Cell counting is a vital practice in the maintenance and manipulation of cell cultures. It is a crucial aspect of assessing cell viability and determining proliferation rates, which are integral to maintaining the health and functionality of a culture. Additionally, it is critical for establishing the time of infection in bioreactors and monitoring cell culture response to targeted infection over time. However, when cell counting is performed manually, the time involved can become substantial, particularly when multiple cultures need to be handled in parallel. Automated cell counters, which enable significant time reduction, are commercially available but remain relatively expensive. Here, we present a machine learning (ML) model based on YOLOv4 that is able to perform cell counts with a high accuracy (>95%) for Trypan blue-stained insect cells. Images of two distinctly different cell lines, Trichoplusia ni (High Five^TM; Hi5 cells) and Spodoptera frugiperda (Sf9), were used for training, validation, and testing of the model. The ML model yielded F1 scores of 0.97 and 0.96 for alive and dead cells, respectively, which represents a substantially improved performance over that of other cell counters. Furthermore, the ML model is versatile, as an F1 score of 0.96 was also obtained on images of Trypan blue-stained human embryonic kidney (HEK) cells that the model had not been trained on. Our implementation of the ML model comes with a straightforward user interface and can image in batches, which makes it highly suitable for the evaluation of multiple parallel cultures (e.g. in Design of Experiments). Overall, this approach for accurate classification of cells provides a fast, bias-free alternative to manual counting.

To counteract hand, foot, and mouth disease-causing viruses such as enterovirus A71 and coxsackievirus A6, virus-like particles (VLPs) have emerged as a leading contender for the development of a multivalent vaccine. However, VLPs have shown rapid conversion from a highly immunogenic state to a less immunogenic state and low particle integrity lifetimes compared to inactivated virus vaccines, thus raising concerns about their overall stability. Here, we produce VLPs to investigate capsid stability using cryogenic electron microscopy (cryo-EM), mass spectrometry (MS), biochemical assays, and atomic force microscopy (AFM). In contrast to prior studies and prevailing hypotheses, we show that insect-cell produced enterovirus VLPs include encapsidated RNA fragments with viral protein coding sequences. Our integrated approach reveals that CVA6 VLPs do not undergo viral maturation, in contrast to virions; that they can encapsidate RNA fragments, similarly to virions; and that despite the latter, they are more brittle than virions. Interestingly, this indicates that CVA6 VLP stability is more affected by lack of viral maturation than the presence of RNA. Our study highlights how the development of VLPs as vaccine candidates should encompass probing for unwanted (viral) RNA content and establishing control of their maturation to enhance stability.

The baculovirus expression vector system (BEVS) is commercially used for producing recombinant (glyco)proteins, gene therapy vectors, and virus-like particles (VLPs) in insect cells. A major challenge in VLP downstream processing (DSP) is the removal of similarly sized baculovirus particle contaminants. Herein we engineered a genetic ON/OFF thermoswitch in the baculovirus vector by mutating the essential gp41 gene. This temperature-sensitive (ts) baculovirus expression system (BacFree^ts) produces high baculovirus titers at 27°C, while contaminating baculovirus production switches off at elevated temperatures. We demonstrate scalable VLP production for chikungunya virus (CHIKV), West Nile virus (WNV), coxsackievirus A6, and foot-and-mouth disease virus (FMDV) at 33°C, concomitant with baculovirus particle reduction up to 99.97%. However, elevated temperature moderately affected the yield of two out of four VLP designs. BacFree^ts is compatible with commonly used insect cells and scalable infection strategies. BacFree^ts effectively reduces baculovirus contaminants in the upstream VLP production phase, which will facilitate DSP of human and veterinary VLP-based vaccines.

Hand, foot, and mouth disease (HFMD) represents an emerging health concern whose main causative agents are Coxsackievirus A6 (CVA6) and enterovirus A71 (EV71). The lack of a CVA6 vaccine and the rise of new HFMD-causing strains due to the containment of established HFMD-causing viruses necessitates the search for alternative vaccine technologies, including virus-like particle (VLP) vaccine candidates. While studies have demonstrated that production of enterovirus-like particles in various organisms can be achieved by expression of the viral P1 structural proteins and the 3CD protease, optimization based on the interplay between the three most commonly altered infection parameters (multiplicity of infection (MOI), viable cell density at the time of infection (VCD), and the infection period) is often not investigated. To address this challenge, we have performed Design of Experiments (DoE) to optimize the production of CVA6 and EV71 VLPs. Our results indicate that CVA6 VLP production peaks at high MOI, high VCD, and long infection periods. Our subsequent downstream purification processes yielded 38 mg and 158 mg of purified CVA6 and EV71 VLPs from 1 L crude harvest, respectively. This translates into thousands of potential vaccine doses and highlights the economic potential of enterovirus-like particles for vaccine purposes.