Functional Characterization of a Venom Protein Calreticulin in the Ectoparasitoid Pachycrepoideus vindemiae

Venom proteins act in the immunological interactions between parasitoids and their host insects. The effect of venom proteins on host immunity is not fully understood in pupal parasitoids. We identified the functions of a venom protein, calreticulin (PvCRT), in the pupal ectoparasitoid Pachycrepoideus vindemiae. Here, we report that PvCRT features a signal peptide and two conserved “calreticulin” domains. Multiple sequence alignments show that PvCRT shares 83.54% amino acid identity with CRT from both Pteromalus puparum and Nasonia vitripennis, which infers a close relationship among these three species. Using qPCR analysis, we found a lower expression level of PvCRT (0.27-fold) in the venom apparatus compared to the corresponding carcass. Immunohistochemical localization revealed that PvCRT was ubiquitously expressed in venom gland. The expression of the PvCRT gene in Drosophila transgenic lines via the UAS/Gal4 binary expression system reduced the self-encapsulation phenotype of tu(1)Sz1 mutants. Additionally, studies on humoral immunity indicate that PvCRT does not affect the antimicrobial immune responses of the host. This work on an ectoparasitoid will increase our understanding of venom–mediated host-parasitoid interactions.


Introduction
Parasitoids are a valuable group of natural enemies used to manage arthropod pests. In the typical case, female wasps lay eggs either inside (endoparasitoid) or on the surface (ectoparasitoid) of their hosts where their larvae develop [1]. Ultimately, the host dies during the late stage of parasitoid development. The variations in the developmental strategies of endoparasitoids and ectoparasitoids suggest that parasitoids manipulate the physiology of their hosts in different ways [2]. Parasitic factors, including venom proteins [3,4], polydnaviruses (PDVs) [5], virus-like particles (VLPs) [6] and ovarian secretions [7,8] are injected into the host while the eggs are laid. Depending on the specific host/parasitoid relationship, some or all of these factors operate to manipulate host development and immunity, which are necessary for successful parasitism.
Venom proteins suppress host immunity, regulate host physiological process and change host metabolism and behavior [1,3]. Venoms are rich in bioactive proteins and peptides [1]. Despite the

cDNA Synthesis and Quantitative Real-Time PCR
The first-strand complementary DNA (cDNA) was synthesized from total RNA using PrimeScript™ RT Reagent Kit with gDNA Eraser (Takara, Beijing, China). qPCR was carried out using the ChamQ TM SYBR ® qPCR Master Mix (Vazyme, Nanjing, China) and run on a CFX96™ Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA) following the manufacturer's instructions. The specific qPCR primers were designed using AlleleID 6 software (PREMIER Biosoft, Palo Alto, CA, USA) (Supplementary Materials, Figure S1), gene expression levels were normalized to the reference gene (28S rRNA) [35]. The qPCR programs were set as following: enzyme activation at 95 • C for 30 s, followed by 40 cycles with denaturation at 95 • C for 5 s, annealing at 60 • C for 30 s, and melting curve analysis. The mRNA expression levels were determined by the comparative 2 − CT method [36].

Gene Cloning
cDNA from P. vindemiae venom apparatus was used as a template to clone the PvCRT. The PCR reaction included 5 µL cDNA template (20 ng/µL), 1.5 µL of each gene-specific primer (10 µmol/L), 25 µL KOD One TM PCR Master Mix (Toyobo, Shanghai, China) and 17 µL sterile water. PCR primers were designed using the Primer Premier 6 software (PREMIER Biosoft International, Palo Alto, CA, USA) (Supplementary Materials, Figure S1). The reaction conditions were 35 cycles of 98 • C for 10 s, 60 • C for 5 s, and 68 • C for 7 s. The pUASTattB plasmid (GenBank: EF362409.1), a cloning vector for producing Drosophila transgenic lines, was digested by EcoRI and KpnI (Thermo, Carlsbad, CA, USA). PCR amplification product was analyzed on 1.0% agarose gel, and recombined into the enzyme-digested vector using ClonExpress Ultra One Step Cloning Kit (Vazyme, Nanjing, China). Finally, positive cloning was verified by DNA sequencing.

Gal4-Driven Expression of PvCRT
Transgenic lines were generated by the Drosophila Resources and Technology Platform (Shanghai Institute of Life Sciences, Chinese Academy of Sciences). Appropriate crosses were performed to obtain a homozygous line carrying two copies of PvCRT (UAS-PvCRT). To explore the functions of PvCRT on host immunity, Gal4-driven expression in immune tissues (fat body and hemocytes) was initiated by crossing the Cg-Gal4 (Cg) lines to UAS-PvCRT lines and the offspring was denoted Cg > UAS-PvCRT [37]. The hybrids between Cg lines and WT lines (Cg/WT), and the crossed offspring between WT lines and UAS-PvCRT lines (WT/UAS-PvCRT) were used as control. All crosses were performed at 25 • C. We established a positive control of encapsulation in pupal Drosophila using the temperature-sensitive mutant tu(1)Sz 1 . At 28 • C tu(1)Sz 1 larvae form melanotic tumors, involving the encapsulation of abnormal caudal fat body regions [24]. In vivo self-encapsulation experiments were conducted by crossing virgin temperature-sensitive mutant tu(1)Sz 1 to Cg > UAS-PvCRT lines. Cg/WT lines were used as control. Once crossed, offspring were maintained at 28 • C for further self-encapsulation observation. Larvae and pupae were then scored for the tu(1)Sz 1 phenotype as described [38].

Genomic DNA Extraction and PCR Detection
Sets of ten D. melanogaster pupae were collected into 1.5 mL centrifuge tubes followed by genomic DNA extraction according to the manufacture's protocol using Fastpure Cell/Tissue DNA Isolation Mini Kit (Vazyme, Nanjing, China). Total RNA of Drosophila pupae was extracted and cDNA was synthesized as described. PCR amplification was conducted using PvCRT gene-specific primers (Supplementary Materials, Figure S1). PCR reactions were performed as described and amplification products were analyzed on a 1.0% agarose gel, cloned into TA/Blunt-Zero Vector (Vazyme, Nanjing, China) and then sequenced.

Western Blotting
Venom proteins from P. vindemiae were collected as described [39], and total proteins of the carcass, whole body of P. vindemiae or D. melanogaster were separately extracted using Minute™ total protein extraction kit for animal cultured cells/tissues (Invent Biotechnologies, Beijing, China). The protein concentration was determined by a modified Bradford protein assay Kit (Sangon Biotech, Shanghai, China). Protein extracts (1 µg) were denatured in sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) sample-loading buffer and separated on 12% SDS-PAGE gels, then transferred onto polyvinylidene fluoride membranes (Merck Millipore, Darmstadt, Germany). A 5% non-fat milk solution was used for blocking the membranes and the primary antibody was incubated at 4 • C overnight. Rabbit polyclonal antibodies against PvCRT, β-actin were diluted 1:2000 and 1:4000, respectively. Goat anti-rabbit IgG-horseradish peroxidases conjugate diluted 1:4000 (GenScript, Nanjing, China) was used as the secondary antibody for 1 h incubation. The signal was developed using the super signal west dura extended duration substrate (Thermo Scientific, Waltham, MA, USA) and detected with a UVP ChemiDoc-It Imaging System (Upland, UVP Company, CA, USA).

Immunofluorescence Staining
Immunofluorescence staining was conducted as described previously [40]. Briefly, the venom apparatus of 4-day old female wasps was dissected and fixed with 4% polyformaldehyde for 30 min, followed by rinsing with 0.3% phosphate buffered saline with Triton-X 100 solution (PBST) three times. Subsequently, samples were blocked for 1 h with 5% goat serum (Sangon Biotech, Shanghai, China), diluted in PBST and washed for 20 min with PBST three times. The rabbit anti-PvCRT primary antibody (diluted 1:100 in 5% goat serum) was incubated for 48 h at 4 • C. Rabbit serum that had not been immunized was used as control. Samples were washed for 20 min with PBST three times and incubated with the Dylight 488-conjugated goat anti-rabbit secondary antibody

Bacterial Infection and Survival Analysis of D. melanogaster Adults
We inoculated 1 mL LB medium with Pseudomonas aeruginosa or Staphylococcus aureus, and allowed them to grow at 37 • C until OD 600 nm reached 0.03 and 0.1, respectively. We centrifuged 1 mL of each culture for 3 min at 5000× g and discarded the supernatant. The bacterial pellets were washed 3× with 1 mL PBS, re-suspended in 1 mL PBS and stored at 4 • C for inoculation. We collected 5-7 day old male flies and anesthetized them with CO 2 . A thin metal needle was sterilized with ethanol before dipping into the bacterial solution and the lateral side of the fly's thorax was pricked. We transferred the pricked flies with a brush to a sterile vial containing corn-meal fly medium and incubated them at 25 • C. Flies were collected at indicated time points. Uninfected flies were used as controls. Total RNA extraction and qPCR were performed as described above. Primers for diptericin (Dpt) and drosomycin (Drs) genes were used to quantify the expression levels of antimicrobial peptides, ribosomal protein L32 (RpL32) was used as internal control [41]. The survival graphs represent one representative experiment out of three independent repeats with at least 20 flies per genotype.

Data Analysis
The gene expression levels for qPCR were analyzed by unpaired two-tailed Student's t Test and one-way ANOVA followed by Tukey's multiple comparison tests. The penetrance data were analyzed by the Fisher Exact test. All statistical analyses were carried out by the data processing system (DPS) package version 9.50 and statistical significances were marked with asterisks (ns: no significant difference, *: p < 0.05, **: p < 0.01, ***: p < 0.001) or different letters (p < 0.05) [42]. All figures were plotted using GraphPad Prism 7.0 (GraphPad, San Diego, CA, USA).

Sequence Analysis of PvCRT
The coding DNA sequence (CDS) of PvCRT obtained by searching the transcriptome of venom apparatus was validated by gene cloning and sequencing. Our analysis showed the full-length CDS of the PvCRT gene is 1224 bps, encoding 408 amino acids with a predicted molecular weight of 47.65 kDa and an isoelectric point of 4.39. In agreement with previous reported CRTs, PvCRT features a signal peptide (1-19 aa) in the N-terminus and two conserved "calreticulin" domains ( Figure 1), which may be crucial for its functional performance. However, unlike the PpCRT, [11] we did not identify the "coiled-coil" domain in PvCRT.
Insects 2020, 11, x FOR PEER REVIEW 5 of 15 The gene expression levels for qPCR were analyzed by unpaired two-tailed Student's t Test and one-way ANOVA followed by Tukey's multiple comparison tests. The penetrance data were analyzed by the Fisher Exact test. All statistical analyses were carried out by the data processing system (DPS) package version 9.50 and statistical significances were marked with asterisks (ns: no significant difference, *: p < 0.05, **: p < 0.01, ***: p < 0.001) or different letters (p < 0.05) [42]. All figures were plotted using GraphPad Prism 7.0 (GraphPad, San Diego, CA, USA).

Sequence Analysis of PvCRT
The coding DNA sequence (CDS) of PvCRT obtained by searching the transcriptome of venom apparatus was validated by gene cloning and sequencing. Our analysis showed the full-length CDS of the PvCRT gene is 1224 bps, encoding 408 amino acids with a predicted molecular weight of 47.65 kDa and an isoelectric point of 4.39. In agreement with previous reported CRTs, PvCRT features a signal peptide (1-19 aa) in the N-terminus and two conserved "calreticulin" domains ( Figure 1), which may be crucial for its functional performance. However, unlike the PpCRT, [11] we did not identify the "coiled-coil" domain in PvCRT. The PvCRT shared the highest amino acid identity (83.54%) with venom CRT from N. vitripennis (BlastP, E-value = 0) and P. puparum (BlastP, E-value = 0). Multiple sequence alignments based on the "calreticulin" domains demonstrated that PvCRT has highly conserved properties with other hymenopteran insects ( Figure 2). The PvCRT shared the highest amino acid identity (83.54%) with venom CRT from N. vitripennis (BlastP, E-value = 0) and P. puparum (BlastP, E-value = 0). Multiple sequence alignments based on the "calreticulin" domains demonstrated that PvCRT has highly conserved properties with other hymenopteran insects ( Figure 2).   [32,44]. The conserved amino acid residues were highlighted, red means the entirely conserved amino acid residues and yellow represents the 70-100% conservation in amino acid residues.

Phylogenetic Analysis
We chose 17 CRTs for phylogenetic reconstruction [15,45], which indicated that CRTs from hymenopteran insects clustered together and diverged from other orders. Among the clustered hymenopteran CRTs, PvCRT is more closely related to N. vitripennis and P. puparum, as seen in the sequence analyses ( Figure 3).

Phylogenetic Analysis
We chose 17 CRTs for phylogenetic reconstruction [15,45], which indicated that CRTs from hymenopteran insects clustered together and diverged from other orders. Among the clustered hymenopteran CRTs, PvCRT is more closely related to N. vitripennis and P. puparum, as seen in the sequence analyses ( Figure 3).

Transcriptional Profiles of PvCRT
The transcriptional profiles of the venom apparatus and carcass were compared ( Figure 4A). We recorded lower PvCRT expression in the venom apparatus (0.27-fold) relative to the carcass. Western blotting also showed the expression of the gene in the venom apparatus and carcass. The accumulation of mRNAs encoding PvCRT over 1-7 days post-eclosion in female adults shows fairly stable expression on days 2-3 and 5-7 post-eclosion ( Figure 4B). The mRNA level appeared to reach a peak on day 4, although it only differed significantly from day 1. The small numerical differences do not suggest a biological significance ( Figure 4B). Immunoblotting also indicated the presence of PvCRT in the female adults over days 1-7 post-eclosion.

Transcriptional Profiles of PvCRT
The transcriptional profiles of the venom apparatus and carcass were compared ( Figure 4A). We recorded lower PvCRT expression in the venom apparatus (0.27-fold) relative to the carcass. Western blotting also showed the expression of the gene in the venom apparatus and carcass. The accumulation of mRNAs encoding PvCRT over 1-7 days post-eclosion in female adults shows fairly stable expression on days 2-3 and 5-7 post-eclosion ( Figure 4B). The mRNA level appeared to reach a peak on day 4, although it only differed significantly from day 1. The small numerical differences do not suggest a biological significance ( Figure 4B). Immunoblotting also indicated the presence of PvCRT in the female adults over days 1-7 post-eclosion.

Immunohistochemical Visualization of PvCRT in Venom Apparatus
Immunoblotting shows that PvCRT is expressed in the venom apparatus. We localized the PvCRT antigen within the venom apparatus, shown as green staining in the venom gland (Vg) and, with less staining in the venom reservoir (Vr) ( Figure 5).

Immunohistochemical Visualization of PvCRT in Venom Apparatus
Immunoblotting shows that PvCRT is expressed in the venom apparatus. We localized the PvCRT antigen within the venom apparatus, shown as green staining in the venom gland (Vg) and, with less staining in the venom reservoir (Vr) ( Figure 5).

Immunohistochemical Visualization of PvCRT in Venom Apparatus
Immunoblotting shows that PvCRT is expressed in the venom apparatus. We localized the PvCRT antigen within the venom apparatus, shown as green staining in the venom gland (Vg) and, with less staining in the venom reservoir (Vr) ( Figure 5).

Construction of PvCRT Transgenic Drosophila
We produced Drosophila transgenic lines carrying a UAS transgene encoding the PvCRT protein (UAS-PvCRT) and confirmed the presence of PvCRT in Drosophila by PCR. Figure 6A shows a specific band of about 1200 bp in UAS-PvCRT line, but not in the WT. The PvCRT was validated by DNA sequencing. The expression of the PvCRT was initiated by crossing the UAS-PvCRT lines to Cg driver (Cg > UAS-PvCRT) [11][12][13][14][15][16][17][18]. We validated the transcription of PvCRT by PCR using cDNA from Cg > UAS-PvCRT, Cg/WT and WT/UAS-PvCRT lines as temples. We found a specific band was amplified in Cg > UAS-PvCRT lines and confirmed the PvCRT by DNA sequencing ( Figure 6B). This shows that PvCRT was transcribed in Cg > UAS-PvCRT lines. We recorded a specific band corresponding to the predicted size in both Drosophila transgenic lines and controls by western blotting ( Figure 6C). We speculate this is due to the high similarity in amino acid sequence (80.5%) between PvCRT and D. melanogaster CRT (Supplementary Figure S1), implying the high potential for cross-reactivity.

Construction of PvCRT Transgenic Drosophila
We produced Drosophila transgenic lines carrying a UAS transgene encoding the PvCRT protein (UAS-PvCRT) and confirmed the presence of PvCRT in Drosophila by PCR. Figure 6A shows a specific band of about 1200 bp in UAS-PvCRT line, but not in the WT. The PvCRT was validated by DNA sequencing. The expression of the PvCRT was initiated by crossing the UAS-PvCRT lines to Cg driver (Cg > UAS-PvCRT) [11][12][13][14][15][16][17][18]. We validated the transcription of PvCRT by PCR using cDNA from Cg > UAS-PvCRT, Cg/WT and WT/UAS-PvCRT lines as temples. We found a specific band was amplified in Cg > UAS-PvCRT lines and confirmed the PvCRT by DNA sequencing (Figure 6B). This shows that PvCRT was transcribed in Cg > UAS-PvCRT lines. We recorded a specific band corresponding to the predicted size in both Drosophila transgenic lines and controls by western blotting ( Figure 6C). We speculate this is due to the high similarity in amino acid sequence (80.5%) between PvCRT and D. melanogaster CRT (Supplementary Figure S1), implying the high potential for cross-reactivity.

PvCRT Inhibits Encapsulation of the Host
In vivo encapsulation assays showed a higher rate of self-encapsulation in larvae and pupae of the Cg/WT Drosophila parental lines, whereas the tu(1)Sz 1 phenotype was significantly rescued in Cg > UAS-PvCRT parental lines (Figure 7). We conclude that PvCRT inhibited encapsulation of the host.

PvCRT Inhibits Encapsulation of the Host
In vivo encapsulation assays showed a higher rate of self-encapsulation in larvae and pupae of the Cg/WT Drosophila parental lines, whereas the tu(1)Sz 1 phenotype was significantly rescued in Cg > UAS-PvCRT parental lines (Figure 7). We conclude that PvCRT inhibited encapsulation of the host.

Construction of PvCRT Transgenic Drosophila
We produced Drosophila transgenic lines carrying a UAS transgene encoding the PvCRT protein (UAS-PvCRT) and confirmed the presence of PvCRT in Drosophila by PCR. Figure 6A shows a specific band of about 1200 bp in UAS-PvCRT line, but not in the WT. The PvCRT was validated by DNA sequencing. The expression of the PvCRT was initiated by crossing the UAS-PvCRT lines to Cg driver (Cg > UAS-PvCRT) [11][12][13][14][15][16][17][18]. We validated the transcription of PvCRT by PCR using cDNA from Cg > UAS-PvCRT, Cg/WT and WT/UAS-PvCRT lines as temples. We found a specific band was amplified in Cg > UAS-PvCRT lines and confirmed the PvCRT by DNA sequencing (Figure 6B). This shows that PvCRT was transcribed in Cg > UAS-PvCRT lines. We recorded a specific band corresponding to the predicted size in both Drosophila transgenic lines and controls by western blotting ( Figure 6C). We speculate this is due to the high similarity in amino acid sequence (80.5%) between PvCRT and D. melanogaster CRT (Supplementary Figure S1), implying the high potential for cross-reactivity.

PvCRT Inhibits Encapsulation of the Host
In vivo encapsulation assays showed a higher rate of self-encapsulation in larvae and pupae of the Cg/WT Drosophila parental lines, whereas the tu(1)Sz 1 phenotype was significantly rescued in Cg > UAS-PvCRT parental lines (Figure 7). We conclude that PvCRT inhibited encapsulation of the host.

The Roles of PvCRT in Antimicrobial Immunity of the Host
Bacterial challenge led to multi-fold increases in mRNAs encoding Dpt at 6 h post-infection by gram-negative bacteria, P. aeruginosa, but there were no differences between transgenic flies and control ( Figure 8A). Similar challenges with the gram-positive S. aureus led to increased mRNAs encoding Drs at 24 h. Again, there were no differences between transgenic lines and control ( Figure 8B). We further tested the involvement of PvCRT in host survival after bacterial infection. In line with the above results, transgenic flies did not show an increased susceptibility to these infections compared to the control ( Figure 8C,D). Thus, we infer that PvCRT is not involved in the antibacterial immunity of the host.

The Roles of PvCRT in Antimicrobial Immunity of the Host
Bacterial challenge led to multi-fold increases in mRNAs encoding Dpt at 6 h post-infection by gram-negative bacteria, P. aeruginosa, but there were no differences between transgenic flies and control ( Figure 8A). Similar challenges with the gram-positive S. aureus led to increased mRNAs encoding Drs at 24 h. Again, there were no differences between transgenic lines and control ( Figure  8B). We further tested the involvement of PvCRT in host survival after bacterial infection. In line with the above results, transgenic flies did not show an increased susceptibility to these infections compared to the control ( Figure 8C,D). Thus, we infer that PvCRT is not involved in the antibacterial immunity of the host. Drosophila adults were infected with bacteria and monitored for survival (n ≥ 18). One-way ANOVA between different groups were performed followed by Tukey's multiple comparison tests, and the same letters mean no significant difference.

Discussion
As a cosmopolitan wasp, P. vindemiae attacks the puparia of many cyclorrhaphous Diptera. They successfully develop from the hosts previously parasitized by other parasitoids, indicating that P. vindemiae has the potential for expanding its host range [46]. Females of P. vindemiae parasitize the invasive spotted wing drosophila, which provides one of the most promising biological resources to control D. suzukii [47]. The efficiency of biological control on spotted-wing drosophila can be improved through conservation or augmentative releases of P. vindemiae [48]. These explorative works illustrate that P. vindemiae could be developed into a biological control agent.
Female P. vindemiae inject venom proteins into their hosts while laying eggs. The venom proteins act in host immune regulation [23]. In general, venom proteins act in inhibiting immunity, interrupting development, and regulating the metabolism of parasitoid hosts [4]. Parasitoids evolve venom proteins to suppress host immune reactions such as hemocyte spreading, encapsulation and Drosophila adults were infected with bacteria and monitored for survival (n ≥ 18). One-way ANOVA between different groups were performed followed by Tukey's multiple comparison tests, and the same letters mean no significant difference.

Discussion
As a cosmopolitan wasp, P. vindemiae attacks the puparia of many cyclorrhaphous Diptera. They successfully develop from the hosts previously parasitized by other parasitoids, indicating that P. vindemiae has the potential for expanding its host range [46]. Females of P. vindemiae parasitize the invasive spotted wing drosophila, which provides one of the most promising biological resources to control D. suzukii [47]. The efficiency of biological control on spotted-wing drosophila can be improved through conservation or augmentative releases of P. vindemiae [48]. These explorative works illustrate that P. vindemiae could be developed into a biological control agent.
CRTs suppress host cellular and humoral immunity [10][11][12]. In the C. rubecula/P. rapae model, venom CRT inhibits hemocyte spreading and encapsulation of their host in a dose-dependent manner [12]. Similarly, PpCRT suppressed the cellular immunity of P. rapae hemocytes [11]. In the endoparasitoid C. plutellae, CRT inhibited nodule formation in its host P. xylostella [15]. Siebert et al. reported that CRT from N. vitripennis venom inhibits melanization of its host Sarcophaga bullata [10]. Analysis of the calculated PvCRT amino acid sequence shows a putative signal peptide of 22 amino acids in the N-terminal region and two highly conserved "calreticulin" domains analogous to other parasitoid CRTs. The coiled-coil motif does not occur in PvCRT. The coiled-coil domain from PpCRT is responsible for its entry into host hemocytes [11]. On the basis of multiple sequence alignments, PvCRT features higher homology with CRTs from N. vitripennis and P. puparum, from which we infer similar protein functions. Phylogenetic analysis indicates PvCRT clustered with P. puparum and N. vitripennis. These proteins may have evolved from the same ancestor. More distant relationships indicate that PvCRT diverged from other insect orders, implying their functional differences.
CRT is a protein with multiple cellular functions in invertebrates and vertebrates [9]. PvCRT expression occurred in the carcass and venom, as seen in other parasitoids [10][11][12], indicating it probably acts in host and parasitoid. Despite their lower abundance, PvCRT may act in inhibiting host immune response, as reported for PpS1V in P. puparum [55]. The expression pattern of PvCRT both in venom and non-venom apparatus imply that a novel functionalization of PvCRT occurred during the co-evolution between P. vindemiae and its hosts. We thus suspect that it was recruited into venoms to perform venom functions [58], and the questionnaire would have been more useful if it had asked participants about the origin of PvCRT.
The Drosophila transgenic lines integrated PvCRT, which was transcribed and expressed by the Cg driver. Crossing the Cg/WT Drosophila lines with tu(1)Sz 1 mutants led to encapsulation of their own tissue in a manner similar to wasp egg encapsulation [38]. The self-encapsulation phenotype of tu(1)Sz 1 was significantly rescued in Drosophila transgenic lines in contrast to control flies. These results agreed with the suppression of host hemocytic encapsulation observed in PpCRT [11]. We speculate that PvCRT created a calcium burst in a similar manner with the SERCA, which inhibits encapsulation [38]. Infection either with P. aeruginosa or S. aureus led to high levels of mRNAs encoding antimicrobial peptides, albeit with no differences between Cg > UAS-PvCRT and Cg/WT, WT/UAS-PvCRT Drosophila lines. Our additional survival experiments did not reveal a role for PvCRT in increasing susceptibility to these bacterial infections. We infer that PvCRT does not act in host antimicrobial immunity.

Conclusions
While endo-and ectoparasitoids are closely related at the phylogenetic level, their life histories and embryonic development differ. Grbic and Strand reported that embryonic development of the ectoparsitoid Bracon hebetor is similar to free-living insects, with relatively large, yolky eggs, while the endoparasitoid Aphidius ervi lays small, yolk-free eggs [59]. Hence, ectoparasitoids differ from endoparasitoids in fundamental ways. The PvCRT facilitates studies of another axis of ecdo-and endoparasitism-the influence of the parasitoids on host immunology. Our work on this axis indicates considerable differences in the biological significance of PvCRT and probably many other proteins in venoms from endo-and ectoparasitoids. Continued work along this axis will yield new, fundamental insights into parasitoid histories.