Enhancing the Insecticidal Potential of Baculovirus by Overexpressing the Mammalian Β-Galactosyl Binding Protein Galectin-1


 Developing bio-pesticides is an important area of research in agriculture for which viruses are an essential tool. Infection by entomological pathogenic viruses kills agricultural pests, and viral progenies are disseminated to infect more pests, eventually achieving long-term pest control in the field. Of the current virus-based pest control models, Autographa californica multiple nucleopolyhedrovirus (AcMNPV) is the most studied. AcMNPV belongs to the family Baculoviridae and can infect many lepidopterans. Although AcMNPV has been previously demonstrated as a potential pest-control tool, its long virus infection cycle has made field application challenging. To overcome this hurdle, we generated a recombinant baculovirus that can express mammalian galectin-1, which is a-galactoside-binding protein that binds to the peritrophic matrix in the midgut of pest lepidopterans and induces perforation of the membrane. Galectin-1 has been shown to affect immune responses in mammals, including humans, but to our knowledge, the effect of galectin-1 on insect immune systems has not been previously reported. Hosts infected with this recombinant virus exhibited reduced appetite and died sooner in both laboratory and small-scale field studies, suggesting that the overexpression of galectin-1 can more efficiently eliminate pest hosts. In addition to disrupting the integrity of the peritrophic matrix, the immune system of hosts infected with recombinant baculovirus carrying the galectin-1 gene was suppressed, making hosts more vulnerable to secondary infection. These results demonstrated that the potential of baculovirus for pest control can be improved by using a recombinant baculovirus that overexpresses mammalian galectin-1 in hosts.


Introduction
Pest control has been important to conventional agriculture, and chemical-based pesticides are widely used for this purpose. Even though chemical pesticides are highly e cient at eliminating pests, the widespread use of pesticides causes environmental pollution and, due to a lack of target speci city, may cause adverse reactions in animals that ingest it while eating the crops (Bonmatin et al., 2015, Goulson, 2013, Henry et al., 2012, vanderWerf, 1996. Bio-pesticides, such as bacteria, fungi, and viruses, have become alternative tools for pest control because they are less likely to be biohazards in the environment and have a more speci c host range (Copping and Menn, 2000). However, bio-pesticides are unable to completely replace chemical pesticides due to their lack of e ciency and underdevelopment of industrialized production. One method for improving bio-pesticide e ciency is to genetically modify pathogens to overexpress toxins, hormones, and speci c enzymes to expedite the infection process (Copping and Menn, 2000).
Autographa californica multiple nucleopolyhedrovirus (AcMNPV) belongs to the family Baculoviridae. Baculoviruses are double-stranded DNA viruses with a genome size ranging from 88 to 153 kb (Burgess, 1977). These viruses mainly infect invertebrates and have been reported to infect more than 600 insect species (Martignoni, 1986). There are two phenotypes in the life cycle of baculoviruses: budded virus (BV) and occluded virus (ODV). After being ingested by larvae, occlusion bodies are passed down to the midgut, where they are dissolved by the alkaline condition of the gut, releasing ODVs. These ODVs are taken up by microvilli and subsequently enter host cells to commence infection to produce BV progenies (Blissard, 1996). BVs then infect cells in blood and other tissues, such as fat, muscle, and epithelium, resulting in systemic infection (Granados andLawler, 1981, Blissard andRohrmann, 1990). Owing to their speci c host range, baculovirus used as bio-pesticides would not pose a threat to the environment or nontarget organisms. Nevertheless, it takes 7 days for baculoviruses to complete their infection cycle and kill the host, which is an issue limiting their applicability as bio-pesticides.
Galectin-1 is an animal lectin that can bind to β-galactoside, which is present in muscle and fat tissues in humans; these cells have receptors for β-galactoside on their surfaces (Barondes et al., 1994, Cho andCummings, 1995). Galectin-1 mediates several physiological functions of these cells, including growth, differentiation, and adhesion (Camby et al., 2006, Adams et al., 1996, Ahmad et al., 2004. The toxicity of galectin-1 in lepidopteran larvae has also been demonstrated (Chen et al., 2009). When diamondback moth (Plutella xylostella) larvae were fed with galectin-1-coated leaves, they became anorexic and exhibited 60% mortality in 4 days. Galectin-1 was able to bind to chitins on the peritrophic matrix in the midgut, disrupting the integrity of the midgut. This disruption not only resulted in anorexia in P. xylostella larvae but also allowed pathogens to enter and invade the hemocoel. Galectin-1 suppresses immune responses in mammals (Allione et al., 1998, Blaser et al., 1998 and can induce apoptosis in the human T cell line HPB-ALL by binding to a cluster of differentiation proteins (CDs) located on the T cell surface with the highest a nity toward CD45 (Perillo et al., 1995). Binding of CD45 by galectin-1 has been shown to induce apoptosis in immune cells (Ong et al., 1994, Stillman et al., 2006. Nevertheless, to the best of our knowledge, the effect of galectin-1 on the immune system of insects has not been reported. To enhance the e ciency of baculoviruses as bio-pesticides in the eld, a recombinant baculovirus carrying the galectin-1 gene under the control of the polyhedrin promoter was constructed (Maeda, 1989, Miller, 1988, Weyer et al., 1990, Bonning and Hammock, 1996. We assessed the e ciency of this recombinant virus as a bio-pesticide against Spodoptera litura (tobacco cutworm) larvae. Single feeding of the recombinant virus to the larvae could decrease larval appetite, and continuous feeding was su cient to cause 70% mortality in 5 days. Gene expression analysis suggested that the expression of genes involved in the Toll pathway was suppressed, increasing the infection e ciency of the recombinant virus. Tests in simulated eld trials and other common pests also con rmed that recombinant viruses combined with Gal-1 have high insecticidal abilities.

Cells and viruses
The S. frugiperda IPLB-Sf-21 (Sf-21) cell line was cultured in TC-100 insect medium containing 10% heatinactivated fetal bovine serum. Sf-21 monolayers were used for propagation of wild type AcMNPV or recombinants baculovirus. The titers of the virus clones were estimated by both conventional TCID 50 and quantitative PCR methods .

Generation of recombinant virus
Gal-1 gene fragment was ampli ed from Chinese hamster ovary cells (CHO cells) and subsequently cloned into T&A™ Cloning Vector using T&A™ Cloning Vector Kit (Yeastern Biotech). T&A™ cloning vector containing Gal-1 gene was digested with XhoI and XbaI and the insert was cloned into pBacPAK8 (Takara), generating pABpRpGal1. Plasmid pABpRpGal1 was puri ed using High Pure Plasmid Isolation Kit (Roche) and co-transfected with BacPAK viral DNA (Takara) into S. frugiperda cell line IPLB-Sf21 using lipofectamine (Invitrogen) as transfection reagent, generating vABpRpGal1. Recombinant baculovirus vABpRpGal1 was ampli ed in IPLB-Sf21 cells and the titre was determined using quantitative real-time polymerase chain reaction, qPCR).
Western blot analysis of Galectin-1 expression S. litura cell line SL1A cells (1 x 10 5 cells/well) were seeded in a 24-well plate and infected with recombinant virus at MOI (multiplicity of infection) = 1. Forty-eight hours after infection, supernatant was removed and cells were lysed with 100 mL RIPA buffer. Cell lysate was mixed with equal volume of Laemmli sample buffer (Bio-rad) and sample was resolved in 10% SDS-polyacrylamide gel using Wet/Tank Blotting Systems (Bio-rad). Resolved sample was transferred onto a PVDF membrane (Millipore) . Expression of Galectin-1 was detected using either rabbit anti-6xHis antibody (LTK) or anti-Galectin-1 antibody as primary antibody, followed by secondary goat anti-rabbit IgG antibody (Merck) or anti-mouse IgG antibody. Signal was detected using Western HRP Substrate (Millipore) and quanti ed using UVP ChemStudio (analytikjena).
Maintenance of S. litura S. litura (Tobacco cutworms) were purchased from a farm which breed these insects in the laboratory for generations (New Taipei City, Taiwan). The arti cial feed for these insects was prepared as follows: a) mix 90 g of kidney bean powder, 36 g of yeast extract, 33 g of wheat germ powder and 480 mL of sterile water in a blender for 5 minutes; b) dissolve 036 g L-cycteine (PanReac AppliChem), 3.6 g L-ascorbic acid (PanReac AppliChem), and 0.225 g streptomycin sulfate (Bioshop) in 60 mL of sterile water; c) pour b) into blender containing a) and mix well; d) dissolve 9.9 g agar completely in 300 mL of water by boiling; and e) pour d) into c) and homogenize for 5 minutes. The content was poured into container and allowed to hardened and kept at 4°C until use. Tobacco cutworms were kept in round and transparent plastic boxes (8cm(L) x 5.5 cm(H)) at 26±1°C with alternating light : dark cycle at 12 hrs : 12 hrs. Feed was refreshed daily. (Tang et al., 2021) Measurement of survival rate, appetite and body length Second-instar S. litura larvae were divided into 4 groups, control, wild-type baculovirus (wt AcMNPV), recombinant vABpRpGal1 (vABpRpGal1) group and commercially available biological insecticides (Bacillus thuringiensis (0.3 kilograms per hectare)). Larvae in wt AcMNPV and vABpRpGal1 groups were fed with 1 x 10 6 pfu virus/larva. For single feeding experiment, larvae were fed with virus on day one of the experiment. For continuous feeding experiment, xed sum of viruses was given daily. Larvae were observed for 7 days and mortality and body length of experiment larvae in each group were recorded daily. For appetite measurement, every larva which survived in the single feeding experiment was placed on one leaf and the area ingested by the larva after 24 hours was recorded.
Relative gene expression analysis SL1A cells were seeded in a 24-well plate and infected with wild-type AcMNPV or recombinant vABpRpGal1 at MOI = 1. Infected cells were harvested 48 hours post infection and total RNA was extracted using Total RNA Mini Kit (Geneaid) following manufacturer's protocol. RNA concentration was determined using NanoDrop 2000 (Thermo Fisher Scienti c). First-strand cDNA was obtained using PrimeScriptÔ RT reagent kit (Takara). Sample was prepared to include: 2 mL of 5x PrimeScript buffer, 0.5 mL of PrimeScript RT enzyme mix, 0.5 mL of oligo dT primer, 2 mL of random hexamer, 100 ng of RNA and RNase-free water. Sample was incubated at 37°C for 15 minutes, followed by 5 seconds at 85°C and then cooled down to 4°C. Relative expression level of target gene in samples was measured by quantitative real-time polymerase chain reaction. Each sample was set up to include 10 mL of SYBR green (Bioline), 0.8 mL of forward primer, 0.8 mL of reverse primer, 1 mL of cDNA as template, and 7.4 mL of sterile double-distilled water. Reaction was carried out with the following program: initial denaturation at 95°C for 10 min, followed by 40 cycles of denaturation at 95°C for 10 sec and extension at 60°C for 30 sec . Sequences of the primer pairs used are listed in table 1.

Phagocytosis activity assay
Fourth instar larvae were injected with phosphate buffered saline (PBS, control), 1 x 10 6 pfu wild-type AcMNPV (wt), or 1 x 10 6 pfu recombinant vABpRpGal1 virus (vABpRpGal1). Forty-eight hours after injection, hemocytes were collected from hemolymph and seeded in a 96-well plate (4 x 10 6 cells/well). FITC-conjugated E. coli were added per well and the plate was incubated in the dark for 1 hr. Cells were washed 2 -3 times with phosphate buffered saline. Staining solution containing 50 mL of 0.4% trypan blue and 50 mL of TC100 culture medium was added per well and cells were stained for 10 min. After washing with PBS, cells were stained with 300 mM DAPI (Thermo Fisher) for 30 minutes. Stained cells were washed 2 -3 times with PBS and re-suspended in 100 mL of TC-100 culture medium. They were observed under uorescence microscope to determine the ratio of E coli to hemocytes (Wan et al., 1993, Chang et al., 2020).

Semi-Field trial
Eight to ten lettuce seedlings were planted with equal spacing in a containing (57cm x 37cm x 19cm). Wild-type AcMNPV and vABpRpGal1 solution containing 1% Tween20 was sprayed onto seedlings in wt and vABpRpGal1 group (1 x 10 9 pfu virus/larva), respectively, and seedlings in control group were sprayed with sterile water. One hour after spraying, 40 or 80 second-instar S. litura larvae were placed on each seedling and observed for 2 or 3 days. The number of surviving larvae was recorded daily and unfound larva was counted as dead. Leaves of each seedling were photographed after experiment. (Cory et al., 1994) Statistical analysis Statistical difference among groups in survival rate was analyzed using Log-rank test. Quantitative PCR (qPCR) data in gene expression analysis was converted to relative expression level (2 -DDCt ) and statistical difference in relative expression level was analyzed using student's t-test. A p value less than 0.05 (p < 0.05) was considered statistically different and marked with an asterisk (*) (Hu et al., 2018).

Generation of recombinant baculovirus carrying Gal-1 gene
To investigate whether the overexpression of galectin-1 could improve the biopesticide potential of baculovirus, a recombinant baculovirus, vABpRpGal1, carrying the Gal-1 gene under the control of the very-late polyhedrin promoter and dsRed gene as a reporter gene was generated. A control virus carrying only the dsRed gene was also generated (Fig. 1A). IPLB-Sf-21 cells infected with vABpRpGal1 exhibited red uorescence 48 h post infection (Fig. 1B), and the expression of Gal-1 in infected cells was con rmed by western blot analysis (Fig. 1C).
Recombinant baculovirus carrying gal-1 gene signi cantly induced higher mortality in S. litura larvae To evaluate the toxicity of recombinant vABpRpGal1 in S. litura larvae, the virus was added once to the arti cial feed on day one of the experiment, larvae were observed for 7 days, and the mortality was recorded daily. There was no difference between the vABpRpGal1 and control virus groups in the survival rate (over 95% for both) (Fig. 2A). Larvae fed with vABpRpGal1 were smaller (Fig. 2B), and therefore, the surviving larvae were placed on sweet potato leaves (one leaf per larva) to assess their appetite. The larvae from the control virus group ingested more than 50% of the leaves, whereas the larvae from the vABpRpGal1 group ingested less than 50% of the leaves (Fig. 2C and 2D), indicating that vABpRpGal1 reduced larval appetite.
To evaluate the effect of continuous vABpRpGal1 exposure in larvae, viruses were added to the arti cial feed daily for 7 days. The larvae fed with the control virus had a survival rate of 80% on day 7, whereas larvae fed with vABpRpGal1 had a survival rate of less than 40% on day 5 and less than 20% on day 7 (Fig. 3A). The analysis of body length showed that, starting on day 3 of the experiment, larvae in virus groups exhibited slower growth rates compared to those in the control group, and larvae in the vABpRpGal1 group started to show signs of growth arrest on day 5 and had the shortest body length on day 6 ( Fig. 3B and 3C). These results suggested that galectin-1 could negatively in uence the appetite and growth of S. litura larvae.
Galectin-1 increases the vulnerability of S. litura larvae to baculovirus by suppressing the immune system Galectin-1 was shown to bind to receptors on the T cell surface in mammals and subsequently induce apoptosis, resulting in a suppressed immune system. This is also a tactic used by cancer cells to avoid immune system attacks. To evaluate the effect of vABpRpGal1 infection on the immune response of S. litura, the expression of genes involved in the Toll pathway was measured as pathogen-induced humoral immunity in insects mediated mainly via the Toll pathway. Upstream toll-like receptors and downstream anti-bacterial peptide cecropin expression levels increased in larvae infected with the control virus, suggesting that pathogen infection elicited an immune response. In contrast, the expression levels of upstream toll-like receptors (Toll), Myd88, Pelle, and downstream anti-bacterial peptide cecropin were lower in the vABpRpGal-1 group (Fig. 4A), suggesting that Gal-1 had an inhibitory effect on the Toll pathway. Similar results were obtained when the expression of genes involved in the IMD pathway, including pgrp, fadd, and relish, was analyzed (Fig. 4B), indicating that Gal-1 suppressed the expression of genes involved in the immune response of host larvae.
Phagocytosis is the primary cell-mediated immune response in insects. To evaluate the effect of Gal-1 on cell-mediated immunity, the phagocytic activity of hemocytes collected from S. litura larvae 48 h after infection with baculovirus was measured using FITC-labeled E. coli. Phagocytosed E. coli emitted green uorescence when observed under a uorescence microscope, while the uorescence from unphagocytosed E. coli was masked by trypan blue dye. Fluorescence microscopy images showed green uorescence in the control and control virus groups, whereas signi cantly less green uorescence signals were detected in the vABpRpGal1 group (Fig. 4C, left). Phagocytic activity (phagocytosis) was determined based on the ratio of the number of uorescent cells to the total number of cells. The phagocytic activity was inhibited following viral infection, and the activity was further inhibited by the expression of Gal-1 (Fig. 4C, right). Combining these results, we concluded that Gal-1 exerted an inhibitory effect on both humoral and cell-mediated immune responses.

Recombinant baculovirus expressing Gal-1 has potential as a bio-pesticide in the eld
To evaluate the potential of recombinant baculoviruses expressing Gal-1 as bio-pesticides, a eld trial was conducted. Eight to ten lettuce plants were planted in a box (40 cm × 60 cm, Fig. 5A), and the leaves were sprayed with a solution containing control virus, vABpRpGal1, or sterile water for the control group. In total, 40 or 80 S. litura larvae were evenly distributed on the leaves. To evaluate the toxicity of recombinant vABpRpGal1 in S. litura larvae, leaves were sprayed with virus once on day 1 of the experiment. Larvae were observed for 1 to 7 days, and the mortality was recorded daily. The survival rate was measured and photographs were taken of the leaves ingested by the larvae after 2 days (cabbage) and 3 days (lettuce). Ingested lettuce photographs showed that larvae without viral infection ingested almost all the leaves, whereas larvae infected with the control virus ingested approximately half of the leaves, and larvae infected with vABpRpGal1 ingested the fewest leaves (Fig. 5B). The survival rate for the control and control virus groups was 95% and 85%, respectively, whereas it was only 52.5% for the vABpRpGal1 group (Fig. 5C). The other two lettuce replicates showed the same trend ( Fig. 5D and 5E).
Similar results were obtained when cabbage was used as the food source, with a survival rate of 97.5% for the control, 82.5% for the control virus, and 70% for the vABpRpGal1 groups (data not shown). These results indicated that overexpression of Gal-1 increased the lethality of baculovirus in S. litura larvae. To evaluate the bio-pesticide potential of recombinant vABpRpGal1 in other pests, a eld experiment was conducted using fall armyworm (Spodoptera frugiperda) and diamondback moth larvae, and similar results were obtained as those with S. litura larvae ( Fig. 6A and 6B). When the experiment was conducted using insects with economic value, such as honeybees, recombinant vABpRpGal1 did not exert any toxicity (Fig. 6C), suggesting its safety in the eld for non-host insects. Generally, using viruses to control pests results in worse outcomes compared to using bacteria. The insecticidal e ciency of recombinant baculovirus expressing Gal-1 was compared with another bio-insecticide, Bacillus thuringiensis, and the results showed that the insecticidal effects (growth and survival rate) of the Gal-1 virus did not differ signi cantly from those of B. thuringiensis (Fig. 6D). Thus, the constructed recombinant virus expressing Gal-1 exhibits potential for practical application as a bio-pesticide.

Discussion
The use of baculoviruses as bio-pesticides has been proposed and tested for over two decades (Moscardi, 1999, Szewczyk et al., 2006. The occluded form of baculovirus is stable in the environment, and virus particles are released after it is ingested by the host to initiate infection (Braunagel et al., 2003). However, a major shortcoming of baculoviruses as bio-pesticides is that they cannot eliminate pests as e ciently as chemical pesticides. Through genetic engineering, recombinant baculoviruses can be generated to carry gene(s) that can help increase the e ciency of the baculovirus as a bio-pesticide (Kitts and Possee, 1993, Summers, 2006, Inceoglu et al., 2001, Stewart et al., 1991. In the present study, a recombinant baculovirus carrying the mammalian galectin-1 gene and expressing the Gal-1 protein vABpRpGal1 was generated, and its potential as an e cient bio-pesticide was evaluated. Single infection with vABpRpGal1 did not increase the mortality of S. litura larvae, but did affect the appetite of the larvae (Fig. 2D). This result may be due to the disruption of the peritrophic matrix in the midgut mediated by galectin-1, which resulted in the larvae being smaller (Fig. 2C) and possibly dying before metamorphosis.
In addition to disrupting the structure of the peritrophic matrix in the midgut, galectin-1 has been shown to suppress the activation of T cells in mammals, a strategy utilized by cancer cells to escape attack from immune response, was suppressed ( Fig. 4A and 4B). Furthermore, a phagocytic activity assay showed that the cell-mediated immune response was also signi cantly suppressed in larvae infected with vABpRpGal1. The mechanism underlying phagocytosis inhibition mediated by galectin-1 in insects is yet to be elucidated, but it may be similar to the inhibition of T cell activation. Suppression of the host immune system not only expedites the baculovirus infection cycle in S. litura larvae but may also increase their vulnerability to other pathogens in the eld. Galectin-1 has been shown to bind to various glycoproteins on the cell surface (Ahmad et al., 2004), and it is likely that physiological functions other than pest immune response are also affected by galectin-1. Further research is needed to fully understand the overall effect of overexpressing galectin-1 on insects.
To increase their e ciency as bio-pesticides, recombinant baculoviruses were engineered to express proteins that would help achieve this goal, and several toxin proteins were tested. A recombinant baculovirus carrying AalT, a scorpion toxin that acts on the nervous system, was able to paralyze larvae and stop them from eating (Mccutchen et al., 1992). Scorpion toxin is a good candidate for enhancing baculovirus e ciency (Maeda et al., 1991, Popham et al., 1998. However, it poses a biohazard risk because the dead larvae may be ingested along with crops, which could induce allergic reactions in humans. In addition to toxins, proteins that are toxic only to insects have also been tested. A recombinant baculovirus expressing the kernel mitochondrial protein URF13 not only suppressed larval growth but also induced death sooner (Korth KL, 1993). Proteins that could disrupt physiological barriers were also tested. Recombinant baculoviruses that express proteases ScathL, from Sarcophaga peregrina, and keratinase, from Aspergillus fumigatus, can degrade tissue structure causing the infected larvae to become anorexic (Gramkow et al., 2010). However, humans that ingest infected larvae may also be affected by these proteases. The midgut is the target for insecticidal crystal proteins of B. thuringiensis, and a recombinant baculovirus expressing insecticidal crystal proteins was shown to enhance the lethality of the baculovirus in host larvae (Ribeiro and Crook, 1993, Chang et al., 2003, Pang et al., 1992, Bai et al., 1993. In most of these previous studies, recombinant baculoviruses were directly injected into the host larvae to ensure infection, and virus infection was initiated in the hemolymph. In the present study, the recombinant baculovirus was administered orally via the ingestion of leaves to simulate eld conditions. When administered orally, the virus must pass through the esophagus to the gut and nally reach the hemolymph. Nevertheless, a mortality rate of over 50% was recorded on day 5 post-infection (Fig. 3A), suggesting that recombinant baculoviruses expressing galectin-1 have potential as biopesticides. Further research is required to determine the effect of galectin-1 on humans and other nontarget organisms once ingested.
The goal of genetically modifying baculoviruses is to replace chemical-based pesticides/insecticides in the eld. In the present study, the recombinant baculovirus expressing galectin-1 reduced the appetite of S. litura, S. frugiperda, and P. xylostella larvae, increased their mortality rate in the eld trial (Figs. 5 and 6), and showed no effect on honeybees. These results strongly suggested that incorporating the galectin-1 gene into the baculovirus genome can signi cantly improve the bio-pesticide potential of baculoviruses in the eld.

Data availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Code availability
Software applications are mentioned in the methods, and programs are publicly available.

Competing interests
The author(s) declare no competing interests.   Effect of 7-day feeding on the appetite, growth, and survival rate of S. litura larvae. (A) Survival rate of S. litura larvae after continuous feeding of vABpR and vABpRpGal1. Second instar larvae were fed with approximately 1 × 106 pfu/larva of vABpR, vABpRpGal1, or sterile water daily for 7 days. Survival rate among experimental groups was analyzed using the log rank test. There were 60 larvae per group on day 1 of the experiment. Control: larvae were fed with sterile water; −Gal-1: larvae were infected with vABpR; +Gal-1: larvae were infected with vABpRpGal1. (B) Body length of larvae from day 1 to day 6. The body length of 10 larvae from each group was measured. The data are presented as mean SD, and body length among groups was analyzed using student's t-test. * = p < 0.05 was considered signi cantly different. (C) Representative larvae from each group on day 1 and day 6 of the experiment. blue uorescence: hemocytes. Quanti cation of phagocytic activity of hemocytes is shown on the right. Hemocytes were collected from ve larvae from each group for the phagocytic activity assay. Student's ttest was used to analyze the difference among experiment groups. * = p < 0.05; ** = p < 0.01. presented as mean SD, and survival rate among groups was analyzed using student's t-test. * = p < 0.05 was considered signi cantly different.