Synergistic host-parasitoid antimicrobial peptide interactions ensure the survival of parasitized hosts

ABSTRACT

Host–parasite interaction is always involved in the immune response. It is well known that parasitic wasps introduce various factors to suppress host immune responses, thereby facilitating parasitoid progeny development while compromising host survival. However, little is understood about how parasitic wasps, especially endoparasitoids, survive pathogen threats when they develop in hosts. Antimicrobial peptides (AMPs) are important immune effectors that are active against pathogens and essential for the survival of organisms. Here, we investigated the immune response of the host Plutella xylostella and its endoparasitic wasp Cotesia vestalis during the process of parasitism, focusing on AMPs. We found that the suppression of parasitism on host AMPs diminished as the process went on and selected AMPs responded differently in parasitized hosts following bacterial challenges. We then demonstrated that the host-expressed gloverin was extremely sensitive to pathogen infection after parasitism and was vital for host survival. Furthermore, we found that parasitoid teratocytes expressed antimicrobial peptides, especially hymenopteacin, which were significantly upregulated following bacterial infection. With a broad antimicrobial spectrum, the presence of hymenopteacin in the hemolymph of parasitized hosts significantly decreased bacterial load and increased the survival of the parasitized host. Our study provides a systematic perspective on host-parasitoid immune interaction by highlighting the significance of AMPs.

Introduction

Parasitism is a type of symbiosis in which the parasite benefits at the expense of the host. Parasitoid wasps, referring to hymenopteran insects, lay eggs inside or on the body of other invertebrates. As a natural enemy, the development of parasitoid offspring is lethal to the hosts. Theoretically, the wounding caused by female parasitoid stings and the introduced parasitoid egg would activate the immune response of the host. However, parasitoids collectively evade or suppress the host immune response determined by a set of parasitoid-derived factors, including venom, symbiotic polydnavirus (PDV, recently termed as polydnaviriformes), teratocytes, which are derived from extraembryonic cells of parasitoid eggs, and even larval secretions [1–4]. While immune evasion parasitoids have less impact on the host’s immune system, parasitoids actively suppress host immune response using multiple mechanisms, for example, disruption of cellular immune responses by reducing hemocyte number and inhibiting encapsulation and inhibition of humoral immune response by interfering with host phenoloxidase activation, blocking the nuclear factor-κB (NF-κB) signaling pathways and impairing antimicrobial peptides (AMPs) secretion, etc [5–8]. Considering this, parasitized hosts with weakened immune systems would be more susceptible to opportunistic microbes, which is harmful to the survival of parasitoid offspring. However, the survival strategy of parasitoid progeny in the face of immunological challenges remains largely unknown.

AMP, a kind of small, positively charged peptide, kills microbes by targeting cell membranes or certain intracellular sites and contributes greatly to the innate immune response. Since the first discovery of AMPs, significant progress has been made in identifying novel AMPs and characterizing their potential function beyond immunomodulators [9–13]. Most AMP genes are induced by bacterial infection, regulated by the Toll and Imd NF-κB pathway. Thus, AMPs are often used as indicators to monitor the activity of these immune pathways, especially in insects [11]. Correspondingly, the upregulated AMPs play a crucial role in enhancing the survival rate of various hosts exposed to infections [14–16]. While some AMPs, i.e. defensin and cecropin, are well conserved in insects, other AMPs are taxonomically distinct, such as diptercin and hymenoptaecin, suggesting the diversity of insect AMPs [17,18]. Genetic variants in AMPs correlate with susceptibility against specific pathogens. Functional studies on Drosophila AMPs indicate that a single amino acid difference between diptericin and metchnikowin between individuals may cause phenotypic differences [17,19]. In host-parasitoid interactions, wasp infestation leads to the transcription alteration of AMP genes in many insect hosts, either induction or suppression. For example, while AMPs were persistently downregulated post-parasitization in the Galleria mellonella-Pimpla turionellae system, envenomation by the Habrobracon hebetor resulted in the upregulation of specific AMPs [20,21]. There is growing evidence that host AMPs play a role in defense against parasitoids. A very recent study showed that overexpressing AMPs in Drosophila causes wasp gut microbiota dysbiosis and thereby confers resistance to wasp [22]. Given their role in antibacterial defense, AMPs appear to function as a critical immunological checkpoint in parasitoid–host interactions, maintaining a delicate balance between supporting parasitoid larval survival during secondary infections, while potentially mediating wasp-induced host mortality [23]. However, how AMPs contribute to host immune defense against secondary infections remains poorly understood.

Plutella xylostella (Lepidoptera; Plutellidae), the diamondback moth (DBM), is a notorious herbivore of many economically important cruciferous vegetables (Brassicaceae). Managing the DBM is quite challenging as various field populations have acquired resistance to a wide range of insecticides, including Bacillus thuringiensis toxins [24]. Parasitoids are effective biological control agents for suppressing DBM populations. Cotesia vestalis Haliday (Hymenoptera: Braconidae) is a primary larval endoparasitoid of DBM in China [25]. Several investigations demonstrated that C. vestalis parasitism suppresses the cellular immune response as well as the humoral immune response of DBM [7,26,27]. The immunosuppression is involved in reducing host hemocytes, triggering apoptosis, suppressing melanization, and downregulating the production of certain AMPs. Two immune effectors of C. vestalis, i.e. Cotesia vestalis bracovirus (CvBV)-derived c-type lectin and teratocyte-expressed serpin, have recently been demonstrated to exhibit antibacterial activities for successful parasitism [5,7]. Our previous study has also demonstrated that the teratocytes, cells dissociated from C. vestalis egg when it hatched, express defensins and exhibit anti-microbial activity [28]. In light of these findings, we speculate that the significance of AMPs, as main factors in combating a broad range of microbes, for the survival of parasitoid offspring has been underestimated. In this study, we focused on the response of AMPs in the P. xylostella - C. vestalis system to provide a more comprehensive understanding of the parasitoid-host immunological interaction.

Materials and methods

Insect rearing

Cultures of DMB and C. vestalis were maintained under laboratory conditions. Briefly, DMB larvae were fed with an artificial diet at 25°C and 65% relative humidity under a 14:10 light:dark cycle. Adult P. xylostella and C. vestalis were fed with a 20% honey/water (V/V) solution. For parasitization, the middle stage of 3^rd instar DMB larva was exposed to a newly emerged and mated female C. vestalis until sting behavior was observed.

Sample collection

The parasitized hosts were dissected to obtain wasp larvae and washed in PBS at least 3 times for further analysis. For the teratocyte collection, the parasitized hosts were dissected to release teratocytes into a serum-free medium (Thermo Fisher Scientific, USA). The teratocytes, unlike DMB hemocytes, are larger and non-adhesion, and would suspend at the bottom of the Petri dish and facilitate collection. The collected teratocytes were washed with a serum-free medium 5 times to ensure that there was no contamination of any host tissue debris.

Gene cloning and qPCR

Total RNA was isolated from collected teratocytes, C. vestalis larvae and P. xylostella larvae using TRIzol Reagent (Invitrogen, California, USA) and the first strand cDNA was synthesized using SuperScript III Reverse Transcriptase (Invitrogen). The coding sequences were obtained from the C. vestalis genome [25]. Finally, the full-length cDNA was obtained by PCR. The specific primers used are given in Table S1. The fragment was ligated into the vector pGEM-T easy vector (Promega, Wisconsin, USA), and the recombined vector was then transformed into competent E. coli TG1 cells.

For qPCR, total RNA was extracted and then reverse transcribed into cDNA using the ReverTra Ace qPCR RT kit (Toyobo) according to the manufacturer’s protocol. qPCRs were performed in a CFX Connect real-time system (Bio-Rad) with THUNDERBIRD qPCR Mix (Toyobo). Reactions were carried out for 60 s at 95°C, followed by 40 cycles of 15 s at 95°C and 30 s at 60°C. To analyze the expression profile of the P. xylostella gene, β-actin of P. xylostella (GenBank Acc. No. NM_001309101) and β-tubulin (GenBank Acc. No. EU127912) were used as endogenous controls. Both Cv-β-tubulin (GenBank acc. No.: MT459787) and Cv-18S rRNA (GenBank acc. No.: JX399880.1) were used as internal reference genes for C. vestalis gene expression analyses. The relative expression levels were determined using the 2^−∆∆Ct method. All the primers used for qPCR in this study are listed in Table S1.

Sequence and phylogenetic analysis

The signal peptides were identified by the SignalP 6.0 (https://services.healthtech.dtu.dk/services/SignalP-6.0/), and the proprotein and furin cleavage sites were predicted with ProP 1.0 [29]. The molecular weight and theoretical isoelectric point (pI) were calculated in ProtParam tool (https://web.expasy.org/protparam/). Homologous sequences from other representative insects were retrieved from publicly available genomic data from the NCBI genome database (https://www.ncbi.nlm.nih.gov/guide/genomes-maps/) and InsectBase 2.0 (http://v2.insect-genome.com/). Phylogenetic analysis was performed on IQ-Tree by Maximum likelihood with a bootstrap value of 1000 [30], and amino acid sequence data were shown in Table S2. The tree and protein domain were visualized using TBtools [31]. Gene structures were visualized by the gggenes (v 0.5.1) package.

Insect microinjection

For RNA interference, dsRNA of Gloverin and GFP were produced and purified using a T7 RiboMAXTM Express kit (Promega, Madison, WI, USA). Each DMB larva was injected with ~ 500 ng of dsRNA. Microinjections were performed by Eppendorf Femtojet (Eppendorf, Germany) with a microcontroller (Narishige, Japan) under a Stemi 2000-C microscope (Zeiss, Germany).

For bacteria injection, E. coli and S. aureus were grown for 4–5 h at 37°C with shaking at 280 rpm/min, their OD values were between 0.6 and 0.8. Then, cultures were centrifuged at 1000× g. The bacteria were resuspended in sterile PBS to an OD of 0.01. Monoparasitized P. xylostella larvae (3 days post parasitization, dpp) and non-parasitized ones with the same physiological state were injected with 0.1 μl bacterial solution of E. coli or S. aureus, respectively. The injection of an equal volume of sterile PBS was used as the negative control. P. xylostella larvae were transferred to a new container after injection and reared at 25°C until they were sampled.

Expression and purification of hymenoptaecin and antibody production

After removing the signal peptide and propeptide sequences, the fragment with the mature Hymenoptaecin (mHym) sequence was amplified by specific primers with restriction sites (Table S1) for vector construction. Recombined vector pET-28(a) (Promega) was transformed into competent E. coli ROSETTA (DE3). A positive bacterial colony was grown to OD₆₀₀ = 0.6–0.8 at 37°C. Then isopropyl-d-thiogalactoside (IPTG) (1 mM) was added to induce protein expression at 37°C. Bacterial cells were harvested by centrifugation at 8,000× g for 10 min at 4°C 6 h later. The bacterial pellet was lysed, denatured and then subjected to 12% SDS-PAGE. The induced bacteria were sonicated for 20 min at 100W on an ice bath and then centrifuged. The supernatant and pellet were analyzed with 12% SDS-PAGE. The recombinant mHym was purified by cOmplete™ His-Tag Purification Resin (Roche, Swiss). Since mHym were expressed as inclusion bodies, the purified proteins were then dialyzed for 4 h at 4 °C in a succession of buffers containing 5% glycerol, 1% L-arginine, 2% glycine with different concentrations of urea (6 M, 4 M, 3 M, 2 M, 1 M, 0 M) and then and then ultra-filtered with Amicon Ultra-4 (Millipore). To cleave the His-tag, the refolded proteins were incubated with thrombin (final concentration 1 Unit/ml) for at least 40 min at room temperature, and p-Aminobenzamidine – Agarose (Sigma) was used to remove Thrombin. Thereafter, the supernatant was collected and analyzed by a 16% SDS-PAGE. To obtain the antiserum against Hymenoptaecin, the purified mHym was used as the antigen, and the New Zealand white rabbit was used for polyclonal antibody preparation.

Antimicrobial activity assay

The antimicrobial activity of mHym was tested against two gram-negative bacteria (E. coli and Salmonella typhimurium CMCC50071), two gram-positive bacteria (Micrococcus luteus ATCC49732 and Staphylococcus aureus ATCC25923) and one fungal strain (Monilia albican ATCC12031), which, except for E. coli, were purchased from the BeNa Culture Collection, Jiangsu, China. Bacteria were cultured to the logarithmic growth phase in nutrient broth at 37 °C, and M. albican was grown in potato dextrose agar at 28 °C. Microbes were diluted with either a nutrient agar medium or a PDA medium to a final concentration of OD₆₀₀ = 0.01. The mixture was spread on a Petri dish and then dried at room temperature. The double-round filter paper was used as a carrier. Each dot contains 10 μl mHym, while the same volume of acetic acid or antibiotic (Sigma) (50 ng/ml) was served as the negative and positive control, respectively. Plates were incubated at the appropriate temperature overnight and the growth of the microbe was recorded.

We then determined the minimum inhibitory concentrations of mHym against selected microbes. Fresh cultures with different concentrations of Hym were incubated at 37 °C or 28 °C for 24 h, and OD₆₀₀ was measured by SpectraMax iD5 multi-mode microplate reader (Molecular Devices, USA). Bacteria cultured with 50 ng/ml of antibiotic or 0.01% acetic acid served as controls. The MIC value was set at the lowest concentration, which caused at least 95% growth inhibition. The growth curve of selected microbes treated with mHym (MIC) was measured by a plate reader every 30 mins at OD₆₀₀ for 12 h. Antibiotic (Sigma) (50 ng/ml) and PBS with 0.01% acetic acid served as positive and negative controls, respectively. Each treatment was replicated at least three times.

Immunohistochemistry

The parasitized DMB larvae were dissected to obtain teratocytes, which were divided into an experimental group and a control group. In the experimental group, Anti-Hym (1:5000), Anti-Rabbit 494 and DAPI were added successively. Anti-Rabbit 494 and DAPI were added to the control group successively. The expression of Hym was observed by a laser confocal microscope (Zeiss LSM 780, Carl Zeiss).

Western blotting

Total hemolymph protein from approximately parasitized and non-parasitized P. xylostella larvae was extracted by Total Protein Extraction Kit for Insects (Invent) for western blot according to the manufacturer’s protocol. The protein sample was mixed with the protein loading buffer at a ratio of 4:1, boiled in water for 10 min, cooled rapidly on ice, and centrifuged at 12,000× g for 5 min, and the supernatant was taken as the sample. According to the size of the target protein, PAGE glue and protein Marker with appropriate concentration were selected to add samples to the sample hole of PAGE glue. Each well was about 50 μg of total protein. Proteins were separated in a denaturing polyacrylamide gel and transferred to a polyvinylidene difluoride membrane. After blocking and washing, membranes were then incubated with primary antibodies against anti-hym (1:5000) at room temperature for 2 h. After three washes, membranes were then incubated with secondary antibody Anti-Rabbit 494 (1:5000) at room temperature for 2 h. After five washes, membranes were then incubated with the enhanced chemiluminescence western blotting substrate for imaging (Promega).

Survival rate assay

E. coli and S. aureus were resuspended in sterile PBS at an OD of 0.01. 0.1 µL of resuspended bacteria was injected into parasitized P. xylostella larvae after dsGloverin treatment with an Eppendorf FemtoJet 4i Microinjector (Eppendorf) and a microcontroller (Narishige). Each treatment was performed in triplicate (at least 30 individuals per replicate). P. xylostella larvae were then reared at 25°C and the survival rate was recorded every 6 h.

To determine the antibacterial role of hymenopteacin in vivo, mHym was injected into unparasitized P. xylostella larvae following Wu et al. [5]. Briefly, 0.1 µL mHym (16.25 µg/mL) with S. aureus (OD₆₀₀ = 0.1) or E. coli (OD₆₀₀ = 0.1) were injected into unparasitized late third instar P. xylostella larvae. An equal dose of PBS or corresponding bacteria was used as a blank or positive control, respectively. To estimate the bacterial load, the surface of the treated larvae was disinfected with 75% alcohol 24 h post-injection and then homogenized after the midgut was removed. This homogenate was then diluted 100 times with sterile PBS before being incubated on LB agar at 37°C for 24 h, and then the colony-forming units (CFU) were counted. In the meantime, DNA was extracted from the above-treated larvae using FastPure Cell/Tissue DNA Isolation Mini Kit (Vazyme, China) to determine the relative levels of 16S rRNA by qPCR. Additionally, the death rate was recorded every 12 h.

Statistical analysis

All statistical analyses were performed using SPSS 20.0 software. All the experiments were performed with at least three independent replicates. Gene expression data were presented as means ± SEM and were examined by the Student’s t-test or One-way ANOVA (Tukey’s test), with a p value of 0.05 as the significance threshold. Survival analysis was performed using the log-rank (Mantel-Cox) test with GraphPad Prism 8.

Result

Two different ways of parasitized P. xylostella AMPs respond to microbial infection

Four kinds of AMPs, i.e. cecropin, defensin, gloverin and moricin, have been characterized in P. xylostella [32–35]. To assess the role of AMPs in the host-parasitoid system, the expression levels of five selected AMPs in parasitized and non-parasitized P. xylostella were detected by qPCR. The results showed that the expressions of all five AMPs were significantly suppressed at one-day post-parasitization (1 dpp), and then recovered or upregulated at the late stage of parasitism, with Px-cecropin2 (Px-Cec2) and Px-cecropin3 (Px-Cec3) being upregulated since two dpp, gloverin upregulated since three dpp, and defensin and morcin upregulated at five dpp (Figure 1(A-E)). To figure out the immune response of the parasitized host to bacterial challenge, the host larvae were infected separately with E. coli and S. aureus three days post-parasitization. As shown in Figure 2(A-E), all five AMPs were sensitive to bacterial infection in non-parasitized hosts, with Px-Cec2 and Px-Cec3 showing the most elevations. In parasitized hosts, the trend of these genes was divided: the expressions of Px-Cec2 and Px-Cec3 were relatively stable or slightly increased, while the other three genes were dramatically upregulated, suggesting that defensin, gloverin and morcin were sensitive in parasitized hosts.

Figure 1.

The expression of five antimicrobial peptide genes of P. xylostella following C. vestalis parasitization. Px-β-tubulin and Px-β-actin were used as the internal reference genes. At least six biological replicates were conducted, and error bars represent SEM. A significant analysis was performed by Student’s t-test: *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001.

Figure 2.

The expression of five antimicrobial peptide genes of P. xylostella following bacterial infection. px-β-tubulin and Px-β-actin were used as the internal reference genes. At least six biological replicates were conducted, and error bars represent SEM. A significant analysis was performed by Student’s t-test: *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001.

Parasitized hosts outlive unparasitized ones following bacterial challenges

To further examine the role of AMPs in parasitized P. xylostella, the expression of Px-gloverin (Px-Glo) was knocked down using dsPx-Glo at two dpp before its expression was upregulated. After 24 h, the expression level of Px-Glo was significantly suppressed and the RNA interference efficiency was about 85%. The response of Px-Glo to bacterial stimulation was also weakened (Figure 3(A,B)). The survival rate of P. xylostella in response to bacterial challenge was recorded after dsRNA treatment. Compared with the dsGFP-treated group, Px-Glo knockdown significantly decreased the survival rate of both non-parasitized and parasitized host larvae (Figure 3(C-E)). Notably, the dsPx-Glo-treated non-parasitized hosts had the lowest survival rate after PBS treatment, suggesting a crucial role of gloverin in the immune response of P. xylostella. We also noted that the survival rate was higher in dsPx-Glo-treated parasitized hosts than in non-parasitized hosts. A similar phenomenon was observed after E. coli and S. aureus infection.

Figure 3.

The effect of PxGloverin on the survival rate of parasitized P. xylostella larvae. (A) the RNA interference efficiency of dsGloverin in parasitized larvae. The dsGloverin was injected into P. xylostella larvae at 2 d post-parasitization and the larvae were sampled 24 h later. dsGFP was used as the negative control. Values represent the mean ± SEM of 6 independent experiments. (Student’s t-test, *** p ≤ 0.001). (B) the expression of gloverin in parasitized P. xylostella larvae following bacterial infection. E. coli or S. aureus was injected into parasitized larvae 24 h after dsGloverin-treatment (3 d post-parasitization) and the larvae were collected 6 h later for qPCR analysis. PBS was used as the negative control. Values represent the mean ± SEM of three independent experiments. (C-E) the survival rate of P. xylostella larvae following a bacterial challenge. NP-dsGFP-S: non-parasitized hosts with injecting dsGFP and S. aureus; NP-dsG-S: non-parasitized hosts injected with dsGloverin and S. aureus; P-dsGFP-S: parasitized hosts injected with dsGFP and S. aureus, P-dsG-S: parasitized hosts injected with dsGloverin and S. aureus. Experiments were performed with three independent replicates, using at least thirty larvae for each replicate. Differences between groups were analyzed by the log-rank test (mantel–Cox, *p < 0.05; **p < 0.01; ***p < .001).

Hymenoptaecin is a primary immune effector in the P. xylostella - C. vestalis system

It is known that teratocytes express immune-related genes and exhibit antimicrobial activity [2]. Two kinds of AMPs (Defensin and hymenoptaecin) have previously been identified in C. vestalis [28,36]. In total, four characterized defensins (Cv-def-1, -2, -3 and CvT-def-1) and one putative hymenoptaecin (Hym) were selected for further analysis. The four defensins displayed varying expression patterns in different developmental stages of C. vestalis (Figure 4(A-D)): Cv-def1 and Cv-def2 exhibited high expression levels in male adults, while Cv-Def-3 showed prominent expression in 3^rd instar larvae. CvT-def-1 was highly expressed since the 3^rd instar larvae and peaked in male adults. Unlike the defensins, Hym was specifically expressed in teratocytes and the trend increased as they matured (Figure 4E).

Figure 4.

The expression level of C. vestalis AMP genes in different development stages and teratocytes. 1^st, in 1^st instar of wasp larvae; 2^ndE larvae: early stage of 2^nd instar of wasp larvae; 2^ndM, middle stage of 2^nd instar larvae; 2^ndL, late stage of 2^nd instar stage; 3rd, 3^rd instar larvae; Tera1d, 1-day-old terotocytes (when wasp larvae are 2^ndE); Tera3d, 3-day-old teratocytes (when larvae are 2^nd instar). Cv-18S rRNA and Cv-β-tubulin were used as internal references. Three biological repetitions were performed and error bars represent the standard deviation.

To test whether the larvae of C. vestalis would participate in the host-parasitoid wasp system when the parasitized host was challenged, the expression of AMP genes in C. vestalis larvae and teratocytes was detected 6 h post-E. coli or S. aureus infection. Except for CvT-def-1, the infection of pathogenic bacteria did not stimulate the expression of the other three defensins in C. vestalis larvae (Figure 5A), while the transcript of Hym was not detected (data not shown). In teratocytes, the immune response to bacterial challenge was much more complicated (Figure 5B). Compared with the control group, Cv-def-1 was upregulated after being infected by E. coli, Cv-def-3 was induced by S. aureus, and there was a minor drop in CvT-def-1 expression in response to E. coli. Both Cv-def-2 and Hym were stimulated by both bacteria, with Hym showing a two- to threefold rise compared to the control group.

Figure 5.

The expression of AMP genes in C. vestalis larvae and teratocytes in response to bacterial infection. (A and B) E. coli or S. aureus was injected into P. xylostella larvae at 3 d post-parasitization. C. vestalis larvae (A) and teratocytes (B) were separately collected 6 h later for qPCR analysis. Values represent the mean ± SEM of 3 independent experiments. A significant analysis was performed by one-way ANOVA (Turkey’s test, *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001). (C and D) expression of CvT-def-1 and Hym in teratocytes following bacterial infection of hosts that had been treated with dsGloverin.

We next detected the response of Cv-def-2 and Hym in immune-deficient P. xylostella after infection with bacteria. Cv-def-2 transcript was significantly induced following S. aureus infection, but it is insensitive to E. coli (Figure 5C). The infection of bacteria strongly induced the transcripts of Hym with about 7 ~ 9-fold higher than that in the control groups (Figure 5D). Given that parasitized hosts exhibited a higher survival rate in dsGloverin-treated P. xylostella after a bacterial challenge, the induction of teratocytes expressed AMPs, especially the teratocyte-specific expressed hymenoptaecin, might contribute greatly to the survival of the parasitized host.

Characterization of Hymenoptaecin

Despite the four defensins, the sequence and antimicrobial activity of hymenoptaecin have not been characterized yet. A similarity search of the C. vestalis genome identified two hymenoptaecin-like peptides encoding identical amino acids. The open reading frame (ORF) of Hym is 480 nucleotides, encodes 159 amino acid (aa) residues, with a signal peptide (20 aa), a proline-rich propeptide (36 aa, 38% proline) and a glycine-rich mature peptide (Fig. S1). As one kind of glycine-rich AMP, the glycine content in the mature peptide of Hym is 10%. The predicted molecular weight of the mature peptide is 15.75 kDa and has a pI of 9.52. Similar to hymenoptaecin from the other hymenopteran species, the putative propeptide and mature peptide of Hym are separated by a conserved furin recognition motif, RX(K/R)R, which may serve as a putative recognition cleavage site for intracellular furin for zymogen activation (Fig. S1). In addition, the propeptide shares detectable similarities (~30%) with other known Pro-rich AMPs, including drosocin and apidaecin [18].

A maximum likelihood phylogenetic tree of hymenoptaecin was constructed (Fig. S2) based on homologous sequences from Hymenoptera, especially parasitic wasps. The result showed hymenoptaecin is conserved in Hymenoptera and that from the braconid wasps forms a cluster. The genomic structure revealed that most hymenoptaecins contain at least one intron (Fig. S3). Hymenoptaecins in Cotesia share features of inverted duplication, while duplication of hymenoptaecin in other species, such as N. vitripennis and Leptopilina heterotoma is tandem duplication, suggesting independent duplication of hymenoptaecin in Cotesia.

Hymenoptaecin shows antimicrobial peptide activity in vitro and in vivo

To detect the activity of hymenoptaecin, recombinant hymenoptaecin (mHym) was heterologously expressed with E. coli BL21 and purified (Fig. S4A). The mHym with His-tag is about 20 kDa. After removing the His-tag by thrombin, a ~ 16 kDa protein was obtained (Fig. S4B).

We then examined the effect of purified mHym against three Gram-negative bacteria (E. coli and S. typhimurium), three Gram-positive bacteria (M. luteus and S. aureus) and one fungal strain (M. albican), with antibiotic served as the positive control and PBS containing 0.01% acetic acid as the negative control. Inhibitor zone results showed Hym was active against gram-positive bacteria (M. luteus and S. aureus) and fungi (M. albican). The minimum inhibitory concentrations (MIC) of Hym against M. luteus and M. albican were 8.12 and 16.25 μg/ml, respectively. In addition to the three positive bacteria, the growth curves showed that hymenoptaecin showed activity against E. coli with a concentration of 16.25 μg/ml, but showed no activity against S. typhimurium (Figure 6(A-D)), Fig. S4C).

Figure 6.

The antimicrobial activity of teratocytes-expressed Hym. (A-D) the antimicrobial activity of purified Hym against E. coli (A), S. aureus (B), M. luteus (C), and M. albican (D) in vitro. The left panels show the inhibition zone as indicated by dotted circles. The right panels show the growth curve of selected bacteria incubated with purified Hym (with a concentration of 16.25 μg/ml). Three biological repetitions were performed, and the error bars represent the standard deviation. PBS was used as the negative control, and an antibiotic with a 1:10000 dilution was used as the positive control. (E). Immunohistochemistry of Hym in teratocyes. The nucleus is shown in blue (DAPI); Hym is shown in red (Anti-Rabbit 494). A1-A3, teratocytes were incubated with Anti-Hym, DAPI in order; B1-B3, teratocytes were incubated with serum without Anti-Hym antibody; A1 & B1 show the Hym channel; A2 & B2 show the DAPI channel; A3 & B3 show merged signals of both Hym & DAPI. The scales in the figure are 20 μm. (F) western blot of Hym in parasitized P. xylostella hemolymph. P, hemolymph of parasitized DMB; NP, hemolymph of non-parasitized DMB; M, marker. (G) bacterial loads in DMB larvae post bacterial challenges. Early fourth instar DMB larvae were coinjected with selected bacteria with purified Hym (1.625 ng/insect). Px-β-actin and Px-β-tubulin were used as internal references, and experiments were performed with at least three independent replicates, using at least ten DMB larvae for each replicate. Values represent the mean ± SD of six independent experiments (student’s t-test, *p < 0.01; **p < 0.001; ns, not significant). (H) the survival rate of DMB larvae following bacterial challenges. Experiments were performed with three independent replicates, using at least 30 DMB larvae for each replicate. Differences between groups were analyzed by the log-rank test (mantel–Cox, ***p < 0.05).

In vivo, the expression of Hym in teratocytes was confirmed by immunohistochemical (Figure 6(E)) and western blotting using a polyclonal antibody against mHym (Figure 6(E,F)), demonstrating the secretion of hymenoptaecin into the hemolymph of parasitized host larvae. The co-injection of mHym significantly reduces the bacterial load, as evidenced by the CFU counts and the 16S rRNA level of both E. coli and S. aureus in the hemolymph of P. xylostella (Figure 6(G) and Fig. S4D). The survival tests also showed larvae co-injected with bacteria and mHym had a considerably higher survival rate than those injected with only E. coli or S. aureus (Figure 6H). These results suggest that teratocytes-expressed hymenoptaecin functions as an antibacterial factor in the immune response of the P. xylostella - C. vestalis system.

Discussion

Antimicrobial peptides are important effector factors in the immune system. They exist widely in the biological world and play an extremely important role in the survival of insects that lack adaptive immunity [37]. Several AMPs have been characterized in DMB and divided into four families according to their sequence similarity: cecropin, moricin, gloverin, and defensin. With a broad spectrum of antimicrobial activity, these AMPs are induced by bacterial infection and may be regulated by the conserved NF-kB signaling pathway, suggesting these AMPs participate in the immune response of DMB [33,38]. Our study showed that all four kinds of AMPs were disabled at the beginning of C. vestalis parasitization. Unlike nonparasitized hosts, the regulation of AMP expression in parasitized hosts following infection was complex, involving the effect of host immune response and parasitic factors (BV and teratocyes). Recent studies have shown that BV-carried parasitoid wasps inhibit the NF-κB activation and suppress the expression of AMPs in the host via BV-expressed ankyrin [8,39,40]. CvBV expressed six ankyrin genes, which share the conserved amino acid sequence with Ank-H4 and Ank-N5 in MdBV [41]. A previous study has also shown that C. vestalis (= C. plutellae) bracovirus disables the expression of cecropin in DMB. These pieces of evidence suggest that CvBV participates in disabling the conserved NF-kB immune signaling of DMB. Surprisingly, the expression of AMPs in parasitized larvae was recovered as parasitism progressed and was even upregulated at some time points, possibly due to the waning inhibitory effect of BV, since most of BV genes exhibit spatiotemporal variation in expression [32,34,35].

The relatively stable expression of two cecropins and the dramatic induction of moricin, defensin and gloverin in parasitized host larvae following bacterial infection suggested a precise regulation of AMP gene expression by parasitism [33,42]. Among these AMPs, the expression of gloverin was extremely sensitive to stimulation, including PBS injection, gram-positive and gram-negative bacteria challenging, in parasitized host larvae. In B. mori and M. sexta, gloverin is regulated by the Toll pathway, while gloverin in DMB midgut responds to IMD interference, suggesting gloverin may be regulated by multiple pathways [42–45]. Though its effect in parasitized hosts was marginally smaller than in unparasitized ones, the higher mortality of dsGloverin-treated DMB larvae suggests that gloverin plays a vital role in the host immune response.

As C. vestalis parasitism reduces the overall number of hemocytes and inhibits the melanization of DMB larvae, the parasitized hosts would have a higher death rate following bacterial challenges [5,26]. However, the higher mortality of unparasitized DMB larvae (NP-dsGloverin) was consistent with parasitized hosts lacking teratocytes, suggesting that teratocyte-related factors may compensate for the immune suppression and increase the survival of parasitized hosts [28]. As a host-parasitoid wasp system, bacterial infections triggered AMP production in teratocytes and the host, but the transcripts in wasp larvae were relatively stable. Previous studies have shown that teratocytes exhibit antimicrobial activity by expressing AMPs and neofunctionalizing serpins [5,28,46]. These pieces of evidence show that teratocyte-expressed immune effectors are important for the host-parasitoid wasp system to outcompete for bacterial infection.

Hymenoptaecin has been well-characterized in bees and ants [47]. It is first identified in Apis mellifera and can form pores in the bacterial envelope, which inhibits the viability of gram-negative and gram-positive bacteria [48]. It is a glycine-rich antimicrobial peptide found mainly in Hymenoptera insects [48,49]. Plenty of investigations have shown that hymenoptaecin responds to pathogen infection and environmental stress, i.e. heat and pesticide treatment [50–55]. In Microplitis mediator adults, the expression levels of two hymenoptaecins are upregulated following bacterial challenges, while the role of hymenoptaecin during parasitism has not been reported yet [56]. In this study, hymenoptaecin from C. vestalis contains one proline-rich domain near the N-terminal and one glycine-rich domain near the C-terminal, showing high similarity with hymenoptaecin-1 from N. vitripennis [49]. Being active against microbes (gram-negative and gram-positive bacteria and fungi), the dramatic increase of hymenoptaecin in teratocytes and a higher survival rate of mHym-treated hosts suggest that it plays a crucial role in the survival of parasitized hosts, compensating for the BV-mediated immunosuppression [23]. Previous studies have demonstrated potentiating interactions between hymenoptaecin and other AMPs, such as abaecins [57]. We also observed that the proline-rich propeptide of Cvhym is highly conserved with pronavicin derived from Nasonia vitripennis hymenoptaecin, which exhibits antibacterial activities [18]. This suggests that naturally occurring hymenoptaecin may possess enhanced antimicrobial activity. Regarding the role of host AMPs, hymenoptaecin expression in teratocyes is dramatically increased in gloverin-knockdown hosts. This suggests that a higher dosage of hymenoptaecin is required to combat bacterial infection and that hymenoptaecin may cooperate with host gloverin to establish a defensive line against pathogens in parasitized hosts. The genomic structure and phylogenetic analysis further suggest that the duplication event occurred independently in Cotesia after separation. A duplicated hymenoptaecin gene may enhance the production of antimicrobial peptides, aiding Cotesia wasps in adapting to an endoparasitoid lifestyle.

In conclusion, we investigated the immune regulation of host-parasitic wasps during the process of parasitism focusing on AMPs. Unlike the persistent inhibition of host melanization, we found that the effect of parasitism on AMP inhibition is temporary and the host AMP expression is finely regulated. Furthermore, we figured out that teratocytes expressed hymenopteacin, which exhibits antimicrobial activities, are crucial for host survival when encountering bacterial infection. Thus, we speculated that AMPs from the host and parasitoid wasp teratocytes play essential roles in the host-parasitoid system. Our study provides a systematic perspective on host-parasitoid immune interaction by highlighting the significance of AMPs.

Funding Statement

This research was funded by the Fundamental Research Funds for the Central Universities [226–2024–00070] and the National Natural Science Foundation of China [32272607 and U22A20485].

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data supporting the findings of this study and the supplementary materials are openly available in Figshare at https://doi.org/10.6084/m9.figshare.28594571.v1.

Ethical statements

Research animals in this study do not require ethical approval.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/21505594.2025.2566243