Introduction

Since the development of the baculovirus expression vector system (BEVS) (Smith et al., 1983), insect cells (IC) have been established as a powerful platform for recombinant protein production. The IC-BEVS is scalable, capable of producing proteins with complex post-translational modifications (Ailor and Betenbaugh, 1999; Fernandes et al., 2022a), and induces high expression of transgenes leading to reduced production times (Correia et al., 2020; Correia et al., 2022; Fernandes et al., 2022b). These advantages position IC-BEVS as a potential alternative to traditional systems in the production of complex products like virus-like particles (VLPs) and viral vectors such as recombinant adeno-associated virus (AAV). Nevertheless, the lack of a comprehensive understanding of the severe impact of the viral infection on the host cell machinery still poses major challenges to further improve this expression system.
Efforts have been made to characterize the insect cell host response to virus infection and/or heterologous gene expression. Technologies such as metabolomics (Carinhas et al., 2010; Monteiro et al., 2012; Monteiro et al., 2016), transcriptomics (Chen et al., 2014; Koczka et al., 2018; Silvano et al., 2022; Wei et al., 2017 and Virgolini et al., 2022) and proteomics (Carinhas et al., 2011; Nayyar et al., 2017; Yu et al., 2016) have been employed by our group and others to shed light on the underlying biological mechanisms of cultured insect cells and BEVS. Exploring the response to virus infection on a gene expression level might be especially interesting, as the baculovirus has shown to take-over the cells gene expression machinery, activating stress response mechanisms such as apoptosis, and impacting protein folding and translation mechanisms, among others (Koczka et al., 2018; Wei et al., 2017; Virgolini et al., 2022).
Recent studies have provided genome and transcriptome references forSpodoptera frugiperda and Trichoplusia ni , (Chen et al., 2019; Xiao et al., 2020) and such advancements could thereby be a milestone in the study of gene expression alterations in Sf9 and High Five insect cells during baculovirus infection and/or heterologous protein expression. Indeed, an emerging number of transcriptomic studies in both cell lines can be observed (Koczka et al., 2018; Silvano et al., 2022; Virgolini et al., 2022), highlighting the interest in understanding host cell response to guide rational design of targeted cell line development and process engineering approaches. Nevertheless, these studies are limited by their approach of assessing cell populations instead of single cells, thus ignoring potential heterogeneity within the production platform. Moreover, single-cell RNA sequencing (scRNA-seq) would have the added benefit of identifying sub-populations of cells with distinct gene expression profiles, which might result in phenotypically beneficial traits (Ke et al., 2022).
Single-cell transcriptomics technologies have matured rapidly, allowing the study of gene expression patterns of tens of thousands of single cells in an accurate and cost-effective manner. ScRNA-seq has become standard in evaluating the transcriptome profiles of different cell types within tissues. Nevertheless, first applications in clonally derived cell lines as well as virus infection processes have been shown, each indicating significant heterogeneity within the respective cell population (Russell et al., 2018; Sun et al., 2020; Tzani et al., 2021). While this could indicate the potential complexity and/or heterogeneity in production processes using insect cells and BEVS, the use of single-cell transcriptome analysis to characterize this expression system is so far inexistent.
In this study, we implemented for the first time scRNA-seq in Sf9 insect cells during the production of AAV using a low multiplicity of infection (MOI), dual-baculovirus infection process, assessing population heterogeneity and gene expression profiles prior to and along infection.