Risky MERSr-CoV research in Wuhan
In a recent on BioRxiv several colleagues and I were able to identify the complete genome for a novel Betacoronavirus from contaminating reads in agricultural rice sequencing datasets.
We identified a novel HKU4-related Merbecovirus in a rice dataset sequenced by the Huazhong Agricultural University in Wuhan in early 2020.
The novel HKU4r-CoV assembled from a rice sequencing dataset is in an infectious clone format with a CMV promoter upstream of the 5’-end and a Hepatitis D virus ribozyme followed by a bgH polyA signal downstream of the 3’-end of the genome (Fig. 1). We modelled binding of the RBD of the novel CoV to hDPP4 and found comparable binding free energy as BatCoV HKU4 to hDPP4 (Wang et al. 2014). The novel HKU4r-CoV sequence was submitted to NCBI (Genbank: OK560913.1).
Also in the cotton sequencing dataset (PRJNA602160) we identified a near complete MERS spike gene sequence and found this spike had been inserted into the HKU4r-CoV backbone. While we don’t have access to experiment results, this is likely and almost certainly intended to be ePPP (Gain of function research of concern) research.
While the reverse genetics system used by the WIV for the NIH funded EHA project (discussed below) utilized a MERS-CoV backbone and swapped in HKU4r-CoV components, the WIV (presumably the WIV given their publication history and as we know the WIV outsourced sequencing in Wuhan) appears to have been simultaneously using a HKU4r-CoV backbone for independent research.
I was quite shocked then to see details of the 5 year EHA bat coronavirus project emerge in October detailing MERSr-CoV research at the WIV.
EHA Project
Details about a 5 year grant “Understanding the Risk of Bat Coronavirus Emergence” awarded to EcoHealth Alliance (EHA) by the National Institute of Health (NIH) were made public by The Intercept in Sept. 2021. It took the NIH one year to (apparently begrudgingly) release documentation. You can read the legal case here.
The main cause for concern regarding the EHA project were details about chimeric coronavirus (CoV) contruction using a MERS-CoV backbone.
Middle East respiratory syndrome coronavirus (MERS-CoV)
MERS-CoV is a lineage C Betacoronavirus which can be transmitted zoonotically from camels to humans. The virus uses human and bat dipeptidyl peptidase 4 (DPP4) as entry receptors and replicates efficiently in human airway epithelial cells. As at Feb 2020 there have been a total of c. 2400 cases of MERS-CoV infection, with a case fatality rate of c. 35%. In humans DPP4 is expressed by multiple organs and tissues. Initial symptoms of infection may be gastrointestinal, followed by a more severe systemic and respiratory syndrome. α2,3-sialic acids, CEACAM5 and GRP78 may function as attachment factors for MERS-CoV.
The danger of MERS-CoV to humans led to a Gain-of-Function funding (GOF) pause on experiments involving MERS-CoV by the US in 2014 (only to be lifted in 2017).
EHA Project Year 5: Research of Concern
Work subcontracted by EHA to the WIV conducted in grant award year 5 (06/01/2018–05/31/2019) included generating chimeric full-length infections clones of MERS-CoV with Receptor Binding Domain (RBD) replaced by those from various HKU4r-CoV and assessing infectivity on human cells.
“Taking a similar reverse genetics strategy that we used in SARSr-CoV studies, we constructed the full-length infectious clone of MERS-CoV, and replaced the RBD of MERS-CoV with the RBDS of various strains of HKU4-related coronaviruses previously identified in bats from different provinces in southern China. The full-length MERS-CoV and chimeric viruses with RBDs of HKU4r-CoVs were then rescued. Immunofluorescence assay showed that these chimeric MERS-HKU4rRBD coronaviruses were able to Infect human cells from different tissues including lung, liver, intestine and kidney (Fig. 14 Left). Moreover, efficient replication of the chimeric HKU4r-CoVs were detected by real-time PCR in HeLa cells that express human DPP4 receptor (Fig. 14 Right). The results suggest potential risk of the bat HKU4r-CoVs for cross-species infection in humans.”
Inserting novel RBD’s with unknown affinity for human DPP4 into a known highly human pathogenic MERS-CoV backbone is clearly risky reaserch. Indeed comments by multiple virologists obtained by The Intercept show a shared concern for the danger involved.
But another thing that is striking about the year 5 report (released 2 years late by the way) is that transgenic mice that express human DPP4 had been constructed and were being bred in year 3 at the WIV. The stated aim of year 4 research was to use these mouse models for in vivo MERSr-CoV research. Yet none of this work was reported in the year 4 or year 5 period.
We will go back to year 3 and 4 of the project to look at what had been proposed and completed.
EHA Project Year 3: S gene cloning
In Year 3 (06/01/2016–05/31/2017) MERSr-CoV research processed with the following results “The full-length infectious cDNA clone of MERS-CoV has been successfully constructed. The full-length S gene of 12 different novel bat MERS-related coronaviruses have been amplified and cloned into the T-vectors.”. Transgenic mice that express human DPP4 had been constructed and were being bred in this period at the WIV.
The genomes for the 12 novel MERSr-CoVs as at 11/05/2021 have not been released.
EHA Project Year 4: Mouse model hDPP4 studies proposed
In Year 4 (06/01/2017–05/31/2018) the MERS research aim was stated as “Using the full-length infectious cDNA clone of MERS-CoV, chimeric viruses with the spikes of newly identified MERSr-CoVs will be constructed. The pathogenesis of these MERSr-CoVs will be tested on the human DPP4-expressing mouse model that has already been developed and validated in Y4”
Although research using a human DPP4-expressing mouse model was the proposed aim for year 4, only cell-infectivity experiments were reported.
“Two novel MERSr-CoVs, BtCoV/li/GD/2013–845 and BtCoV/li/GD/2014–422, were identified from great evening bats (la io) in Guangdong province.”. “in the receptor-binding domain (RBD) of the spike protein, the two novel MERSr-CoVs are identical to MERS-CoV at six out of the 13 residues that directly interact with human DPP4 receptor, making them more similar to MERS-CoV than any other known lineage C BetaCoVs (Fig. 34a). Protein-protein interaction assays demonstrated that the spike proteins of the novel MERSr-CoVs bind to both human and bat DPP4 (Fig. 34b). Moreover, bat cells exogenously expressing human DPP4 support the entry of the retrovirus pseudotyped with BtCoV/li/GD/2014–422 spike, while the pseudovirus fails to enter cells that do not express DPP4. The results demonstrate that the spike protein of the newly identified MERSr-CoV recognizes the human DPP4 receptor.”.
Indeed the entire single paragraph pertaining to the novel MERSr-CoV research appears to have been summarized from Luo et al. (2018) which was received by the Journal of Virology on 22 January 2018.
Missing mouse model experiment results
What is striking is the lack of discussion of experiments or results from hDPP4 mouse model experiments. The EHA report clearly states that transgenic mice that express human DPP4 had been constructed and were being bred in year 3 at the WIV. The stated aim of year 4 research was to use these mouse models for in vivo MERSr-CoV research. Yet none of this work was reported in the year 4 or year 5 period. The NIH funded research with the stated aim of infecting hDPP4 mice with MERSr-CoVs. Did the NIH ask EHA where the results of the work they paid for are?
MERS-CoV research outside of EHA/NIH programs
What is clear is that undocumented research using HKU4r-CoV infectious clones was underway in Wuhan (likely at the WIV given it was a ‘CoV lab’) in late 2019/early 2020, just after the end of the Year 5 “Understanding the Risk of Bat Coronavirus Emergence” NIH sponsored EHA project.
Indeed, our findings indicate MERS-related research by the WIV and associated institutions in Wuhan may have been more extensive than has been documented (eg Wu et al. 2016; Luo et al. 2018).
SARS-CoV-2
A key point made by multiple virologists is that SARS-CoV-2 could not have been bio-engineered in a lab because only the WIV1 reverse genetic system had been documented prior to SARS-CoV-2.
“one would first have to demand that scientists have built a completely new reverse genetics system from a previously unknown virus and then introduced a Furin site there.” Drosten, 2021.
Clearly, a re-think is required as to the possibility that undocumented SARSr-CoV reverse genetics systems may have been in active research by the WIV in 2019. More answers need to be requested from the NIH, and the EcoHealth Alliance as to what samples and sequences they have and what they know about undocumented CoV research at the WIV.
