|Home | About | Journals | Submit | Contact Us | Français|
Toxoneuron nigriceps (Hymenoptera, Braconidae) is an endophagous parasitoid of the larval stages of the tobacco budworm, Heliothis virescens (Lepidoptera, Noctuidae). The bracovirus associated with this wasp (TnBV) is currently being studied. Several genes expressed in parasitised host larvae have been isolated and their possible roles partly elucidated. TnBVank1 encodes an ankyrin motif protein similar to insect and mammalian IκB, an inhibitor of the transcription nuclear factor κB (NF-κB). Here we show that, when TnBVank1 was stably expressed in polyclonal Drosophila S2 cells, apoptosis is induced. Furthermore, we observed the same effects in haemocytes of H. virescens larvae, after TnBVank1 in vivo transient transfection, and in haemocytes of parasitised larvae. Coimmunoprecipitation experiments showed that TnBVANK1 binds to ALG-2 interacting protein X (Alix/AIP1), an interactor of apoptosis-linked gene protein 2 (ALG-2). Using double-immunofluorescence labeling, we observed the potential colocalization of TnBVANK1 and Alix proteins in the cytoplasm of polyclonal S2 cells. When Alix was silenced by RNA interference, TnBVANK1 was no longer able to cause apoptosis in both S2 cells and H. virescens haemocytes. Collectively, these results indicate that TnBVANK1 induces apoptosis by interacting with Alix, suggesting a role of TnBVANK1 in the suppression of host immune response observed after parasitisation by T. nigriceps.
Endophagous parasitoids belonging to the order of Hymenoptera develop during juvenile stages in the body of their hosts. During oviposition, the endoparasitoid female injects into the host several factors that alter its physiology, suppressing the host immune system, the endocrine balance, as well as reproductive activity or metabolism1. Among parasitoid secretions, viruses belonging to the family of Polydnaviridae (PDVs) play a key role in the success of parasitism2,3. PDVs are obligate symbionts of ichneumoid and braconid wasps attacking exclusively larval stages of their lepidopteran hosts4. Polydnavirus DNA is integrated into the genome of the parasitoid and therefore vertically transmitted through the germline4. The viral particles, containing circular dsDNA molecules of different sizes, replicate only in the epithelium of parasitoid female ovarian calyx, and are injected into the host, along with ovarian proteins, venom and the egg, during oviposition. PDVs express genes in several target tissues of the host and, together with other maternal or embryonic factors, lead to several functional alterations of their host, providing a suitable nutritional environment to the parasitoid larvae5. Toxoneuron nigriceps (Viereck) (Hymenoptera, Braconidae) is an endoparasitoid wasp belonging to the Braconidae family, parasitising the larval stages of tobacco budworm Heliothis virescens (Fabricius) (Lepidoptera, Noctuidae). Toxoneuron nigriceps Bracovirus (TnBV) encodes several gene families6,7 including ankyrin motif proteins (ANKs)8. The TnBVANK proteins show high levels of amino acid identity to the ankyrin motif of the Drosophila NF-κB inhibitor (IκB)-related protein cactus8,9. Three ankyrin-like open reading frames (ORFs) were identified in the TnBV genome and were denoted TnBVank1-38. TnBVank genes are characterised by ankyrin motif sequence repeats but the usually phosphorylatable serine residues at the N-terminus are absent8. Moreover, none of putative proteins encoded by the TnBVank genes has the C-terminal PEST domain present in the protein Cactus/IκB, which is necessary for signal-mediated degradation10–12 and for the control of protein turnover13. The predicted structure of these proteins suggested that TnBV ankyrin proteins may bind NF-kB/Rel transcription factors of the tumour necrosis factor (TNF)/Toll immune pathway altering the signal transduction cascade8. Indeed, it was demonstrated that in parasitised H. virescens larvae, after bacterial challenge, the nuclear import of NF-κB was inhibited8. Moreover, transfection experiments in human HeLa cells demonstrated that a TnBVank1 gene product reduced the efficiency of the TNF-α-induced expression of a reporter gene under NF-κB transcriptional control, confirming that viral IκB-like proteins are involved in the suppression of the insect immune response8,14,15.
TnBVank1 and TnBVank3 are expressed as early as three hours and six hours, respectively, after parasitisation in haemocytes of H. virescens last instar larvae, but the transcripts are present also several hours after parasitisation reducing over the time. Although expression of TnBVank1 and TnBVank3 is high in H. virescens haemocytes, a much more limited expression level can also be found in the fat body8. Moreover, it is reported that several hours after natural parasitism, the granulocytes, a subpopulation of haemocytes, showed a number of structural damages, as evidenced by actin cytoskeleton disruption and loss of adhesion properties, with general morphological changes which suggest the occurrence of apoptosis16. This further suggests a possible role of parasitism in early immune response suppression.
In insects the processes of cell death are involved in development, but also play an important role in allowing, maintaining and regulating many host-parasitoid interactions. Accordingly, it was shown that parasitoids are able to induce or regulate cell death phenomena in host tissues16–20. Here, we report that TnBVANK1 induces cell death in polyclonal Drosophila S2 cells that stably express TnBVank1. To better understand the role of TnBVANK1, coimmunoprecipitation and coimmunolocalisation experiments were carried out in polyclonal Drosophila S2 cells and the interaction between TnBVANK1 and ALG-2 interacting protein X (Alix) was investigated. Alix is a multifunctional protein, it is ubiquitously expressed and concentrated in phagosomes and exosomes and it was first characterised as an interactor of apoptosis-linked gene protein 2 (ALG-2) that is a Ca2+-binding protein necessary for cell death21. Here, we show that upon Alix silencing by RNA interference (RNAi), TnBVANK1 was no longer able to induce apoptosis in S2 cells.
These data were further confirmed by in vivo transient transfection of TnBVank1 and by silencing of Alix transcript by RNAi in H. virescens larval haemocytes.
We analysed the profile of polyclonal S2 cells stably expressing TnBVank1 (S2[ank1]) by flow cytometry. We used S2 cells and polyclonal S2[ank1] cells 20h after treatment with the cytotoxic alkaloid camptothecin as positive controls and untransfected S2 cells as negative control. All the cell samples were harvested at the same splitting passages and were doubly stained with Annexin V–FITC and PI (Propidium Iodide). Results revealed a significant increase of apoptotic cells in polyclonal S2[ank1] cells line compared to the negative control, indeed the percentage of apoptotic cells, amounted to 39.56%, was not significantly different from the positive controls (S2+CAM and S2[ank1]+CAM) (Fig. 1a).
To further investigate the involvement of TnBVank1 in cell death pathway, we analysed also the caspase-3 activity. Caspase activity detection is recognised as a standard method for assessing apoptosis22. Caspase-3 is activated early during apoptosis. We assayed its activity in polyclonal S2 cells stably expressing TnBVank1 by using the fluorogenic effector caspase substrate Ac-DEVD-ACM. Cell lysates were harvested from S2 cells stably expressing TnBVank1, untransfected S2 cells (negative control), as well as both the positive controls. Caspase-3 activity was significantly increased in S2[ank1] cells compared to the negative control S2 but comparable to positive controls (S2+CAM and S2[ank1]+CAM) (Fig. 1b). This result suggests that TnBVank1 stimulates the activation of caspase-3 indicating a role of TnBVANK1 in induction of apoptosis in S2 cells.
Moreover, the nuclear morphology, examined by staining with Hoechst 33258, showed intact nuclei appearing light blue in colour in the negative control (S2 cells) while polyclonal S2 cells exhibited bright blue indicating condensed or fragmented nuclei, typical morphological characteristics of apoptosis (see Supplementary Fig. S1).
To test whether the production of TnBVANK1 decreases over time in the stable polyclonal cell line, as a consequence of apoptotic cell death, a western blot time-course analysis was performed on cell lysates harvested at subsequent splitting passages (2, 4, 10, 20, 26, 39 passages). Western blot analysis showed an effective reduction of TnBVank1 gene expression over time, indicating that the population of cells, stably expressing TnBVank1 within the stable polyclonal cell line, progressively decreases with a concomitant increase of cell number not expressing TnBVank1 (Fig. 2). This result strongly suggest that the expression of the viral gene TnBVank1 induces cell death in S2 cells.
Haemocytes exhibited transient expression of TnBVank1 when the recombinant vector pIZT/V5-His-ANK1, mixed with a transfection reagent, was injected into the haemocoel of fifth instar H. virescens larva (day one). After trying different transfection reagents, and different time after injection, we selected Lipofectin as the most efficient one and 56h as the best time for in vivo transient expression, at least for our experimental conditions. The expression of TnBVank1 transcripts in transfected haemocytes was detected by reverse transcriptase-polymerase chain reaction (RT-PCR) 56h after injection of pIZT/V5-His-ANK1 using specific primers for TnBVank1 (Fig. 3a-i). The expression of TnBVANK1 protein was assessed by western blot analysis using the anti-V5 antibody (Fig. 3a-ii) and detecting the epifluorescence signal of GFP reporter and the TnBVANK1 protein expression (red signal) by immunofluorescence (Fig. 3b). No signals were detected in RT-PCR, western blot and immunofluorescence experimental controls (haemocytes transfected with the empty pIZT/V5-His vector) except in this last case in which the epifluorescence signal of GFP reporter was detected. Different haemocyte types were successfully transfected and the in vivo transfection efficiency was calculated by counting fluorescent haemocytes in the control because of the signal of GFP protein, and in haemocytes in vivo transfected (see Supplementary Fig. S2). Approximately 80% of haemocytes expressed the GFP reporter (Hemo) or both the GFP reporter and the viral gene TnBVank1 (Hemo[ank1]), respectively (Fig. 3c).
We investigated the effect of natural parasitism and of in vivo transfected haemocytes expressing TnBVank1 on apoptosis by flow cytometric analysis. Cells undergoing early and late stages of apoptosis were single positive for Annexin V and double positive for Annexin V and PI respectively. Apoptotic cells were detected in haemocytes deriving from parasitised larvae (Hemo[para]) and in haemocytes from in vivo transfected larvae (Hemo[ank1]), as well as in the positive control (haemocytes treated with camptothecin). As negative control haemocytes transfected with the empty pIZT/V5-His vector were used and as positive controls, camptothecin (1μM) was injected into haemocoel of the larvae with the vector pIZT/V5-His either with or without TnBVank1 (Fig. 4a).
To further investigate apoptosis with same haemocytes samples, the caspase-3 activity was measured, using the fluorogenic effector caspase substrate Ac-DEVD-ACM. Haemocytes transfected with TnBVank1 (Hemo[ank1]) and haemocytes deriving from parasitised larvae (Hemo[para]) showed a strong increase of caspase-3 activity compared to haemocytes transfected with the empty pIZT/V5-His vector (negative control) (Hemo) (Fig. 4b).
Moreover, phalloidin staining showed actin filaments grouped in bundles under the membrane both in haemocytes from parasitised larvae and in TnBVank1 transfected haemocytes (Fig. 4c). The percentage of apoptotic cells, the caspase-3 activity and the phalloidin staining were comparable in parasitised haemocytes and in TnBVank1 transfected haemocytes. These results suggest that the apoptosis does not seem to be due to an overexpression of the gene TnBVank1 after in vivo transfection respect to its expression in haemocytes after parasitisation. Indeed, when a qPCR on haemocytes collected after in vivo transfection with TnBVank1 and after parasitisation was performed, results did not show significant differences of TnBVank1 transcript expression level after parasitisation and after in vivo transfection (see Supplementary Fig. S3).
Also the nuclei staining with Hoechst 33342 revealed the occurring of apoptosis in H. virescens haemocytes transfected with TnBVank1 (Hemo[ank1]). The cells in the haemocytes transfected with the empty pIZT/V5-His vector (negative control) (Hemo) showed roundshaped and homogeneously stained nuclei, whereas haemocytes transfected with TnBVank1 (Hemo[ank1]) showed accumulation of fluorescent dye indicating obvious chromatin condensation and fragmentation (see Supplementary Fig. S1).
Isolation of TnBVANK1 complexes was performed on a total protein extract from S2 cells stably expressing a V5-tagged form of the protein. Following pre-cleaning treatment, the unbound fraction was immunoprecipitated with an anti-V5 specific antibody in order to capture TnBVANK1 functional complexes. The occurrence of TnBVANK1 protein within the immunoprecipitated material was detected by western blot using anti-V5 antibody (see Supplementary Fig. S4a). Immunoprecipitated complexes were fractionated by SDS-PAGE (see Supplementary Fig. S4b) and protein bands, excised from the gel, were identified by LC-MS/MS analysis. Protein bands from the control S2 cells were also analysed as control. Putative TnBVANK1 protein interactors were obtained by differential analysis comparing the proteins identified in S2 cells stably expressing TnBVank1 and those from the control S2 cells.
The results of this functional proteomic approach are summarised in Supplementary Table S1. As the experiment was performed on the Drosophila S2 cell line, putative interactors were annotated against the Drosophila melanogaster genome and then compared to Homo sapiens proteins. Among interactor proteins, NF-Κb p110 subunit was identified, confirming previous studies showing interaction between TnBVANK1 and the transcription factor NF-κB8.
Moreover, we found several proteins involved in the apoptotic pathway, as the transcription factor Stwl23, the mitochondrial carrier homolog 1 (Mtch)24, ALG-2 interacting protein X21 and the GTPase Rab125,26. Among these proteins, we focused our attention on the ALG-2 interacting protein X, also called Alix. This protein was chosen on the basis of its involvement in programmed cell death, for subsequent analyses.
To further support the interaction between TnBVANK1 and Alix, the localization of both these proteins was examined in polyclonal S2 cell line by immunofluorescence labeling using both anti-Alix and anti-TnBVANK1 antibodies. Figure 5(ii,iii) shows that both TnBVANK1 (green signal) and Alix (red signal) are widely distributed throughout the cytoplasm of polyclonal S2 cell line and the merged picture (Fig. 5-iv) indicates that the two proteins seem to colocalise, as there is a clear overlap of red and green signals of analysed proteins. This result further supports the potential to interact of the two proteins as demonstrated by functional proteomics experiments. In the control, represented by Drosophila S2 cells, as expected, no signal of TnBVANK1 was detected while Alix signal is present in the cytoplasm as in polyclonal S2 cells. This finding provides a further element supporting the hypothesis of the interaction between these two proteins.
In order to assess whether the apoptosis induced by TnBVank1 is mediated by the interaction with Alix, we silenced the Alix gene transcript by RNAi both in the S2 cell line stably expressing TnBVank1 and in H. virescens circulating haemocytes transiently expressing TnBVank1.
The success of RNAi was verified by quantitative Real time PCR (qPCR) and western blot performed on polyclonal S2 cells stably expressing TnBVank1, harvested at different time points (0, 6, and 10h) (Fig. 6a and b) and on H. virescens haemocytes transiently expressing TnBVank1 (Fig. 7a and b) both treated with Alix dsRNA. The cells in which Alix was silenced (S2[ank1] Alix Si and Hemo[ank1] Alix Si) showed that the available transcript of Alix gene was significantly reduced compared to its expression in S2 cell line, in S2 cell line stably expressing TnBVank1 (S2[ank1]) and in S2 cell line stably expressing TnBVank1 treated with the dsApisOBP3 (S2[ank1] OBP3 Si) as well as in circulating haemocytes transiently transfected with the empty pIZT/V5-His vector (Hemo), in haemocytes transiently expressing TnBVank1 (Hemo[ank1]) and in haemocytes transiently expressing TnBVank1 treated with the dsApisOBP3 (Hemo[ank1] OBP3 Si) (Fig. 7a).
Same samples were tested by qPCR for their expression of TnBVank1 transcript demonstrating its expression in all described conditions except in the case of S2 cell line and in circulating haemocytes transiently transfected with the empty pIZT/V5-His vector (Figs 6a and and7b7b).
Caspase-3 assay performed on same samples analysed by qPCR as listed above indicated that in S2 cell line stably expressing TnBVank1 and silenced for Alix transcript, caspase-3 activity was highly reduced compared to the S2 cell line stably expressing TnBVank1 (Fig. 6c). Similar results were obtained in vivo, since caspase-3 activity measured in H. virescens haemocytes cotransfected with Alix dsRNA and TnBVank1 was significantly reduced compared to haemocytes transfected with TnBVank1 alone (Fig. 7c). Taken together, the in vivo and in vitro experiments suggest that Alix/TnBVANK1 interaction strongly contributes in inducing apoptosis both in S2 cell line and in circulating haemocytes.
The bracovirus associated with the wasp Toxoneuron nigriceps (TnBV) is one of parasitic regulatory factors of maternal origin involved in immunosuppression and alteration of endocrine balance of the host Heliothis virescens 1,4,12. A large number of studies have been addressed to the elucidation of parasitised host developmental syndromes6,7,18,27. Briefly, in this experimental model, parasitised last instar larvae fail to pupate, allowing a suitable nutritional substrate for parasitoid larvae over an extended time period28. Here, we report experimental evidence demonstrating the role of TnBVank1, a viral gene member of the TnBV ankyrin (ank) gene family8, in inducing apoptosis processes in a stably TnBVank1 transfected polyclonal insect cell line (Drosophila Schneider’s S2 cells) and in circulating haemocytes of H. virescens larvae transiently transfected with TnBVank1 gene, thus reproducing the apoptotic effects observed in haemocytes of parasitised H. virescens larvae. TnBVank1, together with TnBVank2 and TnBVank3, belongs to the TnBVank family8 and displays significant sequence similarity with members of the IκB family, that are proteins generally involved in the control of NF-κB signaling pathways both in insects and in vertebrates29.
IκB members are characterised by repeated sequences of ankyrins that bind to transcription factors NF-κB/Rel, masking their nuclear localization signal. Upstream sequences of ankyrin serine residues can be phosphorylated by the IκB kinase (IκK) in response to a number of extracellular signals29. Once phosphorylated, IκB can be released and degraded via the proteasome, allowing translocation of the transcription factor NF-κB/Rel in the nucleus and subsequent activation of target genes29. In Drosophila, the IκB-like protein regulates various cellular responses triggered by nuclear transfer protein NF-κB/Rel, such as the embryonic development and the release of defense-related antimicrobial peptides29.
The gene family of viral ankyrins (also called vankyrin), is widespread among both Bracovirus and Ichnovirus, and each genome has several copies of these genes8,15,30–32. All known polydnavirus (PDV) ankyrin genes lack the regulatory sites, serine residues at the N-terminus and PEST domain at the C-terminal end13,15,33–36. These peculiar features strongly suggest that viral ankyrins may act as inhibitors by irreversibly binding and retaining NF-κB in the cytoplasm thus affecting a variety of cellular responses in parasitised hosts, including alteration of innate immune defenses8,14,15. This mechanism, proposed to explain the alteration of the host immune response induced by the parasitoid, is very similar to the strategy used by ASFV (African swine fever virus) to inhibit the activation of NF-κB transcription factors in mammals. The ASFV IκB-like protein irreversibly binds NF-κB, preventing its entry into the nucleus37, similarly to the effect observed for TnBVank18.
The genome of Campoletis sonorensis Ichnovirus (CsIV) contains seven copies of ankyrin genes named vankyrin33. Among these vankyrins, P-vank-1 is an interesting viral factor, as it, when is stably expressed in eukaryotic Sf9 insect cells, seems to have a role in protecting cells from apoptosis34. The role of vankyrin P-vank-1, which belongs to the same family as TnBVank1, led us to hypothesise a possible involvement of TnBVank1 in the regulation of programmed cell death. Here we have shown that, unlike vankyrin P-vank1, the viral gene TnBVank1 induces apoptosis. The different biological roles of these proteins are surprising but not unusual; it is reported that the vankyrin I2-vank-3 of Campoletis sonorensis Ichnovirus (CsIV), although highly similar to P-vank-1 (with 83% amino acid sequence identity), plays a different role36. This could be also explained by their differential expression and localisation in the host tissues.
In the polyclonal S2 cell line stably expressing TnBVank1, a significant increase of apoptotic cell number, compared to untreated S2 cells (negative control), was observed by flow cytometry when doubly stained with Annexin V–FITC and PI. Moreover, compared to negative control, in the polyclonal S2 cell line stably expressing TnBVank1 a significantly higher caspase-3 activity and apoptotic nuclei were detected. A previous work16 revealed typical morphological modifications of apoptosis in parasitised H. virescens granulocytes (a subpopulation of haemocytes). Here we confirmed that haemocytes undergo apoptosis after natural parasitism, by flow cytometry analysis, caspase-3 assay and phalloidin staining. This result led us to deepen the role of TnBVank1 in apoptotic pathway in haemocytes of non parasitised larvae by in vivo transient transfection that is an useful method to determine the biological function of a foreign gene. We observed an apoptotic profile by flow cytometry, a significantly increased caspase-3 activity and nuclear morphological changes in H. virescens haemocytes after TnBVank1 in vivo transfection. These results demonstrate that TnBVank1 is one of the Polydnavirus genes responsible for apoptosis observed in H. virescens haemocytes after parasitisation16. A qPCR performed on both haemocytes derived from parasitised host and TnBVank1 in vivo transient transfected haemocytes evidenced that the expression level of TnBVank1 after transient expression was comparable to its expression after natural parasitism, thus demonstrating that the observed apoptotic effect is not due to the result of an unnatural overexpression of the protein.
TnBVank1 would be the first viral ankyrin that, besides preventing the nuclear translocation of NF-κB8 thus disrupting cell pathways under its control15,29,38, is also able to directly induce apoptosis. TnBVank1 is not the first TnBV gene that induces apoptosis. Lapointe and colleagues18 showed that the gene TnBV1 induces apoptosis when expressed in cell lines derived from Spodoptera frugiperda (SF21) and Tricoplusia ni (High Five). However, when the authors tried the in vivo expression of TnBV1-recombinant baculovirus, no apoptotic effect was detected, leaving the role of TnBV1 expression in parasitised H. virescens larvae elusive. TnBV1 protein may interact with other TnBV proteins during parasitism to exert its function in the apoptotic pathway18. In contrast, our results indicate that a single viral gene, TnBVank1, is able to induce apoptosis, in vitro and in vivo, without the requirement of other virus-encoded genes as interaction partners. The induction of apoptosis is one of the most common strategies adopted by various PDVs to suppress the immune response39. However, the parasitism do not affect all the cells of the immune system16 probably as a strategy to allow the host to respond to a possible further pathogen invasions ensuring the development of wasp embryo39,40.
The molecular mechanisms of apoptosis induction exerted by TnBVank1 were more closely investigated by functional proteomic approaches. Multiprotein complexes comprising TnBVANK1 were isolated from polyclonal S2 stably expressing TnBVank1 by immunoprecipitation and the individual protein components were identified by mass spectrometry. Among the various TnBVANK1 protein partners, we focused our attention on Alix (ALG-2 interacting protein X) as this multifunctional protein is involved in several protein interaction networks, including programmed cell death, endocytic membrane trafficking and cytoskeletal remodeling41,42. To confirm the possible interaction between TnBVANK1 and Alix proteins, polyclonal S2 cell line was examined by coimmunolocalisation experiments, using anti-TnBVANK1 and anti-Alix antibodies. Observations by immunofluorescence microscopy revealed that TnBVANK1 are both present in the cytoplasm and seems to colocalises with Alix protein.
The role of TnBVANK1-Alix interaction in apoptosis was further investigated by silencing experiments both in vitro and in vivo. Silencing of Alix expression in TnBVank1 stably transfected S2 cells resulted in a significant reduction of apoptosis assessed by caspase-3 activity. Similarly, silencing Alix in circulating haemocytes of H. virescens larvae transiently transfected with TnBVank1 drastically reduced caspase-3 activity. In both cases, the TnBVank1 transcript expression level evaluated by qPCR remained constant.
These in vitro and in vivo results suggest that TnBVank1 induces apoptosis through interaction with Alix.
The interaction between TnBVANK1 and Alix was also reported by a recent study showing their colocalisation in prothoracic glands of a Drosophila melanogaster mutated strain expressing TnBVank1. It is reported that TnBVANK1 impairs the ecdysone biosynthesis and causes the developmental arrest by altering the vesicular traffic of ecdysteroid precursors in the prothoracic gland steroidogenic cells of Drosophila melanogaster. The authors showed that the TnBVank1 affects the sterol delivery interacting with Alix that is also involved in controlling the cholesterol export from endosomes16,43.
Although the molecular mechanisms are unclear, Alix cooperating with ALG-2 is involved in the induction of apoptosis through both intrinsic and extrinsic apoptotic pathways41. A possible way of induction of apoptosis by the protein Alix is via TNF-R1 (TNF receptor-1), which contains a DD (Death Domain) capable of activating caspase 8, an initiator of apoptotic process. The interaction TNF-R1/Caspase 8 initiates the apoptotic cascade. This mechanism requires that Alix forms a complex with TNF-R144.
Our data show that TnBVANK1 strongly induces apoptosis interacting with Alix, via the mitochondrial apoptotic pathway. Moreover, the previously demonstrated role of TnBVank1 in the control of the NF-κB signaling8 preventing its nuclear translocation, could also further increase the sensitivity of the cell to the cell death signals38.
This seems like an effective strategy of the virus to disable the host immune system and facilitate successful development of its associated wasp.
Drosophila Schneider’s S2 cells (Life Technologies, Carlsbad, CA, USA) were cultured in Schneider’s Drosophila medium (WVR, Radnor, PA, USA) supplemented with 10% heat inactivated fetal bovine serum (Life Technologies, Carlsbad, CA, USA) and were maintained in 75cm2 tissue culture flasks (Corning, NY, USA) at 27°C.
The TnBVank1 cDNA, obtained as previously described by Falabella and colleagues8, was amplified with specific primers containing XbaI and SacII enzyme restriction sites (underlined):
The obtained amplimer was cloned into TOPO vector (Life Technologies, Carlsbad, CA, USA) and sequenced. The insert was then released by XbaI and SacII restriction enzyme digestion (underlined) and cloned into the expression vector pIZT/V5-His (Life Technologies, Carlsbad, CA, USA). This vector uses the OpIE2 promoter from the Orgyia pseudotsugata baculovirus for constitutive expression of the gene of interest and encodes a zeocin-green fluorescent protein (GFP) gene fusion under control of the OpIE1 promoter. The 3′ ends of the Ank1 fragments were cloned in frame with a “V5 epitope” and with a tail of six histidines. The obtained construct was sequenced, verified and transiently expressed in S2 cells by cationic lipid-mediated transfection. 24h prior transfection S2 cells were seeded at 70 to 80% confluency (0.5×106) in 12-well culture plates (Corning, NY, USA) and subsequently transfected using 8μl of Cellfectin (Life Technologies, Carlsbad, CA, USA) and 1μg of DNA per 1ml of medium and incubated at 27°C for 6h. After incubation, the medium was removed and cells were re-incubated with 1ml of Schneider’s Drosophila complete medium containing 10% FBS at 27°C for 48h. Stably transformed cells were selected by adding 400μg/ml of zeocin (Life Technologies, Carlsbad, CA, USA). Populations of antibiotic-resistant cells were amplified to obtain polyclonal S2 cell lines stably expressing TnBVank1.
Heliothis virescens larvae were maintained on an artificial diet as previously described28. Toxoneuron nigriceps was reared in the laboratory, on larval stages of its host, H. virescens, according to Vinson et al.45. Rearing temperature was 29±1°C for the host, parasitised host larvae and cocoons. T. nigriceps adults were kept at 25±1°C and fed with water and honey. A 16h light photoperiod was adopted and the relative humidity was 70±5%. H. virescens last instar larvae were staged according to Webb & Dahlman46 and synchronised as reported by Pennacchio et al.47.
For experimental procedures, described below, H. virescens larvae were parasitised at the day one of the last larval instar.
For in vivo transfection of pIZT/V5-His-ANK1, Lipofectin reagent (Life Technologies, Carlsbad, CA, USA) was used. pIZT/V5-His-ANK1 (0.5µg) was mixed with 4µl of transfection reagent and incubated for 20min at room temperature for the formation of DNA-lipid complex before injection into the haemocoel of H. virescens fifth instar larvae (day one) using a Hamilton syringe (Sigma-Aldrich St Louis, MO, USA). Before injection, larvae were anaesthetised by immersion in sterile ice water. Haemocytes were collected as described by Ferrarese and colleagues16 56h post-transfection to perform all the biological assays. This time was selected as the most successful for TnBVANK1 expression, tested after a time course at 24h, 48h and 56h post injection (data not shown).
Success of the haemocytes transfection was analysed by reverse transcriptase-polymerase chain reaction, and the protein expression was detected by western blot analysis using anti-V5 antibody (Life Technologies, Carlsbad, CA, USA) and by immunofluorescence experiments. To assess whether TnBVANK1 induces apoptosis in vivo, the caspase-3 assay was performed on haemocytes as describe below.
Total RNA was extracted from H. virescens haemocytes, previously transfected with the vector pIZT/V5-His (as negative control) or recombinant vector expressing TnBVank1, by using TRI Reagent (Sigma-Aldrich St Louis, MO, USA). Concentration and purity of the total RNA samples were measured using the NanoDrop ND-1000 Spectrophotometer (ND-1000, Thermo Scientific, Waltham, MA, USA). To remove DNA contamination, the samples were treated with 1U of DNase I (Life Technologies, Carlsbad, CA, USA) per microgram of RNA, for 15min at room temperature. The reaction was stopped by adding 1µl of 25mmol/L EDTA and incubating at 65°C for 10min. The complementary DNA was synthesised by SuperScript III Reverse Transcriptase (Life Technologies, Carlsbad, CA, USA) according to the manufacturer’s protocol. The cDNA was used as a template for PCR amplification using the following primers:
To confirm no vector DNA contamination was present after RNA extraction, an aliquot of the isolated RNA was directly used as a template for a test PCR (i.e. without reverse transcription), which showed no PCR product.
Heliothis virescens haemocytes were isolated 56h after injection of the vector pIZT/V5-His with or without TnBVank1 as described above. Haemocytes were fixed for 10min at room temperature in 4% formaldehyde in 1X phosphate buffer saline (PBS) pH 7.5 and then permeabilised for 30min with PBS containing 0.1% Triton X-100. The cells were blocked with 1% bovine serum albumin (BSA) for 30min and, subsequently, only the haemocytes transfected with TnBVank1 were incubated with the primary antibody, anti-TnBVANK148 diluted 1:200 in BSA 1%, in a humidified chamber at 4°C overnight. After three washes in PBS 1X, the cells were incubated with the secondary antibody TRITC-conjugated (Sigma-Aldrich St Louis, MO, USA), 1:1000 in 1% BSA for 1h at room temperature in dark condition. After washing in PBS 1X three times, the slides were mounted with FluoroshieldTM with DAPI, histology mounting medium (Sigma-Aldrich St Louis, MO, USA).
To measure the total cell fluorescence, the slides were observed microscopically with Nikon Eclipse 80i equipped with a Nikon Plan Fluor 100x/0.5–1.3 Oil Iris objective and the images were recorded with Nikon Digital Sight DS-U1 camera. The intensity of the cell fluorescence was measured using the ImageJ software (http://rsbweb.nih.gov/i j/).
The GFP protein expression was analysed directly by counting fluorescent haemocytes compared to total haemocytes. A hundred haemocytes from a randomly selected field of view at 40x magnification were analysed for each measurement. Each treatment was independently replicated three times.
To evaluate TnBVank1 expression in haemocytes both in vivo transfected and after parasitism, quantitative RT-PCR (qPCR) experiments were carried out. RNA extraction and cDNA synthesis were conducted as previously described. PCR amplification was performed using GoTaq qPCR Master Mix (Promega, Fitchburg, WI, USA) on an ABI PRISM® 7500 Fast Real-Time PCR System Thermal Cycler (Applied Biosystems, USA) and in a 20μl final volume. Sequence specific primers, each in a concentration of 0.3μM, were:
All qPCR analyses were performed for a set of 3 biological replicates. Quantification analysis of amplification was done using the comparative CT (ΔCT) method. The efficiencies of the three amplicons were approximately equal (Ank1=0.892159; EF1α=0.914590; Rp13=0.896174).
Apoptosis was evaluated by flow cytometric analysis of Annexin V and Propidium Iodide (PI)-stained cells using fluorescein isothiocyanate (FITC) Annexin V Apoptosis Detection kit I (Becton Dickinson, Biosciences, San Jose, CA, USA). The assay was carried out on S2 cells (negative control), on S2 cells stably expressing TnBVank1 and on H. virescens haemocytes after TnBVank1 in vivo transfection and after parasitism. S2 cells and polyclonal S2 cells were used in passages 4–10. For the positive control, apoptosis was induced by treating the cells with 1μM camptothecin, a cytotoxic alkaloid inhibitor of Topoisomerase I, known to induce apoptosis50,51 (Sigma-Aldrich St Louis, MO, USA), or injecting into haemocoel of larvae directly 1μM camptothecin for 20 h34. As negative control for the in vivo assay, H. virescens haemocytes after injection of the vector pIZT/V5-His without TnBVank1 were used. Cell samples were washed twice and incubated with Annexin V-FITC/PI for 15min as manufacturer’s protocol. Stained samples were acquired using Navios flow cytometer and analysed by Kaluza Analysis 1.3 software (Beckman Coulter). Single positive for Annexin V and double positive for Annexin V and PI cells were interpreted as signs of early and late phases of apoptosis respectively.
The caspase-3 fluorimetric assay was carried out on S2 cells (negative control), on S2 cells stably expressing TnBVank1 and on H. virescens haemocytes after TnBVank1 in vivo transfection. S2 cells and polycnonal S2 cells were used in passages 4–10. For the positive control, apoptosis was induced by treating the cells with 1μM camptothecin50,51 (Sigma-Aldrich St Louis, MO, USA) or injecting into haemocoel of larvae directly 1μM camptothecin for 20 h34. As negative control for the in vivo assay, H. virescens haemocytes after injection of the vector pIZT/V5-His without TnBVank1 were used. As a substrate of caspase-3 the Ac-DEVD-AMC caspase-3 fluorogenic substrate (BD Biosciences Pharmingen, San Diego, CA, USA) was used. The assay protocol requires primarily lysis of cells. Cells were centrifuged at 1000g for 10min and then resuspended with Cell Lysis Buffer (10mM Tris-HCl, 10mM NaH2PO4/NaHPO4, pH 7.5, 130mM NaCl, 1% Triton-X-100 and 10mM NaPPi) in suitable amounts such as to obtain 3×105 cells/100µl. The peptide Ac-DEVD-AMC (20µM) and 1ml of Protease Assay Buffer (20mM HEPES pH 7.5, 10% glycerol, 2mM DTT) were added to each reaction. The reaction was then incubated at 37°C for 1h. The amount of AMC released was measured by the spectrofluorometer (Cary Eclipse Fluorescence Spectrophotometer, Agilent Technologies, Walnut Creek, CA, USA) exciting at 380nm and detecting in a range between 430–460nm.
Haemocytes were isolated 56h after injection of the vector pIZT/V5-His with or without TnBVank1 as described above and from parasitised larvae. Cells were fixed, permeabilised and blocked as previously described. Subsequently, haemocytes were incubated with TRITC-conjugated phalloidin (Sigma-Aldrich St Louis, MO, USA), diluted 50µg/ml in 1% BSA for 2h at room temperature in dark condition. After washing in PBS 1X three times, the slides were mounted with glycerol (Sigma-Aldrich St Louis, MO, USA).
The slides were observed microscopically with Nikon Eclipse 80i equipped with a Nikon Plan Fluor 100x/0.5–1.3 Oil Iris objective and the images were recorded with Nikon Digital Sight DS-U1 camera.
S2 cells (negative control), S2 cells stably expressing TnBVank1 and H. virescens haemocytes isolated 56h after injection of the vector pIZT/V5-His with or without TnBVank1 were stained with Hoechst 33258 (Sigma-Aldrich, St Louis, MO, USA) to detect nuclear morphology. Cells were fixed with 4% paraformaldehyde for 10min, washed with PBS, and stained with 10μg/mL Hoechst 33258 at room temperature for 10min in the dark. The cells were washed with PBS 1X three times and the slides were mounted with glycerol for morphologic observation by fluorescence microscopy (NIKON Eclipse 80i) equipped with a Nikon Plan Fluor 40x and 100x/0.5–1.3 Oil Iris objective and the images were recorded with Nikon Digital Sight DS-U1 camera.
The TnBVANK1 protein production over time in the polyclonal S2 cell line that stably expresses TnBVank1 was evaluated by western blot analysis. For proteins extraction, S2 cell line expressing TnBVank1, harvested at different successive splitting (2, 4, 10, 20, 26, 39 passages), was resuspended, transferred to sterile 1.5ml tubes (Eppendorf, Hamburg, Germany) and centrifuged at 1000g for 5min at room temperature. The cell pellet was then resuspended in lysis buffer (50mM Tris-HCl, pH 7.8, 150mM NaCl, 1% Nonidet P-40) to which 1% protease inhibitor cocktail (Sigma-Aldrich, St Louis, MO, USA) was added. After incubation at 37°C for 10min, the lysed cells were centrifuged at 16000g for 10min at room temperature and the supernatant containing soluble proteins was transferred to a new 1.5ml tube (Eppendorf, Hamburg, Germany). The protein concentration was determined by the Bradford method52, using BSA as standard and 100µg of protein lysate were used for each SDS PAGE gel lane. The viral protein TnBVANK1 was detected using the murine anti-V5 antibody (Life Technologies, Carlsbad, CA, USA) as primary antibody (diluted 1:5000 in 5% milk), and anti-mouse conjugated to horseradish peroxidase (Life Technologies, Carlsbad, CA, USA) (diluted 1:5000 in PBS), as secondary antibody. α-actin was chosen as endogenous control, using the anti- α -actin antibody (Sigma-Aldrich, St Louis, MO, USA) (diluted 1:5000 in 5% milk) and the secondary antibody anti-rabbit conjugated to horseradish peroxidase (Life Technologies, Carlsbad, CA, USA, USA) (diluted 1:15000 in PBS). Finally, the Western blot Chemiluminescent HRP Substrate (LiteAblot, Euroclone, Milan, Italy) to detect target proteins was used.
Functional proteomic experiments were performed on protein lysates of S2 cells stably expressing V-5-tagged TnBVANK1. A total protein extract was subjected to pre-cleaning treatment on underivatised agarose beads and the unbound fraction was immunoprecipitated by incubation onto anti-V5 agarose-conjugated antibody overnight under gentle stirring. An aliquot of the non specific bound material was eluted and used as control.
Beads were collected by centrifugation (3,000rpm for 5min) and extensively washed with lysis buffer supplemented with 150mmol/l NaCl. Elution was performed by competition with V-5 peptide in elution buffer. The presence of TnBVANK1 in the eluted fraction was assessed by Western-blot using anti-V5 antibody.
The eluted proteins were precipitated in methanol/chloroform and then loaded onto a 10% SDS-PAGE. Protein bands were excised, reduced with 10mM DTT and carboxyamidomethylated with 55mM iodoacetamide. Tryptic digestion was carried out with 12.5 ng/μl at 4°C for 2h in 10mM NH4HCO3 buffer, pH 7.8 and then for 16h at 37°C. Peptides were then extracted with 10mM NH4HCO3 and 1% formic acid in 50% acetonitrile at room temperature. Gel slices from the negative control S2 cells were also analysed.
Peptide mixtures were analysed by LC-MS/MS using a CHIP MS 6520 QTOF equipped with a capillary 1200 HPLC system and a chip cube (Agilent Technologies, Palo Alto, CA, Italy). Peptide analysis was performed using data-dependent acquisition. Mass spectral data were used to search for the UniProt/SwissProt protein database using an in house version of MASCOT (Matrix Science, Boston, USA).
For the immunofluorescent labeling, S2 cells were fixed for 10min at room temperature in 4% formaldehyde in 1X phosphate buffer saline (PBS) pH 7.5. After permeabilisation for 30min in 0.1% Triton X-100 solution and washing in PBS 1X, the cells were incubated in 1% BSA for 30min and, subsequently, with the mixture of two primary antibodies, anti-TnBVANK145 and anti-Alix53, 1:200 in BSA 1%, in a humidified chamber at 4°C overnight. After three washes in PBS 1X, the cells were incubated with the mixture of two secondary antibodies, which were raised in different species (with two different fluorochromes, i.e. TRITC-conjugated against mouse and FITC-conjugated against rabbit) (Sigma-Aldrich St Louis, MO, USA), 1:1000 in 1% BSA for 1h at room temperature in dark condition. After washing in PBS 1X, the slides were mounted with FluoroshieldTM with DAPI, histology mounting medium (Sigma-Aldrich St Louis, MO, USA).
To measure the total cell fluorescence, the slides were observed microscopically with Nikon Eclipse 80i equipped with a Nikon Plan Fluor 100x/0.5–1.3 Oil Iris objective and the images were recorded with Nikon Digital Sight DS-U1 camera. The intensity of the cell fluorescence was measured using the ImageJ software (http://rsbweb.nih.gov/ij/).
RNA interference (RNAi) experiments both in vivo and in vitro were performed using double-stranded (ds)RNA of the Alix gene (Accession Number Q9VB05) (see Supplementary Fig. S6). As negative control dsRNA of the ApisOBP3 gene (Accession Number 001160057), which is not expressed in S2 cells and in H. virescens haemocytes, was used.
Part of the coding sequences for Alix and ApisOBP3 proteins were amplified by PCR with specific primers containing the T7 binding site at their 5′ ends (underlined):
PCR products were gel purified (Quantum Prep™ Freeze’N Squeeze DNA Gel Extraction Spin Columns, Biorad, Hercules, CA, USA) and Alix dsRNA and ApisOBP3 dsRNA were prepared using the RNAi MEGAscript® kit (Ambion, Austin, TX, USA) following the manufacturer’s instructions. The quality of dsRNA obtained was observed on an agarose gel and the quantity was measured using a NanoDrop (ND-1000, Thermo Scientific, Waltham, MA, USA).
For in vitro RNAi, dsRNAs (16µg for each well) were directly added to the culture medium. A number of cells equal to 4×106 cells in 2ml were plated 24h before silencing in a 6-well plate. Wells contained S2 cells and S2 cells stably expressing TnBVank1, respectively. The same procedure was performed in control cells, but no dsRNA was added. After 2hours the medium containing the dsRNA was replaced with fresh medium without dsRNA and at different time points (T0, T6 and T12 h) 2×106 cells were harvested for RNA extraction and 3.5×105 cells for the post silencing caspase-3 assay.
For in vivo RNAi, 0.5µg of both dsRNAs were mixed with 0.5µg of pIZT/V5-His-ANK1 and 3µl of transfection reagent (Lipofectin) and incubated for 20min at room temperature before injection into the haemocoel of fifth instar H. virescens larvae (day one). Treated larvae were processed for qPCR analysis 56h after the dsRNA administration.
In order to verify the Alix silencing, its transcript expression level was detected by qPCR and, on the same samples, the TnBVank1 transcript level expression was determined. RNA was extracted at different time points from cells (T0, T6 and T10 h after adding dsRNA) using TRI Reagent (Sigma-Aldrich St Louis, MO, USA) and RNA quality was evaluated on an agarose gel. For qPCR, 1µg of total RNA was treated with DNase (Deoxyribonuclease I Amplification Grade, Life Technologies, Carlsbad, CA, USA) and then it was reverse transcripted using Super Script ™ ІІІ First-Strand Synthesis System for RT-PCR (Life Technologies, Carlsbad, CA, USA, USA), with oligo-dT primers, following the manufacturer’s protocol. qPCR was performed using GoTaq® qPCR Master Mix (Promega, Fitchburg, WI, USA) on an ABI Prism® 7500 Fast Real-Time PCR System Thermal Cycler (Applied Biosystem, USA). Sequence specific primers, each in a concentration of 0.3µM were:
As a reference gene, the gene coding for the Drosophila ribosomal protein L32 (RPL32, Accession Number P04359)54 was used. Primers were:
All qPCR analyses were carried out in triplicate. Quantification analysis of amplification was done using the comparative CT (ΔCT) method. The efficiencies of the two amplicons were approximately equal (Alix=0.744509; Ank1=0.902369; Rpl32r=0.841033).
In order to verify the Alix silencing, its transcript expression level was detected by qPCR also in haemocytes and on the same samples the TnBVank1 transcript level expression was determined. RNA was extracted at 56h after injection using TRI Reagent (Sigma-Aldrich St Louis, MO, USA), the quality and quantity were determined using NanoDrop spectrophotometer (ND-1000, Thermo Scientific, Waltham, MA, USA) and potential DNA contamination was eliminated with DNase (Deoxyribonuclease I Amplification Grade, Life Technologies, Carlsbad, CA, USA). cDNA was obtained as described above. qPCR was performed using GoTaq® qPCR Master Mix (Promega, Fitchburg, WI, USA) in a ABI Prism® 7500 Fast Real-Time PCR System Thermal Cycler (Applied Biosystem, USA). Sequence specific primers, each in a concentration of 0.3µM, were:
TnBVank1 and reference gene primer sequences are reported above. All qPCR analyses were carried out in triplicate. Quantification analysis of amplification was done using the comparative CT (ΔCT) method. The efficiencies of the three amplicons were approximately equal (Alix=0.931310; Ank1=0.898875; EF1α=0.874059; Rp13=0.894415).
All data were presented as mean±SD of three independent biological replicates and were compared with analysis of variance (ANOVA) and Tukey’s test using GraphPad Prism 6 software, La Jolla, California, USA.
This work was supported by University of Basilicata (RILfunds). We thank Tab Consulting s.r.l. for financing a Ph.D grant. We also thank Dr. Toshiro Aigaki for kindly giving us anti-Alix antibody, Dr. Mario Lippolis, Dr. Luciana Possidente, Giovanni Petrullo and Francesco Petrullo for their technical support in immunofluorescence experiments and Dr. Giuseppe Cugno and Dr. Pellegrino Musto. This manuscript is in memory of Dr. Franco Graziani of the Institute of Genetics and Biophysics - CNR (Naples, Italy).
P.F. designed the experiments, wrote and critically revised the paper; M.C., H.V., P.P., S.B.V., D.N. and S.A.B. contributed to the data interpretation and critically revised the paper; R.S. carried out experimental work and analysed data. A.A. and C.G. performed functional proteomic experiments. C.S. collected RNA samples. M.N. and R.S. performed coimmunolocalisation experiments. G.G. and R.S. performed qPCR. R.S., I.L. and V.R. performed flow citometry analysis. All authors read and approved the final manuscript.
The authors declare that they have no competing interests.
Electronic supplementary material
Supplementary information accompanies this paper at 10.1038/s41598-017-11939-x.
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.