AcMNPV P74 is cleaved at R325 and R334 by proteinases of both OB and BBMV to expose a potential fusion peptide for oral infection

ABSTRACT

Baculoviruses enter insect midgut epithelial cells via a set of occlusion-derived virion (ODV) envelope proteins called per os infectivity factors (PIFs). P74 of Autographa californica multicapsid nucleopolyhedrovirus (AcMNPV), which was the first identified PIF, is cleaved by an endogenous proteinase embedded within the occlusion body during per os infection, but the target site(s) and function of the cleavage have not yet been ascertained. Here, based on bioinformatics analyses, we report that cleavage was predicted at an arginine and lysine-rich region in the middle of P74. A series of recombinant viruses with site-directed mutants in this region of P74 were generated. R325 or R334 was identified as primary cleavage site. In addition, we showed that P74 is also cleaved by brush border membrane vesicles (BBMV) of the host insect at R325 or R334, instead of R195, R196, and R199, as previously reported. Simultaneous mutations in R195, R196, and R199 lead to instability of P74 during ODV release. Bioassays showed that mutations at both R325 and R334 significantly affected oral infectivity. Taken together, our data show that both R325 and R334 of AcMNPV P74 are the primary cleavage site for both occlusion body endogenous proteinase and BBMV proteinase during ODV release and are critical for oral infection.

IMPORTANCE

Cleavage of viral envelope proteins is usually an important trigger for viral entry into host cells. Baculoviruses are insect-specific viruses that infect host insects via the oral route. P74, a per os infectivity factor of baculoviruses, is cleaved during viral entry. However, the function and precise cleavage sites of P74 remain unknown. In this study, we found that R325 or R334 between the N- and C-conserved domains of P74 was the primary cleavage site by proteinase either from the occlusion body or host midgut. The biological significance of cleavage seems to be the release of the potential fusion peptide at the N-terminus of the cleaved C-terminal P74. Our results shed light on the cleavage model of P74 and imply its role in membrane fusion in baculovirus per os infection.

INTRODUCTION

Baculoviruses are a large group of insect-specific viruses that naturally infect insect hosts through the oral route. Infection is initiated when insect larvae ingest food contaminated with baculovirus occlusion bodies (OBs). Virions embedded within OBs [occlusion-derived viruses (ODVs)] are released from the crystalline structure of OBs under alkaline conditions in the midgut (pH 9–11). ODVs then access highly differentiated columnar epithelial cells and initiate infection by binding and fusing with the microvilli of the columnar cells (1). This process is called per os infection. To date, 10 ODV envelope proteins have been identified as essential for oral infection and are defined as per os infectivity factors (PIFs), including PIF0–9 (2–11). The absence of any PIFs results in the complete loss of oral infectivity. Nine of the 10 PIFs, but not PIF5, form a complete complex of approximately 500 kDa (11). PIF0 (also called P74), PIF1, and PIF2 function in ODV binding to host midgut cells (5, 12), while PIF3 and PIF5 may not participate in ODV binding and fusion (5, 7). However, the specific functions of each PIF during infection remain unclear.

PIF0, the first identified per os infectivity factor, is named P74 because of its 74 kDa molecular mass (2). It is conserved in baculoviruses, and its homologs exist in several distantly related large DNA viruses (13–15). R18-labeled ODV fluorescence dequenching and enhanced green fluorescent protein (EGFP)-labeled ODV visualization assays demonstrated that P74 was associated with ODV binding (12, 16). P74 undergoes essential trypsin cleavage during the ODV release process in the midgut. In Autographa californica multicapsid nucleopolyhedrovirus (AcMNPV), P74 was cleaved by trypsin in insect midgut epithelial cells, and multiple mutations in the region R195/R196/R199 eliminated the cleavage by brush border membrane vesicles (BBMV) and impaired per os infectivity (17, 18). Peng et al. showed that P74 was first cleaved into two subunits by an OB-associated endogenous protease during ODV release and suggested a two-step cleavage model in which P74 is cleaved by OB-endogenous enzymes and host midgut trypsin successively (19). However, the target cleavage site of the endogenous protease was not identified, and the biological function of this cleavage remains unknown.

To address these questions, we first confirmed the endogenous proteinase cleavage of AcMNPV P74 and predicted the potential cleavage location using bioinformatics and AlphaFold2 analyses. Recombinant viruses with mutations in potential cleavage sites were generated, and their impact on the cleavage of P74 and oral infectivity was tested. Our data revealed the primary cleavage sites of the OB-endogenous proteinase and showed that they are important for oral infectivity. In addition, we detected the cleavage of P74 by BBMV. Surprisingly, we found that BBMV cleaved P74 at the same sites as the OB-endogenous proteinase but not as previously reported at R195/R196/R199 (17). We showed that mutations in R195/R196/R199 resulted in the instability of P74 during ODV release and consequently affected oral infectivity. Finally, a revised two-step cleavage model for P74 was developed.

RESULTS

AcMNPV P74 is specifically cleaved by OB-endogenous protease during ODV release

To study the function of P74, AcMNPV-Syn (20) was used as the control virus (control), and recombinant viruses with a p74 deletion (Δp74) and p74 repaired (Rep) were constructed. Western blot analyses were used to detect the cleavage status of P74 in control OBs during the ODV release process, and the OBs derived from Δp74 and Rep-infected larvae were taken for comparison. The OB-endogenous protease was first inactivated by heating OB suspensions at 80°C for 1 h. The ODVs were released by alkaline treatment and subjected to sodium dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and western blot analysis using a monoclonal antibody (mAb) 21A7 against P74. As shown in Fig. 1A, the full-length P74 band (~74 kDa) was observed in the control and Rep samples, but not in that of Δp74, suggesting that P74 remained uncleaved in the heated ODVs. Similar experiments were performed without heating the OBs, and the P74 antibody detected a specific band of approximately 36 kDa in the control and Rep samples, but no bands were detected for Δp74 (Fig. 1B), suggesting that P74 was cleaved by OB-endogenous proteinase during the ODV release process. Since the P74-specific antibody mAb 21A7 recognizes an epitope at the N-terminus of P74 between residues S76 and M95 (Fig. S1), the results indicated that during the ODV release process, an N-terminal 36 kDa product of P74 was generated.

Fig 1.

AcMNPV P74 is cleaved by OB-endogenous proteinase during the ODV release process. OBs were purified from dead Spodoptera frugiperda larvae infected with the indicated viruses. To inactivate endogenous proteinase, OBs were heated at 80°C for 1 h. ODVs from heated (A) or unheated (B) OBs were released by DAS treatment for 5 min and subjected to SDS-PAGE, followed by western blotting with the primary antibodies P74 mAb 21A7 or anti-VP39. The full-length and cleaved P74 is indicated by arrows. The major nucleocapsid protein VP39 was used as a loading control. M, standard molecular marker.

Trypsin inhibitor blocks the cleavage of P74 by the OB-endogenous protease

To investigate what kind of endogenous protease mediates P74 cleavage during ODV release, we added a soybean trypsin inhibitor (SBTI) at different time points during the ODV release process and analyzed the cleavage of P74. Heated or unheated OBs without SBTI treatment were used as cleavage-negative and cleavage-positive controls, respectively. Consistent with the above results, a ~36 kDa cleaved P74 band was visible in the unheated OBs sample, whereas a full-length P74 band was observed in the heated OBs (Fig. 2). When SBTI was added at the beginning of the alkaline treatment (0 min), only full-length P74 was observed, suggesting that the trypsin inhibitor completely blocked the endogenous protease activity. When SBTI was added 1 min after alkaline treatment, the amount of full-length P74 was significantly reduced, and two cleaved bands (~36 and ~38 kDa) were detected. When SBTI was added 3 min after alkaline treatment, full-length P74 was almost undetectable, and the signal of the ~36 kDa band was strong, while that of the ~38 kDa band became very weak. When SBTI was added 5 min after alkaline treatment, only the 36 kDa band was observed (Fig. 2), suggesting that it is a more completely digested product. Collectively, these data suggest that the endogenous proteinase is a trypsin-like proteinase that can quickly (within 1 min in our test) cleave P74 into an N-terminal 36-kDa product under complete digestion and can be inactivated by SBTI.

Fig 2.

Trypsin inhibitor blocks P74 cleavage by OB-endogenous proteinase. For the inhibition assay, SBTI was added to the OB dissolving solution to a final concentration of 1 mg/10⁹ OBs at the indicated time points. The released ODVs were collected and subjected to western blot analysis with the primary antibodies P74 mAb 21A7 or anti-VP39. The arrows indicate the full-length P74; the bracket indicates the N-terminal cleavage product of P74. VP39 was used as a loading control.

Bioinformatic analyses reveal a structural flexible R/K-rich region conserved in P74 homologs

Bioinformatics analysis was performed to predict the cleavage site(s) of P74. Sequence alignment analysis showed that AcMNPV P74 contained two domains: the Baculo_p74-N superfamily (P74-N) and the Baculo_p74 superfamily (P74-C) (Fig. 3A). These two domains are conserved in all baculoviral P74 homologs. Between the two conserved domains is a region rich in arginine (R) and/or lysine (K), which could be potential cleavage sites for trypsin digestion (Fig. 3B). The three-dimensional structure of AcMNPV P74 predicted by AlphaFold2 showed that P74 is shaped like an inverted “L,” anchoring in the viral membrane with two transmembrane regions (I591–H610 and I614–I631） (Fig. 3C). A short stem region links the transmembrane and stacked P74-N and P74-C domains. P74-N and P74-C interact closely with each other, and almost the entire P74-C domain is covered by P74-N, with a flexible loop connecting these two domains. The P74-N contains 14 α-helices (α1–α14) and 6 sets of anti-parallel β-sheets (β1–β12), whereas the P74-C contains 11 α-helices (α15–α25) and 2 transmembrane domains (α26 and α27). The R/K-rich region appears to be a flexible loop in between P74-N and P74-C (Fig. 3C and D). The previously reported BBMV cleavage site, R195/R196/R199 (17), was buried within the α8 of P74-N (Fig. 3C and D). In AcMNPV, the R/K-rich region contains four arginine residues: R319, R323, R325, and R334. Thus, these sites were subsequently further analyzed to identify possible cleavage sites.

Fig 3.

Bioinformatics and AlphaFold2 analyses prediction of the potential cleavage sites of P74. (A) Schematic illustration of P74 showing the conserved domains and R/K-rich region. The AcMNPV P74 aa sequence was compared using the National Center for Biotechnology Information database. The graph shows the two conserved domains: Baculo_p74_N superfamily (P74-N) and Baculo_p74 superfamily (P74-C). (B) Sequence alignment of selected P74 homologs showing the R/K-rich region. The alignments were performed using the CLUSTAL W method. The R and K residues are highlighted in red. The virus names are indicated on the left and the virus classifications are indicated on the right. (C) The structure of AcMNPV P74 predicted by AlphaFold2. P74-N, P74-C, and transmembrane regions are colored in blue, pink, and green, respectively. The dotted region is enlarged at the right bottom corner with the red arrows indicating the sites of R319, R323, R325, and R334. The orange arrows indicate the sites of R195, R196, and R199. (D) Diagram showing the topology of secondary structural elements within P74. The arrows denote β-strands. The cylinders denote α-helices. Blue indicates the P74-N domain; pink indicates the P74-C domain; green indicates transmembrane domains; R319, R323, R325, and R334 are marked with bold red lines; R195, R196, and R199 are marked with orange lines. CpGV, Cydia pomonella granulovirus; CuniNPV, Culex nigripalpus nucleopolyhedrovirus; HearNPV, Helicoverpa armigera nucleopolyhedrovirus; NeleNPV, Neodiprion lecontei nucleopolyhedrovirus.

Both R325 and R334 of AcMNPV P74 are the primary cleavage site of OB-endogenous proteinase

In order to investigate whether the above R/K-rich region is involved in P74 cleavage, a series of recombinant AcMNPVs with site-directed P74 mutations were constructed by substituting arginine residues with glutamine (Q) residues (Fig. 4A). R319Q, R323Q, R325Q, and R334Q were mutants with single substitutions, R319/323Q and R325/334Q had double mutations, and R319/323/325/334Q contained four substitutions. Sf9 cells were individually transfected with recombinant bacmid DNAs to rescue the viruses. As shown in Fig. 4B, the infection assay showed successful infection in all samples at 96 h post infection, indicating that all mutations could produce efficient progeny budded viruses (BVs) (Fig. 4B). This was consistent with the per os function of P74, which is dispensable for BV production.

Fig 4.

Mutation analyses show that R325 and R334 are primary cleavage sites. (A) Construction of recombinant bacmids. All the recombinant bacmids were generated using the backbone of AcBac-Syn (control), which contains EGFP at the hr4 locus (20). The p74 mutant bacmids were generated by Tn7-mediated transposition based on a constructed p74 knockout bacmid. (B) Transfection and infection assays of p74 mutants. Sf9 cells transfected or infected with indicated recombinants and observed by fluorescence microscopy for EGFP expression at 96 h p.t. and 96 h p.i. (C) The mutant OBs were produced by injection of late fourth instar S. frugiperda larvae. Purified ODVs were analyzed by western blotting using P74 mAb 21A7. p.i., post infection; p.t., post transfection.

The OBs of all recombinant viruses were produced by injecting Spodoptera frugiperda larvae with BV and then purified. Results of western blotting using mAb 21A7 are shown in Fig. 4C. For comparison, two Control samples were loaded: one in the middle panel shows incomplete cleavage with both 36 kDa and 38 kDa products, whereas the other in the right panel shows complete cleavage with only the 36 kDa product. Rep, as well as the mutants R319Q, R323Q, R325Q, and R334Q, produced a cleavage product of ~36 kDa, indicating that these single amino acid mutations do not block OB-endogenous proteinase digestion. The double mutant R319/323Q, in the middle panel, produced a cleavage product of ~38 kDa, suggesting that R319/323 was related to the final ~36 kDa product. In contrast, the double mutant R325/334Q in the middle panel and quadruple mutant R319/323/325/334Q in the right panel did not show any digested products with only full-length P74 being detected (Fig. 4C). These results suggest that both R325 and R334 are required as a primary cleavage site of P74.

BBMV cleaves P74 at the same sites as the OB-endogenous proteinase

Previous studies have suggested that AcMNPV P74 is sequentially cleaved twice during ODV entry. The first cleavage of P74 occurred during the release of ODV from OBs and was mediated by an OB-endogenous proteinase, whereas the second cleavage of P74 occurred at R195/196/199 by an unknown protease from BBMV. The second cleavage was essential because the oral infectivity of the R195/196/199 mutant decreased considerably (17). To confirm this, BBMV was collected from the midguts of Spodoptera exigua larvae, and the quality was verified by alkaline phosphatase activity (Fig. S2).

First, we analyzed the cleavage of ODV P74 by BBMV using western blotting. The ODVs were released from OBs by alkaline treatment and treated with different concentrations of BBMV. As shown in Fig. 5A, without BBMV, P74 was cleaved to an ~36 kDa band, and some intact P74 remained. After different amounts of BBMV were incubated with ODVs, we did not observe the expected second cleavage band (~25 kDa), even at a high dose of BBMV.

Fig 5.

Both R325 and R334 are also the primary cleavage site of BBMV. (A) Western blot analyses of P74 cleaved with BBMV. ODVs were released by the alkaline hydrolysis of control OBs, and cleavage of P74 was detected by adding different concentrations of BBMV. (B) Western blot analyses of P74 cleavage by BBMV when OB-endogenous proteinase is inhibited by SBTI. Control ODVs were obtained by adding SBTI during the ODV release. SBTI was removed by centrifugation, and the ODVs were treated with 200 µg trypsin and different amounts of BBMV, respectively. (C and D) Mutations in R325 and R334 block BBMV cleavage of P74. The ODVs of R325/334Q (C) and R319/323/325/334Q (D) were purified and incubated with 200 µg trypsin and different amounts of BBMV, respectively. Western blot analysis of P74 cleavage with P74 mAb 21A7. AcPIF5 was used as a loading control.

To verify whether BBMV alone can cleave P74, we detected the cleavage of P74 by inhibiting OB-endogenous proteinase with SBTI. As shown in Fig. 5B, P74 of ODVs appeared to be intact after inhibition with SBTI. After the incubation of ODVs with 200 µg trypsin, the cleaved fragments of ~36 kDa were detected and a band of intact P74 was barely detectable. After BBMV was incubated with ODVs, two cleaved bands of ~36 and ~38 kDa were detected. Cleavage of P74 was detected only at 50 µg BBMV and higher (Fig. 5B), and as the BBMV increased above 50 µg, less and less of full-length P74 was detected. These results suggest that BBMV proteinase cleaves P74 at the same sites as trypsin and OB-endogenous proteinase.

We further analyzed BBMV cleavage using the P74 mutants R325/334Q and R319/323/325/334Q, in which the OB-endogenous proteinase cleavage sites were mutated. As shown in Fig. 5C and D, these mutants did not produce obvious cleaved bands after incubation with BBMV (or trypsin). Therefore, the R325/334Q mutation blocked not only endogenous protease cleavage but also the cleavage by protease(s) from BBMV.

Mutations in R195, R196, and R199 result in degradation of P74 during ODV release

Our results showing that BBMV cleaves P74 at R325/R334 are different from those reported by Slack et al. (17), in which they suggested that trypsin cleaved AcMNPV EGFP-tagged P74 at R195, R196, and R199, and that mutation of these sites compromised per os infectivity. To verify these results, we constructed a mutant virus, R195/196/199Q, in which the R195, R196, and R199 of AcMNPV P74 were substituted with Q (Fig. 6A). Transfection and infection experiments showed that the mutant virus was successfully rescued and the progeny BVs were produced efficiently (Fig. 6B). The P74 cleavage by OB-endogenous proteinase was then tested by western blotting using heated and unheated OBs. As shown in Fig. 6C, when the OBs were heated, full-length P74 was detected in both control and R195/196/199Q, suggesting successful assembly of the mutated P74 into the ODV. However, although a band at 36 kDa was found for unheated control OBs consistent with cleavage at R325/R334, no P74 or smaller peptides were detectable for unheated R195/196/199Q OBs (Fig. 6C). This indicates that the R195/196/199Q mutation results in instability and degradation of P74 (at least for the N-terminus of P74) during ODV release.

Fig 6.

Mutations in R195, R196, and R199 result in instability of P74 during ODV release. (A) Construction of recombinant bacmid R195/196/199Q. The mutant bacmid was generated by Tn7-mediated transposition based on a p74 knockout bacmid. (B) Transfection and infection assays of R195/196/199Q. Sf9 cells transfected or infected with R195/196/199Q recombinants were observed by fluorescence microscopy for EGFP expression at 96 h p.t. and 96 h p.i. (C) Western blot analysis of P74 of control and R195/196/199Q ODVs. OBs of control and R195/196/199Q were treated with or without heating. Purified ODVs were used for western blotting with P74 mAb 21A7.

The absence of P74 cleavage significantly affected the oral infectivity

Per os infection with the mutant viruses was performed using the droplet method, and the lethal concentration 50 (LC₅₀) was determined (Table 1). The mutant viruses were separated into two groups for the bioassays, and Rep was used as a positive control for both groups. The LC₅₀ values of Rep in these two experiments were similar (3.67 × 10⁴ OBs/mL vs 4.23 × 10⁴ OBs/mL). PoloPlus analysis showed no significant difference between Rep and the four single-point mutations, R319Q, R323Q, R325Q, and R334Q. The LC₅₀ value of Rep was significantly different from those of R325/334Q and R319/323/325/334Q. The LC₅₀ value of R325/334Q was approximately 10 times that of Rep, and the LC₅₀ value of R319/323/325/334Q was approximately 15 times that of Rep. This suggests that the absence of P74 cleavage significantly affects the per os infectivity of AcMNPV.

TABLE 1.

LC₅₀ of P74 mutants in the third instar S. exigua larvae^a

Virus	LC50 (95% Clb) (104 OBs/mL)	Potency ratio (95% Cl)
Rep	3.67 (1.41–10.69)	–
R319Q	1.75 (0.79–3.59)	2.101 (0.905–4.881)
R325/334Q	35.13 (22.65–55.43)	0.105 (0.056–0.194)
R319/323/325/334Q	56.65 (25.65–126.55)	0.065 (0.028–0.152)
Rep	4.23 (2.66–6.92)	–
R323Q	6.03 (3.57–10.97)	0.703 (0.339–1.456)
R325Q	5.95 (3.34–11.65)	0.711 (0.328–1.542)
R334Q	3.76 (1.20–14.54)	1.126 (0.555–2.283)
R319/323Q	4.48 (2.42–11.14)	0.946 (0.400–2.233)

^a Potency ratio was calculated by dividing the LC₅₀ value of mutants by that of Rep. Potency ratios with confidence levels that include 1 indicate that the LC₅₀ values of the viruses compared are not significantly different from each another.

^b CI, confidence interval.

DISCUSSION

In AcMNPV, the current model suggests that the cleavage of P74 occurs in two steps (19). The first step of cleavage is conducted by the OB-endogenous protease during ODV release under alkaline conditions, and the second cleavage step is mediated by a proteinase from midgut epithelial cells. The first cleavage was suggested to occur between the conserved Baculo_p74_N domain (amino acid [aa] 5–312) and the Baculo_p74 superfamily domain (aa 351–583), generating an N-terminal 35 kDa fragment and an ~40 kDa C-terminal fragment (19). The second cleavage sites were predicted to be R195, R196, and R199 of P74 based on a previous study (17). In this study, we identified the primary cleavage site (R325 or R334) of P74 mediated by the endogenous protease and showed that proteases residing in BBMV target the same cleavage sites of AcMNPV P74. Thus, we refined the previous P74 cleavage model in which both endogenous and midgut proteases cleave P74 at the same sites to ensure the inevitable cleavage of P74 (Fig. 7A). A bioassay showed that the R325/334 mutation significantly impaired viral oral infectivity, indicating that cleavage of P74 is important for its biological function during oral infection.

Fig 7.

Cleavage model of P74 and prediction of potential fusion peptides. (A) Cleavage model of P74. P74 is cleaved by OB-endogenous and host proteases at R325 and R334. (B) Cleavage of P74 exposes the potential fusion peptide at the N-terminus of P74-C. The potential fusion peptides of four representative species of baculovirus genera were predicted. Each circle represents an amino acid. Gray represents hydrophobic amino acids; yellow represents hydrophilic polar amino acids; red represents hydrophilic non-polar amino acids.

The difference between the current study and the previous cleavage model is that there is only one functional cleavage site, R325/334, and the one suggested at R195/196/199 by Slack et al. (17) is not actually a cleavage site, although this region is important for oral infectivity. To be noted, the method we used here is different from that of the previous report. We used a specific mAb 21A7 (epitope: S76–M95) to detect P74 and its mutants, while Slack et al. used an antibody against EGFP to detect P74-EGFP fusion protein. The fusion of EGFP to the P74 C-terminus might alter the conformation of P74 and result in different pattern of P74 cleavage. Our data confirmed the loss of oral infectivity of the R195/196/199Q mutant (Table S1), which was likely due to the instability of P74 during ODV release, as shown in Fig. 6C.

Although we identified both R325 and R334 as the primary cleavage site, cleavage also occurred at R319 and R323. The predicted N-terminal product size should be 36.3, 36.8, 37.0, and 38.1 kDa if P74 is cleaved at R319, R323, R325, and R334, respectively. Two cleavage bands at ~36 and ~38 kDa were often observed after the cleavage of P74, and the ~36 kDa band appears to be a completely digested product (Fig. 2, 4, and 5). Being an incomplete digestion band, the occurrence of the 38 kDa bands appears to be associated with the protease activity. It could be detected when the activity of OB-endogenous proteinase is weak, as in the case shown in the middle panel of Fig. 4C. When R319/R323 were mutated, only the ~38 kDa band was observed (Fig. 4C), suggesting that the cleavage of R319/R323 generated the ~36 kDa product. When R325/R334 were mutated, no cleavage bands were observed (Fig. 4C). Therefore, OB-endogenous proteases and host proteases may preferentially cleave R325 or R334 and subsequently expose R319 or R323 for further cleavage. N-terminal sequencing of the cleaved fragments in the future might clarify exactly where the actual cleavage site is.

Certain viral entry factors undergo cleavage and conformational changes during entry. One of the significant events in the activation of class I membrane fusion proteins is cleavage. For example, the membrane fusion protein hemagglutinin (HA) of the influenza virus is produced as a precursor protein (HA0) in the virion envelope. During the entry process, the receptor-binding domain HA1 binds to the cellular receptor sialic acid, and trypsin cleaves HA0 to produce receptor-binding (HA1) and fusion (HA2) subunits. HA2 is allosteric at low pH, exposing the fusion peptides (21).

To date, it has been reported that P74 may be involved in ODV binding to host midgut cells (12). We found that P74 contains certain features resembling those of class I membrane fusion proteins. First, it is cleaved into P74-N and P74-C during ODV release. Second, the 18 aa (D338–V355) at the N-terminus of P74-C are amphipathic helices in the predicted secondary structure, which is a potential fusion peptide sequence (Fig. 7B). We also analyzed the predicted P74-C domains of representative viruses from the four genera of baculoviruses and found that all contained a predicted amphiphilic helical secondary structure at the N-terminus of P74-C (Fig. 7B), suggesting that the exposure of potential fusion peptides by cleavage is likely to be a conserved activation process in P74 homologs.

In conclusion, we found that during ODV invasion, OB-endogenous proteases and host proteases mainly cleave R325 or R334 of P74. Based on the analysis and prediction of the structure of P74, we hypothesized that P74 may perform membrane fusion functions. Cleavage of P74 may expose the fusion peptides located at the N-terminus of P74-C, initiating the membrane fusion process. Of course, this hypothesis is awaited to be verified in future studies.

MATERIALS AND METHODS

Cell lines, viruses, and insects

The Sf9 cell line, derived from S. frugiperda, was maintained in Grace’s medium supplemented with 10% fetal bovine serum at 27°C. vAcBac-Syn (20) was used as the control virus. S. frugiperda larvae and S. exigua larvae were reared on an artificial diet at 28°C.

Western blot analysis of P74 cleavage

OBs were purified from the infected larvae. The production and purification methods for OBs were as previously described by Li et al. (22). To inactivate the endogenous protease, OBs were heated at 80°C for 1 h. Heat-treated or untreated OBs were dissolved in an alkaline buffer, dilute alkaline saline [DAS: 0.1 M Na₂CO₃, 0.15 M NaCl, 10 mM EDTA (pH 11)], at 25°C for 5 min, followed by neutralization with a 1/10 vol of Tris-HCl (500 mM, pH 7.5). After removing non-dissolved debris by centrifugation at 2,400 × g for 5 min, the ODVs were collected by centrifugation at 20,000 × g for 30 min at 4°C. The purified ODV proteins were separated by 12% SDS-PAGE and analyzed by western blotting using a house-made anti-P74 mAb 21A7, followed by incubation with a horseradish peroxidase-conjugated secondary antibody against mouse IgG (Sigma-Aldrich). The nucleocapsid protein VP39 or ODV envelope protein PIF5 was used as an internal reference. Polyclonal antibodies against VP39 and mAbs against PIF5 were prepared in the laboratory (22). Signals were detected using SuperSignal West Pico chemiluminescent substrate (Pierce).

To verify whether P74 was cleaved by a trypsin-like proteinase, an SBTI (Sigma-Aldrich, St. Louis, MO) was added to DAS at a final concentration of 1 mg/10⁹ OBs at the indicated time points during the ODV release process. More specifically, for the 0 min samples, SBTI was added first, and then the OBs were dissolved in the DAS solution. For the other samples, the OBs were treated with DAS for 1, 3, or 5 min, followed by the addition of SBTI. At 5 min after DAS treatment, neutralization was conducted and the ODV samples were processed for western blot analysis, as described above.

To detect the cleavage of P74 by BBMV, the first cleavage by endogenous proteinase was either not inhibited or inhibited using SBTI. SBTI was then removed by pelleting ODV at 20,000 × g for 30 min. The purified ODV sample was incubated with BBMV. As a cleavage-positive control, 200 µg trypsin was added to cleave P74 at 28°C for 30 min. In addition, 5, 10, 50, 100, 150, and 200 µg BBMVs were added to cleave P74, respectively. The cleavage of P74 was detected by western blotting.

Computational analysis of P74 homologs

The P74 homologs used in this analysis were obtained from GenBank. Multiple sequence alignments were performed using MEGA7 ClustalW and manually edited using GeneDoc software. The structure of AcP74 was predicted using AlphaFold2 (23) (https://colab.research.google.com/github/sokrypton/ColabFold/blob/main/AlphaFold2.ipynb). The helical wheel of amphiphilic helices from the N-terminus of P74-C was generated with the use of the online tool heliQuest (24) (https://heliquest.ipmc.cnrs.fr/cgi-bin/ComputParamsV2.py). Output visualization was used for figure generation in CorelDRAW, and rotation of the resulting helix was performed for better figure presentation.

Generation of p74 mutant recombinant viruses

AcBac-Syn was used as a template to amplify the upstream and downstream homologous arms of p74 using the primers listed in Table S2. The chloramphenicol resistance gene (cm^R) was amplified using pKS-cm^R. The fragments were then connected and amplified using overlapping PCR. The p74 of AcBac-Syn was replaced with cm^R by homologous recombination in Escherichia coli to generate the p74 deletion bacmid bAcBacΔp74.

To prepare a P74 repaired bacmid, the p74 gene was amplified from AcBac-Syn by PCR using the primers Acp74-F and Acp74-R. To add an S tag (KETAAAKFERQHMDS) to the C-terminus of P74, the coding sequence of p74 was amplified by PCR using the primers Acp74-F and Acp74-S tag-R (Table S2). Eight P74 mutants, R319Q, R323Q, R325Q, R334Q, R319/323Q, R325/334Q, R319/323/325/334Q, and R195/196/199Q, were generated by changing R to Q. All PCR amplicons and the vector pFastBacDual (pFBD) were digested with restriction enzymes BamHI and HindIII and then linked with T4 DNA ligase. DH10B cells were transformed individually with the plasmids pFBD-p74 and pFBD-p74mut with bAcBacΔp74 bacmid and the helper plasmid expressing transposase (Bac-to-Bac Baculovirus Expression System, Gibco-BRL). Transposition-positive bacmids were screened using PCR with M13F and M13R (Bac-to-Bac).

Sf9 cells were transfected with the recombinant bacmids to produce recombinant viruses. Transfection was performed using Cellfectin (Invitrogen) according to the manufacturer’s protocol. Sf9 cells were cultured at 27°C in Grace’s insect medium supplemented with 10% fetal bovine serum. The cells were observed using an Evos fluorescence microscope (Thermo Fisher Scientific) every 24 h after transfection or infection. The BV supernatant was collected 96 h after transfection.

BBMV preparation and verification

Our methodology was an adaptation of the procedure described by Slack and Lawrence (18). Midguts were dissected from third instar S. exigua larvae. Gut contents including peritrophic membranes were removed, and the guts were stored at −80°C. Fifty midguts were suspended in 1.2 mL buffer A [300 mM mannitol, 5 mM ethylene glycol tetraacetic acid (EGTA), 20 mM Tris-HCl, pH 9] and 300 µL 600 mM MgCl₂ solution. The midgut suspension was homogenized using a glass homogenizer and incubated on ice for 20 min. Homogenates were centrifuged at 3,000 × g for 15 min at 4°C. Supernatants were collected and centrifuged at 20,000 × g for 30 min at 4°C. BBMV pellets were suspended in 500 µL buffer A, aliquoted into Eppendorf tubes, and stored at −80°C. BBMV concentration was determined using the Bradford protein concentration assay. Different amounts of BBMV were diluted in 96-well plates with 150 µL buffer A. Substrate nitroblue tetrazolium (NBT)/5-bromo-4-chloro-3-indolyl phosphate (BCIP) (50 µL) was added to each well to react with alkaline phosphatase in BBMV, and the OD₅₇₀ was determined.

Bioassays

S. frugiperda larvae in the late fourth or early fifth instar were injected with the BV of each mutant to obtain OBs. To examine the in vivo effects of the predicted P74 cleavage site mutants, bioassays were carried out on early third instar S. exigua larvae using the droplet method. Briefly, OBs of Rep, R319Q, R323Q, R325Q, R334Q, and R319/R323Q were diluted with feeding buffer [1% (wt/vol) brilliant blue and 10% (wt/vol) sucrose] to final concentrations of 1 × 10³, 5 × 10³, 2.5 × 10⁴, 1.25 × 10⁵, and 6.25 × 10⁵ OBs/mL. OBs of R325/334Q were diluted to final concentrations of 1 × 10⁴, 5 × 10⁴, 2.5 × 10⁵, 1.25 × 10⁶, and 6.25 × 10⁶ OBs/mL. OBs of R319/323/325/334Q were diluted to final concentrations of 5 × 10⁴, 2.5 × 10⁵, 1.25 × 10⁶, 6.25 × 10⁶, and 3.125 × 10⁷. The feeding assay was divided into two groups, and each group was independently carried out twice. The LC₅₀ values and potency ratio of the recombinant viruses were calculated using Probit analysis (25) in PoloPlus (26). In addition, OBs of Δp74 and R195/196/199Q were diluted to a final concentration of 1 × 10⁸ OBs/mL and fed to early third instar S. exigua larvae using the droplet method.

DATA AVAILABILITY

The accession numbers for the P74 sequence of presented baculoviruses are NP_054168.1 (AcMNPV), NP_075089.1 (HearNPV), NP_148844.1 (CpGV), YP_025247.1 (NeleNPV), and NP_203378.1 (CuniNPV). Source data are provided within the article and supplemental files.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/jvi.00235-24.

Supplemental material. jvi.00235-24-s0001.docx.

Tables S1 and S2; Figures S1 and S2.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.