA). Currently, dsRNA administration methods include immersion, oral feeding, and microinjection [24][25][26][27][28]. The specificity of RNAi makes these methods environmentally safer than other approaches currently, minimizing toxicity to non-target species and reducing the likelihood of resistance in insect populations [29][30][31].
In the current study, a low-cost mosquito insecticide alternative utilizes transgenic microalgae to control mosquito populations in suburban waters. Since chitin mainly exists in fungi and invertebrates, it has minimal effects on vertebrates. So, it is considered a safe target for human health and safety. This study also explores the impact of recombinant microalgae on phytoplankton and zooplankton populations in suburban waters via high-throughput sequencing, laying a foundation for the safe use and evaluation of recombinant microalgae.

Mosquito maintenance
The rearing of mosquitoes was performed as previously reported [32,33]. Ae. aegypti Rockefeller strain mosquitoes were donated by Professor Han Qian of Hainan University. The local wild type Ae. albopictus were captured in Haikou, China, and propagated for 5 generations in the laboratory. Both mosquito species were kept in an insectary at the Chinese Academy of Tropical Agricultural Science. The mosquitoes were maintained at 26˚C and 70-80% relative humidity.
Additionally, the mosquito cages contained a sponge soaked with sugar water, and chicken blood was used as the blood meal for female mosquitoes. After the female mosquitoes laid eggs on wet filter paper, the eggs were collected and kept dry. The eggs were then placed in incubation water (containing rat food) for hatching.

Microalgae strains and plasmids
The Chlorella vulgaris HOC5 strain was isolated from local waters in Hainan, China, and cultured in BG11 medium. Liquid cultures were maintained at 25˚C in a shaker (200 rpm) under continuous light [34]. Next, plasmid pMD18-T was locally constructed by our research group and kept in our laboratory. The pMaa7 IR / XIR RNAi expression vector was purchased from the Chlamydomonas center (Duke University, Durham, NC, USA).

Microalgae transfection
The Chlorella was grown and collected by centrifugation and resuspended in an electroporation buffer. For transfection, 2 μg of pMaa7IR/CHSAIR was mixed with the recipient Chlorella and incubated on ice for 15 min. The mixtures were then transferred into an electroporation cuvette and subjected to an electric pulse in an electrometer (BTX, ECM 630, USA) with a 1600 V/cm voltage and a pulse time of 1 ms. Then, the suspension was transferred to BG11 medium containing 60 mmol/L sorbitol and allowed to recover overnight. Finally, the cells were collected by centrifugation (10000 g, 3 min) and plated on BG11 agar plates containing 10 μg/mL paromomycin until the algal colonies appeared after 5-10 days [36,37].

Mosquito feeding experiments
After the RNAi vector, pMaa7 IR/CHSAIR had been transferred into Chlorella. The transgenic microalgae were characterized using PCR to determine if the chsa inverted repeat expression cassette had been integrated into the Chlorella genome. The positive clones were used to feed the mosquitoes in the laboratory. As previously described, ten and 300 mosquito feeding experiments were performed in the insectary [33]. In the 10 mosquito feeding experiment, The mosquitoes were divided into experimental and control groups. Each group contained 10 L1 larvae in 5 mL water supplemented with 5 mg fresh microalgae. The larvae in the experimental groups were fed the chsa RNAi transgenic Chlorella strains (CHSA1-5), respectively. In contrast, the larvae in the control groups were fed wild-type Chlorella HOC5, water, fodder, and the empty plasmid pMaa7IR/XIR transgenic strain. The experiments were performed in triplicate, recording the mortality, pupation, and adult emergence rates.
In the 300 mosquito feeding experiment, 300 L1 larvae were placed into 50 mL of water supplemented with 20 mg of fresh microalgae. The larvae feeding the recombinant Chlorella CHSA5 were set as the test treatment, whereas the larvae fed with Chlorella HOC5, water, and fodder were used as controls. The triplicate experiments included records of mortality, pupation, and adult emergence rates.

Water quality and Chlorella growth in the target water area
Water samples (10 L) were collected from the Meishe river, Shapo reservoir, and Hongcheng lake in Haikou City, Hainan, China. A model DR3900 Laboratory Spectrophotometer (HACH, CO, USA) was used to measure the concentrations of nitrogen, phosphorus, ammonia nitrogen, nitrite nitrogen, chemical oxygen demand (COD), and silicate. After centrifuging the water samples at 5000 rpm for 5 min, the algal species were observed and identified by light microscopy. Subsequently, 30 mL of 2x10 7 Chlorella HOC5 was inoculated into the 10 L water samples, and the Chlorella growth kinetics were observed and recorded.

Simulated-field trials
Simulated field trials were performed following the Mysore et al. with modification using 1.5 m 3 volume and 0.4 mm aperture polyester mesh breeding mosquito cages [40]. Four 1000 L buckets for microalgae culture were placed in cages with LED lights to facilitate algal photosynthesis in water. A ventilation pump was used to ensure the continuous circulation of the water. Then, 10 L of the cultured microalgae was poured into the buckets. Aeration and sufficient light were maintained for the culture of Chlorella from 10 L to 200 L. Then, Meishe river water was added to adjust the algal volume to 800 L at 2x10 6 -2x10 7 cells / mL. The aeration pump pressure was adjusted to make the water flow in the barrel. Approximately 1000 L1 larvae were placed in each cage. A 10% sucrose solution was prepared for male adults and chicken blood for egg-laying females. The number of adult Aedes mosquitoes was counted weekly similar to the abundance of Chlorella determined in the water. One day before mosquito counting, the blood meal and sugar water were discontinued (to reduce the physical strength of mosquitoes temporarily). The counting period was between 3:00 and 6:00 a.m. When counting, a dim LED flashlight was used to illuminate the mosquito cages.

Sample preparation and DNA extraction of test water
An 18S high-throughput sequencing analysis was performed on the water from the simulatedfield trials to detect the effect of Chlorella on the phytoplankton and zooplankton in the test water. After feeding Aedes for 28 days, a 1 L water sample was collected from the Aedes-containing bucket and filtered through a 0.40 μm polycarbonate membrane (Millipore, Burlington, MA, USA) under a 30 kPa vacuum to collect the plankton. The membranes were stored at −80˚C until the time of analysis. Genomic DNA (gDNA) was extracted using a modified Winnepenninckx et al. method [41].

Amplification and sequencing of the V4 hypervariable region
The V4 hypervariable region of the 18S rDNA was selected as the target for high-throughput sequencing analysis. The targeted region was amplified using a pair of eukaryotic universal primers (D514, 5 0 -TCCAGCTCCAATAGCGTA-3 0 , and B706R, 5 0 -AATCCRAGAATTT-CACCTCT-3 0 ) [42,43]. Sequence libraries were constructed using the NEB Next Ultra DNA Library Prep Kit for Illumina (Illumina, CA, USA) as described in the manufacturer's protocol. The library quality was assessed using the Qubit Fluorometer and Agilent Bioanalyzer 2100 system (Agilent, CA, USA). Finally, the libraries were sequenced using an Illumina HiSeq2500 platform. Low-quality reads were filtered to generate high-quality clean tags using QIIME (Version 1.7.0). Singletons were removed from the dataset, and sequences with > 97% similarity were clustered into operational taxonomic units (OTUs) using UPARSE (Version 7.0.1001) [44,45]. Assignments of the taxonomic OTUs were completed using the Silva and Nucleotide Databases in National Center for Biotechnology Information (NCBI) (https://blast.ncbi.nlm. nih.gov) [46]. Sample diversity was determined using QIIME (Version 1.7.0), and Shannon indexes were calculated following the Whittaker [47] protocol.

Statistical analyses
Data analysis was performed using Statistical Package for the Social Sciences (SPSS) and presented as mean ± S.D. (standard deviation). Duncan's multiple ranges and t-test tests were performed to examine significant differences between means. Values with p-value < 0.05 were considered statistically significant. Asterisks indicate the statistical significance in all cases: � P <0.05, �� P <0.01. Error bars show the standard deviation.

chsa RNAi transgenic Chlorella are lethal to Ae. aegypti
The RNAi target gene is the chsa gene (AAEL002718) of Ae. aegypti. The target region for silencing by the RNAi was nucleotides 331 to 621 in the CDS. cDNA of Ae. aegypti was used as a template to amplify the chsa RNAi interference fragment, and an amplicon of approximately 290 bp was obtained. This fragment was then cloned into pMaa7IR/XIR in forward and reverse directions to produce the recombinant RNAi vector of pMaa7IR/CHSAIR. After the RNAi vector was transformed into C. vulgaris HOC5, 79 positive recombinant algal clones were identified by PCR, and five were used for subsequent experiments. In the 10 mosquito feeding experiment, the larvae fed recombinant Chlorella clones CHSA3 and CHSA5 began to die on the second day. Larvae fed clones CHSA1, CHSA2 and CHSA4 began to die on the third day and all the larvae fed recombinant Chlorella died within 13 days. None of the larvae died when fed with water and fodder, but only 6.7 and 23.3% died within 15 days after feeding on Chlorella HOC5 and empty-plasmid Maa7IR/XIR transgenic Chlorella. Orally feeding Aedes larvae with chsa RNAi recombinant Chlorella is lethal (Fig 1A).
Only 6.67% of Aedes mosquito larvae fed recombinant Chlorella CHSA1 and CHSA3 pupated, and none pupated after feeding on recombinant Chlorella CHSA2, CHSA4, and CHSA5. In the control groups, 100% of the larvae pupated when fed on fodder. 96.7 and 76.7% of pupation rates were observed in larvae fed on Chlorella HOC5 and empty plasmid Maa7IR/ XIR transgenic Chlorella, respectively. However, no larva pupated when fed on water alone ( Fig 1B).
None of the pupae developed into adults after feeding on any of the recombinant Chlorella strains. In contrast, 100, 96.67, and 76.67% of control pupae emerged into adults when fed on fodder, Chlorella HOC5, and empty plasmid Maa7IR/XIR transgenic Chlorella, respectively ( Fig 1C).

Feeding experiment of 300 Ae. aegypti mosquitoes
Approximately 300 L1 Ae. Aegypti larvae in each treatment were assessed using feeding experiments for 30 days. The larvae fed recombinant Chlorella CHSA5 began to die on the second day, and 88.5% of larvae died within 30 days. In the controls, only 3.8, 0.3, 0.8, and 10.3% of the larvae died after feeding on water, fodder, Chlorella HOC5, and Maa7IR/XIR transgenic Chlorella, respectively (Fig 2A).
The larvae fed on fodder started pupating on day 4 and 99.7% of larvae pupated by day 12. The larvae fed Chlorella HOC5 started pupating on day 5, and by day 20, 94.7% pupated. Besides, the larvae fed on Maa7IR/XIR transgenic Chlorella began to pupate on day 5, and 86.3% pupated by day 19. In contrast, larvae fed on recombinant Chlorella CHSA5 began to pupate on day 6, and by day 15, 10.0% pupated (Fig 2B).
The fodder-fed Aedes fully emerged into adults on day 15, and 98.3% of the pupae emerged within 30 days. The Aedes fed Chlorella HOC5 fully emerged into adults by day 22, with 94.0% of the pupae emerging into adults. However, only 8.7% of Aedes pupae emerged into adults when fed on recombinant Chlorella CHSA5 (Fig 2C).

Ae. albopictus mosquito feeding experiment
Similar experiments to those described above were conducted on Ae. Albopictus mosquitoes. Larvae fed on recombinant Chlorella CHSA5 had the highest mortality (100%) on the 10 th day, followed by 86.7 and 83.3% mortality rates in larvae fed on recombinant Chlorella CHSA3 and CHSA1, respectively. None of the larvae died when fed on water and fodder. Only 6.7 and

PLOS NEGLECTED TROPICAL DISEASES
23.3% of the larvae died when fed on control Chlorella HOC5 or Maa7IR/XIR transgenic Chlorella, respectively (Fig 3A). In the 300 Ae. albopictus mosquito feeding experiment, 9.7% Aedes pupae emerged into adults when fed on recombinant Chlorella CHSA5. However, 93.8 and 88.7% of the control pupae emerged into adults within 30 days of feeding on fodder and

PLOS NEGLECTED TROPICAL DISEASES
Chlorella HOC5, respectively (Fig 3B). These results show that recombinant Chlorella is more lethal and effective on Ae. aegypti than Ae. albopictus larvae.

Expression of chsa mRNA in Aedes larvae fed recombinant Chlorella
Real-time PCR was used to examine the expression of the chsa gene in Aedes larvae fed on recombinant Chlorella strains using Ae. Aegypti larvae fed with Chlorella HOC5 as controls. The expression of chsa gene was significantly lower in Ae. Aegypti larvae fed on recombinant Chlorella than in the control larvae (Fig 4). The lowest rate (97.6%) was observed in larvae fed

Simulated-field evaluation of recombinant Chlorella CHSA5 activity
The recombinant Chlorella CHSA5 activity was assessed under simulated-field conditions in preparation for future field studies. First, the water quality and local dominant algal strains were tested in three areas of Haikou City to detect whether Chlorella CHSA5 is suitable for growth in the above waters. The nitrogen, phosphorus, ammonia, nitrate, nitrite, and chemical oxygen demand (COD) levels in the Meishe river, Shapo reservoir, and Hongcheng lake exceeded the recommended limits of a clean water source. Pollutants and eutrophication were observed in the water bodies. Moreover, the salinity of Hongcheng lake was too high and reached 1.5% (S1 Table).
Chlorella CHSA5 was cultured in 10 liters of water from the above-described water sources. Chlorella CHSA5 grew normally in the Meishe river and Shapo reservoir water samples. Moreover, algal cells reached peak levels on day 9 in the Meishe river water and on day 7 in the Shapo reservoir. The high salinity of the Hongcheng lake water negatively impacted Chlorella CHSA5 growth (S1 Fig). We chose water from the Meishe river for the simulated-field trial because the trial was located near the water collection point, and Chlorella was the dominant algal strain in the Meishe river.
A simulated field release study was conducted to infer the potential environmental impacts of this technology. An idle factory building located 5 kilometers from the nearest residential area in the suburbs of Haikou City was selected for the experiment to reduce the impact of the trial on residents (Fig 5A, 5B and 5C). However, the experiment was performed using Ae. Albopictus instead of Ae. aegypti after considering the predominant dengue virus transmission in Hainan by Ae. albopictus. The number of Aedes mosquitoes fed the Meisher river water increased from 1000 to a maximum of 4825 individuals after 63 days and slowly decreased to approximately 4300 (112 days). The Aedes mosquitoes raised in the Meisher river water supplemented with wild-type Chlorella HOC5 increased to a maximum of 6844 individuals after 70 days and slowly decreased to approximately 6500 individuals (112 days). In the treatment group, Aedes mosquitoes raised in the Meisher river water supplemented with recombinant Chlorella CHSA5 decreased from 1000 to 0 (105 days) (Fig 5D).
The data indicate that Chlorella CHSA5 can suppress Aedes mosquito populations under simulated field conditions. However, the growth potential of Chlorella was limited, as the number of Chlorella in Meishe River was only 66900/mL. Generally, Aedes larvae feed on microalgae or microorganisms in the water. Adding Chlorella HOC5 significantly increased the number of Aedes mosquitoes in the Meishe river because Chlorella HOC5 increased the food density for Aedes (Fig 5E).

Sequencing results of 18S V4 hypervariable region in the test water
An 18S high-throughput analysis was performed to understand the impact of recombinant Chlorella CHSA5 on biological populations of the test water. There were 85,340 tags qualified tags after quality control and 85,036 high-quality tags after removing unclassified and unique tags from the dataset. The total number of assigned OTUs was 2,021. The rarefaction curves of OTUs were prepared for all the samples, and each sample had a weak slope at the end of the curves (S2 Fig). Twenty groups of eukaryotic microalgae were identified in the Meishe river (MSH) (at the class level), including Chlorophycea, Chrysophyceae, Cryptophyceae, and Bacillariophyceae ( Fig  6A, S2 Table). The MSH groups contained 386 OTUs within 118 genera. Chlorophycea had the largest number of OTUs (75) within 32 genera. Chrysophyceae, the second largest class, contained 69 OTUs within 13 genera. Cryptophyceae, Bacillariophyceae, Mediophyceae, and Dinophyceae also hosted a relatively large number of OTUs (S2 Table). In contrast, Phaeophyceae, Mamiellophyceae, Nephroselmidophyceae, Dictyochophyceae, Prymnesiophyceae, Raphidophyceae, Pedinophyceae, Ulvophyceae, and Rhodellophyceae had much lower OTU numbers (S3 Fig, S2 Table).

Variation of phytoplankton community in the test water
The MSH group had the highest species diversity. Adding excess Chlorella HOC5 to the Meishe river water increased the relative abundance of Chlorella from 18.6 to 51.6% (MSHH group). In the CHSA5 treatment group, the relative abundance of Chlorella rose to 75.2%. Moreover, the number of phytoplankton in the MSHH group decreased by 44 at the genus level and 149 OTUs due to the rapid increase of Chlorella relative to the dynamics observed in the MSH group. The number of phytoplankton in the CHSA5 group decreased by 51 at the genus level and 180 OTUs (Fig 7A, S3 Fig, S4 Fig, S2 Table). The relative abundance of other major species decreased in the MSHH and CHSA5 groups compared to the MSH group, except for Chlorella, Tetradesmus, Poterioochromonas, and Coelastrella (Figs 7A and 8A). Furthermore, the MSH group had a higher Shannon diversity index (5.15) than the MSHH (2.04) and CHSA5 groups (1.48). The Shannon index of the MSH group was significantly different from that of the MSHH and CHSA5 groups (P �0.01). There was also a significant difference in the Shannon indexes between the MSHH and CHSA5 groups (P �0.01) (Fig 9A). The above results show that rapidly increasing the wild-type or recombinant Chlorella is related to the death and disappearance of other microalgae. Since the relative abundance of recombinant Chlorella in the CHSA5 group was higher than that of wild-type Chlorella in the MSHH group, the species decreased the most, and the diversity index was the lowest in the CHSA5 group.

Variation of zooplankton community in the test water
A total of 29, 37, and 29 groups of protozoa were identified at the class level in the MSH, MSHH, and CHSA5 groups, respectively. Spirotrichea, Oligohymenophorea, Maxillopoda, Prostomatea, Litostomatea, Bigyra, and Choanoflagellata were the majority of the observed protozoans (Fig 6B, S3 Table). Among these, there were 207 OTUs among 53 genera in the MSH group, 239 OTUs within 54 genera in the MSHH group, and 201 OTUs within 50 genera in the CHSA5 group (S3 Table).
In contrast to the zooplankton species in the MSH group, the number of zooplankton species increased by 1 genus and 32 OTUs when wild-type Chlorella HOC5 was added to the MSHH group, while the zooplankton species in the CHSA5 group decreased by 3 genera and 6 OTUs (Fig 7B, S3 Table). In the CHSA5 group, the relative abundance of the main genera (Paramecium, Cryptocaryon, Strobilidium, Bicosoeca, Tintinnidium, and Halteria) decreased compared to the MSH and MSHH groups. In contrast, the relative abundance of Tropocyclops,

PLOS NEGLECTED TROPICAL DISEASES
Blepharisma, Mesocyclops, and Epistylis increased (Figs 7B and 8B). The Shannon diversity indexes were 4.52, 5.76, and 4.59 in the MSH, MSHH, and CHSA5 groups, respectively. The Shannon index of the MSHH group was significantly higher than that of the MSH and CHSA5 groups (P �0.01) but not significantly different between the MSH and CHSA5 groups ( Fig  9B). These results indicate that adding wild-type Chlorella to the MSHH group result in algal blooms and increase the number of zooplankton species feeding on Chlorella. However, adding recombinant Chlorella to the CHSA5 group slightly decreased the number of zooplankton species, probably due to the chsa dsRNA fragment carried by the recombinant Chlorella. Since chitin is present on protozoan surface coats [48], the chsa dsRNA in recombinant Chlorella possibly inhibited the expression level of chsa in protozoa, thus, decreasing the number of protozoan species in the CHSA5-treated water.

Discussion
Aedes mosquitoes are vectors of epidemic diseases that affect public health worldwide, such as yellow fever, dengue fever, Zika virus, and chikungunya virus [49][50][51][52]. These diseases cause a large number of deaths every year. Without effective therapies or vaccines, vector control is the most effective strategy to prevent the spread of these diseases. Numerous studies have shown that RNAi is a powerful tool for controlling insect populations and may be less susceptible to insect resistance than other strategies. In this study, Chlorella, a natural food for mosquito larvae, served as the exogenous RNAi carrier. Recombinant Chlorella can be directly released into suburban waters as a biological pesticide. This biological "insecticide" can selfreplicate, lowering production costs.
In this study, the Aedes mosquito chsa gene was used as the target for RNAi. Since this gene primarily exists in invertebrates and fungi, it is likely to have a limited impact on humans, higher animals, and plants. However, environmental impacts are unclear [48][49][50][51][52][53][54][55][56][57]. Chlorella, a dominant algal strain widely in local waters, was selected as the host for genetic manipulation. The genetically modified mosquito-killing Chlorella adapted to environmental conditions, allowing for rapid growth to reach a desired level of relative abundance. Some of the Ae. aegypti larvae fed on recombinant Chlorella died the next day, and all the larvae died within 13 days (Fig 1A). In the feeding experiment of 300 Aedes mosquitoes, 88.5% of the larvae fed on recombinant Chlorella died within 30 days. These results indicate that oral administration of chsa RNAi recombinant Chlorella is lethal to Ae. aegypti larvae (Fig 2A). Subsequently, Ae. albopictus were fed the recombinant Chlorella. The results showed that the recombinant Chlorella had a lethal effect on Ae. albopictus (Fig 3A and 3B). The 290 bp target sequence homology of Ae. aegypti and Ae. albopictus chsa coding region sequence was 91.3%, which was expected.
Considering that the main vector of the dengue virus (Ae. Albopictus) in Hainan, China, this species was used in the simulated field experiment study [58,59]. The waters of the Meishe river in Haikou City were used for seeding using recombinant Chlorella, one of the predominant microalgae strains in the local ecosystem. In addition, the eutrophication of the Meishe river, with excessive N, P, ammonia, and COD (S1 Table), was favorable for the survival and reproduction of the recombinant Chlorella (S1 Fig).
The test results showed that the recombinant Chlorella reduced the number of Aedes mosquitoes from 1000 to 0 within 105 days. The Ae. albopictus population was effectively suppressed (Fig 5D). 18S high-throughput sequencing analysis was performed to understand the impact of recombinant Chlorella on biological populations in the test water. Adding excess recombinant Chlorella to the Meishe River increased its relative abundance by 76.6%. However, it decreased the number of phytoplankton by 51 (at the genus level) and the number of OTUs by 179 (Fig 7, S3 and S4 Figs). This outcome is the result of nutritional competition. The rapid reproduction of Chlorella, similar to the phenomenon of algal blooms in natural waters, where Chlorella primarily consumed nutrients in the water, resulted in the death of a portion of the naturally occurring phytoplankton (Fig 7A, Fig 8A). The proliferation of Chlorella has also led to an increase in the number of zooplankton species that feed on wild Chlorella. However, the chsa dsRNA carried by the recombinant Chlorella may inhibit the expression of chsa genes in several species of protozoans with high homology, resulting in a decrease in the number of protozoan species (Fig 7B, Fig 8B) in the CHSA5 group.
Although some progress has been made in this study, the impact of exogenous DNA fragments carried by recombinant Chlorella, especially antibiotic resistance genes, on environmental organisms, including the organisms in direct and indirect food chains, needs to be further evaluated. The research aimed to develop low-cost, environmentally friendly mosquito control technologies. Although this self-propagating microalgae insecticide has been validated in the laboratory, further assessment of the effects of blooms caused by recombinant Chlorella after release into waters is necessary. There have been many reports on harmful algal blooms [60][61][62] mainly caused by microalgae-containing toxins. These toxin-containing microalgae, including Alexandrium species, Gymnodinium catenatum, Prorocentrum lima, Dinophysis acuta, Anabaena, and Cylindrotheca, can be transmitted to mammals and humans through fish and shellfish, threatening human health [63]. Non-toxic algal blooms have little impact on ecosystems and are rarely studied and discussed. Non-toxic recombinant Chlorella should be kept at levels that can control mosquito populations and have little impact on other plankton species, making it acceptable to the public. While promising, this approach will require refinement, including in vitro gene editing technology, shRNAi technology, etc., to minimize the impact of exogenous DNA fragments carried by recombinant microalgae so that the technology can be employed with a high degree of safety. Propinquity to saturation is denoted by weak slopes at the end of rarefaction curves. Sequences with �97% similarity were assigned to an OTU. MSH1-3: the mosquito was living in water from the Meishe river alone. MSHH1-3: the mosquito was living in water from the Meishe river supplemented with wild Chlorella HOC5. CHSA1-3: the mosquito was living in water from the Meishe river supplemented with recombinant Chlorella CHSA5.