RNA interference technology to improve the baculovirus-insect cell expression system

Biotechnology Advances xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Biotechnology Advances journal homepage: www.elsevier.com/locate/biotechadv

Research review paper

RNA interference technology to improve the baculovirus-insect cell expression system Cuitlahuac Chavez-Pena, Amine A. Kamen

⁎

Viral Vectors and Vaccines Bioprocessing Group, Department of Bioengineering, McGill University, 817 Sherbrooke Street West, Room 387, Montreal, Quebec H3A 0C3, Canada

A R T I C L E I N F O

A B S T R A C T

Keywords: Baculovirus expression system Recombinant protein RNA interference siRNA Vaccines manufacturing

The baculovirus expression vector system (BEVS) is a popular manufacturing platform for the production of recombinant proteins, antiviral vaccines, gene therapy vectors, and biopesticides. Besides its successful applications in the industrial sector, the system has also played a significant role within the academic community given its extensive use in the production of hard-to-express eukaryotic multiprotein complexes for structural characterization for example. However, as other expression platforms, BEVS has to be continually improved to overcome its limitation and adapt to the constant demand for manufacturing processes that provide recombinant products with improved quality at higher yields and lower production cost. RNA interference, or RNAi, is a relatively recent technology that has revolutionized how scientist study gene function. Originally introduced as a tool to study biological and disease-related processes it has recently been applied to improve the yield and quality of recombinant proteins produced in several expression systems. In this review, we provide a comprehensive summary of the impact that RNAi-mediated silencing of cellular or viral genes in the BEVS has on the production of recombinant products. We also propose a critical analysis of several aspects of the methodologies described in the literature for the use of RNAi technology in the BEVS with the intent to provide the reader with eventually useful guidance for designing experiments.

1. Introduction BEVS is a powerful tool for the production of recombinant proteins widely used in several areas of research and industry (Palomares et al., 2005). It is particularly favored over others platforms for the manufacturing of vaccines (Mena and Kamen, 2011), gene therapy vectors (Carinhas et al., 2009; Cecchini et al., 2008) and biopesticides (Reid et al., 2013). Like other platforms for recombinant protein expression, BEVS has been improved through the years thanks to the efforts of many research groups. From a bioprocess standpoint, most of those efforts have been directed to increase the production capabilities of the system. Research focusing on this goal is still ongoing, and revolves around several aspects including, mode of operation of the cultures (Elias et al., 2000; Petricevich et al., 2001; Wu et al., 1989), development of cell culture media (Chan and Reid, 2016) and understanding the biology and engineering host cells and viruses (Hitchman et al., 2010a, 2010b; van Oers, 2011). Prior to the establishment of BEVS as an expression system, the molecular biology of both insect cells and baculovirus was an active

area of research, which has been crucial for the inception and evolution of the system. Noteworthy, while working on baculovirus deletion mutants, Gale Smith and Max D. Summers came with the idea of replacing the coding sequence of the non-essential viral gene polyhedrin with the nucleotide sequence for a protein of interest, a concept that constitutes the core of the BEVS technology (Smith et al., 1983). Thenceforth, significant work has been focused, on the one hand, in modifying the backbone of the baculovirus to optimize recombinant protein production (Hitchman et al., 2010a, 2010b), achieve expression of multiple recombinant proteins from a single virus (Bieniossek et al., 2012) or enhance the stability of the vector. On the other hand, several groups have genetically modified insect host cells to improve the quality of the recombinant proteins by, for example, modifying the expression of cellular genes involved in the posttranslational processing of proteins (Harrison and Jarvis, 2006) or apoptotic responses (Lin et al., 2007). Attempts to change the expression of cellular or viral genes in the BEVS are routinely accomplished by inserting or deleting gene coding sequences or modifying their regulatory units and evaluating the

Abbreviations: BEVS, Baculovirus expression vector system; CHO cells, Chinese hamster ovary cells; HEK, Human embryonic kidney cells; MIARE, Minimum Information About an RNAi Experiment; miRNA, Micro RNA; RNAi, RNA interference; shRNA, Short hairpin RNA; siRNA, Small interfering RNA ⁎ Corresponding author. E-mail address: [email protected] (A.A. Kamen). https://doi.org/10.1016/j.biotechadv.2018.01.008 Received 18 September 2017; Received in revised form 11 December 2017; Accepted 13 January 2018 0734-9750/ © 2018 Published by Elsevier Inc.

Please cite this article as: Chavez-Pena, C., Biotechnology Advances (2018), https://doi.org/10.1016/j.biotechadv.2018.01.008

Biotechnology Advances xxx (xxxx) xxx–xxx

C. Chavez-Pena, A.A. Kamen

of the Argonaute superfamily, confers RISC with its catalytic activity. Loading of siRNA or miRNA into RISC requires the participation of several cofactors which varies from species to species and depends on the type of molecule being loaded. In the canonical pathway, protein R2D2 is responsible for loading siRNAs while another protein, loquacious, does it for miRNAs. In both cases, these RNA-binding proteins sense thermodynamic asymmetries in the strands of the siRNA/miRNA and bind to the strand with the more stable 5′-end, thus defining which strand will be loaded into Ago. This strand, called the guide strand, remains bound to Ago, while the other is degraded (Schwarz et al., 2003). The activated RISC targets mRNAs with a complementary sequence to the one in the guide strand. Typically, siRNA-directed RNAi results in degradation of the mRNA, while miRNA-directed gene silencing is established by translational inhibition (Carthew and Sontheimer, 2009). Although the core components of RNA-mediated gene silencing are well conserved among eukaryotes, there are some species-specific deviations from the canonical pathway. Many are related to the number of paralogs of ago and dicer genes in each organism and also the extent to which these paralogs specialize in using siRNA or miRNA to guide the silencing. In other cases, the importance of particular RISC-related proteins seems not to be the same across different species. For example, Drosophila r2d2 null mutants are unable of mounting a strong RNAi interference response, while similar mutants in mouse derived cell lines can effectively silence the expression of a target upon dsRNA treatment. RNAi mechanisms in insect species, and their derived cell lines, used in the baculovirus expression system have its particularities too; knowing and understanding them is essential for the efficient use and further improvement of RNAi in the BEVS.

impact that such gene has on the baculovirus infection process or production of recombinant proteins. Although reliable, this approach can be laborious, expensive and unsuitable for high throughput screenings (Novina and Sharp, 2004). Fortunately, during the last few years, new tools for the study of genes functions have emerged which represent valuable alternatives to the more traditional approach previously described. Of particular interest is RNA interference(RNAi), a tool derived from the 1998 report by Fire and Mello where they described that injection of exogenous double-stranded RNA (dsRNA) into Caenorhabditis elegans led to the silencing of the endogenous homolog mRNA (Fire et al., 1998). The discovery of RNAi did not only changed the paradigm of the biological role of RNA but soon proved to be a powerful tool in several research areas including developmental biology, medicine, and biotechnology. RNAi pathway is well conserved among eukaryotes; hence, it has been possible to expand its use to several eukaryotic-based expression systems, such as Chinese Hamster Ovary (CHO cells) or Human Embryonic Kidney(HEK) cell lines (Hebert et al., 2009a, 2009b). In those, it has been shown that production of recombinant proteins can be enhanced by introducing or expressing dsRNA with a homologous sequence to a specific gene, for example, transcriptional regulators or apoptosis mediators. Gene silencing through RNA interference has been extensively used in lepidopteran cell lines and baculovirus for different purposes. For example, to explore and better understand the different processes involved in the viral infection (Terenius et al., 2011); to protect insects of commercial importance from viral infections (Valdes et al., 2003); to construct baculovirus-based vectors for delivery of dsRNA in mammalian cells (Suzuki et al., 2008) and as a metabolic engineering tool to increase yields in BEVS. In this review, we focus on those reports were RNA interference was used as a mean to improve the quality or quantity of recombinant proteins produced in the BEVS.

2.2. Particular aspects of RNAi pathway in cell lines used in the BEVS Cell lines employed in BEVS originate from Lepidopterans, an order of arthropods in which systemic RNAi in-vivo has been difficult to achieve consistently (Terenius et al., 2011), although this is not the case for reports of RNA-mediated silencing performed in vitro in cell cultures. This difficulty has given rise to numerous studies aiming to identify limiting factors of the RNAi pathway in Lepidopterans (Swevers et al., 2011; Terenius et al., 2011). For in-vitro cell cultures, the limiting factor in RNAi experiments remains the delivery of dsRNA into the cell cytoplasm, which means uptake of exogenous RNA from the environment, i.e. the culture media (Yu et al., 2013). The mechanism of cellular uptake of dsRNA has been thoroughly studied in C. elegans. The cells of this nematode express a protein called SID-1 which is capable of transmembrane channelmediated uptake of dsRNA into cells (Wynant et al., 2014). Because of SID-1, dsRNA can enter into C. elegans cells without the need of transfection or microinjection; a technique called “RNAi by soaking.” sid-1 homologous sequences have been identified in several lepidopterans and other insects. However, it is unclear if their expression products are in full involved in the transport of dsRNA since silencing of sid1-like sequences in lepidopterans like T. castaneum had no adverse effect on the RNAi response (Tomoyasu et al., 2008). An alternative mechanism for the uptake of dsRNA in lepidopterans could be one similar to that found in D. melanogaster. Drosophila S2 cells lack sid-1-like sequences but can up-take dsRNA from the media by a receptor-mediated endocytosis process, in which a scavenger receptor plays an essential role. Although the actual mechanism of dsRNA uptake in insect cells is not well understood yet, several research groups are attempting to make this step more efficient. In two different reports, the Spodoptera frugiperda derived cell line Sf9 (Xu et al., 2013b), and the Bombyx mori derived line cell Bme21 (Kobayashi et al., 2012) were transformed to express C. elegans SID-1 ectopically. Transformed stable cell lines were capable of uptaking “naked” dsRNA from the media without the use of transfection reagents, and silencing efficiencies of exogenous and endogenous genes increased to 90%. However, some of their clones

2. The RNAi pathway 2.1. General considerations Since the discovery of RNAi, a lot of efforts have been dedicated to understand its biological role and to identify its main components. Enabled by the purification of key enzymes, the development of RNA interference cell-free systems and the mapping of genes associated with RNAi pathway, a broad understanding of RNAi mechanism, function, and limitations has been achieved. This section will briefly cover the general pathway, and basic component associated with the RNA mediated silencing; whereas several excellent reviews are available for a more detailed understanding of the RNA-mediated silencing (Carthew and Sontheimer, 2009; Ghildiyal and Zamore, 2009; Liu and Paroo, 2010). The canonical RNA interference pathway is shown in Fig. 1. RNAi is triggered in response to exogenously introduced or endogenously transcribed dsRNA; in the first case, dsRNA can be microinjected or transfected into cells, while in the second, dsRNA is synthesized by the cell using as templates genomic sequences from viruses or plasmids, in the form of inverted repeats or self-annealing transcripts. Once dsRNA is in the cytoplasm, Dicer - a type III RNA endonuclease- cleaves it to produce short 21–25 nt long fragments, paired in a way that leaves two nucleotide overhangs at the 3′ ends of both strands (Ghildiyal and Zamore, 2009). Dicer family of enzymes can produce two kinds of cleavage products: micro (mi)RNAs and small interfering (si)RNAs. In some species, a single Dicer can produce both types of short RNAs, while in others the task is divided between two different enzymes: Dicer-1 for miRNAs and Dicer-2 for siRNAs. These short RNA fragments are loaded into the RNA-mediated silencing complex- or RISC- and are the active, sequence-specific mediators of RNAi (Elbashir et al., 2001). RISC is a macromolecular complex with endonuclease activity responsible for degrading the targeted transcript; Ago, a protein member 2

Biotechnology Advances xxx (xxxx) xxx–xxx

C. Chavez-Pena, A.A. Kamen

Fig. 1. A simplified model for the canonical RNAi pathway.1)RNAi is initiated by either the introduction of dsRNA or siRNA into the cytoplasm or (1′) the expression of shRNA from plasmids or nuclear DNA. 2)Large dsRNA or shRNA are processed by dicer into small double stranded RNA fragments with 5′-end overhangs. 3) Accessory proteins like R2D2 help to load the guide strand of the siRNAs into Ago processing enzyme which leads to the formation of the RNA silencing complex RISC. 4)The guide strand in RISC recognizes the target mRNA. 5) RISC degrades the mRNA which prevents translation.

the parental cell line or simply to boost its expression to enhance the RNA-mediated silencing. We already mentioned the work performed in the Laboratory of Silkworm Science (Kyushu University, Japan) in which heterologous expression of C. elegans sid-1 has enabled transformed cell lines to uptake dsRNA from the media without the use of transfection reagents (Kobayashi et al., 2012; Xu et al., 2013a, 2013b). Another example is the overexpression of Bm-ago2 in B. mori larvae and derived cell lines that significantly facilitates the dsRNA-mediated silencing when targeting an exogenous reporter gene (DsRed) as well as an endogenous one (Bmblos-2) (Li et al., 2015). Also, in an attempt to increase a poor Dicer-like activity previously reported in insect cell line extracts (Flores-Jasso et al., 2004), Lee and co-workers transformed Bm4 cells with a plasmid containing the E. coli RNase III gene which codes for an enzyme that belongs to the Dicer-like family of nucleases. Upon dsRNA treatment, their modified cell line silenced the expression of the reporter gene in 80% compared with the 60% observed in the parent cell line (Lee et al., 2013). So far, some of these transformed cell lines have become excellent platforms to expand the understanding of lepidopteran cell biology as well as the baculovirus infection process by using RNAi technology. Most certainly, these cell lines can also represent an excellent tool to study the effect of gene silencing on the production of recombinant products in BEVS.

showed low sensitivity to RNAi by soaking, which the authors explain as a consequence of the low levels of expression of SID-1 (Xu et al., 2013a). In insect cell lines, experiments seeking to validate the expression of the RNAi associated factors, have shed light about differences with the canonical RNAi pathway. These experiments have helped to understand inconsistencies observed when silencing genes in Lepidopterans and promote the development of transformed cell lines with enhanced RNAi activity. Genome-wide expression analysis in S. frugiperda (Ghosh et al., 2014) and B. mori (Swevers et al., 2011) derived cell lines, and S. exigua larvae have confirmed that most RNAi core factors (R2D2, Loquaceous, Ago1, Ago2, Dicer1, and Dicer2) are encoded in Lepidopterans. However, the number of paralogs seems to differ from species to species, and expression levels vary depending on the tissue and the development stage. For example, in Bm5 cells, a B. mori derived cell line, four different Ago genes have been identified (Wang et al., 2013), while in S. litura and S. frugiperda the number of identified ago genes is three and two respectively (Ghosh et al., 2014; Gong et al., 2015). Both B. mori and S. litura have orthologs of ago-1 and ago-2 genes found in D. melanogaster while ago-2 is absent in S. frugiperda. This finding is interesting since cell lines derived from S. frugiperda can effectively conduct silencing when transfected with siRNA thereby suggesting a crosstalk between the miRNA and siRNA pathways in these cells. A similar situation is observed with RNA binding protein R2D2, which in the canonical pathway is necessary for loading siRNA into RISC. In this case, the r2d2 gene is present in both S. litura and S. frugiperda while, surprisingly, Bm5 cell line is devoid of R2D2 expression (Swevers et al., 2011); however, despite the absence of R2D2, Bm5 cells were capable of specific gene silencing when dsRNA was transfected into the cells. One possible explanation for the efficient RNA-mediated silencing in Bm5 cell line despite the lack of R2D2 is that another dsRNA-binding protein, for example, Loquacious which is ubiquitously expressed, can compensate for the absence of BmR2D2 as has been observed in Drosophila. These findings led several groups to develop cell lines (Table 1) in which RNAi-related genes were introduced to compensate their lack in

3. RNAi-mediated silencing for improving recombinant protein production in BEVS Genetic engineering of cell lines has played a fundamental role in the optimization of platforms used for the production of recombinant proteins. There are numerous examples (comprehensively reviewed in (Xiao et al., 2014) and (Fischer et al., 2015)) of cases in which either introduction and overexpression of beneficial genes or deletion of disadvantageous genes, caused a significant improvement in cell performance during manufacturing processes. Thanks to the coming of age of RNAi technology and availability of whole genomic and transcriptomic sequence for cell lines used in the production of recombinant proteins 3

Biotechnology Advances xxx (xxxx) xxx–xxx

C. Chavez-Pena, A.A. Kamen

Table 1 List of Lepidoptera cell lines and larvae strains with genetic modification on the RNAi machinery. Transformed organism/cell line

Modified Gene

Promoter

Reported effect

Ref

B. mori Larvae Bme21 Sf9 BmN4 Bm5

B. mori Ago-2

OpiE-2

C. elegans SID-1

OpiE-2

40% increase in silencing efficiency Modified cells can efficiently uptake dsRNA from media without the addition of transfection reagents

Tribolium R2D2

Ie1

Li et al. (2015) Xu et al. (2013a) (Xu et al. (2013b) Kobayashi et al. (2012) Swevers et al. (2011)

BmN4

E. coli RNase III

Actin A3

No differences in silencing were observed compared with parental cell line that does not express R2D2 Increase of silencing efficiency up to 90%

(like CHO, HEK293, High Five®), it is possible to explore the effect of transient knockdown of selected genes on the productivity of these expression platforms. In contrast to gene knock-out approaches, RNA-mediated silencing can be applied to genes which normal expression is necessary for the regular functioning and propagation of cells and can be done at a specific time point during manufacturing processes. The latter possibility is particularly attractive in the case of BEVS since replications of baculovirus and production of recombinant proteins are both strongly dependent on the stringent temporal expression pattern observed for baculoviral genes. Given the nature of BEVS, we consider appropriate to discuss separatly the research done on silencing cellular or viral genes and its effect on the performance of the system during the production of recombinant products. A summary of these studies is provided in Table 2.

Lee et al. (2013)

2013; Yu et al., 2008). These proteins can block the activation of caspase-1, the principal and most studied effector caspase of Lepidoptera (Lin et al., 2007) and thus inhibit the onset of the apoptotic response during the early phases of the infection process. However, Lin et al. (2007), and Hebert et al. (2009a, 2009b) found that infected cells 48 to 72 h post infection exhibit signs of apoptosis, including DNA fragmentation, membrane blebbing, and increased caspase activity, suggesting that expression levels of baculoviral anti-apoptotic proteins could no longer prevent caspase activation. Following previous reports of antiapoptosis engineering in CHO cells, Kim and coworkers explored the use of RNA against caspase-1 in Sf9 cells. They found that after baculovirus infection or stimulation with Actinomycin D (a known apoptosis inducer), cells transfected with anti-caspase-1-dsRNA exhibited a significant decrease in caspase activity and substantially less DNA fragmentation than non-treated cells (Kim et al., 2007). Several groups have gone one step further and taken advantage of the simplicity and robustness of in-vivo expression of dsRNA to establish High Five (Hebert et al., 2009a, 2009b), BmN (Wang et al., 2016) and Sf9 (Lai et al., 2012) derived cell lines capable of stably down-regulating expression of caspase-1 through RNAi. These transformed cell lines are more resistant to apoptosis upon virus infection at different multiplicities of infection (MOI) and have demonstrated superior viability than the parental cell lines, prolonging the productive life of the culture. For example, two days post-infection viability of control and transformed Sf9 cells infected at MOI = 5 were 55% and 78% respectively (Lin et al., 2007); this demonstrated that silencing of caspase-1 is an effective way for maintaining viability after infection. RNAi-mediated silencing of caspase-1 also had a positive impact on the production of recombinant proteins upon baculovirus infection. When transformed cell lines were infected with recombinant baculovirus expressing GFP (Hebert et al., 2009a, 2009b), SEAP (Lai et al., 2012) and the fusion protein Tim4-Fc (Wang et al., 2016), expression levels were 100%, 100%, and 400% respectively higher than their parent cell lines.

3.1. Silencing insect cell genes on the production of recombinant proteins Increased understanding of insect cell biology has enabled several groups to design engineering strategies targeting different bottlenecks known to affect the production of recombinant proteins. Examples include cell growth capacity, apoptosis resistance, glycosylation and folding of proteins and the secretory pathway (Xiao et al., 2014). Within works reporting the use of RNAi to engineer insect cells, silencing of apoptosis-related genes is probably the best-studied case. Apoptosis is a necessary physiological function which helps eliminate unhealthy cells. However, its onset during bioreactor operation is undesirable since it lowers the fraction of viable cells and affects yield and quality of recombinant products (Jadhav et al., 2013). In insect cells, apoptosis can be triggered by nutrient limitation, shear stress, oxidative stress or baculovirus infection itself, the latter despite the fact that baculoviruses carry genes coding for antiapoptotic proteins like P35, P49 and IAP's (inhibitor of apopotosis) family of proteins (Rohrmann, Table 2 Targeted viral and cellular genes for engineering of the BEVS using RNAi technology. Target gene

Function

Type of silencing agent

Reported effect

Ref

caspase-1

Pro-apoptotic

dsRNA/shRNA

Cells remain viable for a longer period of time after infection

β-N-Acetylglucosaminidase

Glycosylation pathway

shRNA

Increase in the fraction of complex-type glycans in glycosylated proteins

Cyclin E

Cell cycle regulator

dsRNA

Calnexin Calreticulin Protein disulfide isomerase v-cath gp64

Protein folding

dsRNA

Arrest of cell in G1 acompanied by an increased yield of recombinant protein Depletion of these chaperones did not have a negative impact on the production of secreted recombinant proteins

Hebert et al. (2009a, 2009b) Wang et al. (2016) Lai et al. (2012) Nagata et al. (2013) Nomura et al. (2015) Wu et al. (2013)

Proteolysis of host-cells Viral fusion protein

siRNA siRNA/dsRNA

Reduction in the proteolysis of recombinant protein Reduction in the production of residual baculovirus

vp80

Structural protein

dsRNA

Reduction in the production of residual baculovirus

orf34

Unknonkn

siRNA

Enhanced heterologus gene expression

4

Imai et al. (2015)

Kim et al. (2007) Lee et al. (2015) Quadt et al. (2007) Schultz and Friesen (2009) Salem et al. (2013)

Biotechnology Advances xxx (xxxx) xxx–xxx

C. Chavez-Pena, A.A. Kamen

a more abundant protein translational machinery. On the second case, Imae and coworkers overexpressed and silenced five endoplasmic reticulum chaperones (BiP, calnexin, calreticulin, ERp57, and protein disulfide isomerase) in Bme21 cells which were subsequently infected with baculovirus expressing different mammalian glycoproteins. Unexpectedly, they found that neither depletion or overproduction of ER chaperones had a significant impact on the secretion or cellular distribution of the recombinant proteins (Imai et al., 2015). This finding contradicts the traditional idea that co-expression of chaperones can improve the folding of recombinant proteins. However, as pointed out by one author, “it is possible that overexpression of chaperones is not needed for the majority of the proteins, but it may be helpful for proteins with complex structures, such as those with multiple transmembrane domains” (van Oers et al., 2014).

Interestingly the effect was not as large when cells were infected with baculovirus expressing intracellular proteins like CAT or Luciferase since only 30% increase in protein yield was observed. Nevertheless, taken together, these results indicate that suppression of caspase-1 activity by RNAi provide an efficient method of increasing recombinant protein production in the BEVS. Engineering of the glycosylation pathway in insect cell lines represents another active field of research in which RNAi technology has been applied. The glycosylation pattern of proteins produced with BEVS differs from that observed in proteins produced in mammals; such difference can have an impact on the efficacy of therapeutic glycoproteins (Geisler and Jarvis, 2009). The main difference between glycoproteins produced in mammalian expression system and BEVS is the abundant presence of complex N-glycans (multiply branched, terminally sialylated) in the former, whereas in the latter, most N-glycans are simpler paucimannose structures (Palomares et al., 2005). In principle, insect cells have the capacity to produce complex N-glycoforms. However, the lack or poor expression of certain glycosyltransferases and the high enzymatic activity of glycosidase favor the formation of paucimannose structures. One strategy to overcome this difference is the development of stably transformed insect cell lines overexpressing genes encoding enzymes needed for the synthesis of complex mammalian-like glycosylation (Harrison and Jarvis, 2006). However, this approach faces a major challenge, since infection with recombinant baculovirus induces shut-off of host gene expression thus limiting the beneficial effects of the modifications inserted into the host genome (van Oers, 2011). An alternative approach based on RNAi technology is the silencing of fused lobes(fdl), an insect cell gene encoding the enzyme β-N-Acetylglucosaminidase (Léonard et al., 2006). In insect cells, the high specific activity of β-N-Acetylglucosaminidase removes the terminal N-Acetylglucosamine from the core oligosaccharide added to proteins during the first steps of the glycosylation pathway, thus preventing further elongation of the oligosaccharide branches to form complex-type N-linked glycans (Geisler and Jarvis, 2009). In 2008, Geisler demonstrated that transfection of Sf9 cells with plasmids encoding an inverted repeat sequence targeting Sf-fdl gene reduced the activity of β-N-acetylglucosaminidase (Geisler et al., 2008). This finding encouraged several groups to explore the effect of reduced β-N-acetylglucosaminidase activity on the recombinant proteins glycosylation in BEVS. Nagata et al. (2013) reported that treatment of the soaking RNAisensitive cell line BmN4-SID1 with anti-fdl dsRNA led to almost complete depletion of its transcript levels and specific glycosidase activity. Analysis of glycoproteins produced by baculovirus infection of the repressed cell line showed the almost complete conversion of the paucimannosidic structures into complex-type structures. Similarly, transgenic silkworm larvae in which a DNA fragment coding for an RNA hairpin targeting fdl was inserted showed a 50% decrease in glycosidase activity, and the amount of GlcNAc-type N-glycan in recombinant ALP (produced by baculovirus infection) increased significantly following the suppression of Bmfdl (Nomura et al., 2015). Thus far, RNAi suppression of β-N-acetylglucosaminidase seems a step in the right direction to achieve a more mammalian-like glycosylation in recombinant proteins produced in BEVS. However, this approach most likely will have to be complemented with the heterologous expression of other glycosyltransferases (Kim and Cha, 2015). RNAi-mediated gene silencing has also been used in BEVS to enhance recombinant protein production by cell cycle modulation and to investigate the role of chaperones in the expression of secreted proteins. In the first case, downregulation of cyclin E (a positive regulator of G1to-S phase transition) in High Five cells using dsRNA caused a dosedependent shift in the cell cycle distribution from G2/M to G1 (Wu et al., 2013). The authors demonstrated an almost two-fold increase in recombinant GFP expression when silencing was induced briefly before baculovirus infection. A similar effect was observed in other expression platforms like CHO (Bi et al., 2004) and Drosophila S2 (March and Bentley, 2007) cells, and it was attributed to G1 being characterized by

3.2. Silencing baculovirus genes on the production of recombinant proteins In 2003, the use of dsRNA to silence the expression of a baculovirusencoded gene was reported for the first time (Kramer and Bentley, 2003). Shortly after a second report describing the treatment of larvae with dsRNA against structural and non-structural baculoviral genes to confer resistance to infection was published (Valdes et al., 2003). The results from both studies (sharp reduction in gene expression and increased survival rate for treated larvae) positioned RNAi as an efficient tool to down-regulate baculoviral genes and opened the door to look for universal strategies to achieve RNAi-mediated silencing in the BEVS (Salem and Maruniak, 2007). Since then, RNAi-mediated silencing has been actively used i) to develop novel strains of silkworms with increased resistance against baculovirus infection (Subbaiah et al., 2013), ii) to further investigate the baculovirus-insect cell interactions during the infection process (Nguyen et al., 2013), and more recently iii) to investigate the impact that transient silencing of baculoviral genes has on the production of recombinant products. Posttranslational silencing of the baculovirus gene v-cath, which encodes for a cysteine protease, was investigated in an attempt to decrease the proteolysis of recombinant proteins produced in BEVS (Kim et al., 2007). Infected cells in which the gene was silenced exhibited higher viability, reduced proteolysis and up to three-fold increase in recombinant GFP than control cells. This finding confirmed an earlier discovery by Kaba et al. (2004) who suggested that v-cath was not essential for the propagation of baculovirus in cell cultures leading to the development of novel baculovirus vectors in which v-cath and the genes coding for chitinase, p26, p10 and p74, were removed (Hitchman et al., 2010a, 2010b). These multiple-deletion viruses have shown enhanced levels of protein production for both secreted and cytoplasmic proteins. Gene deletion is indeed a useful strategy for the generation of modified baculoviruses to improve recombinant protein production; however, it is limited to genes not essential for baculovirus propagation in cell culture. In contrast, post-translational silencing mediated by RNAi does not have this limitation. Taking advantage of this characteristic, some papers have reported silencing of essential genes during the production of recombinant proteins. An example of this approach is the downregulation of gp64, a highly expressed gene coding for AcMNPV fusion glycoprotein, a transmenbrane protein essential for cell entry and cell-to-cell viral transmission (Oomens et al., 1995), which forms part of the baculoviral envelope and is essential for the cell to cell transmission of infection (Monsma et al., 1996). A limited silencing of gp64 using siRNA (Lee et al., 2015) or dsRNA (Schultz and Friesen, 2009) has proven to reduce residual baculovirus contaminants and increase in 30% the yield of a recombinant protein. The reduction of residual baculovirus contaminants has the added benefit of simplifying the downstream purification process. Other groups have also investigated the possibility of reducing residual baculovirus contaminants generated during recombinant protein production by targeting the essential genes vp80 (Marek et al., 2011) and DNA binding protein (dbp) (Lee et al., 2015; Quadt et al., 2007). In both cases, titers of baculovirus 5

Biotechnology Advances xxx (xxxx) xxx–xxx

C. Chavez-Pena, A.A. Kamen

infectious particles were significantly reduced, although in the later, silencing of dbp also severely compromised the yields of recombinant protein. More recently, Salem et al. demonstrated that silencing of ORF34, a transcriptional unit without known function but essential for baculovirus spread, enhance heterologous gene expression (Salem et al., 2013). 4. Resources and guidelines for RNAi experiments in BEVS RNAi interference represents a valuable tool for metabolic engineering. Like all techniques, successful RNAi experiments require careful experimental design to enable accurate interpretation of the results. The RNAi Global Informatics Work Group1 have developed the MIARE (Minimum Information About an RNAi Experiment) reporting guidelines, a set of informal rules that describe the minimum information that should be reported for RNAi experiments to enable the unambiguous interpretation and reproduction of the results. In this section, we briefly discuss, based on the MIARE reporting guidelines, some of the technical details of RNAi experiments in BEVS reported in the literature. A proper description of the targeted gene is among the most critical information that should be provided when reporting silencing experiments. During the preparation of this manuscript, we found many instances in which the targeted gene was described by simply indicating the gene name only accompanied by a prefix corresponding to the species of origin. This approach can lead to confusions and difficulties when comparing results given the existence of discrepancies and changes in gene nomenclature. A better approach is proposed in the GenomeRNAi database guidelines (Schmidt et al., 2013), to include for each targeted gene also the GeneID identifier from the Entrez Gene database, thus providing an unambiguous identification system. For cases in which GeneID identifiers are not yet available, NCBI Reference Sequence (RefSeq) accession numbers can be used as identifiers. Due to the non-reductant nature of RefSeq collection, these also provide a clear identification system (Pruitt et al., 2002). If neither of those identifiers is available, for example when silencing a new putative gene, providing the targeted sequence is recommended. For an excellent example of how to describe a silencing experiment of a putative gene, we recommend the paper by Geisler et al. (2008). Similarly, RNAi reagents should be thoroughly described. Three types of reagents can be used: dsRNAs, chemically synthesized siRNAs, and short-hairpin RNAs (shRNAs) (Mohr et al., 2010). The selection of one over the other depends on several factors including cost, required duration of the effect, the susceptibility of the cell line to transfection, availability of validated libraries, etc. According to our review of 54 reports, dsRNAs are the most commonly used type of reagents for gene silencing in the BEVS (Fig. 2). This can be easily explained by the fact that they represent the most cost-effective alternative and can be easily generated from PCR products or a plasmids using T7 promoters and T7 polymerases (Ramadan et al., 2007). A significant disadvantage of dsRNAs is that their enzymatic processing by Dicer can generate nonspecific siRNAs which can cause off-target effects (Kulkarni et al., 2006). Off-target effects are the result of nonspecific knockdown of other genes caused by sequence homology and can lead to cell apoptosis or cytotoxicity (Lee et al., 2015). To overcome this limitation, chemically synthesized siRNAs can be used instead. These can be designed to minimize the potential number of off target sites following available design rules (Birmingham et al., 2007; Vert et al., 2006). Chemically synthesized siRNA are very popular for gene silencing in CHO cells given the availability of validated libraries. Although we have found that their use in insect cells accounts only for 13% of reported experiments (Fig. 2), the recent introduction of Spodoptera frugiperda in the Zoonome Custom siRNA Libraries (GE Healthcare 1

Fig. 2. Frequency of use of different types of RNAi reagent for gene silencing in insect cell lines used for the expression of recombinant proteins. Using diverse search criteria in NCBI-PubMed we identified a total of 54 reports that relied on either dsRNA, siRNA or shRNA to silence the expression of baculoviral or induced cell genes. Each one of the query entries included the words “RNA interference” plus one of the following: baculovirus, insect cell, Spodoptera or Bombyx mori.

Dharmacon Inc) will most probably encourage their use for future experiments. dsRNAs and siRNAs are appropriate for transient silencing, their effect usually lasting less than 7 days. However, when long-term RNA-mediate silencing is required, neither dsRNA or siRNA are up to the challenge. For these cases, a vector containing sequences for the expression of short hairpin RNA (shRNAs) are required. A shRNA is a strand of ribonucleic acid with two self-complementary 19–22 bp long sequences linked by a short loop of 4–11 nt; this loop is similar to the hairpin found in naturally occurring miRNA (Taxman et al., 2010). shRNAs are usually expressed from polIII promoters and function as precursors of siRNAs. Properly designed vectors used for the expression of shRNAs are capable of DNA integration and can be used to generate stably transformed insect cell lines that effectively silence the expression of target genes (Kim et al., 2012). At this point, it is still unclear if a given type of reagent (dsRNA, siRNA or shRNA) causes more potent RNAi response in insect cells. Some reports suggest that chemically synthesized siRNAs at doses as low as 9 nM are more potent than dsRNAs used at doses up to five times higer (Agrawal et al., 2004); this was attributed to the poor processing of the dsRNA by dicer enzyme in insect cell lines (Lee et al., 2013). However, in some cases efficient silencing using siRNA requires concentration as large as 100 nM (Zhao et al., 2013). Even more, some studies indicate that treatment of insect cells with shRNAs, which also have to be cut by dicer enzyme to become active, is as effective as treatment with siRNA (Kim et al., 2012). There is some evidence indicating that these inconsistencies may be a consequence of the size of the long dsRNA used as a precursor. According to two independent reports that used either dsRNAs (Wu et al., 2013) or shRNAs (Huang et al., 2007) of different lengths to respectively silence the expression of cycline E and EGFP in insect cells, ribonucleic acid precursors larger than 800 nt were more efficient than smaller ones (around 500 nt). These results are in agreement with the trend observed by plotting information regarding silencing efficiency and length of dsRNA provided in the supplementary material of the broad survey of RNAi experiments in lepidopteran insects conducted by Terenius et al. (see Fig. 3) (Terenius et al., 2011). However, this observation might be questionable, since neither of these reports accounts for differences in ribonucleic acid dose, and in the case of the large survey, for the method used for the administration of the silencing reagent. To facilitate comparison between studies, researchers should consider that,

http://www.rnaiglobal.org/Home/

6

Biotechnology Advances xxx (xxxx) xxx–xxx

C. Chavez-Pena, A.A. Kamen

genes for which silencing produce a well defined phenotype (Ramadan et al., 2007). Several vendors offer reagents to target standard, endogenously expressed housekeeping genes such as GAPDH, cyclophilin B and lamin A/C. Unfortunately, these have not been validated yet for use in insect cell lines, and further confirmation is still required. In the case of silencing negative controls, “scrambled” dsRNAs/siRNAs or reagents against reporter genes (e.g., LacZ or GFP dsRNAs) that do not target any gene expressed in the cell system can be used (Echeverri and Perrimon, 2006). These thend to be more useful than ‘mock transfection’ negative controls in which the silencing reagent is excluded, as these might have effects that are completely unrelated and irrelevant to the experimental samples (Ramadan et al., 2007). As normalization controls, it is preferable to select genes for which expression is constant through the length of the experiment. In the case of insect cell lines, it has been demonstrated that genes coding for the cell cycle regulator ecdysoneless and ribosomal subunit 28S are more adequate than genes commonly used in other species because their expression levels remain constant even after baculoviral infection (Salem et al., 2014).

Fig. 3. The frequency of use of dsRNA reagents of different length and its observed efficiency in Lepidopterans. The graph was constructed using the complementary information included in the large survey from Terenius et al. (2011) in which researchers from 43 institutions provided a detailed description of RNA interference experiments. The length of dsRNAs was calculated from their sequences and grouped as shown in the graph. The fraction of experiment with high silencing efficiencies as reported by the author of each experiment are shown.

5. Conclusion Research focused on improving the baculovirus expression system remains an active and important area of interest. This endeavor will benefit from better understanding the underlying biological process governing the interaction of the host cell with the virus and the role of specific genes in the expression of recombinant proteins. RNA interference technology has already demonstrated to be a feasible alternative to traditional methods for manipulation of gene expression in insect cell lines, for research puposes or for the ultimate goal of producing a more robust and efficient platform for the expression of recombinant proteins. RNA interference technology, among other technologies available for the regulation of gene expression in insect cells, has still a number of limitations to overcome. However, its use in insect cells will most probably become more prominent now that siRNAs libraries against a model lepipdopteran species is available and with the development of more cost-effective methods to efficiently deliver siRNA reagents in insect cells, and particularly in Spodoptera frugiperda cell line. It is also believed that by adhering to well-established guidelines for designing RNAi experiments and reporting results, the exchange of information between groups working with insect cell technology will be facilitated and will contribute to an accelerated improvement of the BEVS expression system.

regardless of the type of molecule used to elicit a RNAi response, it is important to properly select and report the region of the gene that is being targeted as well as the sequence and dose of the reagent being used. To this end, the approach taken by research institutions associated with RNAi Global seems very appropriate. It consists of creating individual, well-identified entries in the Open chemistry database (PubChem) each one including significant information (Schmidt et al., 2013). As of August 2017, the database contains more than 2 millions entries (https://www.ncbi.nlm.nih.gov/pcsubstance query siRNA/ dsRNA/shRNA). One of the early steps required for successful RNA-mediated silencing experiments is identifying the proper conditions for efficient introductions of RNAi reagents. Techniques for the introduction of silencing reagents include chemical introduction (transfection), electroporation, and the bathing method (Echeverri and Perrimon, 2006). Transfection is the most common technique since not all insect cell lines can directly take dsRNA or siRNA from the media (bathing method), and electroporation requires expensive equipment. There are several types of reagents for the transfection of siRNA/dsRNA: cationic polymers, cationic lipids, calcium phosphate. Several companies have developed specialized products, mainly cationic lipids, which are purposely designed to enhance the formation of RNA-lipid complexes to accomplish maximum transfection efficiency. However, as demonstrated by Fisher et al. (Fischer et al., 2013) delivery efficiency can significantly vary from reagent to reagent depending on the cell type and culture media used during transfection, regardless of the efficiency claimed by the reagent manufacturer. Also, when selecting transfection reagents, cytotoxicity and effects on cell health should be carefully evaluated to prevent false positive while testing silencing efficiency (Haney, 2007). A 2010 report that evaluated silencing efficiencies in the insect-derived cell line CiE1 using several transfection reagents suggests that Lipofectamine (Invitrogen) is highly effective (as described in (Yu et al., 2013)). Although this is not necessarily the case for insect cell lines used in the BEVS, this information can be used as a starting point for the optimisation of transfection protocols. Finally, it is difficult not to overemphasize the relevance of properly selecting read-out options to evaluate silencing efficiency, both at messenger and protein level. This involves not only choosing the proper technique (microarray, qPCR, fluorescence reported-base assays, western blot, etc.) but also the right combination of positive and negative silencing controls, as well as normalization controls. As a positive control, it is preferable to use reagents of known efficiency that target

Aknowledgements Authors would like to acknowledge in-kind contribution from the National Research Council of Canada and financial support from the Natural Sciences and Engineering Research Council of Canada (RGPIN2015-05132 and STPGP-462995-14) as well as Canada Research Chair (CRC-2403940). C.C-P. is a recipient of Conacyt grant for graduate studies. References Agrawal, N., Malhotra, P., Bhatnagar, R.K., 2004. siRNA-directed silencing of transgene expressed in cultured insect cells. Biochem. Biophys. Res. Commun. 320, 428–434. http://dx.doi.org/10.1016/j.bbrc.2004.05.184. Bi, J.X., Shuttleworth, J., Al-Rubeai, M., 2004. Uncoupling of cell growth and proliferation results in enhancement of productivity in p21CIP1-arrested CHO cells. Biotechnol. Bioeng. 85, 741–749. http://dx.doi.org/10.1002/bit.20025. Bieniossek, C., Imasaki, T., Takagi, Y., Berger, I., 2012. MultiBac: expanding the research toolbox for multiprotein complexes. Trends Biochem. Sci. 37, 49–57. http://dx.doi. org/10.1016/j.tibs.2011.10.005. Birmingham, A., Anderson, E., Sullivan, K., Reynolds, A., Boese, Q., Leake, D., Karpilow, J., Khvorova, A., 2007. A protocol for designing siRNAs with high functionality and specificity. Nat. Protoc. 2, 2068–2078. http://dx.doi.org/10.1038/nprot.2007.278. Carinhas, N., Bernal, V., Yokomizo, A.Y., Carrondo, M.J.T., Oliveira, R., Alves, P.M., 2009. Baculovirus production for gene therapy: the role of cell density, multiplicity of

7

Biotechnology Advances xxx (xxxx) xxx–xxx

C. Chavez-Pena, A.A. Kamen

suppression of β-N-acetylglucosaminidase and expression of β -1,4-galactosyltransferase. In: Castilho, A. (Ed.), Glyco-Engineering: Methods and Protocols. Springer New York, New York, NY, pp. 179–191. http://dx.doi.org/10.1007/978-14939-2760-9_13. Kim, E.J., Kramer, S.F., Hebert, C.G., Valdes, J.J., Bentley, W.E., 2007. Metabolic engineering of the baculovirus-expression system via inverse “shotgun” genomic analysis and RNA interference (dsRNA) increases product yield and cell longevity. Biotechnol. Bioeng. 98, 645–654. http://dx.doi.org/10.1002/bit.21353. Kim, N.Y., Baek, J.Y., Choi, H.S., Chung, I.S., Shin, S., Lee, J.I., Choi, J.Y., Yang, J.M., 2012. Short-hairpin RNA-mediated gene expression interference in trichoplusia ni cells. J. Microbiol. Biotechnol. 22, 190–198. http://dx.doi.org/10.4014/jmb.1108. 08045. Kobayashi, I., Tsukioka, H., Kômoto, N., Uchino, K., Sezutsu, H., Tamura, T., Kusakabe, T., Tomita, S., 2012. SID-1 protein of Caenorhabditis elegans mediates uptake of dsRNA into Bombyx cells. Insect Biochem. Mol. Biol. 42, 148–154. http://dx.doi.org/ 10.1016/j.ibmb.2011.11.007. Kramer, S.F., Bentley, W.E., 2003. RNA interference as a metabolic engineering tool: potential for in vivo control of protein expression in an insect larval model. Metab. Eng. 5, 183–190. http://dx.doi.org/10.1016/S1096-7176(03)00027-2. Kulkarni, M.M., Booker, M., Silver, S.J., Friedman, A., Hong, P., Perrimon, N., MatheyPrevot, B., 2006. Evidence of off-target effects associated with long dsRNAs in Drosophila melanogaster cell-based assays. Nat. Methods 3, 833–838. http://dx.doi. org/10.1038/nmeth935. Lai, Y.-K., Hsu, J.T., Chu, C.-C., Chang, T.-Y., Pan, K.-L., Lin, C.-C., 2012. Enhanced recombinant protein production and differential expression of molecular chaperones in sf-caspase-1-repressed stable cells after baculovirus infection. BMC Biotechnol. 12, 83. http://dx.doi.org/10.1186/1472-6750-12-83. Lee, J.M., Kojin, Y., Tatsuke, T., Mon, H., Miyagawa, Y., Kusakabe, T., 2013. Coexpression of Escherichia coli RNase III in silkworm cells improves the efficiency of RNA interference induced by long hairpin dsRNAs. Insect Sci. 20, 69–77. http://dx.doi.org/ 10.1111/j.1744-7917.2012.01569.x. Lee, H.S., Lee, H.Y., Kim, Y.-J., Jung, H.-D., Choi, K.J., Yang, J.M., Kim, S.S., Kim, K., 2015. Small interfering (Si) RNA mediated baculovirus replication reduction without affecting target gene expression. Virus Res. 199, 68–76. http://dx.doi.org/10.1016/j. virusres.2015.01.015. Léonard, R., Rendić, D., Rabouille, C., Wilson, I.B.H., Préat, T., Altmann, F., 2006. The Drosophila fused lobes gene encodes an N-acetylglucosaminidase involved in Nglycan processing. J. Biol. Chem. 281, 4867–4875. http://dx.doi.org/10.1074/jbc. M511023200. Li, Z., Zeng, B., Ling, L., Xu, J., You, L., Aslam, A.F.M., Tan, A., Huang, Y., 2015. Enhancement of larval RNAi efficiency by over-expressing argonaute2 in bombyx mori. Int. J. Biol. Sci. 11, 176–185. http://dx.doi.org/10.7150/ijbs.10235. Lin, C.-C., Hsu, J.T., Huang, K.-L., Tang, H.-K., Shu, C.-W., Lai, Y.-K., 2007. Sf-Caspase-1repressed stable cells: resistance to apoptosis and augmentation of recombinant protein production. Biotechnol. Appl. Biochem. 48, 11–19. http://dx.doi.org/10. 1042/BA20070044. Liu, Q., Paroo, Z., 2010. Biochemical principles of small RNA pathways. Annu. Rev. Biochem. 79, 295–319. http://dx.doi.org/10.1146/annurev.biochem.052208. 151733. March, J.C., Bentley, W.E., 2007. RNAi-based tuning of cell cycling in Drosophila S2 cells - effects on recombinant protein yield. Appl. Microbiol. Biotechnol. 73, 1128–1135. http://dx.doi.org/10.1007/s00253-006-0560-x. Marek, M., Van Oers, M.M., Devaraj, F.F., Vlak, J.M., Merten, O.W., 2011. Engineering of baculovirus vectors for the manufacture of virion-free biopharmaceuticals. Biotechnol. Bioeng. 108, 1056–1067. http://dx.doi.org/10.1002/bit.23028. Mena, J.A., Kamen, A. a, 2011. Insect cell technology is a versatile and robust vaccine manufacturing platform. Expert Rev. Vaccines 10, 1063–1081. http://dx.doi.org/10. 1586/erv.11.24. Mohr, S., Bakal, C., Perrimon, N., 2010. Genomic screening with RNAi: results and challenges. Annu. Rev. Biochem. 79, 37–64. http://dx.doi.org/10.1146/annurevbiochem-060408-092949. Monsma, S.A., Oomens, A.G., Blissard, G.W., 1996. The GP64 envelope fusion protein is an essential baculovirus protein required for cell-to-cell transmission of infection. J. Virol. 70, 4607–4616. Nagata, Y., Lee, J.M., Mon, H., Imanishi, S., Hong, S.M., Komatsu, S., Oshima, Y., Kusakabe, T., 2013. RNAi suppression of B-N-acetylglucosaminidase (BmFDL) for complex-type N-linked glycan synthesis in cultured silkworm cells. Biotechnol. Lett. 35, 1009–1016. http://dx.doi.org/10.1007/s10529-013-1183-9. Nguyen, Q., Nielsen, L.K., Reid, S., 2013. Genome scale transcriptomics of baculovirusinsect interactions. Viruses 5, 2721–2747. http://dx.doi.org/10.3390/v5112721. Nomura, T., Suganuma, M., Higa, Y., Kataoka, Y., Funaguma, S., Okazaki, H., Suzuki, T., Kobayashi, I., Sezutsu, H., Fujiyama, K., 2015. Improvement of glycosylation structure by suppression of β-N-acetylglucosaminidases in silkworm. J. Biosci. Bioeng. 119, 131–136. http://dx.doi.org/10.1016/j.jbiosc.2014.07.012. Novina, C.D., Sharp, P. a, 2004. The RNAi revolution. Nature 430, 161–164. http://dx. doi.org/10.1038/430161a. van Oers, M.M., 2011. Opportunities and challenges for the baculovirus expression system. J. Invertebr. Pathol. 107 (Suppl), S3–15. http://dx.doi.org/10.1016/j.jip. 2011.05.001. van Oers, M.M., Pijlman, G.P., Vlak, J.M., 2014. Thirty years of baculovirus-insect cell protein expression: from dark horse to mainstream technology. J. Gen. Virol. http:// dx.doi.org/10.1099/vir.0.067108-0. Oomens, A.G., Monsma, S.A., Blissard, G.W., 1995. The baculovirus GP64 envelope fusion protein: synthesis, oligomerization, and processing. Virology 209, 592–603. http:// dx.doi.org/10.1006/viro.1995.1291. Palomares, L.A., Estrada-Mondaca, S., Ramirez, O.T., 2005. Principles and applications of

infection and medium exchange. Appl. Microbiol. Biotechnol. 81, 1041–1049. http:// dx.doi.org/10.1007/s00253-008-1727-4. Carthew, R.W., Sontheimer, E.J., 2009. Origins and mechanisms of miRNAs and siRNAs. Cell 136, 642–655. http://dx.doi.org/10.1016/j.cell.2009.01.035. Cecchini, S., Negrete, A., Kotin, R.M., 2008. Toward exascale production of recombinant adeno-associated virus for gene transfer applications. Gene Ther. 15, 823–830. http://dx.doi.org/10.1038/gt.2008.61. Chan, L.C.L., Reid, S., 2016. Development of serum-free media for lepidopteran insect cell lines. In: Murhammer, W.D. (Ed.), Baculovirus and Insect Cell Expression Protocols. Springer New York, New York, NY, pp. 161–196. http://dx.doi.org/10.1007/978-14939-3043-2_8. Echeverri, C.J., Perrimon, N., 2006. High-throughput RNAi screening in cultured cells: a user's guide. Nat. Rev. Genet. 7, 373–384. http://dx.doi.org/10.1038/nrg1836. Elbashir, S.M., Elbashir, S.M., Lendeckel, W., Lendeckel, W., Tuschl, T., Tuschl, T., 2001. RNA interference is mediated 1- and 22-nucleotide RNAs. Genes Dev. 15, 188–200. http://dx.doi.org/10.1101/gad.862301.vents. Elias, C.B., Zeiser, A., Bédard, C., Kamen, A.A., 2000. Enhanced growth of Sf-9 cells to a maximum density of 5.2 × 10(7) cells per mL and production of beta-galactosidase at high cell density by fed batch culture. Biotechnol. Bioeng. 68, 381–388. Fire, A., Xu, S., Montgomery, M.K., Kostas, S.A., Driver, S.E., Mello, C.C., 1998. Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 391, 806–811. http://dx.doi.org/10.1038/35888. Fischer, S., Wagner, A., Kos, A., Aschrafi, A., Handrick, R., Hannemann, J., Otte, K., 2013. Breaking limitations of complex culture media: functional non-viral miRNA delivery into pharmaceutical production cell lines. J. Biotechnol. 168, 589–600. http://dx.doi. org/10.1016/j.jbiotec.2013.08.027. Fischer, S., Handrick, R., Otte, K., 2015. The art of CHO cell engineering: a comprehensive retrospect and future perspectives. Biotechnol. Adv. 33, 1878–1896. http://dx.doi. org/10.1016/j.biotechadv.2015.10.015. Flores-Jasso, C.F., Valdes, V.J., Sampieri, A., Valadez-Graham, V., Recillas-Targa, F., Vaca, L., 2004. Silencing structural and nonstructural genes in baculovirus by RNA interference. Virus Res. 102, 75–84. http://dx.doi.org/10.1016/j.virusres.2004.01. 018. Geisler, C., Jarvis, D., 2009. Insect cell glycosylation patterns in the context of biopharmaceuticals. In: Post-Translational Modification of Protein Biopharmaceuticals, pp. 165–191. http://dx.doi.org/10.1002/9783527626601.ch7. Geisler, C., Aumiller, J.J., Jarvis, D.L., 2008. A fused lobes gene encodes the processing βN-acetylglucosaminidase in Sf9 cells. J. Biol. Chem. 283, 11330–11339. http://dx. doi.org/10.1074/jbc.M710279200. Ghildiyal, M., Zamore, P.D., 2009. Small silencing RNAs: an expanding universe. Nat. Rev. Genet. 10, 94–108. http://dx.doi.org/10.1038/nrg2504. Ghosh, S., Kakumani, P.K., Kumar, A., Malhotra, P., Mukherjee, S.K., Bhatnagar, R.K., 2014. Genome wide screening of RNAi factors of Sf21 cells reveal several novel pathway associated proteins. BMC Genomics 15, 775. http://dx.doi.org/10.1186/ 1471-2164-15-775. Gong, L., Wang, Z., Wang, H., Qi, J., Hu, M., Hu, Q., 2015. Core RNAi machinery and three Sid-1 related genes in Spodoptera litura (Fabricius). Int. J. Agric. Biol. 17, 937–944. http://dx.doi.org/10.17957/IJAB/15.0005. Haney, S. a, 2007. Increasing the robustness and validity of RNAi screens. Pharmacogenomics 8, 1037–1049. http://dx.doi.org/10.2217/14622416.8.8.1037. Harrison, R.L., Jarvis, D.L., 2006. Protein N-glycosylation in the baculovirus-insect cell expression system and engineering of insect cells to produce “mammalianized” recombinant glycoproteins. Adv. Virus Res. 68, 159–191. http://dx.doi.org/10.1016/ S0065-3527(06)68005-6. Hebert, C.G., Valdes, J.J., Bentley, W.E., 2009a. In vitro and in vivo RNA interference mediated suppression of Tn-caspase-1 for improved recombinant protein production in High Five cell culture with the baculovirus expression vector system. Biotechnol. Bioeng. 104, 390–399. http://dx.doi.org/10.1002/bit.22411. Hebert, C.G., Valdes, J.J., Bentley, W.E., 2009b. Investigating apoptosis: characterization and analysis of Trichoplusia ni-caspase-1 through overexpression and RNAi mediated silencing. Insect Biochem. Mol. Biol. 39, 113–124. http://dx.doi.org/10.1016/j.ibmb. 2008.10.009. Hitchman, R.B., Possee, R.D., Crombie, A.T., Chambers, A., Ho, K., Siaterli, E., Lissina, O., Sternard, H., Novy, R., Loomis, K., Bird, L.E., Owens, R.J., King, L.A., 2010a. Genetic modification of a baculovirus vector for increased expression in insect cells. Cell Biol. Toxicol. 26, 57–68. http://dx.doi.org/10.1007/s10565-009-9133-y. Hitchman, R.B., Possee, R.D., Siaterli, E., Richards, K.S., Clayton, A.J., Bird, L.E., Owens, R.J., Carpentier, D.C.J., King, F.L., Danquah, J.O., Spink, K.G., King, L.A., 2010b. Improved expression of secreted and membrane-targeted proteins in insect cells. Biotechnol. Appl. Biochem. 56, 85–93. http://dx.doi.org/10.1042/BA20090130. Huang, Y., Deng, F., Hu, Z., Vlak, J.M., Wang, H., 2007. Baculovirus-mediated gene silencing in insect cells using intracellularly produced long double-stranded RNA. J. Biotechnol. 128, 226–236. http://dx.doi.org/10.1016/j.jbiotec.2006.09.016. Imai, S., Kusakabe, T., Xu, J., Li, Z., Shirai, S., Mon, H., Morokuma, D., Lee, J.M., 2015. Roles of silkworm endoplasmic reticulum chaperones in the secretion of recombinant proteins expressed by baculovirus system. Mol. Cell. Biochem. 409, 255–262. http:// dx.doi.org/10.1007/s11010-015-2529-5. Jadhav, V., Hackl, M., Druz, A., Shridhar, S., Chung, C.Y., Heffner, K.M., Kreil, D.P., Betenbaugh, M., Shiloach, J., Barron, N., Grillari, J., Borth, N., 2013. CHO microRNA engineering is growing up: recent successes and future challenges. Biotechnol. Adv. 31, 1501–1513. http://dx.doi.org/10.1016/j.biotechadv.2013.07.007. Kaba, S.A., Salcedo, A.M., Wafula, P.O., Vlak, J.M., Van Oers, M.M., 2004. Development of a chitinase and v-cathepsin negative bacmid for improved integrity of secreted recombinant proteins. J. Virol. Methods 122, 113–118. http://dx.doi.org/10.1016/j. jviromet.2004.07.006. Kim, Y.K., Cha, H.J., 2015. Engineering N-glycosylation pathway in insect cells:

8

Biotechnology Advances xxx (xxxx) xxx–xxx

C. Chavez-Pena, A.A. Kamen

Methods, Methods in Molecular Biology. Humana Press, Totowa, NJ, pp. 139–156. http://dx.doi.org/10.1007/978-1-60761-657-3_10. Terenius, O., Papanicolaou, A., Garbutt, J.S., Eleftherianos, I., Huvenne, H., et al., 2011. RNA interference in Lepidoptera: an overview of successful and unsuccessful studies and implications for experimental design. J. Insect Physiol. 57, 231–245. http://dx. doi.org/10.1016/j.jinsphys.2010.11.006. Tomoyasu, Y., Miller, S.C., Tomita, S., Schoppmeier, M., Grossmann, D., Bucher, G., 2008. Exploring systemic RNA interference in insects: a genome-wide survey for RNAi genes in Tribolium. Genome Biol. 9, R10. http://dx.doi.org/10.1186/gb-2008-9-1r10. Valdes, V.J., Sampieri, A., Sepulveda, J., Vaca, L., 2003. Using double-stranded RNA to prevent in vitro and in vivo viral infections by recombinant baculovirus. J. Biol. Chem. 278, 19317–19324. http://dx.doi.org/10.1074/jbc.M212039200. Vert, J.-P., Foveau, N., Lajaunie, C., Vandenbrouck, Y., 2006. An accurate and interpretable model for siRNA efficacy prediction. BMC Bioinforma. 7, 520. http://dx.doi. org/10.1186/1471-2105-7-520. Wang, G.H., Jiang, L., Zhu, L., Cheng, T.C., Niu, W.H., Yan, Y.F., Xia, Q.Y., 2013. Characterization of Argonaute family members in the silkworm, Bombyx mori. Insect Sci. 20, 78–91. http://dx.doi.org/10.1111/j.1744-7917.2012.01555.x. Wang, Q., Zhou, Y., Chen, K., Ju, X., 2016. Suppression of Bm-Caspase-1 expression in BmN cells enhances recombinant protein production in a Baculovirus expression vector system. Mol. Biotechnol. 58, 319–327. http://dx.doi.org/10.1007/s12033016-9931-4. Wu, J., King, G., Daugulis, A.J., Faulkner, P., Bone, D.H., Goosen, M.F.A., 1989. Engineering aspects of insect cell suspension culture: a review. Appl. Microbiol. Biotechnol. 32, 249–255. http://dx.doi.org/10.1007/BF00184969. Wu, H.-C., Hebert, C.G., Hung, C.-W., Quan, D.N., Carter, K.K., Bentley, W.E., 2013. Tuning cell cycle of insect cells for enhanced protein production. J. Biotechnol. 168, 55–61. http://dx.doi.org/10.1016/j.jbiotec.2013.08.017. Wynant, N., Santos, D., Vanden Broeck, J., 2014. Biological Mechanisms Determining the Success of RNA Interference in Insects, 1st ed. International Review of Cell and Molecular Biology Elsevier Inc.http://dx.doi.org/10.1016/B978-0-12-800178-3. 00005-1. Xiao, S., Shiloach, J., Betenbaugh, M.J., 2014. Engineering cells to improve protein expression. Curr. Opin. Struct. Biol. 26, 32–38. http://dx.doi.org/10.1016/j.sbi.2014. 03.005. Xu, J., Mon, H., Kusakabe, T., Li, Z., Zhu, L., Iiyama, K., Masuda, A., Mitsudome, T., Lee, J.M., 2013a. Establishment of a soaking RNA interference and Bombyx mori nucleopolyhedrovirus (BmNPV)-hypersensitive cell line using Bme21 cell. Appl. Microbiol. Biotechnol. 97, 10435–10444. http://dx.doi.org/10.1007/s00253-0135279-x. Xu, J., Nagata, Y., Mon, H., Li, Z., Zhu, L., Iiyama, K., Kusakabe, T., Lee, J.M., 2013b. Soaking RNAi-mediated modification of Sf9 cells for baculovirus expression system by ectopic expression of Caenorhabditis elegans SID-1. Appl. Microbiol. Biotechnol. 97, 5921–5931. http://dx.doi.org/10.1007/s00253-013-4785-1. Yu, Q., Lin, T., Feng, G., Yang, K., Pang, Y., 2008. Functional analysis of the putative antiapoptotic genes, p49 and iap4, of Spodoptera litura nucleopolyhedrovirus with RNAi. J. Gen. Virol. 89, 1873–1880. http://dx.doi.org/10.1099/vir.0.2008/ 001008-0. Yu, N., Christiaens, O., Liu, J., Niu, J., Cappelle, K., Caccia, S., Huvenne, H., Smagghe, G., 2013. Delivery of dsRNA for RNAi in insects: an overview and future directions. Insect Sci. 20, 4–14. http://dx.doi.org/10.1111/j.1744-7917.2012.01534.x. Zhao, H., Yi, X., Hu, Z., Hu, M., Chen, S., Muhammad, R. ul H., Dong, X., Gong, L., 2013. RNAi-mediated knockdown of catalase causes cell cycle arrest in SL-1 cells and results in low survival rate of Spodoptera Litura (Fabricius). PLoS One 8. http://dx.doi.org/ 10.1371/journal.pone.0059527.

the insect cell-baculovirus expression vector system. In: OzturK, S.S., Wei-Shou, H. (Eds.), Cell Culture Technology for Pharmaceutical and Cell-Based Therapies. CRC Press, pp. 627–692. Petricevich, V.L., Palomares, L.A., González, M., Ramı́rez, O.T., 2001. Parameters that determine virus adsorption kinetics: toward the design of better infection strategies for the insect cell - baculovirus expression system. Enzym. Microb. Technol. 29, 52–61. http://dx.doi.org/10.1016/S0141-0229(01)00323-4. Pruitt, K., Brown, G., Tattiana, T., 2002. The Reference Sequence (RefSeq) Database. [WWW Document]. NCBI Handb. URL. https://www.ncbi.nlm.nih.gov/books/ NBK21091. Quadt, I., van Lent, J.W.M., Knebel-Mörsdorf, D., 2007. Studies of the silencing of baculovirus DNA binding protein. J. Virol. 81, 6122–6127. http://dx.doi.org/10.1128/ JVI.02768-06. Ramadan, N., Flockhart, I., Booker, M., Perrimon, N., Mathey-Prevot, B., 2007. Design and implementation of high-throughput RNAi screens in cultured Drosophila cells. Nat. Protoc. 2, 2245–2264. http://dx.doi.org/10.1038/nprot.2007.250. Reid, S., Chan, L., van Oers, M.M., 2013. Production of Entomopathogenic Viruses, Mass Production of Beneficial Organisms: Invertebrates and Entomopathogens. Elsevierhttp://dx.doi.org/10.1016/B978-0-12-391453-8.00013-3. Rohrmann, G., 2013. Baculovirus Molecular Biology, 3rd editio. ed, Baculovirus Molecular Biology [Internet]. National Center for Biotechnology, Bethesda (MD) (doi:24479205). Salem, T.Z., Maruniak, J.E., 2007. A universal transgene silencing approach in baculovirus-insect cell system. J. Virol. Methods 145, 1–8. http://dx.doi.org/10.1016/j. jviromet.2007.04.012. Salem, T.Z., Zhang, F., Thiem, S.M., 2013. Reduced expression of Autographa californica nucleopolyhedrovirus ORF34, an essential gene, enhances heterologous gene expression. Virology 435, 225–238. http://dx.doi.org/10.1016/j.virol.2012.10.022. Salem, T.Z., Allam, W.R., Thiem, S.M., 2014. Verifying the stability of selected genes for normalization in Q PCR experiments of spodoptera frugiperda cells during AcMNPV infection. PLoS One 9. http://dx.doi.org/10.1371/journal.pone.0108516. Schmidt, E.E., Pelz, O., Buhlmann, S., Kerr, G., Horn, T., Boutros, M., 2013. Genome RNAi: a database for cell-based and in vivo RNAi phenotypes, 2013 update. Nucleic Acids Res. 41, 1021–1026. http://dx.doi.org/10.1093/nar/gks1170. Schultz, K.L.W., Friesen, P.D., 2009. Baculovirus DNA replication-specific expression factors trigger apoptosis and shutoff of host protein synthesis during infection. J. Virol. 83, 11123–11132. http://dx.doi.org/10.1128/JVI.01199-09. Schwarz, D.S., Hutvágner, G., Du, T., Xu, Z., Aronin, N., Zamore, P.D., 2003. Asymmetry in the assembly of the RNAi enzyme complex. Cell 115, 199–208. http://dx.doi.org/ 10.1016/S0092-8674(03)00759-1. Smith, G.E., Summers, M.D., Fraser, M.J., 1983. Production of human beta interferon in insect cells infected with a baculovirus expression vector. Mol. Cell. Biol. 3, 2156–2165. http://dx.doi.org/10.1128/MCB.3.12.2156.Updated. Subbaiah, E.V., Royer, C., Kanginakudru, S., Satyavathi, V.V., Babu, A.S., Sivaprasad, V., Chavancy, G., Darocha, M., Jalabert, A., Mauchamp, B., Basha, I., Couble, P., Nagaraju, J., 2013. Engineering silkworms for resistance to baculovirus through multigene RNA interference. Genetics 193, 63–75. http://dx.doi.org/10.1534/ genetics.112.144402. Suzuki, H., Tamai, N., Habu, Y., Chang, M.O.O., Takaku, H., 2008. Suppression of hepatitis C virus replication by baculovirus vector-mediated short-hairpin RNA expression. FEBS Lett. 582, 3085–3089. http://dx.doi.org/10.1016/j.febslet.2008.07. 056. Swevers, L., Liu, J., Huvenne, H., Smagghe, G., 2011. Search for limiting factors in the RNAi pathway in silkmoth tissues and the Bm5 cell line: the RNA-binding proteins R2D2 and translin. PLoS One 6. http://dx.doi.org/10.1371/journal.pone.0020250. Taxman, D.J., Moore, C.B., Guthrie, E.H., Huang, M.T., 2010. Short Hairpin RNA (shRNA): design, delivery, and assessment of gene knockdown. In: Sioud, M. (Ed.),

9