Comparative analysis of the SARS coronavirus genome: a good start to a long journey.

Published online May 9, 2003 http://image.thelancet.com/extras/03cmt124web.pdf

Severe acute respiratory syndrome (SARS) is the latest in a long list of emerging infectious diseases to gain the attention not only of the scientific community but also of the general public. SARS is characterised by symptoms that include a temperature over 38°C, dry cough, myalgia, and mild sore throat, which progresses to atypical pneumonia. The disease is unusual in its high morbidity and mortality rates; and it mimics and indeed rivals the 1918 influenza pandemic, which still stands as the worst pandemic in history. More disturbingly, SARS seems to transmit through direct contact and extreme infection-control measures are thus needed in affected areas.

Although SARS currently has no absolutely confirmed cause, a first approximation of Koch's postulates has been satisfied for a newly discovered virus within the family Coronaviridae. 3, 4 This newly identified virus has been designated SARS coronavirus (SARS-CoV). Coronaviruses are large enveloped viruses with a positive-sense RNA genome ranging in size from 27 to 30 kb, the largest of any of the RNA viruses. Clinically speaking, coronaviruses are usually associated with the common cold in human beings, but in animals the virus can lead to highly virulent respiratory, enteric, and neurological diseases as well as hepatitis. Human coronaviruses are usually difficult to culture in vitro, whereas most animal coronaviruses and SARS-CoV can be easily cultured in Vero E6 cells.

Recently, in a Canadian collaborative tour de force, the Michael Smith Genome Sciences Centre in Vancouver, British Columbia, and the National Microbiology Laboratory in Winnipeg, Manitoba, were the first to obtain a full sequence of SARS-CoV.8, 9 With 29 751 bases, the genome, denoted Tor-2, possesses a classic coronavirus complement of 11 open-reading frames, S and E glycoproteins, and the matrix, replicase, and nucleocapsid proteins. However, at the aminoacid level, SARS-CoV has minimum homology with any of the three classes of coronavirus; SARS-CoV thus lies in its own group.

In today's Lancet, YiJun Ruan and colleagues take the sequences of five SARS strains of known passage history in human beings and compare them with the Tor-2 sequence as well as other fully sequenced SARS-CoV strains from China and Vietnam. In addition to identifying small clusters of inherited mutations that are lineage-specific and can be used as molecular signatures to track transmission, the results suggest a remarkable genetic conservation of the virus since the outbreak was first documented in February, 2003. Although there are a handful of mutations between isolates, they may be due to adaption in culture. Previous experience with the evolution of influenza A virus shows that such “end-of-lineage” changes are due to mutations selected in laboratory culture. In Ruan's study, most of the mutations in the SARS-CoV sequences probably represent adaption to Vero cells during propagation before sequencing. It can therefore be concluded that as the virus passes through human beings, SARS-CoV is maintaining its consensus genotype and is thus well adapted to the human host. This finding may indeed be a double-edged sword.

SARS-CoV is not likely to change rapidly and thus may not readily mutate to a benign infection, as seen with most other infectious agents transmitted by direct contact and where patients’mobility is crucial to facilitate transfer of disease. Any disease property that enhances transmission to new hosts will be the subject of positive evolutionary selective pressures. Experimental analysis of viral evolution shows that genomic stability in a specific host is a function of viral population size-passage of large populations results in increased replicative fitness and virulence due to the competitive selection of optimising mutations. By contrast, passage of limiting dosages results in decreased replicative fitness and attenuation due to the adventitious trapping of deleterious mutations. The hope with SARS-CoV is that it will attenuate its properties in human beings. Sadly, Ruan and colleagues’results suggest that this attenuation does not seem to be happening at a rapid rate and may indicate a worrisome epidemiological trend.

Yet, there may be a bright side to the evidence reported by Ruan and colleagues. The apparent genetic stability makes a vaccine seem more achievable. However, coronaviruses can transduce genes from foreign genomes, recombine with other coronaviruses, and mutate to drive rapid and extreme evolution-presumably SARS-CoV is a product of such evolution. Furthermore, SARS-CoV was found in the faeces of patients, which indicates more than one tissue tropism. This discovery also raises the slight possibility of transmission through water in which the property of high virulence can be maintained under conditions that allow untreated sewage contamination of surface and ground waters. In addition, animal and human experiences with coronaviruses have illustrated a range of problems, including a lack of protective immunity, persistent infection, and immune enhancement (in which antiviral antibodies increase disease progression). This latter problem may explain in part the relatively high mortality rate. If a vaccine is possible, there is a long way to go.

There is also one more very puzzling aspect to the SARS story, which comes to the heart of the confusion and concern in the scientific and medical communities. Where and how did SARS arise? A great deal of retrospective isolation and genomic sequencing of human and animal coronaviruses, especially in the virological hotbed of southern China, is required for the molecular sleuthing needed to solve this puzzle. SARS-CoV possesses a novel organisation of five open-reading frames that encode proteins with as yet unknown functions which, from experience with other viruses, may encode antagonists of innate and adaptive immunity. Much like the human genome, the SARS-CoV genetic code charts a detailed roadmap through unknown territories. Molecular virologists now need to interpret gene function from this genetic sequence, which will not be easy given that single aminoacid changes can dramatically affect biological function.

As the SARS outbreak looks to be contained in the more technologically advanced regions of the world, complacency must be avoided. That a coronavirus with high virulence has arisen should not be a surprise, because historically the worst epidemics of respiratory disease have come from the influenza A viruses, which ominously share several biological features with coronaviruses. Both viruses are zoonoses, both have instances of dual tropism for respiratory and gastrointestinal tissues, and, most importantly, both possess mechanisms for the generation of extreme genetic variability. Models of influenza might guide the understanding of the SARS coronavirus.

There is a short period of grace in which to combat SARS. The major questions about SARS-CoV control and biology, such as mode of transmission, spread, and mechanisms of virulence, need answering—before the next time SARS rears its ugly corona.