From camels to humans: this is how MERS do it

From the Bedford laboratory, Vaccine and Infectious Disease Division

Middle East respiratory syndrome coronavirus (MERS-CoV) was first identified in 2012 during an outbreak in the Arabian Peninsula. Since then, the virus has been endemic in camels of that area and has infected over 2,000 people and caused more than 700 deaths. Deaths usually occur in high risk individuals, and are often associated with comorbidities, suggesting some cases may have gone undiagnosed leaving epidemiologic data incomplete. In some cases, contact with camels was associated with infection; however, not every human case could be traced back to livestock interaction. In order to further understand past outbreaks and identify possible future threats to the population, scientists from the Bedford laboratory in the Vaccine and Infectious Disease Division at Fred Hutch studied the published genome sequences. Using sequences from both humans and camels, they explored the evolution of the virus and sought to understand the transmission between camels and humans; this work was recently published in eLIFE.  

Figure: Typed maximum clade credibility tree showing both camel and human MERS-CoV genome sequences. Color denotes the confidence of sequence assignment to host with gray being increased uncertainty. Data show the small clusters of human sequences (blue) within the larger more diverse camel (orange) branches.

Using 274 MERS-CoV sequences, the group used a structured coalescent approach to determine that most of the MERS-CoV evolution taking place in camels and human sequences is transient and terminal. Of the 174 sequenced MERS-CoV genomes isolated from humans, the group estimated that there were 56 separate camel-to-human transmissions (see figure). This was demonstrated by the human viruses emerging as clusters of related sequences that were within the diverse (non-clustering) camel sequences. These data suggest that there may be more camel-to-human case introductions than reported by epidemiologic data. Using the posterior distribution of MERS-CoV introduction events, the group modeled seasonality in zoonotic transmission. This led to the identification of a four-month period when transmission was highest (April-July) which coincides with the loss of maternal antibodies in newly calved animals. This suggests that once there are new susceptible animals in the population (calves), there is a rapid increase in camel infection which is spread to the human population due to heightened viral presence in the host population. To determine the reproductive number (Ro, the number and distribution of total cases resulting from a single case) for MERS-CoV in humans, the group used a Monte Carlo simulation. This resulted in a Ro of below 1 which suggests that the virus could not be sustained in the human population (a primary case would transmit to less than 1 person). This fits with the epidemiologic data, when one takes into account the bias towards larger human outbreaks occurring in hospitals, which could be attributed to poor hygienic practices.

            Overall, these data suggest that the main brunt of MERS-CoV infectivity and evolution takes place in the camel population and that humans appear to be dead-end hosts. These data support previous hypotheses that focused on controlling infection in the camel population and thus reducing human introductions and human infection. By focusing on the camel host, especially during the post-calving season, the threat of human transmission could be mitigated.

Dudas G, Carvalho LM, Rambaut A, Bedford T. 2018. MERS-CoV spillover at the camel-human interface. Elife, 7.

Funding was provided by the National Institutes of Health, Pew Charitable Trust, European Commission, Wellcome, and Fred Hutchinson Cancer Research Center.