The Membrane-Anchoring Region of the AcMNPV P74 Protein Is Expendable or Interchangeable with Homologs from Other Species

Baculoviruses are insect pathogens that are characterized by assembling the viral dsDNA into two different enveloped virions during an infective cycle: occluded virions (ODVs; immersed in a protein matrix known as occlusion body) and budded virions (BVs). ODVs are responsible for the primary infection in midgut cells of susceptible larvae thanks to the per os infectivity factor (PIF) complex, composed of at least nine essential viral proteins. Among them, P74 is a crucial factor whose activity has been identified as virus-specific. In this work, the p74 gene from AcMNPV was pseudogenized using CRISPR/Cas9 technology and then complemented with wild-type alleles from SeMNPV and HearSNPV species, as well as chimeras combining the P74 amino and carboxyl domains. The results on Spodoptera exigua and Rachiplusia nu larvae showed that an amino terminal sector of P74 (lacking two potential transmembrane regions but possessing a putative nuclear export signal) is sufficient to restore the virus infectivity whether alone or fused to the P74 transmembrane regions of the other evaluated viral species. These results provide novel information about the functional role of P74 and delimit the region on which mutagenesis could be applied to enhance viral activity and, thus, produce better biopesticides.


Introduction
The Baculoviridae family is composed of insect-specific rod-shaped viruses with large covalently closed circular double-stranded DNA (cccdsDNA), from 81,755 bp (NeleNPV) to 178,733 bp (XecnGV) in length, usually used as crop pest control agents among other biotechnological uses [1,2]. In nature, baculoviruses infect susceptible hosts in the larval state and produce two genetically identical but morphologically distinct viral phenotypes: occlusion-derived viruses (ODVs) and budded viruses (BVs) [3]. The first are those that initiate the infection in the midgut defining the host range (primary infection), and they are characterized by being embedded in a crystalline matrix denominated occlusion bodies (OBs) that physically protect virions against environmental stresses, while the BVs Manipulation through viral genome engineering such as PIF deletions or modifications allows us to assess the protein function and its implications in the baculoviral infectivity. In recent years, genome mutagenesis has become easier with the introduction of clustered regularly interspaced short palindromic repeats/Cas9 or CRISPR/Cas9 technology [48][49][50]. The CRISPR/Cas9 system involves a single guide RNA (sgRNA) that interacts and directs the Cas9 endonuclease to a specific site to perform a double-strand DNA break (DSB) [51]. This triggers DNA repair via one of two major pathways, the error-prone nonhomologous end joining (NHEJ) or the homology-directed repair (HDR) pathway. The latter mechanism requires the provision of donor DNA with homology to the flanking sequence of the DSB. However, in the absence of this template, the endogenous cell repair machinery conducts NHEJ, leaving consequential insertions/deletions (indel) mutations as a scar near the site of the break [52].
In this work, a variant of AcMNPV, the prototypic baculovirus most used for biotechnological purposes, was knocked out in the p74 gene by CRISPR/Cas9 technology and NHEJ repair, and then complemented with sequences that encode P74 domains or P74 chimeras generated by combinations of domains from different baculovirus species (Spodoptera exigua multiple nucleopolyhedrovirus (SeMNPV) and Helicoverpa armigera single nucleopolyhedrovirus (HearSNPV)). The characterization of the P74 protein and the biological effects on susceptible larva add new evidence for understanding the baculovirus PIF complex.

In Vitro Cell Culture and Insect Rearing
The Spodoptera frugiperda Sf21 insect cell line [53] (obtained from GIBCO BRL, Gaithersburg, MD, USA) was cultured on monolayers at 25 • C in Grace's medium (GIBCO BRL, Gaithersburg, MD, USA) supplemented with 10% fetal bovine serum (FBS; GIBCO BRL, Gaithersburg, MD, USA) and antibiotics and antimycotics (GIBCO BRL Gaithersburg, MD, USA). Spodoptera exigua larvae were derived from the established colony at the University of Valencia (Spain), and Rachiplusia nu larvae were obtained from AgIdea S.A. (Pergamino, Buenos Aires, Argentina). Insects were reared with an artificial diet at 25-27 • C with 70% relative humidity and a photoperiod of 16/8 h (light/dark).

Bioinformatics Studies
A nonredundant set of 76 P74 protein sequences was obtained from the reported genomes (22 group I alphabaculoviruses, 28 group II alphabaculoviruses, 22 betabaculoviruses, three gammabaculoviruses, and 1 deltabaculovirus) (Supplementary Table S1) and aligned using ClustalX [54]. The consensus line was used to generate a sequence similarity plot employing a residue-by-residue sliding window strategy (35 nucleotides long), where arbitrary values were assigned to different symbols (+1 for identical residues, +0.5 and +0.25 for different degrees of conservative changes, and −0.2 for nonconservative changes). The algebraic sum was normalized by the window length and allotted to the window central position. Moreover, the evolutionary history was inferred with MEGA7 [55] using the maximum likelihood method based on the LG + G + I model [56], while also considering orthologs to P74 of nudiviruses, hytrosaviruses, and nimaviruses (Supplementary Table S1). Initial trees for the heuristic search were obtained using neighbor-joining and BioNJ algorithms (the matrix of pairwise distances was estimated using a JTT model), and the topology with a superior log likelihood value was selected. A discrete gamma distribution was used to model the different evolutionary rates among sites (two categories (+G, parameter = 1.1482)). The weblogo server was used with default parameters (https://weblogo.berkeley.edu/logo.cgi; last access 1 October 2021) [57], and the hydropathy profiles were generated using a sliding overlapping window strategy (21 residues long displaced from one residue at a time). The putative transmembrane domains, the disulfide bonds, and the nuclear export signal (NES) predictions were performed using webservers (https: //services.healthtech.dtu.dk/service.php?TMHMM-2.0, accessed on 26 November 2021, [58]; http://disulfind.disi.unitn.it/, accessed on 26 November 2021, [59]; http://clavius.bc.edu/ clotelab/DiANNA/, accessed on 26 November 2021, [60]; https://services.healthtech.dtu.dk/ service.php?NetNES-1.1 [61]; last access 1 October 2021) and standard parameters.

P74 Gene Knockout in AcMNPV-bacmid
The AcMNPV-bacmid bMON14272 [62] was depleted in the p74 gene using CRISPR/Cas9 technology. Briefly, the CRISPR RNA (crRNA, 5 AACTGGCTTTCAGCAAGCGC 3 ) used to direct the Streptococcus pyogenes Cas9 nuclease (New England Biolabs, Ipswich, MA, USA) to the initial part of the p74 open reading frame (ORF) was previously designed 207 bp downstream of the ATG codon [50] using the CHOP-CHOP online platform (https://chopchop.rc.fas.harvard. edu accessed on 26 November 2021) and synthesized (Integrated DNA Technologies, Leuven, Belgium). To generate the ribonucleoprotein (RNP) complex, 1 µL of the crRNA (1 µM) was incubated with 1 µL of tracrRNA (1 µM) and 98 µL of nuclease-free duplex buffer (both from Integrated DNA Technologies, Leuven, Belgium) for 5 min at 95 • C. Then, 1 µL of the crRNA-tracrRNA mixture was combined with 6.2 µL of Cas9 (1 mM) and 2.8 µL of nuclease-free water to a final volume of 10 µL (incubation at 37 • C for 5 min). The bacmid DNA was purified following the manufacturer's instructions of Bac-to-Bac system (GIBCO BRL Gaithersburg, MD, USA) and added to the previous mixture (1 µg), which was then transfected in Sf21 cells (70% confluent) using Cellfectin ® II Reagent (Invitrogen, Carlsbad, CA Carlsbad, USA) in Grace's medium without FBS. After 5 h of incubation, the medium was replaced by Grace's medium supplemented with 10% FBS and incubated for 72 h at 25 • C. Later, BVs were isolated from supernatants following the plaque assay protocol [63], and the individual clones were analyzed by amplicon Sanger sequencing using PrepMan ® Ultra reagent (Applied Biosystems, Foster City, CA, USA) and the primers FwP74seq and RvP74seq (Table 1) [50] to detect indel mutations. Once P74 depleted bacmid (bMON14272∆p74) was identified, BVs were replicated in Sf21 cells, and the viral DNA was purified using the standard phenol/chloroform protocol [63] to be sequenced by Illumina (MiSeq System, Laboratorio de Secuenciación Servicio Central de Soporte a la Investigación-SCSIE, Universitat de València, Valencia, Spain). Finally, DH10b Escherichia coli competent cells were transformed with the AcMNPV-bacmid∆p74 DNA and the pMON7124 helper plasmid (GIBCO BRL Gaithersburg, MD, USA) to generate the DH10bac∆p74 strain for further Bac-to-Bac applications.

OB Generation for the Recombinant Baculoviruses
Both bMON14272 and bMON14272∆p74 do not express the polyhedrin gene; therefore, they do not generate OBs in cell culture. Due to this, a co-occlusion strategy was employed to produce infecting OBs using as helper a p74-deficient virus that expresses polyhedrin. Thus, a recombinant baculovirus using bMON14272∆p74 and a pFastBac-Dual carrying mCherry ORF (under p10 promoter) and AcMNPV polyhedrin ORF (under polyhedrin promoter) was generated (Bac∆p74/Cherry+polH), multiplied, and titrated as previously mentioned to produce BVs and OBs. All OB productions were carried out by coinfecting Sf21 cells with each GFP-producing virus (providing variants of P74) with the mCherryproducing virus (providing polyhedrin) at a multiplicity of infection (MOI) = 5 for each one. After 5 days of incubation at 25 • C, the cells were harvested, pelleted by centrifugation at 1000× g for 5 min, and resuspended in 0.1% w/v SDS. OBs were separated from the cellular debris by loading the lysate in a 40% w/v sucrose solution and centrifuging at 30,000× g for 30 min. The resulting pellets were washed with distilled water and centrifuged at 2000× g for 5 min. Finally, after resuspending the OB pellets in distilled water, they were quantified using a Neubauer chamber and an optical microscope, thus generating AcMNPV-∆p74, AcMNPV-p74AcAc, AcMNPV-p74SeSe, AcMNPV-p74HaHa, AcMNPV-p74AcSe, AcMNPV-p74AcHa, AcMNPV-p74SeAc, AcMNPV-p74Ac, and AcMNPV-p74wt. The presence of P74 protein in the OBs was verified by Western blot, first resolving the viral proteomes in a 10% SDS polyacrylamide gel electrophoresis (SDS-PAGE) and then using mouse anti-6×HIS as the primary antibody (BD Biosciences, Franklin Lakes, NJ, USA) and a goat anti-mouse immunoglobulin conjugated with HRP as the secondary antibody (Jackson ImmunoResearch Laboratories INC, West Grove, PA, USA). Protein bands were detected using an enhanced chemiluminescent substrate (ECL; Thermo Fisher Scientific, Watham, MA, USA) and a C-Digit blot scanner (LI-COR, Bad Homburg, Germany).

Bioassays with Recombinant AcMNPVs
Two methodologies were used to assess the infectivity of the recombinant viruses. On the one hand, 16 fourth-instar S. exigua larvae were injected into the hemocoel with 5 µL of 5 × 10 4 BVs of each different AcMNPV in MQ water (Grace's medium was used as a negative control). The infections were confirmed by the visualization of fluorescence (Leica MZ10 F), and larvae were incubated at 26 • C until death. On the other hand, to evaluate the oral infectivity of co-occluded recombinant viruses by the droplet feeding method [67], suspensions of OBs were diluted in a 10% v/v phenol red and 10% w/v sucrose solution to perform bioassays in 16 early third-instar S. exigua or R. nu larvae (1 × 10 7 OBs/mL). Exposed insects were maintained at 26 • C and mortality was recorded every 12 h (S. exigua) or 24 h (R. nu) until all individuals died. The bioassays were performed side-by-side for all the viruses and repeated three times. Mortality curves were assessed using the Kaplan-Meier method and compared using the log-rank analysis (Mantel-Cox test) using GraphPad Prism software (GraphPad Software Inc., San Diego, CA, USA). The hemolymph was recovered from infected larvae to expose Sf21 cells growing in monolayers on six-well plaques and was incubated for 4 days. The supernatants of infection recovered after this time were treated with PrepMan ® Ultra reagent (Applied Biosystems, Waltham, MA, USA) to obtain the viral DNAs that were later used as PCR templates using the primers FwMe53 y RvP74seq (Table 1). These PCR products were sequenced by Sanger method (STAB VIDA, Caparica, Portugal) to confirm the genetic stability of the p74 knockout.

Phylogeny
The baculovirus PIF proteins are structural polypeptides associated with the ODV envelope and functionally essential for primary infection. P74 phylogeny reproduced the organization of Baculoviridae in the four recognized genera [6], showing an ancestral evolution pattern as the viral family diversified (Figure 1). Orthologs to this protein are found in other ancestrally related invertebrate virus families such as Nudiviridae, Hytrosaviridae, and Nimaviridae [36], suggesting the existence of similar functional roles for this protein in those viruses. The P74 proteins considered in this work to evaluate their role in the AcMNPV PIF complex belong to well-differentiated clades within alphabaculoviruses. Furthermore, both AcMNPV and SeMNPV can infect Spodoptera exigua larvae (one of the models to be evaluated with the P74 variants), whereas HearSNPV cannot.

Protein Domains
A P74 similarity study considering baculoviruses from the four genera revealed the occurrence of few conserved motifs, where it was possible to highlight three at the amino terminal region (Nt). Each of these three domains contains two cysteines (probably necessary for folding by putative disulfide bonds), along with one at the carboxyl-terminal region (Ct) containing a putative NES which could also contribute to the protein localization on the virion surface (Figure 2A). In this regard, three transmembrane (TM) regions can also be predicted (NES is contained in TM-1) at the Ct, which would allow the polypeptide to be structured in the ODV envelope exposing the Nt outside of the virion ( Figure 2B-D) as previously reported [38]. This organization could suggest that the P74 protein has a sector with a structural role in the Ct (containing TMs) and another exposing a functional characteristic in the Nt associated with the specific recognition for the host midgut cells. In fact, it is in the Nt where a protease-mediated activation process has been reported [43,44]. Interestingly, TM-2 and TM-3 differ from TM-1 in that they are in a region of minimal similarity; their physicochemical characteristics seem to be more important than the residues found there. By contrast, TM-1 is in one of the most conserved regions for P74 in Baculoviridae. Immediately downstream of the NES (and at the end of the TM-1) appears a conserved amino-acid triad consisting of tryptophan, aspartic acid, and proline (WDP). In this sense, the corresponding nucleotide sequence is "TGG GAT CCN", enabling the occurrence of a BamHI recognition site (-G GAT CC-). This is an opportunity to separate the ORF into two sectors (Nt of 458 vs. Ct of 187 amino acids in AcMNPV) for any baculoviral p74 (without affecting the amino-acid sequence when in-phase chimerization is performed), thus facilitating a better analysis for this important factor.

CRISPR/Cas9 Application
The AcMNPV bacmid was edited by CRISPR/Cas9 technology, generating a knockout in the p74 gene by NHEJ. To this end, the Cas9/RNP complex was guided to the initial part of the p74 ORF (207 bp downstream of ATG codon), and three mutant candidates were selected among 11 viral clones. Two of these affected viruses showed a deletion of 3 bp (nucleotides 205-207 and 203-205), and a third one presented a 4 bp deletion (nucleotides 207-210) in the p74 ORF. Because the latter evidenced a frameshift that produced an early stop codon (Figure 3), it was the candidate of choice to continue the studies. After verifying the absence of off-target mutations by NGS whole-genome sequencing, this mutagenized bacmid was then used to reconstitute the Bac-to-Bac system to carry out complementation assays with p74 ORF variants. In this regard, two donor vectors were used to generate two types of viruses over the p74 knockout backbone: one type expressing P74 variants and GFP; a second type expressing polyhedrin and mCherry. Thus, when cells were simultaneously infected by these recombinant bacmids, orally infective particles could be generated ( Figure S1).

AcMNPV OBs with P74 Variants
Regarding the strict host specificity of P74, the corresponding ORFs from SeMNPV and HearSNPV (both species belonging to the group II Alphabaculovirus genus; Figure 1) were selected along with the Nt and Ct protein regions considering its separation at the natural BamHI site located in WDP, to generate p74 chimeric alleles for their evaluation in the AcMNPV context (group I Alphabaculovirus genus; Figure 1). For the p74 ORF from AcMNPV, it was first necessary to introduce a same-sense mutation to add a BamHI recognition site in the nucleotide sequence that translates into WDP. As shown in the study of relative similarity among P74s of all baculoviral genera (Figure 2A), the protein of AcMNPV has a greater similarity in its Nt domain than in its Ct domain with respect to its orthologs, although both have relatively low identity percentages (Table 2). Another striking fact is the difference in isoelectric point (PI) between the two sectors, with Ct being predominantly acidic (Table S2). Values are expressed as percentages using whole numbers ("similarity" is shaded in light gray and "identity" is indicated by an uncolored background). "Complete" refers to the full protein, while "Nt" and "Ct" are respectively the amino and carboxyl terminal domains separated in the amino-acid triad WDP (containing at the nucleotide level a BamHI recognition site).
Different OB stocks containing variants of P74 (AcMNPV-p74AcAc, AcMNPV-p74SeSe, AcMNPV-p74HaHa, AcMNPV-p74AcSe, AcMNPV-p74AcHa, AcMNPV-p74SeAc, and AcMNPV-p74Ac) and controls (AcMNPV-∆p74 and AcMNPV-p74wt) were generated and produced in Sf21 cells using the knockout bacmid expressing polyhedrin and mCherry to co-occlude the viruses carrying the GFP-P74 versions ( Figure S1 and Figure 4). The green and red fluorescent cells, perceived as yellow and orange cells in the merged image ( Figure 4D), show that the coinfection occurred, and that complemented OBs could be generated. The presence of the P74 recombinant proteins in the OBs was then confirmed by Western blot using the 6×His-tag added at the Nt ( Figure 5).  Figure 5), suggesting that it was probably found in a lower proportion, or that the tag was removed by some proteolytic activity promoted, perhaps, by the absence of the terminal end. This unexpected result was not due to changes in the genetic sequence (this was verified by sequencing); thus, this batch of OBs was also considered in the subsequent assays.

Infectivity of AcMNPV Variants in S. exigua
The infectivity of all recombinant AcMNPVs was primarily validated by BV injections in S. exigua larvae (insect susceptible to oral and intrahemocelic AcMNPV infection) confirming that, for all cases, the secondary infection was not affected (Figure 6A,B). In addition, to control the correct performance of the model under study, OBs of AcMNPV-p74AcAc were employed to per os infect the same invertebrate species (Figure 6C,D). These studies showed that cell-to-cell infection was not affected, and that the 6×Histag (and the introduction of the single mutation to add the BamHI site in p74 ORF from AcMNPV) did not prevent the success of the primary infection.
Comparative bioassays were then carried out to study the influence of the different P74 domains and variants on the viral specificity and pathogenicity (Figure 7). S. exigua larvae showed not to be susceptible to the per os infection with AcMNPV-∆p74, as expected since it does not express any P74, reconfirming its PIF nature as previously reported [18,45]. Thus, the remaining polypeptide of 71 amino acids resulting from NHEJ mutation in p74 ORF (Figure 3) did not generate an infective phenotype, confirming the gene pseudogenization. In addition, complementation with the P74 protein from SeMNPV (AcMNPV-p74SeSe) and HearSNPV (AcMNPV-p74HaHa) did not produce infectious OBs, regardless of whether this replacement was made from viruses to which S. exigua is naturally susceptible (SeMNPV) or not (HearSNPV). In the same direction, AcMNPV-p74SeAc also failed to infect the insects, showing that the Ct domain would not contain the host-restricted functions. However, AcMNPV-p74AcAc, AcMNPV-p74AcSe, AcMNPV-p74AcHa, and AcMNPV-p74Ac were infectious (as well as the positive control, AcMNPV-p74wt). Interestingly, not only were AcMNPV-p74wt and AcMNPV-p74AcAc equally infectious and lethal despite expressing P74 under different promoters (p74 vs. polyhedrin gene promoters, respectively), but also all of those that contained the P74-Nt domain from AcMNPV performed similarly (whether the motif was alone or combined with the Ct region of the other tested species), with no significant differences observed in mortality among them. Although it was not possible to demonstrate the presence of the Nt domain from the AcMNPV-p74Ac OBs by recognizing the fused 6×His-tag by Western blot, this assay showed that it was present to rescue the virus infectivity. In all cases, the viral progeny produced in the larvae where the infection progressed were analyzed to certify the stability of the mutation in p74 gene.

Discussion
Previous studies failed to achieve infective baculoviruses when PIFs were replaced by orthologous sequences [36,68], except for some PIF3 replacements in Helicoverpa armigera nucleopolyhedrovirus (HearNPV) by the homologs from Spodoptera litura nucleopolyhedrovirus (SpltNPV) and Mamestra brassicae multiple nucleopolyhedrovirus (MbMNPV), which retained only partial oral infectivity [47,69], providing evidence of the stringency on the recognition process of the host in the primary infection. This may be due to the stability of the PIF complex in structural terms and/or due to the need for specific interactions with host factors, including cell receptors and activating enzymes such as proteases. This is the first study that analyze the expression of P74 chimeras (combining Nt and Ct domains from different species) during an infection cycle to try to find which region (if not the entire protein) is the one that carries such biological restrictions.
The non-expression of any PIF protein separately nullifies the infectivity of the ODVs in susceptible larvae. This would indicate that there are important structural and functional constraints where each intervening factor is required in an essential way. Three putative transmembrane domains were predicted in P74 protein consistent with previous reports [17,42], although other studies only identified the two closest (TM-2 and TM-3) to the carboxyl terminal end [36,38]. It is not clear whether TM-1 is a real transmembrane region, because truncated versions of P74 (only deleting TM-2 and TM-3 and replacing it with GFP) were soluble and apparently not anchored to the ODV membrane [45]. After being synthesized in the cytoplasm, PIFs must be imported into the nucleus for proper ODV assembly. Although most of the ODV envelope proteins contain an INM-SM sequence which tags them for nuclear import [38,40,70], no similar motif was reported in P74. Nonetheless, in this study, the presence of a NES-like sequence was predicted at the end of TM-1, one of the most conserved regions for P74 in Baculoviridae. These kinds of motifs are rich in hydrophobic residues (e.g., leucine), start with an alpha helix, and are usually involved in the transportation of proteins from the nucleus to the cytoplasm, binding with the export karyopherin known as chromosomal region maintenance 1 (CRM1 or exportin 1) [71]. Given the late appearance of the P74 protein during the infection cycle [18], it could be speculated that this region has the binding role needed for the conformation of the PIF complex in association with the hypertrophied nucleus envelope of the infected cell. This could explain why truncated P74 versions lacking TM-1, TM-2, and TM-3 did not rescue oral infectivity, as demonstrated in previous work [45]. Our results showed that a truncated variant of P74 containing TM-1 was sufficient to rescue the infectivity of a P74 defective AcMNPV. This complementation was successful in the absence of TM-2 and TM-3, as well as when that region was replaced by orthologs from viral species of the same genus but different group, as SeMNPV and HearSNPV are. Remarkably, the bioassays also showed no significant differences in the infectivity or pathogenicity of these AcMNPV-P74 Ct domain replacement or missing variants, when compared to each other or even with AcMNPV-p74wt. This demonstrates that the acidic Ct region of 187 amino acids containing the most important TM regions (TM-2 and TM-3) seems to play a nonessential structural role and can be replaced by orthologous segments from other species (that naturally infect the same host or not) without compromising the P74 activity and the OB virulence.
The maintenance of infectivity does not occur when the Nt is replaced (regardless of whether it comes from a viral species infectious for the host), showing that this domain contains the functional restrictions. It is also interesting to note that the Nt supports peptide fusion without affecting its role, in accordance with previous evidence showing that this region undergoes a proteolysis process for its activation [43,44]. These results together indicate that the region of the first 458 amino acids of P74 protein of AcMNPV is sufficient and essential to consolidate a functional PIF complex in the ODV envelopes and is responsible for the host-related restrictions.
It is important to highlight some methodological aspects associated with functional genomics studies to be considered in this context. The introduction of knockout mutations is an indispensable tool for understanding gene functions. In baculoviruses, this type of research is traditionally performed by replacing viral sequences with marker or reporter genes [46,72]. Although this kind of strategy has made it possible to generate substantial knowledge so far, it carries associated risks, such as the introduction of cryptic promoters or the elimination of sequences with other unknown important roles (e.g., miR genes, origins of replication, expression-modulators of neighboring genes, and psi packaging-like elements) that can impact the viral cycle and the virions that are produced (both morphologically and in their infectious capacities). Many of these functional sequences are usually contained within genes; thus, a knockout by replacement or deletion can simultaneously affect several viral functions [73][74][75]. Therefore, the application of CRISPR/Cas9 technology and NHEJ in baculovirus for these purposes appears as a better option due to its lower probability of unwanted side-effects and should now be the preferred tool for this type of genomic study. In this sense, continuing with the work where the use of gene editing mediated by CRISPR/Cas9 is reported for optimizing the use of baculovirus as a protein expression vector [50], the present study demonstrates its usefulness for baculovirus functional genomics research and proposes a methodological detail for its implementation henceforth.

Conclusions
The P74 protein is an essential factor to support primary baculovirus infection in a susceptible host, without affecting secondary infection. In this work, it was shown that the elimination of the transmembrane region of the terminal carboxyl end of P74 (an acidic part of low amino-acid conservation but structural preservation) does not compromise its function, and that this region can even be replaced by orthologs of species that infect or do not infect the same host. However, considering our results and those of the literature, it seems that a central hydrophobic region (which can be hypothesized as TM and/or as NES) is essential for the activity and presumably necessary to associate this protein with the rest of the PIF complex and ODV envelope. This "hydrophobic core region" requires further research to elucidate its role because it is strikingly conserved in all baculoviral species. Furthermore, this study opens the door for future genetic editing mediated by CRISPR/Cas9 to improve/change the efficiency of the primary infection by replacing regions of the p74 gene in wild-type baculovirus, thus increasing the biopesticide power in strategies of biologic control for crop pests.