Insect immune resolution with EpOME/DiHOME and its dysregulation by their analogs leading to pathogen hypersensitivity

Abstract

Upon immune challenge, recognition signals trigger insect immunity to remove the pathogens through cellular and humoral responses. Various immune mediators propagate the immune signals to nearby tissues, in which polyunsaturated fatty acid (PUFA) derivatives play crucial roles. However, little was known on how the insects terminate the activated immune responses after pathogen neutralization. Interestingly, C20 PUFA was detected at the early infection stage and later C18 PUFAs were induced in a lepidopteran insect, Spodoptera exigua. This study showed the role of epoxyoctadecamonoenoic acids (EpOMEs) in the immune resolution at the late infection stage to quench the excessive and unnecessary immune responses. In contrast, dihydroxy-octadecamonoenoates (DiHOMEs) were the hydrolyzed and inactive forms of EpOMEs. The hydrolysis is catalyzed by soluble epoxide hydrolase (sEH). Inhibitors specific to sEH mimicked the immunosuppression induced by EpOMEs. Furthermore, the inhibitor treatments significantly enhanced the bacterial virulence of Bacillus thuringiensis against S. exigua. This study proposes a negative control of the immune responses using EpOME/DiHOME in insects.

Graphical abstract

1. Introduction

Epoxyoctadecamonoenoic acids (EpOMEs) including coronaric acid (9,10-EpOME) and vernolic acid (12,13-EpOME) are derived from linoleic acid (LA) by the catalytic activities of specific cytochrome P450 monooxygenases (CYPs) belonging to CYP2C and CYP2J subfamilies in mammalian neutrophils and macrophages (Hildreth et al., 2020). EpOMEs mediate relevant inflammatory response upon pathogen infection (Viswanathan et al., 2003). However, excessive EpOMEs are associated with acute respiratory distress syndrome in humans and in rodent models presumably mediated via corresponding diols, 9,10-dihydroxy-octadecamonoenoate (9,10-DiHOME) and 12,13-dihydroxy-octadecamonoenoate (12,13-DiHOME) (Hayakawa et al., 1990; Thompson and Hammock, 2007). The DiHOMEs can be detoxified using selective inhibitors of the mammalian soluble epoxide hydrolase (sEH) (McReynolds et al., 2021; Bergmann et al., 2022).

In insects, these EpOMEs and DiHOMEs have been speculated to play a crucial role in the immune response (Xu et al., 2015). Indeed, they were detected in all the developmental stages of the mosquito, Culex quinquefasciatus (Xu et al., 2015). Feeding sEH inhibitor increased EpOME levels in the mosquito midgut and reduced the total number of bacteria in the lumen, suggesting that EpOMEs were associated with insect immunity (Xu et al., 2017). The role of EpOMEs in insect immune response was further analyzed in the immune-challenged larvae of a lepidopteran insect, Spodoptera exigua, which exhibited relatively high levels of 941.8 pg/g of 9,10-EpOME and 2,198.3 pg/g of 12,13-EpOME (Vatanparast et al., 2020). In the insect species, a specific cytochrome P450 enzyme (CYP, SE51385) and a soluble epoxide hydrolase (Se-sEH) mostly catalyzed the production and hydrolysis of EpOMEs, respectively. These two regioisomers of EpOMEs suppress the cellular and humoral immune responses. Their immunosuppressive effects were further supported by treatment with urea-based sEH inhibitors, which inhibited the cellular and humoral immune responses of S. exigua probably by increasing the levels of EpOMEs in response to immune challenge. Thus, EpOMEs facilitate anti-inflammatory response in the insect. However, little is known regarding the role of EpOMEs and their association with pro-inflammatory agents.

DiHOMEs are also produced in neutrophils in humans following bacterial challenge (Ozawa et al., 1988; Zhang et al., 1995). Although DiHOMEs are the metabolites of EpOMEs, they act as mediators of neutrophil chemotaxis (Totani et al., 2000). High concentrations of DiHOMEs under physiological stress such as following severe skin burn or COVID infection can lead to toxicity by acting as isoleukotoxins, resulting in severe dysfunction of the innate immune responses (Bergmann et al., 2022). DiHOMEs also suppress excessive production of reactive oxygen species in neutrophils (Thompson and Hammock, 2007). However, studies investigating the physiological role of DiHOMEs in insects have yet to be reported.

This study investigated the physiological role of C18 oxylipins in attenuating the immune response in insects by analyzing the functional relationship between EpOMEs and pro-immune mediators as well as EpOME metabolism. The physiological roles of the resulting DiHOMEs in insect immunity were also investigated. The physiological role of C18 oxylipins in immune resolution was determined by analyzing the effects of DiHOME analogs or sEH inhibitors on dysregulation of the immune responses and susceptibility to an insect pathogen.

2. Materials and Methods

2.1. Insect rearing

Larvae of S. exigua were collected from Welsh onion (Allium fistulosum L.) fields in Andong, Korea. The larvae were reared under laboratory conditions (27 ± 1°C, 16:8 h (L:D), and 60 ± 5% relative humidity) with an artificial diet (Goh et al., 1990). They underwent five instars (L1~L5) before pupation. A diet of sugar solution (10%) was fed to adults and an empty Petri dish was provided for oviposition.

2.2. Chemicals

9,10-EpOME, 12,13-EpOME, 9,10-DiHOME, 12,13-DiHOME, LA, and prostaglandin E₂ (PGE₂) were purchased from Cayman (Ann Arbor, MI, USA). A caspase inhibitor Z-DEVD-FMK (DEVD) was purchased from Sigma-Aldrich Korea (Seoul, Korea). Other EpOME derivatives (diEpOME, THOME, and THF-diol) were prepared as described in previous studies involving the synthesis of oxygenated LA metabolites (Moghaddam et al., 1996; Newman et al., 2002). sEH inhibitors (AUDA, CUDA, 1661, 1663, 1659, and 1213) were prepared as reported previously for the synthesis of urea-pharmacophore-based inhibitors of the sEH enzyme (Morisseau et al., 2002; Singh et al., 2021). Four EpOME mimics (A839-A842) were generated via oxymercuration-demercuration, as previously described for the synthesis of arachidonic acid-derived epoxyeicosatrienoic acid (EET) mimics (Kim et al., 2004), using methyl LA with an appropriate alcohol as the solvent (methanol, ethanol, n-propanol or iso-propanol, respectively), followed by purification on silica gel to obtain the corresponding mono-alkoxide as a mixture of four regioisomers. The free acids (A843-A846) were obtained by basic hydrolysis of the corresponding methyl esters (A839-A842, respectively). The synthesis of A839-A846 is described in the Supporting Information (online suppl. Figs. S2 and S3). An anticoagulant buffer (ACB, pH 4.5) was prepared with 186 mM of NaCl, 17 mM of Na₂EDTA, and 41 mM of citric acid. Phosphate-buffered saline (PBS) was prepared with 100 mM phosphoric acid and the pH was adjusted to 7.4 using 1 N NaOH.

2.3. Hemocyte-spreading behavior assay

Two-day-old L4 larvae of S. exigua were used to analyze the hemocyte-spreading behavior. Approximately 250 μL of hemolymph was collected from five larvae and mixed with 750 μL of cold ACB, and incubated on ice for 20 min. The diluted hemolymph was then centrifuged at 800×g for 5 min at 4°C to obtain the pellet, which was re-suspended in 300 μL of filter-sterilized TC-100 insect cell culture medium (Welgene, Gyeongsan, Korea). Next, 9 μL of hemocyte suspension was incubated with 1 μL of a test chemical on a glass coverslip for 40 min at room temperature (RT). After incubation, hemocytes were fixed with 4% paraformaldehyde for 10 min at 25°C, followed by washing three times with filter-sterilized 1× PBS. Hemocytes were then permeabilized with 0.2% Triton-X in PBS for 2 min at 25°C. Hemocytes were then washed three times, followed by incubation with 5% skim milk for 10 min at 25°C and subsequently with fluorescein isothiocyanate (FITC)-tagged phalloidin in PBS for 60 min. After washing three times, the cells were incubated with 4´,6-diamidino-2-phenylindole (DAPI, 1 mg/mL). Finally, after washing twice in PBS, the cells were observed under a fluorescence microscope (DM2500, Leica, Wetzlar, Germany) at 400 × magnification. Hemocyte-spreading was determined based on the extension of F-actin growth beyond the original cell boundary of hemocytes. The cell spread was scored by counting the cells exhibiting F-actin growth in 100 randomly selected cells. Each treatment was performed in triplicate using independent hemocyte preparations.

2.4. RNA extraction, cDNA preparation, and qPCR

To extract RNA, the intestine was removed from the larvae to avoid any contamination due to non-target organisms. Total RNAs were extracted from the larvae using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. The extracted RNA was used to synthesize complementary DNA (cDNA) using RT‐premix (Intron Biotechnology, Seoul, Korea) containing an oligo-dT primer. The cDNA was quantified using a spectrophotometer (NanoDrop, Thermo Fisher Scientific, Wilmington, DE, USA). The cDNA (80 ng per μL) was used as a template for quantitative PCR (qPCR) using gene-specific primers (online suppl. Table S1). The qPCR was performed using SYBR Green real‐time PCR master mixture (Toyobo, Osaka, Japan) as described by Bustin et al. (2009) using a real‐time PCR system (Step One Plus Real‐Time PCR System, Applied Biosystems, Singapore). The reaction mixture (20 μL) contained 10 μL of Power SYBR Green PCR Master Mix, 1 μL of cDNA template (80 ng), and 1 μL each of forward and reverse primers, and 7 μL of deionized distilled water. The qPCR was initiated by heat treatment at 95°C for 10 min, followed by 40 cycles of denaturation at 94°C for 30 s, annealing at 52°C for 30 s, and extension at 72°C for 30 s. A ribosomal protein gene, RL32, was used as an endogenous control. Each treatment was performed in triplicate with independent samples. Expression analysis of qPCR was performed using a comparative CT method (Livak and Schmittgen, 2001).

2.5. AMP gene expression analysis

The levels of 10 antimicrobial peptide (AMP) genes were analyzed using L4 larvae after an immune challenge with Escherichia coli (1.73×10³ cells/larva) injection into the hemocoel. The insects were also injected with test compound(s) along with E. coli to assess any effect on AMP induction. Total RNA was collected as described above 8 h after treatment. After cDNA synthesis, qPCR was performed as described above using gene-specific primers (Table S1). The ribosomal protein gene, RL32, was amplified as the housekeeping gene serving as the internal standard in the qPCR assay. All samples were analyzed in triplicate.

2.6. Cellular immune assay based on hemocytic nodule formation

L4 larvae were used in bioassays to determine hemocyte nodule formation. E. coli was injected at a dose of 1.73×10³ cells per larva through the proleg and incubated at room temperature (RT) for 8 h. The insects were also injected with test compound(s) along with E. coli cells to assess effects on nodulation. After the dissection of test larvae, the melanized nodules were counted under a stereoscopic microscope (Stemi SV11, Zeiss, Jena, Germany) at 50× magnification. Each treatment was performed in triplicate.

2.7. Cytotoxicity test based on cellular blebbing

Cell blebs on the hemocyte membrane are symptoms of cytotoxicity as reported previously by Cho and Kim (2004). A test compound (1 μg per larva) was injected into L4 larvae and incubated at RT for 24 h. The hemolymph was collected into an ACB as described above. Subsequently, a hemocyte suspension (10 μL) was placed on a slide glass. The number of cell blebs were counted from 100 randomly selected cells under a phase contrast microscope (BX41, Olympus, Tokyo, Japan) at 100× magnification. Each assessment was performed in triplicate with independent insect samples.

2.8. Terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) assay of apoptosis

A TUNEL assay to determine apoptosis was performed using an in situ Cell Death Detection kit (Abcam, Cambridge, UK). For this assay, 1 μg of test chemical was injected into L4 larvae and incubated for 24 h under the rearing conditions described above. After incubation, hemolymph was collected into ACB as described above. Subsequently, 10 μL of hemocyte suspension was mixed with 1 μL of 10 μM 5-bromouridine (BrdU) solution containing terminal transferase. The mixture was then placed on cover glass in a wet chamber. After 1 h incubation at RT, the medium was replaced with 4% paraformaldehyde and incubated at RT for 10 min. The cells were washed with PBS, and treated with 0.3% Triton-X in PBS, followed by incubation for 2 min at RT. After blocking with 5% bovine serum albumin in PBS for 10 min, cells were incubated with mouse anti-BrdU antibody (diluted 1:15 in blocking solution) for 1 h at RT. After removing the unbound anti-BrdU antibody, cells were incubated with FITC-conjugated anti-mouse IgG antibody (diluted 1:500 in blocking solution) for 1 h. After washing with PBS, 10 μL of DAPI was added and incubated at RT for 5 min. Cells were then washed with 10 μL of PBS. Glycerol: PBS (1:1) solution (10 μL) was then added to cells on a cover glass, which was then placed on a slide glass for observation under a fluorescence microscope (Leica) in FITC mode. Each treatment was replicated three times. Each replication represented an independent sample preparation.

2.9. Insecticidal bioassay

To determine the bacterial virulence against S. exigua, the insect pathogen, Bacillus thuringiensis ssp. aizawai (Bt), was cultured in tryptic soy broth (TSB) (Difco, Detroit, MI, USA) at 28°C. After 48 h, the cultured Bt suspension was treated at 4°C for another 48 h for sporulation. After centrifugation at 8,000 rpm for 15 min, the cell pellets were harvested and freeze-dried. Bioassay was performed using the leaf-dipping method. Briefly, a piece of cabbage leaf (2 cm²) was dipped into Bt suspension (10⁸ spores/mL) for 5 min. After removing the excess suspension, the treated cabbage was placed on a filter paper in a Petri dish (90 × 15 mm). Mortality was assessed at 4 days after treatment. Each treatment (= 10 larvae) was performed in triplicate.

2.10. Phenoloxidase (PO) activity

PO activity was measured using L-3,4-dihydroxyphenylalanine (DOPA) as a substrate. Each L4 larva was treated with a test chemical (1 μg) along with E. coli (1.73×10³ cells/larva). After 8 h, hemolymph was collected from the treated larva, and the plasma was separated by centrifugation at 800×g for 5 min at 4°C. The reaction mixture (200 μL) consisted of 10 μL of plasma, 4 μL of test compound, 10 μL of DOPA, and 176 μL of PBS. Absorbance was assessed at 495 nm using a VICTOR multi-label Plate reader (PerkinElmer, Waltham, MA, USA). The activity was expressed as change in absorbance per min per plasma (ABS/min/μL). Each treatment was replicated three times.

2.11. Insect sample preparation to quantify EpOMEs and DiHOMEs

Two-day-old L4 larvae of S. exigua were used for extraction of PGE₂, EpOMEs, and DiHOMEs. The tissue samples were collected from the whole body without gut to avoid exogenous contamination. E. coli (1.73 × 10³ cells/larva) was injected into the insect hemocoel through the proleg. An insect body sample was collected from ~0.85 g of larvae at different time points (0 ~ 48 h) into a 15 mL tube and washed three times with chilled PBS. The tissue preparation was performed in triplicate. Each sample was homogenized three times (10 min per cycle) in PBS with an ultrasonicator (Bandelin Sonoplus, Berlin, Germany) at 80% power, and subsequently adjusted to a pH of 4.0 using 1 N HCl. The sample was mixed with 1 mL ethyl acetate to obtain an organic hyper-phase. The aquatic phase was extracted twice again with ethyl acetate. The combined ethyl acetate extracts containing free fatty acids or their derivatives were concentrated to ~500 μL under a gentle nitrogen stream and applied to a small silicic acid column (2 × 90 mm containing 30 mg of Type 60A, 100–200 mesh silicic acid, Sigma-Aldrich Korea). Extracts were sequentially eluted with 250 μL of increasingly polar solvents starting with 100% ethyl acetate, followed by ethyl acetate and acetonitrile (1:1, v/v), acetonitrile and methanol (1:1, v/v), and 100% methanol. All the fractions were used to quantify PGE₂, EpOMEs, and DiHOMEs. Each treatment was replicated three times with independent sample preparation.

2.12. Measurement of PGE₂, EpOMEs, and DiHOMEs using LC-MS/MS

LC-MS/MS was conducted using a QTrap 4500 (AB Sciex, Framingham, MA, USA) equipped with an auto-sampler, a binary pump, and a column oven. The analytical column was an Osaka Soda (Osaka, Japan) C18 column (2.1 mm × 150 mm, 2.7 μm) maintained at 40°C. The mobile phases consisted of 0.1% formic acid in water and 0.1% formic acid in acetonitrile. The linear gradient was: 30% B at 0 min, 30% B at 2 min, 65% B at 12 min, 95% B at 12.5 min, 95% B at 25.0 min, 30% B at 28.0 min, and 30% B at 30 min. The flow rate was 0.40 mL/min. The auto sampler was set at 5°C and the injection volume was 10 μL. LC-MS/MS was equipped with an electrospray ionization (ESI) source. ESI was performed in negative ion mode. The source parameters after optimization were: source temperature, 600°C; curtain gas flow rate, 32 L/min; ion source gas flow rate, 60 L/min; and the adjusted spray voltage, −4,000 V. Analyses were performed in multiple reaction monitoring detection modes using nitrogen as collision gas. MassView1.1 software (AB Sciex, Seoul, Korea) were used for peak detection, integration, and quantitative analysis.

2.13. Statistical analysis

Continuous dependent variables were subjected to one-way analysis of variance (ANOVA) using PROC GLM in the SAS program (1989). Mortality data were transformed by arsine for normalization. All experiments were performed in three independent replicates. The means ± standard errors (SD) were plotted using the Sigma Plot (Systat Software, Point Richmond, CA, USA). The means were compared with the least significant difference (LSD) test involving a Type I error of 0.05.

3. Results

3.1. Sequential synthesis of PGE₂, EpOMEs, and DiHOMEs in immune-challenged larvae of S. exigua

C18 and C20 oxylipins such as EpOME and PGE₂ are synthesized in insects using specific enzymes (Fig. 1A). Bacterial challenge with E. coli induced the expressions of three phospholipase A₂ (PLA₂) genes, EpOME synthase (CYP) gene, and Se-sEH gene in S. exigua larvae (Fig. 1B). Among three PLA₂ genes, two intracellular and calcium-independent PLA₂ (iPLA₂A and iPLA₂B) genes were quickly induced at 0.5 h post-injection (pi) compared to secretory PLA₂ (sPLA₂) gene exhibiting its maximal expression peak at 2 h pi. In contrast, the maximal peaks of CYP and Se-sEH were detected 4 h and 12 h pi, respectively. In addition, CYP was expressed earlier (F = 5.70; df = 1, 9; P < 0.0001) than Se-sEH. Interestingly, only PGE₂ injection induced the gene expression of CYP and Se-sEH (Fig. 1C).

Fig. 1.

Expression profiles of phospholipase A₂ (PLA_2, GenBank accession number: MH061374.1), EpOME synthase (CYP, MT375603.1), and soluble epoxide hydrolase (sEH, MT375604.1). (A) Biosynthetic pathways of C18 and C20 oxylipins. Phospholipid (PL) is catalyzed by PLA₂ to release linoleic acid (LA). LA is then elongated by elongase (ELO) and desaturated by desaturase (DES) to produce arachidonic acid (AA) (Hasan et al., 2019). AA is oxygenated by a specific peroxidase called peroxynectin (Pxt) to produce prostaglandin E₂ (PGE₂). LA is oxidized by a specific cytochrome P450 monooxygenase (CYP) to produce EpOME, which is hydrolyzed by a soluble epoxide hydrolase (sEH) to DiHOME. (B) Inducible expressions of two intracellular and calcium-independent PLA₂ (iPLA₂A and iPLA₂B), a secretory PLA₂ (sPLA₂), CYP, and sEH genes in response to bacterial challenge. E. coli was injected into the larval proleg (1.73 × 10³ cells/larva). After 8 h incubation, the total extracted RNA from the fat body was used to determine the mRNA expression level. Different levels of expression in RT-qPCR samples were normalized using the expression levels of a ribosomal protein, RL32. There were no significant changes in their expression levels in the larvae without the bacterial challenge (Fig. S5A). (C) Inducible expressions of CYP and sEH after PGE₂ injection. PGE₂ was injected into each larva at a dose of 1 μg. There were no significant changes in their expression levels in the larvae in control (Fig. S5B). (D) Quantification of PGE₂ and EpOMEs/DiHOMEs in the whole body without the gut in S. exigua larvae. The samples were collected from the challenged larvae, in which L4 larvae were challenged with E. coli (1.73 × 10³ cells/larva). After injection, the total body except the gut was collected at different time points and used to measure PGE₂, EpOME, and DiHOME levels using LC–MS/MS. Each assessment was replicated with three independent samples.

LC-MS/MS was used to analyze the concentrations of PGE₂, EpOMEs, and DiHOMEs in immune-challenged larvae at different time points (Fig. 1D). To avoid exogenous dietary contamination, the entire gut was removed from the whole body preparation before extracting the oxylipins. The immune challenge significantly increased the levels of PGE₂ and EpOMEs/DiHOMEs. The induced levels of PGE₂ measured by LC-MS/MS were detected earlier (0.5 ~ 2 h pi) than those (> 12 h pi) of EpOMEs/DiHOMEs. However, the up-regulation of PGE₂ level was earlier than the induction time of the PLA₂ expression, which would be responsible for the PGE₂ biosynthesis. This might be the time lag between the sample preparation to extract PGE₂ and LC-MS/MS analysis. In contrast, the RT-qPCR did not need the relatively long time difference between RNA extraction and qPCR. The induced levels of the two EpOMEs did not differ significantly (F = 0.26; df = 1, 9; P = 0.6138). However, the concentrations of the two DiHOMEs showed statistically significant differences (F = 17.46; df = 1, 9; P < 0.0001) following induction, and 12,13-DiHOME was detected at higher levels than 9,10-DiHOME. Even though both C18 oxylipins were detected in a relatively similar late phase (after 8 h post-infection) of the immune responses, 12,13-DiHOME was detected earlier than its precursor, 12,13-EpOME. This might be explained by a rapid turnover of the EpOMEs into DiHOMEs and subsequent accumulation of the inactive products because the DiHOME amounts were more than those of EpOMEs.

3.2. EpOMEs, but not DiHOMEs, suppress immune response

To investigate the role of EpOMEs/DiHOMEs in insect immunity, hemocyte-spreading behavior was assessed by adding them to the hemocyte suspension (left panel in Fig. 2A). Control hemocytes spread via cytoplasmic extensions along with increased levels of F-actin. However, the addition of 12,13-EpOME inhibited hemocyte spread in a dose-dependent manner (Fig. 2B). Treatment with LA, its biosynthetic precursor, also slightly suppressed the hemocyte spread. However, 12,13-DiHOME did not suppress even at the highest test concentration. Compared with 9,10-EpOME, 12,13-EpOME exhibited higher suppression of hemocyte-spreading behavior. However, none of the regioisomeric DiHOMEs inhibited (F = 2.92; df = 1, 4; P = 0.1625) the spread of hemocytes.

Fig. 2.

Contrasting effects of EpOMEs and DiHOMEs on immune responses of S. exigua. (A) Hemocyte-spreading behavior (left) and nodule formation (right) after exposure to test compounds including linoleic acid (LA). Hemocytes were stained with FITC against F-actin and DAPI against nucleus. Nodules are indicated by arrows. The scale bar represents 10 μm. (B) Effect of different doses of test compounds on hemocyte-spreading behavior. The cellular spread was scored by counting the number of cells exhibiting F-actin growth out of cell boundary among 100 randomly selected cells. The right panel compares the effects of a single dose (1 μg) of DiHOME and EpOME in each regioisomer on hemocyte spread. (C) Effect of different doses of test compounds on nodulation. Nodule formation was determined by dissection at 8 h after injection with E. coli (1.73×10³ cells/larva) and/or test compounds. For control (‘CON’) treatment, only E. coli was injected. Right panel shows a comparison of nodule formation after exposure to a single dose as described above. (D) Upregulation of phenoloxidase (PO) activity at 8 h after PGE₂ (1 μg per larva) treatment. (E) Downregulation of PO activity with test compounds (μg) injected along with the bacteria. (F) Effects of EpOMEs and DiHOMEs on the expression of 10 antimicrobial peptide (AMP) genes: apolipophorin III (AL), defensin (Def), gallerimycin (Gal), gloverin (Glv), lysozyme (Lyz), transferrin I (T1), transferrin II (T2), attacin 1 (A1), attacin 2 (A2), and cecropin (Cec). AMP gene expression was determined via RT-qPCR at 8 h after injection with E. coli (1.73×10³ cells/larva) and/or test compounds. Each treatment was replicated three times with individual tissue preparations. ‘NS’ stands for no significant difference between two treatments. Asterisks or different letters above standard error bars indicate significant differences among means with Type I error of 0.05 (LSD test).

Nodule formation (right panel in Fig. 2A) was assessed to test the effect of oxylipin on the cellular immune response of S. exigua. Bacterial challenge resulted in the formation of about 55 nodules per larva in the hemocoel (Fig. 2C). Both regioisomeric EpOMEs significantly suppressed the nodule formation, although 12,13-EpOME was more potent than 9,10-EpOME and exhibited dose-dependent inhibition. LA also suppressed the nodule formation but was less potent. However, none of the DiHOMEs exhibited any inhibitory activity (F = 1.80; df = 1, 4; P = 0.2508) against the cellular immune response.

Melanization occurs in cellular immune responses via upregulation of PO activity in S. exigua (Shrestha and Kim, 2008). PGE₂ treatment induced the upregulation of PO activity in a dose-dependent manner (Fig. 2D). However, 12,13-EpOME significantly suppressed the PO activity, whereas 12,13-DiHOME did not alter PO activity (Fig. 2E)

To test the influence of DiHOMEs on the humoral immune responses, 10 AMP genes induced by immune challenge were assessed (Fig. 2F). Both EpOME regioisomers suppressed the expression of a few AMPs following immune challenge. The regioisomer 9,10-EpOME suppressed attacin 1 and cecropin genes while 12,13-EpOME suppressed attacin 1, attacin 2, and gloverin. Interestingly, both DiHOMEs upregulated the expression of some AMPs, in which 12,13-DiHOME upregulated the four AMPs suppressed by the EpOMEs. In addition, either of two DiHOMEs upregulated the expression of defensin, gallerimycin, or lysozyme genes.

3.3. EpOMEs exhibit hemolytic activity at low concentrations while DiHOMEs do not

The analysis of hemocyte behavior in response to 12,13-EpOME treatment revealed cellular blebbing (Fig. 3A). However, 12,13-DiHOME treatment had limited effect on cell blebbing in hemocytes. The cell blebbing was further analyzed under different concentrations of EpOMEs/DiHOMEs (Fig. 3B). Both EpOMEs significantly induced cell blebbing: 12,13-EpOME was more potent than 9,10-EpOME and exhibited hemolytic activity in a dose-dependent manner. LA also induced cell blebbing but its activity was substantially lower than that of EpOMEs. However, 9,10-DiHOME did not exhibit hemolytic activity. In contrast, 12,13-DiHOME was slightly hemolytic but significantly weaker than LA.

Fig. 3.

Contrasting cytotoxic effects of EpOMEs and DiHOMEs against S. exigua hemocytes. (A) Cytotoxicity manifested by hemocyte membrane blebbing (arrows) after exposure to test compounds including linoleic acid (LA). The scale bar represents 10 μm. (B) Effect of different doses of test compounds on cellular blebbing. The right panel compares the cytotoxic effects of a single dose (1 μg) of DiHOME and EpOME in each of regioisomers. (C) Apoptotic effects characterized by DNA fragmentation (stained by FITC against BrdU) of hemocytes after exposure to test compounds including DEVD (a caspase inhibitor) using TUNEL assay. Nucleus was stained with DAPI. Cells were visualized under differential interference contrast (DIC). (D) Apoptotic effects of different doses of test compounds. The right panel compares the apoptotic effects of a single dose (1 μg) of DiHOME and EpOME in each regioisomer. Apoptosis was scored by counting the number of FITC-stained cells among 100 randomly selected cells. Each treatment was replicated three times with individual tissue preparations. ‘NS’ stands for no significant difference between two treatments. Asterisks or different letters above standard error bars indicate significant differences among means with Type I error of 0.05 (LSD test).

The cytotoxicity manifested by cellular blebbing was further analyzed by TUNEL assays to determine whether the hemolytic activity was caused by apoptosis (Fig. 3C). Incubation of hemocytes with 12,13-EpOME led to DNA fragmentation, which was observed by DNA end labeling with BrdU. The DNA fragmentation increased in a dose-dependent manner following incubation with 12,13-EpOME. Treatment with 9,10-EpOME also induced DNA fragmentation but less than in 12,13-EpOME treatment (Fig. 3D). Treatment with Ac-DEVD-AMC and Caspase-3 inhibitor (Ac-DEVD-CHO) significantly rescued the DNA fragmentation induced by 12,13-EpOME. LA also induced the DNA fragmentation but its activity was substantially lower than that of EpOMEs. Both DiHOMEs also exhibited apoptotic activity at high concentrations but their activities were significantly lower than those of EpOMEs.

3.4. Effect of other LA metabolites on immune response

To analyze the physiological roles of EpOMEs and their hydrolytic products (DiHOMEs) in insects, different LA metabolites were tested by preparing two epoxide-containing diEpOMEs and two different tetrahydroxy metabolites (THOME and THF-diols) (Fig. 4A). As expected, diEpOMEs similar to EpOMEs strongly suppressed cellular immune response such as hemocyte-spreading behavior (Fig. 4B) and nodule formation (Fig. 4C). They also inhibited PO activation (Fig. 4D). The diEpOME significantly suppressed the induction of some AMP genes (Fig. 4E). In addition, diEpOME showed high levels of cytotoxicity against hemocytes by inducing apoptosis (Fig. 4F and 4G). In contrast, THOME, similar to DiHOMEs, did not inhibit the cellular immune response upon immune challenge (Fig. 4B and 4E). Similar to DiHOMEs, it induced humoral immune response by further upregulating the gene expression of some AMPs induced by the bacterial infection (Fig. 4E). Interestingly, THF-diols slightly retained the inhibitory activities against cellular and humoral immune responses but were substantially weaker than those of the epoxide metabolites. Both THOME and the THF-diol were less cytotoxic to hemocytes compared with diEpOME (Fig. 4F and 4G).

Fig. 4.

Effects of three different linoleic acid (LA) metabolites (diEpOME, THOME, and THF-diol) on immune responses of S. exigua. (A) Chemical structures of LA metabolites. (B) Effects of three LA metabolites (1 μg/larva) on hemocyte-spreading behavior. The cellular spread was determined by counting the cells exhibiting F-actin growth (stained by FITC) out of cell boundary among 100 randomly selected cells. (C) The effect (1 μg/larva) on nodule formation at 8 h after bacterial infection (1.73 × 10³ cells/larva) is shown. (D) The effect (1 μg/larva) on phenoloxidase (PO) activity at 8 h after bacterial infection (1.73×10³ cells/larva) is shown. (E) Effect (1 μg/larva) on the expression of 10 antimicrobial peptide (AMP) genes: apolipophorin III (AL), defensin (Def), gallerimycin (Gal), gloverin (Glv), lysozyme (Lyz), transferrin I (T1), transferrin II (T2), attacin 1 (A1), attacin 2 (A2), and cecropin (Cec). Expression of AMP genes via RT-qPCR at 8 h after injection with E. coli (1.73×10³ cells/larva) and/or test compounds. Each treatment was replicated three times with individual tissue preparations. The cytotoxic effects of each treatment (1 μg/larva) on hemocytes was measured based on cellular blebbing (F) and apoptosis using TUNEL assay (G). Each cytotoxicity test was performed by counting the number of cell blebs or FITC-stained cells for apoptosis among 100 randomly chosen cells. Each treatment was replicated three times. ‘NS’ stands for no significant difference between two treatments. Asterisks or different letters above standard error bars indicate significant differences among means with Type I error of 0.05 (LSD test).

3.5. Effect of EpOME alkoxides on immune response

The immunosuppressive activities of EpOMEs were further analyzed using stable alkoxide derivatives (Fig. 5A). Methoxy (A839, A843), ethoxy (A840, A844), propoxy (A841, A845), and iso-propoxy (A842, A846) ethers were used as bioisosteres for the epoxide in the EpOMEs. Each alkoxide derivative was prepared and tested as a regioisomeric mixture of 9-, 10-, 12- and 13-alkoxyoctadecenoic acids or methyl esters. Among the eight alkoxide derivatives, the methyl ester A841 was the most potent suppressor of cellular immune response in a dose-dependent manner (Fig. 5B–5D). The analysis of AMP expression revealed the relative potency of A841 in suppressing the induction of AMP genes (Fig. 5E) compared with those of both EpOMEs (Fig. 2F). A841, similar to 12,13-EpOME, was also cytotoxic to hemocytes and induced cellular blebbing (Fig. 5F) and apoptosis (Fig. 5G).

Fig. 5.

Effects of eight different EpOME alkoxides on immune responses of S. exigua. (A) Chemical structures of the alkoxides. (B) Effects of EpOME alkoxides (1 μg/larva) on hemocyte-spreading behavior. The cellular spread was determined by counting the cells exhibiting F-actin growth (stained by FITC) out of cell boundary among 100 randomly selected cells. Right panel shows the inhibitory activities of different doses of A841 on hemocyte-spreading behavior compared with 12,13-EpOME treatment (1 μg/larva). (C) The effect of treatment (1 μg/larva) on nodule formation at 8 h after bacterial infection (1.73×10³ cells/larva). Right panel shows the inhibitory activities of different doses of A841 on nodulation compared with the effects of 12,13-EpOME (1 μg/larva). (D) The effect of treatment (1 μg/larva) on phenoloxidase (PO) activity at 8 h after bacterial infection (1.73×10³ cells/larva). Right panel shows the inhibitory activities of different doses of A841 on PO activity compared with the effects of 12,13-EpOME (1 μg/larva). (E) Effect of A841 (1 μg/larva) on the expression of 10 antimicrobial peptide (AMP) genes: apolipophorin III (AL), defensin (Def), gallerimycin (Gal), gloverin (Glv), lysozyme (Lyz), transferrin I (T1), transferrin II (T2), attacin 1 (A1), attacin 2 (A2), and cecropin (Cec). The expression of AMP genes via RT-qPCR was evaluated at 8 h after injection with E. coli (1.73 × 10³ cells/larva) and/or test compounds. Each treatment was replicated three times with individual tissue preparations. Cytotoxic effects of A841 (1 μg/larva) on hemocytes were measured in terms of cellular blebbing (F) and apoptosis using TUNEL assay (G). Each cytotoxicity test was performed by counting the number of cell blebs or FITC-stained cells (for apoptosis) among 100 randomly chosen cells. Each treatment was replicated three times. ‘NS’ stands for no significant difference between two treatments. Asterisks or different letters above standard error bars indicate significant difference among means with Type I error of 0.05 (LSD test).

3.6. Effect of sEH inhibitors on immune responses

The immunosuppressive activity of EpOMEs suggested that sEH inhibitors suppress the immune responses of S. exigua by elevating the endogenous EpOME levels following the immune challenge. Six urea-based sEH inhibitors (Fig. 6A) were assessed to test the hypothesis. All sEH inhibitors significantly inhibited the cellular (Fig. 6B–6D) or humoral responses (Fig. 6E). The sEH inhibitors AUDA and CUDA were most potent inhibitors. These two sEH inhibitors were also highly cytotoxic to hemocytes and induced cellular blebbing (Fig. 6F) and apoptosis (Fig. 6G).

Fig. 6.

Effects of six different sEH inhibitors on immune responses of S. exigua. (A) Chemical structures of the inhibitors. (B) Effects of the inhibitors (1 μg/larva) on hemocyte-spreading behavior. The cellular spread was determined by counting the number of cells exhibiting F-actin growth (stained by FITC) out of cell boundary among 100 randomly selected cells. (C) Their effect of inhibitors (1 μg/larva) on nodule formation at 8 h after bacterial infection (1.73 × 10³ cells/larva). (D) The effect of inhibitors (1 μg/larva) on phenoloxidase (PO) activity at 8 h after bacterial infection (1.73 × 10³ cells/larva). (E) The effect of treatment (1 μg/larva) on the expression of three antimicrobial peptide (AMP) genes: gallerimycin (Gal), gloverin (Glv), and cecropin (Cec). AMP genes were analyzed by RT-qPCR at 8 h after injection with E. coli (1.73 × 10³ cells/larva) and/or test compounds. Each treatment was replicated three times with individual tissue preparations. Cytotoxic effects of the inhibitors (1 μg/larva) on hemocytes were measured by counting the number of cellular blebs (F) and determining the degree of apoptosis using TUNEL assay (G). Each cytotoxicity test was performed by counting the number of cell blebs or FITC-stained cells (for apoptosis) among 100 randomly selected cells. Each treatment was replicated three times. ‘NS’ indicates no significant difference between two treatments. Asterisks or different letters above standard error bars indicate significant differences among means with Type I error of 0.05 (LSD test).

3.7. Enhanced Bt virulence following treatment of S. exigua with diEpOME or sEH inhibitors

The immunosuppressive and hemolytic activities of EpOME derivatives (Fig. 7A) and sEH inhibitors suggested their potential role in enhancing the virulence of entomopathogenic microbes. Each of these compounds showed significant insecticidal activities against S. exigua larvae, and diEpOME was more potent than the sEH inhibitor (Fig. 7B). To test the enhanced virulence hypothesis, B. thuringiensis (= an entomopathogenic bacterium) was treated with each of the potent immunosuppressants (Fig. 7C). All three treatments significantly enhanced the bacterial virulence, and AUDA and the EpOME-mimic A841 were the most potent compounds.

Fig. 7.

Immunosuppressive activity of EpOMEs and their role in enhancing entomopathogenic virulence against S. exigua. (A) Diagram illustrating the physiological roles of C18 and C20 oxylipins in modulating insect immunity. (B) Insecticidal activities of the four selected compounds (EpOME, diEpOME, A841, and AUDA) against S. exigua following injection of hemocoel at a dose of 1 μg/larva to L4 larvae. Controls were injected with DMSO. (C) Enhanced virulence of B. thuringiensis aizawai (Bt) against S. exigua. The larvae injected with the test compounds (1 μg/larva) were fed with cabbage soaked in Bt suspension using a leaf-dipping method. Mortality was assessed at 4 days after treatment (‘DAT’). Ten larvae were used in each treatment and the treatment was performed in triplicate. Different letters above standard error bars indicate significant differences among means with Type I error of 0.05 (LSD test).

4. Discussion

Polyunsaturated fatty acids (PUFAs) are fatty acids that contain two or more double bonds in their carbon backbone. The importance of PUFAs as essential fatty acids has been appreciated since the discovery of the health benefits of animal and human diets containing omega-3 and omega-6 fatty acids (Hildreth et al., 2020). In particular, long-chain PUFAs (LC-PUFAs: 20 or more carbon atoms) function as signaling molecules contributing to immunity, development, and reproduction. LC-PUFAs are also rich (7–25%) in aquatic insects, but are rare in terrestrial insects due to evolutionary adaptation to prevent oxidative stress (Kim and Stanley, 2021). However, terrestrial insects are rich in LA (18-carbon fatty acid) present in phospholipids (PLs). For example, phospholipids extracted from the fat body of S. exigua larvae contained 34.4% LA of the total fatty acids (Kim et al., 2016). LA metabolites such as EpOMEs and DiHOMEs are well known for their physiological roles in mammals (Hildreth et al., 2020). However, their physiological role in insects have not received substantial attention. This study assessed the physiological role of LA metabolites in a lepidopteran insect, especially their crucial role in mediating immune responses.

Upon immune challenge, increasing levels of EpOME and DiHOME regioisomers were produced along with PGE₂ production in S. exigua. The immune challenge also sequentially induced the gene expression of PLA₂, EpOME synthase, and sEH. Among three different PLA₂s, this study showed the early induction of iPLA₂ genes compared to sPLA₂ gene and the early induction of iPLA₂ genes well explained the induced PGE₂ level at the early infection stage. This suggests a differential physiological functions of the PLA₂s in S. exigua. Our previous studies indicated that sPLA₂ is functional in a prophylactic immunity and phospholipid digestion in S. exigua (Vatanparast et al., 2018). In contrast, iPLA₂ of S. exigua catalyzes phospholipids to change fatty acid composition in biological membrane and is a main target of a pathogenic bacterium Xenorhabdus nematophila for host immunosuppression (Vatanparast and Kim, 2020). This current study suggests that iPLA₂s acts as acute mediation of the immune responses in S. exigua. More interestingly, the injection of PGE₂ alone induced the gene expression of EpOME synthase and sEH. This suggests that the immune challenge induces PLA₂ gene expression or directly activates its catalytic activity to synthesize C20 oxylipins including PGE₂, which activates cellular or humoral immune responses of S. exigua (Kim and Stanley, 2021). Arachidonic acid and LA in the phospholipids of S. exigua facilitate the catalytic activity of PLA₂ to release the fatty acids, especially LA. LA subsequently undergoes chain elongation and desaturation to produce arachidonic acid, which is used to synthesize various eicosanoids (Hasan et al., 2019). Subsequent induction of EpOME synthase and sEH enzymes utilizes the free LA to generate EpOME and DiHOME isomers presumably to shut down the induced immune response. The relative fatty acid abundance triggers inflammation via prostaglandin synthesis. The inflammation is resolved by LA metabolites.

Parent LA alone potently suppressed immune response induced by bacterial challenge in S. exigua. This inhibitory activity may be explained by the high levels of inhibitory activity of EpOME isomers against cellular and humoral immune responses of S. exigua. Among the two regioisomers, 12,13-EpOME was more potent than 9,10-EpOME. This supports the previous study (Vatanparast et al., 2020) demonstrating the role of EpOMEs as immune suppressors. In addition, the current study showed that DiHOMEs had limited inhibitory activity against cellular immune responses. These data suggest that DiHOMEs are inactive metabolites of EpOMEs due to their catalytic hydrolysis by sEH. However, in humoral immune response, DiHOMEs actively induced selective AMP gene expression including the two attacins, defensin, gallerimycin, gloverin, lysozyme, and cecropin, which include the AMPs inhibited by EpOMEs. The induction of the selected AMP genes by DiHOMEs might be explained by the immunological role of AMPs in the late phase of immune responses in insects. Compared to the acute cellular immune responses with a few hours, the humoral immune responses using AMPs play crucial role in mopping off the remaining pathogens at late phase (Haine et al., 2009). Thus, DiHOMEs may neutralize the suppressive activity of the EpOMEs against AMP gene expressions. Overall, these data suggest that DiHOMEs are specific metabolites of EpOMEs, which prevent excessive or prolonged inhibition of immune response in insects. DiHOMEs are potent immune mediators in mammals because they are synthesized by activated neutrophils and induce chemotaxis of other neutrophils at relatively low concentrations (~10 nM) (Ishizaki et al., 1999; Totani et al., 2000). The negative controls of EpOMEs were further supported by the similar activity of their dual epoxide analog, diEpOME. Further, the two possible DiHOME analogs (THOME and THF-diol) showed limited inhibitory activity against cellular immune responses but induced AMP expression.

Compared with DiHOMEs, EpOMEs were highly cytotoxic against hemocytes of S. exigua. The cytotoxic activity of EpOMEs was attributed to apoptosis based on cellular blebbing and DNA fragmentation in hemocytes exposed to EpOMEs (Elmore, 2007). Treatment with a caspase inhibitor significantly rescued the hemocytes from the cytotoxic activity of 12,13-EpOME. Apoptosis is a form of programmed cell death that removes unwanted or damaged cells (Denton et al., 2013). Some entomopathogens use apoptosis to inhibit host immune response by killing hemocytes, e.g. the entomopathogenic nematode, Ovomermis sinensis inhibits the immune response of Helicoverpa armigera (Jiao et al., 2018); Serratia marcescens and Porphyromonas gingivalis inhibit Bombyx mori (Ishii et al., 2010, 2012); and an ectoparasitoid wasp, Pachycrepoideus vindemiae, inhibits Drosophila (Wan et al., 2020). Apoptosis, however, plays a crucial role in the host immune response especially during viral infection in both Lepidoptera and Diptera, in which a rapid induction of apoptosis restricts viral replication by removing infected cells (Liu et al., 2013). Apoptosis in the midgut epithelium (known as sloughing) effectively blocks the pathogen entry into the insect hemocoel by eliminating infected cells and replacing the epithelium with regenerated cells (Clem, 2005). In lepidopteran and dipteran insects, cell lysis is required for the release of prophenol oxidase (PPO) from specific hemocytes called oenocytoids (in Spodoptera) or crystal cells (in Drosophila) because PPO lacks the signal peptide and is released into plasma for activation by serine proteases (Jiang et al., 2001; Bidla et al., 2007; Shrestha and Kim, 2008). PGE₂ triggers cell death via membrane receptors specific to the hemocytes of S. exigua (Shrestha et al., 2011). However, the role of EpOMEs in cell lysis and activation of PPO remains unknown. Based on the delayed production of EpOMEs compared with the immune activator, EpOMEs may play a crucial role in eliminating infected or damaged cells during late immune response. Our current study also showed that the cell death induced by EpOMEs was attenuated by hydrolysis into less active forms, DiHOMEs, in S. exigua.

The immunosuppressive activity of EpOMEs was further supported by their alkoxide derivatives or agonists such as sEH inhibitors, which prevented their degradation into DiHOMEs. The alkoxides are synthetic derivatives obtained by replacing the epoxide in EpOME with a methoxy, ethoxy, propoxy or (iso)propoxy group. The alkoxy ethers cannot be hydrolyzed by sEH; however, as shown by the insect growth regulator methoprene, alkoxides can mimic epoxides in several biological systems. This significantly modifies the biological activities to suppress different immune responses, suggesting the critical importance of the epoxide structure in the immunosuppressive activity of 12,13-EpOME. Interestingly, A841 (= the methyl ester propoxy EpOME mimic) was the most potent among the eight alkoxides tested and even more potent than 12,13-EpOME itself. A similar screening of the oxylipin alkoxides using kidney cells was conducted to evaluate their role in protecting organs against reactive oxygen species during anticancer drug treatment using cisplatin (Kim et al., 2004). The immunosuppressive activities of EpOMEs were evaluated using sEH inhibitors to upregulate the endogenous EpOME levels by blocking the conversion of EpOMEs to DiHOMEs. Urea-based compounds inhibiting sEH originally designed to inhibit mammalian sEH (Singh et al., 2021) inhibited invertebrate sEHs including nematodes (Harris et al., 2008) and insects (Xu et al., 2016; Vatanparast et al., 2020). Among six sEH inhibitors, AUDA and CUDA were highly potent, presumably by elevating endogenous EpOME levels, which led to the inhibition of immune response in the test insect. The dysregulation of the immune responses induced by the EpOME analogs or agonists might facilitate entomopathogens because the pathogens are usually eliminated by the immune responses. Insect immunity is known to exhibit efficient defense against Bt infection (Caccia et al., 2016). Our current study demonstrates that the EpOME analogs (diEpOME and A841) or agonist (AUDA) significantly enhanced the Bt virulence. This finding supports the role of EpOMEs as negative regulators of insect immunity and DiHOMEs are the their inactive metabolites.

This study demonstrates the immunosuppressive roles of EpOMEs mediated by inhibition of cellular and humoral immune responses in a lepidopteran insect. Further, the EpOMEs also induce apoptosis against immune cells. In contrast, their hydroxylated metabolites, DiHOMEs, are inactive in immunosuppression, suggesting that EpOMEs resolve the immune response induced by immune mediators following pathogen infection in insects (see Fig 7A). Although several immune mediators, such as nitric oxide (NO), biogenic monoamines, cytokines, and eicosanoids, are known to induce various immune responses (Gillespie et al., 1997; Kim et al., 2018), the termination or resolution of the induced immune response in insects has yet to be established. Injury in mammals leads to upregulation of innate and subsequent acquired immune responses to defend against pathogens. The immune response is later resolved at the site of injury by anti-inflammatory mediators such as resolvins, protectins, glucocorticoids, catecholamines, prostaglandins (PGs) of the E-series, NO, and interleukins (Serhan, 2007; Ayala et al., 2003). In addition to the inhibition of cellular and humoral immune responses by EpOMEs, their high cytotoxic activities against hemocytes via apoptosis is crucial in resolving the immune responses in insects. Immune response is ultimately suppressed by the overt induction of apoptosis in immune cells endogenously, and is well known in mammals (Brown and Savill, 1999). The first known mediator suppressing insect immunity was PGI₂ in S. exigua (Ahmed et al., 2021). Interestingly, the genes of EpOME synthase or PGI₂ are expressed at a late stage after the immune challenge probably to terminate the induced immune response. The study findings suggest that EpOMEs play a crucial role in resolving immune response and DiHOMEs are the inactive forms in insects. More broadly, these data suggest that insects like mammals exhibit an active mechanism for resolving as well as initiating inflammation. A comparative analysis of diverse species can provide possible insight into the underlying regulatory mechanisms of initiation and resolution of inflammation.

Data availability

The authors confirm that all data underlying the findings are fully available without restriction. All relevant data are within the paper and its Supporting Information file.