The COVID-19 pandemic has undoubtedly become the defining event of 2020. Since its first identification in December 2019, the viral pandemic has quickly spread to almost every part of the world and has caused a significant impact on our society. As of August 2020, the worldwide death toll of COVID-19 has surpassed 800,000 cases. Our economy is also experiencing the largest recession since the Great Depression. Combatting the disease requires a concerted effort from many different areas, including fast and accurate virus detection, suitable protection gears, and vaccine and antiviral treatment options. The CBE department at KAIST has joined the worldwide effort by actively engaging in researches to combat COVID-19.

Many viruses, including SARS-CoV-2, generate long double-stranded RNAs (dsRNAs) as byproducts of transcription (for DNA viruses) or during replication (for RNA viruses). By targeting long dsRNAs as a common biomarker for viruses, a research team led by Professor Yoosik Kim and Professor Sheng Li developed a reactive polymer-based platform capable of detecting a wide-range of viruses. Furthermore, the dsRNA-targeting antibodies recognize the length of the dsRNAs, not their specific sequences, thus allowing the recognition of viruses without knowing nucleotide sequences of their genome.

Several key technologies were developed and optimized in the design of the platform. The researchers utilized a highly reactive polymer known as poly(pentafluorophenyl acrylate) (PPFPA), allowing the immobilization of many antibodies for efficient dsRNA capture. To enhance the detection sensitivity, a two-step detection scheme was devised. Viral long dsRNAs were first captured by surface-immobilized antibodies, and then visualized using fluorophore-tagged antibodies that also recognize dsRNAs. As multiple fluorophore-tagged antibodies may bind to a single dsRNA target, this approach significantly enhanced the signal-to-noise ratio in the detection of viral long dsRNAs. As a proof of concept, the research team applied the optimized platform to detect a number of different viruses. They showed that by using a single platform, both hepatitis C virus (HCV) and hepatitis A virus (HAV) can be detected.

This technology can be particularly useful during a viral pandemic such as COVID-19. As an effective screening detector, the developed platform can be used to quickly differentiate virus-infected population from non-infected ones, thus saving valuable time and medical resources. More importantly, since the detection is not sequence dependent, it can be used to target the current active virus strand as well as other mutated viruses. This research is recently published in Biomacromolecules.

There are many efforts to develop antiviral drugs to treat COVID-19 patients. As the discovery of new drugs may take years, many of the candidates under investigation are rediscovery of existing drugs. For example, the drug Remdesivir was being developed to target Ebola virus, but it was found to show encouraging results in treating COVID-19 patients. A key hurdle to overcome in the drug development is to understand the potential side effects of the drug as well as negative health effects caused by unexpected drug-drug interactions (DDIs). This is particularly important when treating COVID-19 patients as the ones with the most severe symptoms are people with underlying medical conditions, for example hypertension and diabetes.

To overcome this challenge, Professor Sang Yup Lee's and Professor Hyun Uk Kim's labs developed a computational method to predict the types and severity of DDIs. Specifically, they constructed an algorithm to effectively predict potential adverse effects caused by the simultaneous administration of multiple drugs. This strategy is being applied to identify potential negative health effects when candidate COVID-19 treatment drugs are co-administered with other drugs prescribed for treating underlying medical conditions. Furthermore, the research team is also attempting to suggest safer alternative drugs for treating any underlying diseases when simultaneously taking COVID-19 treatment drugs. Potential drugs currently considered for the COVID-19 treatment are not examined as thoroughly as usual drug development pipelines because of the immense time pressure at the moment. Development of a computational algorithm as in this study will help mitigate unwanted clinical accidents.