Snellenius manilae bracovirus suppresses the host immune system by regulating extracellular adenosine levels in Spodoptera litura

Sufficient energy supply to the host immune system is important for resisting pathogens. Therefore, during pathogen infection, the host metabolism is reassigned from storage, growth, and development to the immune system. Previous studies in Drosophila melanogaster have demonstrated that systemic metabolic switching upon an immune challenge is activated by extracellular adenosine signaling, modulating carbohydrate mobilization and redistributing energy to the hemocytes. In the present study, we discovered that symbiotic virus (SmBV) of the parasitoid wasp Snellenius manilae is able to down-regulate the extracellular adenosine of its host, Spodoptera litura, to inhibit metabolism switching. The decreased carbohydrate mobilization, glycogenolysis, and ATP synthesis upon infection results in the host being unable to supply energy to its immune system, thus benefitting the development of wasp larvae. When we added adenosine to the infected S. litura larvae, we observed enhanced host immune responses that decreased the pupation rate of S. manilae. Previous studies showed that after pathogen infection, the host activates its adenosine pathway to trigger immune responses. However, our results suggest a different model: we found that in S. manilae, SmBV modulates the host adenosine pathway such that wasp eggs and larvae can evade the host immune response.


Immune suppression by parasitoid symbiont virus, SmBV, via extracellular adenosine signaling.
Previous studies have shown that PDVs inhibit the cellular 27 and humoral immunity [28][29][30] of the host. Therefore, we tested the effects of SmBV on cellular immunity in S. litura by phagocytosis. Results showed that the hemocytes collected from SmBV-infected larvae exhibited decreased phagocytic activity compared with the control, indicating attenuated cellular immunity after SmBV infection (Fig. 1A). S. litura is a natural host for AcMNPV, and AcMNPV was used as a control. Results showed that phagocytosis increased in AcMNPV-infected larvae, indicating that AcMNPV infection induces cellular immune response. Phagocytosis quantitation results showed that only 19% of the hemocytes from SmBV-infected larvae phagocytized FITC-labeled E. coli in comparison to 32% of non-infected larvae and 40% of AcMNPV-infected larvae (Fig. 1B). Previous studies reported that viruses predominantly activate humoral immunity via the Toll-7 pathway 31 . Therefore, we quantified the gene expression levels of the Toll-7 receptor and a downstream antimicrobial protein (Cecropin) after infections. Results showed that gene expression levels of Toll and Cecropin decreased in S. manilae-parasitized or SmBV-infected S. litura (Fig. 1C), suggesting that immune responses of S. litura are inhibited by SmBV.
Previous studies have demonstrated that extracellular adenosine signaling is up-regulated upon pathogenic infections. Hence, to confirm whether extracellular adenosine signaling is affected in S. litura after SmBV infection, we first measured AdoR and Adgf expressions, which indicate the status of GPCR signaling and adenosine metabolism, respectively. We found that the gene expression levels of AdoR and Adgf were reduced in S. manilae-parasitized or SmBV-infected hosts (Fig. 1D). As an extracellular adenosine deaminase, expression of Adgf is tightly modulated following the titer of extracellular adenosine 13,32 . Hence, decreased Adgf expression indicates that extracellular adenosine might be decreased after SmBV infection. To confirm this, we quantified extracellular adenosine levels in the hemolymph of infected larvae. The adenosine level in the hemolymph of the controls was found to be 6.24 pmol/μl; these levels were significantly lower in SmBV-infected and S. manilae-parasitized larvae, 5.47 pmol/μl and 4.7 pmol/μl, respectively (Fig. 1E). These results indicate that the adenosine signaling of S. litura is influenced by SmBV infection.

Down-regulation of systematic metabolism switching by wasp and SmBV infections.
Since adenosine signaling is a known pathway for regulating energy switching from development and storage toward immune response upon an immune challenge 4 , and based on the results in the previous section, we hypothesized that the reduction in adenosine level by SmBV infection might alter the host metabolism, resulting in inhibition of the immune responses of S. litura. Quantification of the glycogen levels in the fat body showed that it increased to 0.48 mg/dl after parasitization by S. manilae ( Fig. 2A). Decreased glycogen phosphorylase (glyp) expression also indicated that glycogenolysis is not activated after wasp parasitization (Fig. 2B). These results suggest that glycogenesis remained in the fat bodies of parasitized hosts. In addition, expression of trehalase (tre) in the hemocytes was decreased, while it remained unchanged in the fat body (Fig. 2C). These results suggest that glycolysis in the hemocyte of infected larvae was down-regulated by S. manilae parasitization. In addition, the downregulation of carbohydrate mobilization was also shown in wasp-and SmBV-infected larvae. The glucose level in hemolymph decreased 24 h after parasitization, and the lowest level was detected after 24 h (Fig. 2D). The glucose level in fat body decreased 12 h post-parasitization and reached the lowest level at 24 h (Fig. 2E). These results indicate that circulative glucose in hemolymph decreases after parasitization by S. manilae, resulting in a reduction in glucose supply to immune cells in the hemolymph for immune responses. To further confirm that SmBV alone was responsible for the reduced glucose supply in the hemolymph, SmBVs were injected into fourth-instar larvae; it was found that the glucose levels decreased significantly in the hemolymph (Fig. 2F) and fat body (Fig. 2G). These results demonstrate that SmBVs play a key role in inhibiting carbohydrate mobilization in the host.
Previous studies using Drosophila infected by wasps or bacteria demonstrated that adenosine signaling initiates glycogen breakdown in the fat body and increases carbohydrate mobilization in the hemolymph during an immune challenge 3,4 . Moreover, the expression of glycolytic genes was shown to be decreased in the fat body but increased in hemocytes; this metabolic switching process is critical for supplying sufficient energy to the hemocytes. Notably, and unlike in the Drosophila model, our results reveal an increased glycogen level and decreased carbohydrate mobilization in S. litura after wasp S. manilae or SmBV infection (Fig. 2). To compare our results, we further measured the expression of glycolytic and TCA cycle genes in the hemocyte and fat body of S. litura after S. manilae parasitization (Fig. 3A). The expression levels of glycolytic and TCA enzymes did not show a significant change or even a decrease after parasitization (Fig. 3B). Conversely, the expression levels of the same set of metabolic genes were increased in the fat body. These results indicate that metabolic switching does not initiate after S. manilae parasitization, so the energy distribution is not switched toward immune responses. In addition, we also found that the ATP titer was decreased after S. manilae parasitization (Fig. 3C) or SmBV infection (Fig. 3D), indicating that ATP synthesis is suppressed by SmBV infection. Since extracellular adenosine could be converted from extracellular ATP, these results correspond to our observation of decreased extracellular adenosine titer after SmBV infection (Fig. 1E).
Carbohydrate metabolism regulated by extracellular adenosine signaling is required for immune response. Our results implied that inhibition of the host immune response after SmBV infection might be due to suppression of its carbohydrate metabolism by the down-regulation of adenosine levels.
To confirm the effect of adenosine on carbohydrate metabolism, insects were fed with 5-mM adenosine; the glucose level (Fig. 4A) and ATP level (Fig. 4B) in hemolymph were found to be higher than that of the control. In addition, the expression levels of the genes involved in glycolysis and the TCA cycle showed significantly increased in the hemocytes after feeding with adenosine, indicating that adenosine promotes carbohydrate metabolism (Fig. 4C). To confirm whether adenosine affects immune responses, SmBV-infected fourth-instar larvae were fed with adenosine at concentrations of 1.65 mM and 5 mM, and their hemocytes subjected to a phagocytosis assay with FITC-labeled E. coli. Results showed that phagocytic activity increased in SmBV-infected larvae given 1.65 mM or 5 mM adenosine compared to control larvae (Fig. 5A). Quantitation results showed that the proportion of phagocytosis increased in larvae given 1.65 mM and 5 mM adenosine by 10% and 20%, respectively, compared with larvae that were infected with virus only. We further tested the encapsulation activity of the hemocytes by using beads after adenosine treatment, and the results showed that the encapsulation activity was significantly increased by 35% in larvae given 5 mM adenosine (Fig. 5B). Moreover, the suppression of Toll-7 and Cecropin expressions by SmBV infection were restored after feeding with 5 mM adenosine (Fig. 5C). These results demonstrate that adenosine induces cellular and humoral immune response in SmBV-infected larvae. The growth inhibition by wasp parasitization was also restored after adenosine treatment, suggesting that artificial feed containing adenosine was able to suppress wasp infection (Fig. 5D). In addition, wasp pupation decreased by 40% compared with the control when 5 mM adenosine was given to S. litura (Fig. 5E). These results confirm that decreased adenosine signaling by SmBV or wasp infection causes the immune suppression, and feeding with adenosine can reverse this immune suppression.
SmBV negatively affects host adenosine pathway. To investigate how SmBV decreases extracellular adenosine levels, we quantitated the gene expression levels of the enzymes involved in adenosine metabolism pathway in SmBV-infected larvae, including Ecto-NTPDase and Ecto-5′-nucleotidase, as well as the adenosine transporter, equilibrative nucleoside transporter (ENT) (Fig. 6A). Results showed that the gene expression levels of Ecto-NTPDase, Ecto-5′-nucleotidase, and ENT1 ( Fig. 6B-D) were lower in SmBV-infected larvae compared to the control. This showed that the enzymes for maintaining the adenosine homeostasis were affected after SmBV infection, thus resulting in reducing the extracellular adenosine concentration in infected larvae.

Effects of adenosine pathway and energy synthesis inhibitors on host immune responses.
Our results show that the application of additional adenosine increases both cellular and humoral immune responses which are suppressed by SmBV infection. Since adenosine is an important signal for activating the immune response via modulating the host energy metabolism, we further examined whether inhibition of adenosine signaling and energy synthesis compromise immune responses upon infection. As SmBV is not able to replicate in the host, we used AcMNPV-infected S. litura cell lines to analyze the effects. Our result showed that the AcMNPV titer significantly decreased in infected-cells treated with adenosine (Fig. 7A). Conversely, the applications of the glycolysis or adenosine transportation inhibitors, 2-deoxy-D-glucose (2DG) and dipyridamole (Dipy) increased the virus titers significantly (Fig. 7B). 2DG is a glucose analog and is considered as a competitive inhibitor of glycolysis. In cells, 2DG competes with glucose for hexokinase. However, 2DG forms 2-DG-6-phosphate, which cannot undergo glycolysis and affects energy production. Dipy is an inhibitor of nucleoside transporters on the cell membrane that can inhibit adenosine transport between cells. Our experiments proved that suppression of adenosine transportation as well as glycolysis of SL1A cells compromise the antivirus responses. Together with our pervious in vivo experiments, we conclude that activation of adenosine signaling is essential for metabolism reprogramming which is important for host antivirus responses. The SmBV is capable of suppressing the adenosine signaling in S. litura resulted in a failure of energy switching and decreased host antivirus responses.

Discussion
Our results showed that phagocytosis and Toll and Cecropin gene expression are suppressed after wasp and SmBV infection (Fig. 1); these results correspond to previous studies proving that PDV infection modulates the host growth and metabolism 22,23,33 as well as inhibiting cellular and humoral immune responses [27][28][29]34 . Several studies have demonstrated the importance of adenosine in regulating physiological reactionsupon cell damage and Ct values were obtained and standardized, and Excel was used to plot the heat map. Red represents an increase in gene expression level, while green represents a decrease in gene expression level. The quantitated metabolic enzymes include Pgi, Pfk, Tpi, Gapdh, Pglym, Eno, Ldh, Cs, Idh, and Scs. All values are shown as the mean ± SD of three replicates. The ATP level was measured in the hemolymph of second-instar larvae 0, 12, 24, 36, and 48 h after wasps (C) or SmBV infection (D) (1 × 10 5 copies/larva). All values are shown as the mean ± SD of four replicates for ATP measurements, and P-values were calculated using Student's t-test (*p-value < 0.05, **p-value < 0.01; ***p-value < 0.005).

Figure 4. Addition of adenosine increases carbohydrate metabolism. Analysis of glucose (A) and ATP (B) levels
in the hemolymph of fourth-instar larvae after feeding with artificial diet containing 5 mM adenosine. (C) qPCR was used to quantify the gene expression levels of metabolic enzymes in hemocytes of fourth-instar larvae after feeding with 5 mM adenosine. At least three repetitions were conducted for each group. All values are shown as the mean ± SD, and P-values were calculated using Student's t-test (*p-value < 0.05). (2020) 10:2096 | https://doi.org/10.1038/s41598-020-58375-y www.nature.com/scientificreports www.nature.com/scientificreports/ pathogen infection in mammals and Drosophila 4,35-38 , and it can act as a signal for metabolic switching to provide energy to the immune system and/or cell repair. The function of adenosine in mammals is to modulate the inflammatory and immune processes; ischaemia, hypoxia, inflammation, and oxidative/nitrosative stress can all stimulate its release from any type of cell 39 . In Drosophila, it was found that pathogen infection increased adenosine levels 40 , leading to increased glucose and trehalose levels to provide energy to hemocytes to differentiate into functional immune cells to activate immune responses 4,37 . Our study revealed that after S. litura was infected by SmBV, its extracellular adenosine level decreased and the gene expression levels of AdoR and Adgf decreased significantly (Fig. 1D,E). It has been described that blocking AdoR signaling under non-infectious conditions decreases prohemocyte differentiation in the lymph gland 11 . Taking together, decreasing in adenosine signal might be one of the strategies for SmBV to suppress the host immune response.
In the present study, our resutls showed that glycogen level was higher and expression of GP was lower in the fat body of SmBV-infected larvae in comparing to non-infected larvae ( Fig. 2A). This indicates that the efficiency of converting glycogen to glucose in the S. litura fat body is reduced after SmBV infection 41,42 . In addition, the glucose content in the hemolymph was significantly lower after infection (Fig. 2D). Previous studies on Galleria mellonella showed decreased carbohydrate levels in G. mellonella pupae 24 h after parasitization with PDV-containing Pimpla turionellae 43 . A similar phenomenon was also observed in SmBV-infected S. litura. However, after S. litura was infected by SmBV, the glucose level in both hemolymph and fat bodies was restored after 48 h (Fig. 2D,E). Restoration of the carbohydrate level in the host after initial suppression upon infection may be essential for the growth and development of wasp larvae 44 .
In the present study, the extracellular adenosine level was found to decreases after SmBV infection of S. litura; this suppression leads to reduced immune responses as demonstrated by a phagocytic activity assay of the hemocytes (Fig. 1A). While oral administration of adenosine in infected S. litura increased the phagocytic activity (Fig. 4A). Oral administration was chosen in this study because injection could cause injury, which would interfere with the subsequent physiological analyses. To confirm that the orally-administered adenosine reached the hemolymph, the amount of adenosine in hemolymph was measured; it was confirmed to increase significantly (data not shown).
Adenosine synthesis mainly occurs through two pathways. The first pathway occurs when cells are under metabolic stress (reduction in the intracellular ratio of ATP and AMP) 45,46 , where ATP in cells is metabolized to form adenosine which may further be transported to extracellular space via ENTs 47,48 . The second pathway occurs when ATP/ADP are released from damaged cells upon stresses, they will be converted to adenosine through Ecto-NTPDase and Ecto-5′-NT 49 . However, when S. litura was infected by SmBV, it was found that not only extracellular ATP levels (Fig. 3C,D) but also expression levels of Ecto-NTPDase, Ecto-5′-NT, and transporter ENT1 are decreased (Fig. 6B-D), indicating that SmBV affects host adenosine metabolism. At present, it is still unclear how SmBV affects the host adenosine pathway. Previous studies have indicated that PDVs contain many non-coding regions that can produce miRNAs 50,51 , and that these miRNAs can regulate the host physiological responses 52,53 . Preliminary prediction results of SmBV miRNAs showed that seven miRNAs produced by SmBVs may be capable of inhibiting enzymes involved in the adenosine pathway (Fig. S1). It is therefore hypothesized that SmBVs may  www.nature.com/scientificreports www.nature.com/scientificreports/ produce miRNAs to modulate adenosine metabolic genes of the host, which may contribute to a decreased adenosine level in SmBV-infected S. litura. Further experiments will be performed to test this hypothesis.
Previous studies have demonstrated that PDV is crucial in immunosuppression upon wasp infection; the exact mechanisms are unclear. Our results demonstrate that SmBVs are able to suppress the host carbohydrate metabolism and immune responses by negatively regulating extracellular adenosine signaling (Fig. 8). To our knowledge, this is the first description of the mechanism by which viruses inhibit the host immune system by manipulating adenosine pathway which further compromises the host energy metabolisms.

Methods
Insects. S. litura were reared in cages and kept in an environmental chamber at 29 ± 1 °C with a light:dark cycle of 12:12 h. The formula of artificial diet was from a previous study 54 . S. manilae were reared in growth chambers at 26 ± 1 °C with a light:dark cycle of 12:12 h; 15% glucose was used for feeding 17 . Virus and cell. Purified SmBV was obtained by first resecting the calyxes of adult female S. manilae. The calyxes were then placed in TC-100 medium at 4 °C and homogenized using a pestle and mortar. The homogenate was centrifuged at 3000 rpm for 5 s. Then, the samples were removed, the supernatant was aspirated and a 0.45-μm syringe filter was used for purification. Finally, qPCR was used to quantitate virus copy numbers 55,56 . AcMNPV was cultured using the IPLB-Sf-21 cell line and qPCR was used to quantitate virus titers 57 . S. litura SL1A cells were cultured in TC-100 medium (USBio) containing 10% FBS (Gibco BRL) and cultured in a 26 °C incubator. SL1A cells (2 × 10 5 ) were preincubated for 2 hours with Dipy (20 μM), 2DG (10 mM), or adenosine (100 μM). Subsequently, the cells were infected with AcMNPV at a multiplicity of infection (MOI) of 1. The supernatants were harvested to detect viral titers at 48 hours post-infection. www.nature.com/scientificreports www.nature.com/scientificreports/ Phagocytosis assay. Fourth-instar larvae were injected with SmBV (1 × 10 5 copies/larva). After 36 h, hemocytes (4 × 10 4 cells/well) were collected and FITC-labeled E. coli (4 × 10 4 cells/well) with TC-100 culture medium were added in a total volume of 100 μl. The cells were cultured at room temperature for 60 min and subsequently washed 2-3 times with 10% phosphate-buffered saline (PBS). Then, 50 μl of 0.4% trypan blue was added and the cells stained for 10 min before washing 2-3 times with 10% PBS. Next, 4% formaldehyde was added and incubated for 30 s to fix the samples. Finally, 50 μl of DAPI (Thermo Fisher) was added, and the cells were incubated for 30 min and washed 2-3 times with 10% PBS. The proportion of FITC-labeled E. coli and hemocytes was counted by fluorescence microscopy 58 . Nucleic acid extraction and RT-qPCR. S. litura tissues were extracted using Trizol (Geneaid); detailed procedures have been described previously 59 . Briefly, a PrimeScript TM RT Reagent Kit (Takara) was used for cDNA synthesis; 500 ng of RNA was collected and 2 μl of 5 × PrimerScript TM buffer, 0.5 μl of PrimerScript RT Enzyme Mix I, 0.5 μl of Oligo dT primer, and 0.5 μl of random hexamers were added. A spectrophotometer (Nanodrop 2000) was used to measure the optical density of the RNA (260/280 nm). SYBR green (Bioline) and an ABI Plus One real-time system (StepOnePlus, Applied Biosystems) were used for qPCR. All the gene expression levels were normalized to the expression level of 18S. A list of primer sequences used in this study is given in Table S1.
Glycogen, glucose, and adenosine measurement. The levels of glycogen, glucose, and adenosine were determined in S. litura hemolymph using colorimetric methods with a glycogen assay kit (MET-5022, Cell Biolabs Inc.), glucose assay kit (Cell Biolabs Inc.), and adenosine assay kit (Cell Biolabs Inc.), respectively. Detailed procedures have been described previously 60 .
Encapsulation assay. Hemolymph samples from fourth-instar larvae were collected and each mixed with 1 ml of Pringle's saline 61 . The hemocytes were spun down by centrifugation at 500 × g at 4 °C for 5 min. After washing the hemocytes two times with Pringle's saline, they were resuspended in 1 ml of saline; a hemocytometer was used to calculate the hemocyte density. Fifty microliters of the suspended hemocytes and 5 μl of Sephadex A-25 beads were added to each well of a 96-well plate 62 . Parafilm was used to seal the plates before incubation at 26 °C for observation of the encapsulation status. ATP analysis. The ATP level was quantitated by an ATP Determination Kit (A22066, Invitrogen). Ten microliters of hemolymph and 90 μl of reaction solution containing ddH 2 O, reaction buffer, dithiothreitol (DTT), D-Luciferin, and luciferase were mixed in a total volume of 100 μl. After reaction in the dark for 15 min, a luminometer was used to measure the luminescence at an absorbance wavelength of 560 nm. A standard curve was used to convert luminescence into ATP content.
Adenosine feeding test. Newly hatched S. litura were given an artificial diet containing 1.65 mM or 5mM adenosine. After 7 to 8 days, eighty second-instar larvae were infected by ten S. manilae females for 48 h. The number of S. manilae that successfully pupated was calculated and regarded as viable S. manilae under host immune responses.

Statistical analysis.
The CT values obtained from the qPCR were used to obtain relative expression levels using the 2 −ΔΔCt formula; 18S was used as the reference gene 56,63 . Each treatment group was compared with the control; Student's t-test was used to test for significant differences. A heat map was plotted using the logarithm of the standardized Ct values (Log 10 ) using Microsoft Excel 59 .