Engineering Coronaviruses to Evaluate Emergence and Pathogenic Potential.

A recent study provides a platform for generating infectious coronavirus genomes using sequence data, examining their capabilities of replicating in human cells and causing diseases in animal models, and evaluating therapeutics and vaccines. Similar approaches could be used to assess the potential of human emergence and pathogenicity for other viruses.

The severe acute respiratory syndrome (SARS) epidemic in 2003 and the Middle East respiratory syndrome (MERS) epidemic in the last 3 years have shown that coronaviruses (CoVs) have the capability to cause major epidemics. For the SARS epidemic, a total of >8000 laboratory-confirmed cases with >800 deaths were observed (http://www.cdc.gov/sars/about/fs-sars.html). This horrific epidemic was followed by the publication of >7500 scientific papers on CoVs visible in PubMed, which represents two-thirds of the total number of publications on CoVs in Pubmed. Despite the numerous studies on CoVs, it is still difficult to predict which CoV may have the potential to emerge as the next culprit. A recent study in PNAS by Menachery et al. [1] and another similar study in Nature Medicine published in December 2015 by the same group [2] reported the use of existing sequence data with reverse genetics to engineer SARS-related CoVs and evaluate their potential of emergence and pathogenicity.

Shortly after the emergence of SARS-CoV, SARS-related CoVs were found in civets [3]. However, multiple lines of evidence showed that the civets are just the intermediate or amplification hosts for SARS-CoV. Through intensive surveillance studies in various mammals in Hong Kong, Lau et al. reported the presence of SARS-related CoVs in Chinese horseshoe bats in Hong Kong [4]. A similar observation was also reported by another group in mainland China [5]. Since then, numerous SARS-related CoV sequences were observed in different species of the horseshoe bat. These SARS-related CoV sequences possessed different degrees of sequence identities to the SARS-CoV originally found in humans. Most importantly, it was difficult to predict which SARS-related CoV may have the potential to emerge in humans again, causing another SARS epidemic.

In order to predict whether a SARS-related bat CoV, named WIV1-CoV, discovered in Chinese horseshoe bats in Yunnan [6], had the potential to emerge in humans, Menachery et al. synthesized a full-DNA clone based on the sequence of the virus [1]. The clone was shown to be able to generate infectious virus in primary human airway epithelial cell culture, confirming its ability to efficiently replicate in human cells. When the spike gene of WIV1-CoV was used to replace that of SARS-CoV MA15, a mouse-adapted virus [7], the resulting CoV did not replicate efficiently or cause disease in mice. However, if transgenic mice that expressed the ACE2 receptor were used, WIV1-CoV was able to replicate and cause weight loss. These experiments showed that WIV1-CoV actually has significant potential to emerge in humans.

In the second part of their study, Menachery et al. evaluated various therapeutics and vaccines for possible treatment and prevention of WIV1-CoV infection [1]. They found that antibodies that have been shown to block SARS-CoV might also be used for protection of humans against WIV1-CoV infections. However, they found that the antibodies generated by immunizing animals using inactivated SARS-CoV did not protect WIV1-CoV infection in aged mice. This implied that, if WIV1-CoV emerges to cause another epidemic, immunization of humans using inactivated SARS-CoV may not be useful to control the epidemic.

The platform that Menachery et al. employed can be used for evaluating the emergence and pathogenic potential of other CoVs. Before the SARS epidemic, fewer than 10 CoVs with complete genome sequences were available. After the SARS epidemic, and up to March 2016, there was an addition of more than 40 CoVs with complete genomes sequenced. These CoVs include two human CoVs (HCoVs; HCoV-NL63 and HCoV-HKU1) and at least 30 other mammalian and avian CoVs. Two additional lineages in Betacoronavirus and a novel genus, Deltacoronavirus, have been discovered 8, 9. This diversity of coronaviruses is due to the infidelity of their RNA-dependent RNA polymerases, their high frequencies of homologous RNA recombination, and their large genomes. CoVs are well known for the difficulties in culturing them, and most of the newly discovered CoVs are so far noncultivable. Just from the sequences of these CoVs, it is impossible to predict which one(s) may have the potential in causing the next epidemic. Synthetic full-DNA clones for these CoVs can be made, and systematic evaluation can be performed. Since betaCoVs, including HCoV-OC43, SARS-CoV, and MERS-CoV, are the culprits of all large human epidemics, we should start the systematic evaluation by focusing on the betaCoVs.

In addition to CoVs, this approach can also be extended to study other viruses with the potential for emergence. In recent years, multiple viruses – such as influenza viruses H5N1 and H7N9, SARS-CoV and MERS-CoV, Ebolavirus and Zika virus – have emerged in humans. Some of them have apparently disappeared after the amplification host was segregated from humans, whereas others have persisted for years. Each of these viruses has a lot of closely related viruses or sequences obtained by either conventional sequencing or metagenomics approaches. The platform of Menachery et al. can be used for assessing the potential of these cultivable viruses or viral sequences in causing another epidemic, and the effectiveness of using various therapeutic or vaccination modalities in their treatment or prevention if an epidemic due to these viruses really emerges.

Despite the academic contribution of these research studies, such potentially gain-of-function experiments must be scrutinized and handled with the utmost care. Recently, there has been much discussion and debate on the potential threats of gain-of-function experiments [10]. There are fears that accidental release of these ‘super-virulent’ viruses or their use as biological weapons may lead to uncontrollable epidemics. However, we believe that these gain-of-function experiments should not be banned because of the potential threats. By contrast, each experiment should be carefully examined for the potential benefits weighed against potential threats. In our opinion, the potential benefits are enormous, as exemplified by the works by Menachery et al. 1, 2. It is the laboratory practice and conduct of the researchers that are most important in safeguarding any potential leakage of viruses.