Whoever lacks the capacity to put on blinders, so to speak, and to come up to the idea that the fate of his soul depends upon whether or not he makes the correct conjecture at this passage of this manuscript may as well stay away from science.

Science as a vocation

Table of Contents

Information on the international outbreak of the illness known as Severe Acute Respiratory Syndrome (SARS)

© 2003 and Disclaimer by Antoine Danchin. All Rights Reserved.

This information was provided "as is" during and immediately after the outbreak of SARS.

Since then another lethal coronavirus spread in a more limited way, the MERS (Middle East Respiratory Syndrome) coronavirus.

A regular update of epidemics monitored from a variety of sources

Important links

Mainland China

Official update of the situation in China

Department of Health, HKSAR (Chinese) (English)

Atypical Pneumonia Guidelines for Workplace (Chinese) (English)

World Health Organization

Centres for Disease Control and Prevention, USA

ProMed (Global Electronic electronic reporting system for outbreaks of emerging infectious diseases & toxins, open to all sources)

An excellent bulletin for companies describing the epidemic and updating information about SARS for persons developing business with Hong Kong and Mainland China

A reference treaty, in several languages, about SARS and its treatment

A description of the situation and questions in Hong Kong is found here

Faculty of Medicine The University of Hong Kong
Fight against Atypical Pneumonia

A site created by young people exploring information available on the Internet

SARS, An Open Scar

The epidemic

Note that a day to day remarkably insightful analysis of epidemics in the world (including animal and plants) is proposed at the PROMED site (see also the WHO site for a different account. The present summary incoporates information from forums held on the Internet in Mainland China).

On 11 february 2003 an outbreak of "atypical pneumonia" in Southern China culminated after having developed without much notice since mid-november of the previous year in the Guangdong province. This led the French General Consulate in Guangzhou to close the French School in the city, until more was understood about the disease. Its Japanese counterpart behaved in the same way. The rumor spread that several tens or even hundreds of people had been contaminated, with a high death toll. The most remarkable feature of that particular disease was that medical and paramedical staff of several hospitals had been simultaneously affected, some with very severe symptoms. Unfortunately no reliable information was available at that time and observers had to rely on the testimony of parents or friends of the victims, sometimes in an indirect way. This led to hints that this was only a rumor, perhaps triggered by companies interested in selling antiviral drugs. Indeed, it was likely that the outbreak was caused by a virus, perhaps a new flu strain (but perhaps "ordinary" flu): the H5N1 deadly influenza virus had just reappeared in Hong Kong, and this was a matter of great concern. This outbreak fitted the time of the year, and the possibility of hospital contamination. The significant death toll (rumor gave figures as high as 5 per cent) were consistent with flu-like virus since flu is indeed a deadly, albeit ordinary, disease... In the absence of active medicines, the local population rushed to buy traditional drugs, in particular vinegar, that was used for fumigations, and disappeared rapidly from shops. Several accidents, some fatal, were attributed to unsafe use of fumigation.

On February 22nd, the initial patient in Hong Kong, a medical professor from Guangdong (Pr JL L, of Zhong Shan Da Xue in Guangzhou) who was admitted in serious condition to Kwong Wha hospital warned the staff attending him that he had caught an extremely contagious disease while attending patients at the hospital where he worked. He said he already had symptoms on February 15th, indicating that he would have probably contaminated other people during the past week or so. He was placed in isolation and his health deteriorated until his death on March 4th. There, despite precautions, he contaminated four staff and family members. While other patients with a similar disease had already been placed in hospitals throughout the world , this initial patient was only later found to be the cause of the outbreak because he probably contaminated them while staying at the Metropole Hotel in Kowloon, in the elevator, or waiting for the elevator. One reason of missing the Metropole connection early on was that exactly at the same period a father and his son (the nine year old child had been admitted for treatment on February, 19th) died of H5N1 flu, causing an intense concern that the disease, that had spread from poultry to humans, might shift its host range, and pass from humans to humans. It was however almost immediately recognized that the "atypical-pneumonia" affecting people in Guangdong Province was different from H5N1 flu.

Indeed all index patients for known outbreaks elsewhere in the world (Viet Nam, Singapore and Canada) could be traced back to a common source, the visit or the stay they made to the Metropole Hotel in Kowloon, perhaps significantly in the smoking zone of the Hotel. One of those, who visited the Hotel on February 21st, 22nd and met a sneezing man in the elevalor, was admitted with high fever and pneumonia to the Prince of Wales Hospital in the now infamous ward 8A on March 5th. There he directly or indirectly infected 70 hospital staff, and indirectly eight of their children, 17 medical students and possibly even the Hospital authority Chief Executive William Ho. Several days later his sister feels unwell and is take to Princess Margaret Hospital, after having unfortunately infected at least three of her colleagues. These patients are distributed in other hospitals, including Pamela Youde Nethersole Eastern Hospital, where a staff attending one of the patients transmits the disease to her son... In the mean time a second patient, a tourist from Vancouver who had stayed at the Metropole Hotel is admitted to Saint Paul's Hospital, where he infects three staff. Another guest from the Metropole met a man who sneezed in the elevator where she was. On February 23rd she flies back to her home in Toronto where she infects five relatives, starting the outbreak at the Scarborough Grace Hospital there. She passes away on March 5th, her son passes away on March 13th. Index patients four, five and six were young tourists from Singapore who left the Metropole on February 25th. They later infect 17 persons when they returned home and are admitted to hospital. One of the medical staff, a doctor who leaves for New York after treating the patients, feels ill on the plane and is taken off the plane at Frankfurt, in Germany, where he is treated.

The outbreak is still ongoing in Singapore. Finally, patient seven, a businessman from Shanghai, who also stayed at the 9th floor of the Metropole, flies to Hanoi, where he starts the outbreak there with more that 50 people affected in two hospitals, passing away on March 13th after having returned to Hong Kong to be treated at the Princess Margaret Hospital. As a new unrecogized disease, Severe Acute Pulmonary Syndrome (SARS) was first identified in Viet Nam on 28 February, when Dr Carlo Urbani, an epidemiologist from the Hanoi WHO office examined a patient with a severe form of pneumonia for which no etiology could be found. On March 10th, 22 hospital workers in Hanoi French Hospital were ill with a similar acute respiratory syndrome, and by March 11th similar outbreaks had been reported among hospital workers in Hong Kong. The Italian Medical Doctor who recognized the disease in Hanoi, and triggered the Global Alert issued by the World Health Organisation passes away in a Hospital in Bankgok on March 29th. Three French doctors who caught the disease there while attending patients were forbidden to leave Hanoi at the end of March, to comply international regulation rules controlling the spread of highly contagious diseases. One was in serious condition and died in Hanoi on April 12th.

On March 17th the WHO issues a global alert about the disease.

On March 27th there had already been 516 cases world wide (not counting the cases from the Mainland) and 18 deaths. It took some time to get access to the complete record of the medical data in the Mainland. On March 30th , including the recognized Mainland cases, one already counted more than 1,600 cases and 58 deaths.

April 1st. Application of the standard SIR (Susceptible-Infected-Removed) epidemic model to study the recent spread of the Severe Acute Respiratory Syndrome (SARS) in Hong Kong by scientists at the Department of Mathematics of the University of Hong Kong revealed features that are a matter of some concern. The SIR model studied the spread of SARS at Amoy Gardens, a housing estate in Hong Kong. The SIR model has been successfully applied to model an influenza epidemic in an England Boarding School in 1978. Comparing the parameters estimated in these two cases the mathematicians found that they match with each other (same order of magnitude). Especially the infection parameters in both cases are close to each other. This suggests that both diseases may have similar paths of transmission.. If this is the case, then it suggests that the disease of more contagious than previously suspected. A privileged route might be from face, to hand, to object, to hand, to face, in addition to sneezing and coughing (which to happen in the later phases of the disease) .

It must be stressed that these conclusions are limited to the very specific case of high rise buildings, where elevators, and in particular the "maintain door open" button in elevators is a frequent intermediate contact between persons. Direct generalisation to other situations is not possible. However one must note that all types of behaviours involving hands may be dangerous at some point (e.g. publicly used toilets, including in planes, trains etc). It is essential to remember that the structure of (local) societies is extremely important in the spread of diseases. [PDF of an article written before the present epidemic]

At this time two articles have already appeared in the New England Journal of Medicine, on the epidemiology of the disease, propagating ideas as fast as the virus itself — with some delay, perhaps causing trouble in the interpretation of the facts. In one of these articles, coming from Hong Kong, the conclusion is "SARS appears to be infectious in origin. Fever followed by rapidly progressive respiratory compromise is the key complex of signs and symptoms from which the syndrome derives its name. The microbiologic origins of SARS remain unclear." In the second article, coming from Canada, the conclusion is "SARS is a condition associated with substantial morbidity and mortality. It appears to be of viral origin, with patterns suggesting droplet or contact transmission. The role of human metapneumovirus, a novel coronavirus, or both requires further investigation." One of the interesting lessons we shall need to investigate after the outbreak is over is the role of publications in the reaction to diseases, as well as the underlying motivations of journalists. As on April 12th it cannot be told whether the disease is spreading in a linear or in an exponential fashion. The increase in the recent days is a matter of great concern, since it suggests exponential growth. If the total number of cases reached 5,000 by the end of the month, this would suggest that the disease may have come out of control. Mathematicians at the Department of HKU are investigating the possibility that the disease would result from mutations derived from a rather innocuous highly contagious disease (with oral-faecal transmission route) that would have changed its tropism by mutation. This has already been observed in livestock in the past few years, with coronaviruses in particular. On April 17th the perception of the epidemic was quite different in Guangdong and in Hong Kong. This is for example a mail by a Professor at Zhong Shan University in Guangzhou (not a scientist), reflecting the widespread belief there: "Scientists here have produced the virus and found an easy and also the quickest way to identify SARS. It is almost over as far as Canton is concerned. It is a big problem in Hong Kong where the virus seems to have changed a little bit. In the mass media the whole event seems to me more a political myth than a real disease." In Beijing, curiously, the population appeared to be showing much more anxiety. This points out the extremely important sociological factor in any disease, epidemic in particular. At this date the current evidence remains that respiratory droplet transmission, requiring close contact with a case of SARS, is the predominant mode of transmission. The outbreak at Amoy Gardens is now understood: it is linked to the way the sewage system is arranged at this estate. This epidemiological study points to a strong link between the respiratory virus and virus infecting the gut, as previously suspected. A remarkable feature of the disease is its uneven spread. At the date of May 2nd comparing Shanghai (2 cases) and Beijing (1,636 cases) is extremely surprising. In the same way one must remark the very small number of children involved in the cohort of patients, in particular patients with severe symptoms. This is very unusual.

The virus

Although highly suspected, the actual cause of the disease is not absolutely confirmed on March 30th. It may appear strange to laymen that it took some time to identify the cause of the disease. In fact, progresses have been remarkably rapid. This is only because of the exceptional progresses of basic academic research that scientists were able to do so, in a very short period of time. Lack of response of the patients to antibiotics indicated very early on that the disease was not caused by bacteria (some, in Guangdong were of course afraid of anthrax, because of the scare of terrorism associated with pneumonia after september 11th). The symptoms, and the rapidity of onset and contagion, were compatible with viral transmission. However, they did not fit with the behaviour of known viruses. This period of the year witnesses many pneumonia, often of bacterial origin, but it is most unusual that the medical staff and families would be contaminated at such a high level. The initial symptoms are similar to those of flu, but in this case pneumonia is far from being systematic. Using all means at their disposal (in particular diagnostic tests derived from many years of academic research in immunology and enzymology, and highly sensitive test for the genetic material of the virus) scientists rapidly concluded that the disease was not caused by an influenza virus.... Now, it became difficult to guess what it was, especially if it were a new species. The first method of choice is to use samples extracted from patients (nasal secretions and lung washes) to see, using electron microscopes, whether some object stands out. In a first set of experiments, in particular at the Chinese University of Hong Kong, in Germany and in Canada, viruses of the widely spread family of Paramyxoviridae appeared in the samples collected from a number of patients. This family, that comprises the ubiquitous viruses causing measles or mumps, appeared to be convincing, because of the variety of their targets, and because of their highly contagious nature. Caution was however warned because humans are living with a very large number of viruses of this family, many of which are not highly pathogenic. electron microscopy is not discriminatory: it shows all what is present in the sample and can be stained for this technique.

The original pattern of infection suggested either a mutant of a known virus — but none was recognized — or a new virus type. In this context, the virus identified in the SARS outbreak in Canada (March 22nd), a metapneumovirus (6 identifications out of 8 cases), which is also a virus from the paramyxoviruses family, looked as if it were a good candidate. This new virus had been discovered in 2001 as usually the cause of common cold-like diseases, fitting with the onset of SARS. The World Health Organisation however warned that one should find the same virus in all connected patients to propose that a given virus would be the cause of the disease. It organised an international team of collaborating laboratories, working on viral respiratory diseases, and designed probes (nucleic acid primers that could be used in hybridisation experiments) to identify common viruses in the samples isolated from a variety of patients in the world. One couple of such probes allowed scientists, in particular from the CDC in Atlanta and from the team of Pr Malik Peiris at The University of Hong Kong, to suggest that the virus was, in fact, a coronavirus [History of discovery: PDF]. Coronaviridae constitute still another family of viruses, very well known from scientists investigating respiratory diseases since they cause about one third of the epidemics of common cold every year. Interestingly, among the various primers proposed by the WHO, not all appeared to hybridize with the virus in the patients samples, indicating that it was a new type of coronavirus. One should note however that it is not unusual that severe infections are caused by two different viruses infecting the same patient. On March 28th Pr Peiris in Hong Kong and Pr Fleischer in Hamburg concluded, from samples isolated from different patients that the cause was probably a new type of coronavirus for which they proposed two types of preliminary diagnostic tests, a diagnostic based on immunology, and a diagnostic based on the highly sensitive Polymerase Chain Reaction, the famous PCR. Several teams in the world proceeded to sequence the virus genome, which should be of the order of 30,000 RNA nucleotides. This should allow one to devise extremely accurate tests for diagnosis (a new more accurate PCR test was announced by the CDC on april 14th ) and would be helpful in the future if it were important to create a vaccine against the disease. Here we need a word of caution: it should be borne in mind however that viral diseases are often not well adapted to vaccination (see the repeated failure of finding a vaccine against the AIDS virus) and we do not have, in general, good antiviral agents. Those which exist are not very effective, and when they are they induce a wealth of unwanted dangerous side effects. Some coronaviruses respond partially to treatment, however. The conclusion at this point is that the most important move now is to take appropriate prophylactic measures, and to learn lessons from this outbreak to plan for the next that, undoubtedly, will come sooner or later.

Where does the virus come from? This anecdote, reported by the Journal Ming Bao on march 29 is interesting: "WHO pointed out in earlier days that SARS was originally detected in FoShan and HeYuan county, Guangdong Province. A medicial professor in RenMin hospital of HeYuan county named Xie JinKui does not agree with that assumption. Prof. Xie said that the first case in HeYuan was detected on 17th of Dec, last year (2002). The patient is a 35 year old man. He is a cook in ShenZhen, working in a restaurant where he has close contact with wild animals, such as snakes. According to Prof. Xie, before he visited RenMin hospital in HeYuan, he went to FuTian hospital in ShenZhen on 15th of Dec, 2002. Prof. Xie emphasized that the patient's health was very bad when he was first sent to RenMin hospital, the symptoms were high fever, breathing difficulty, and shadows in chest X-ray test. On the 2nd day, the patient was sent to the GuangZhou Military hospital, where he infected serveral medical staffs. Furthermore, his wife and two sisters were also infected. Prof. Xie said that the first case of HeYuan is from ShenZhen. However, from the medical record, several original patients have close relationships with wild animals. Some of them are cooks and some of them are providers of wild animals to restaurants. Prof. Xie said at last that HeYuan is the first county who reported SARS to Guangdong local medical officials. But he emphasised that this does not mean that SARS is originating from HeYuan. After the talk with Prof. Xie, reporters of Ming Bao also visited the FuTian hospital in Shenzhen. And they did find the record of the patient that Prof. Xie mentioned, but the date was 20th August, 2002, not 15th Dec, 2002. Finally they even found the patient himself: he admitted that he got pneumonia last yeat, but refused to give more details." indicates that exploring the idea of an animal origin for the virus is worth investigating. This is a theme that has very much in common with the spontaneous attitude of Man towards Nature. It must also be well understood that in Guangdong, the habit to eat all kinds of wild food is widely spread, and might have indeed an importance in the transfer of animal viruses to humans. Scientists in the Netherlands were indeed able to switch the host of coronaviruses just by swapping one of their genes. This is the experimental proof that the host range of viruses may be easily changed by recombination: there is no doubt that this will be the sources of new epidemics, and the present one should be a lesson for the future. A further observation demonstrated, in pigs, that coronaviruses infecting the gut (causing gastro-enteritis) would, by simple mutation, change their tropism to the lungs. This makes the hypothesis of a virus with an oral-faecal route of transmission might be the present cause of SARS, perhaps after mutation from a benign strain (gastro-enteritis was frequent in the Hong Kong region during the past few months).

On April 12th it is still not yet known whether the coronavirus that has been identified is the cause, and the only cause of the disease. The role of metapneumoviruses is still explored, and different isolates of coronaviruses are investigated. In addition, apart from PCR (which gives many false negatives) no diagnostic tool is yet able to identify the disease at an early stage. Note that the coronavirus as the (only) cause of the disease is still disputed... Two isolates have been sequenced, by a Canadian team on April 13th, and the CDC in Atlanta on April 14th. This rapid success is due to excellent collaboration between the teams, with efficient sharing of information. The virus is clearly a coronavirus, but not of a previously known type. Its origin is therefore a puzzle. The virulence of the virus suggests that it does not directly come from humans, but rather from animals. One would rather expect warm blooded animals (mammals or birds) rather than cold blooded animals such as reptiles, but this has to be investigated in details. On April 16th a team in Rotterdam, backed by the world-wide consortium created by WHO demonstrated that the new coronavirus isolated from patients with SARS is the cause of the disease, as it was able to infect monkeys and create the symptoms of SARS. According to Professor Rabenau (April 22nd), the virus can stand dry on a surface at room temperature and retain infectivity (2 logs lower) after 48 hours. Under similar conditions HIV would not stand for more than one hour. The virus is remarkably stable for a virus of its category (May 3rd).

Possible origin of the virus. According to the China Post in Taiwan, scientists in Guangdong province traced the virus that causes SARS to the endangered civet cat. They found SARS antibodies in traders of wild animals who did not develop symptoms of the disease. "Of the 5 traders identified, 4 worked with rabbits, cats and other wild animals, while one of them worked with poultry and other wild birds," the report said. None of them, however, developed any of the pneumonia-like symptoms of SARS, He Yaqing, deputy director of the Shenzhen Center for Disease Control (CDC), told Yangcheng Evening News on May 24th. These findings suggest that the form of the coronavirus suspected to have jumped from either the civet cat or the raccoon dog to humans was less virulent than the SARS coronavirus transmitted among humans. After jumping from animals, the SARS virus mutated and became more lethal to humans, the investigator said. This is consistent with the "double epidemic" pattern presented below (since the beginning of April), as genetic studies showed that the coronavirus isolated in wild animals had 29 more nucleotides than were found in some SARS patients in southern China, suggesting a change of tropism by deletion. "The survey of wild animal traders with SARS antibodies shows that these traders once had SARS, but became infected without knowing it and cured themselves without ever manifesting clear symptoms, the report said. The Shenzhen CDC did not say how many traders were tested for SARS antibodies, nor was it clear if the traders were a source of transmission of the global SARS epidemic."

A lesson from outbreaks in domestic animals: a Human Respiratory CoronaVirus (HRCoV) hypothetical scenario

Domestic animals live in strong relationship with humans. They also live as highly concentrated populations, somewhat similar to those that humans have in cities. We are animals. It is therefore extremely interesting to analyse the birth and propagation of epidemics in livestocks: chicken, pigs and cattle in particular. Not only is this interesting as models of sudden microbial outbreaks, but also because viruses and bacteria often change hosts, and shift from an animal host to humans, while becoming for a while highly virulent. Emerging diseases usually appear in animals before reaching humans. Dr Hubert Laude from INRA, in France, has devoted many years of work to viral outbreaks in pig farms (see bibliography at PubMed: sequence, immune response, molecular biology). The present summary of an oubreak of a mild respiratory disease caused by a coronavirus, the Porcine Respiratory CoronaVirus, PRCV, that spread world-wide after 1984, is based, with new interpretations on his review article (published in French in Virologie, 2:305-316, 1998). This does not prove that the present epidemic is of the same type (the origin of the virus was not yet known at the end of April 2003), but it shows how coronaviruses evolve, in real situations. We should be well aware that livestock is permanently infected by old diseases (such as foot and mouth disease, endemic in Asia and recently causing a dramatic outbreak in UK) and new diseases that seem to appear from nowhere.

For example, in 1991, swines were infected by a totally new virus, PRRSV, that spread throughout Europe and the USA. Less dramatic in its consequences, but a lesson for today's apparition of SARS in humans is the apparition in 1984 of a porcine respiratory disease that did not cause deadly symptoms, but that became a concern because it was able to infect the whole of the pig breeds in Europe, within a few months. Until that date pigs were infected by a highly contagious gastroenteritis virus, the Transmissible GastroEnteritis Virus, TGEV, a member of the Coronaviridae. This disease was a matter of great concern because it killed piglets with an efficiency that reached 100% in some farms. A serological test was devised to identify contaminated stocks. In 1983-1985 a puzzling observation was made: many farms where animals tested positive did not display symptoms of the disease. It was therefore apparent that a variant of TGEV had appeared, causing mild symptoms. The virus was isolated in Ghent, Belgium, in the laboratory of M. Pensaert. It was found not to cause gastroenteritis, but, in contrast, mild respiratory symptoms. Rapidly however it was found to be highly related to TGEV, and not only immunologically, but in its genome sequence as well. The gut-tropism had evolved into nasal and lung tropism! This demonstrates that coronaviruses can change tropism by mutation, a very dangerous feature that may happen during the progression of the disease. One should also remember that these viruses are extremely prone to recombination, so that, in case of co-existence of two epidemics, they might shift tropism from one site to another one by simple recombination. A simple scenario would be gut tropism (large S protein) => lung tropism (deletion in S) => back to lung tropism (by recombination)... How could this happen?

Coronaviruses are RNA viruses that mimick messenger RNA and develop in the cytoplasm of target cells. The sequence of their genome can be determined by standard techniques, adapted to the situation of RNA molecules. A single, long, RNA (about 30,000 nucleotides) is translated into 7 to 11 proteins, depending on the strain. The two first very large proteins (1a and 1b) have activities permitting processing of both the RNA, and the proteins into smaller products, as well as permitting replication of the viral genome and its encapsidation into a crown-like envelope (hence the latin name "corona"). The third protein, S, constitutes the spikes that attach to the surface of the host cells, allowing engulfing of the virus into their cytoplasm, where they multiply. In the enterovirulent strain S is made of a long protein, comprising more than 1400 amino acid residues, that fold and assemble in groups of three, making the spike. Remarkably, in the respiratory strain, S was much shorter, about 1200 residues in length. And this resulted from a deletion of a piece of the genome in the S region, corresponding to the amino-terminal end of the protein, just after the signal that targets it to membranes. The result of this deletion is that the spikes are no longer able to attach to their normal host cells in the small intestine, the enterocytes, but, rather, can colonize the internal surface of the nose, the bronchae and the lung alveolae. The new virus is no longer TGEV, but a Porcine Respiratory CoronaVirus...

The story does not stop there (although, quite unfortunately for our knowledge about SARS, most basic research on the disease stopped when it was found that PRCV did not result in a considerable loss in pig production). Scientists discovered that this mild disease appeared at many places in the world, and naturally, they sequenced the corresponding viral genome isolates. The surprise came when they discovered that, rather than one single deletion event, several, probably independent, events caused the same change of tropism from the gut to the lungs. Thus, from an original virus, with gut tropism, many variants appeared, most of them with a large deletion in the amino-terminal end of the spike protein. This indicates that the original TGEV could continuously result in stable genetic variants (mutants) that would infect lungs instead of intestine. In the case of pig breeding this was not too much of a concern, since the gut form was much more severe than the respiratory form. Could not we predict that SARS reflects a similar story, with a different conclusion: the gut form would be the mild one, while the lung form is severe? This is the Human Respiratory CoronaVirus (HRCoV) scenario. If it is substantiated the shape of the outbreak may be easily explained, including variations in local virulence, as well as modes of transmission. The origin of the virus is not known, it may be linked to mammals, but it also may come from birds. It may come from domestic animals, but it may come from wild animals as well, since Guangdong people are known to be fond of all kinds of food. After the sequence of the virus was known it is clear that it differed from all known viruses, except that its spike protein, S, is somewhat similar to that of the coronavirus that causes mouse hepatitis. It could therefore come from a rodent, or be a recombinant between a rodent virus, and that of a cat (cats eat rodents, men eat both rodents and cats...). This is why the pedigree of the virus will be most interesting.

With this hypothesis, can we be optimistic? Yes, we can, although, if we take lessons from what happened to pigs, it seems likely that the disease will spread throughout the world. The first reason for optimism is that immune cross reaction between the gut form and the respiratory form may result in protection against SARS of all those who have first been in contact (sufficiently early on) with the gut form. This would create a pattern of double epidemic, with a somewhat unusual pattern of outbreaks, with regions "naive" for the gut form, the most susceptible to SARS. A second reason for optimism is that the likely target of the spikes is probably known, it is a protein named "Neutral Amino Peptidase", that acts as a receptor on epithelial cells. Knowing the receptor it will be possible to construct molecules that will bind to it and prevent binding of the virus. This will also help in discovering which part(s) of the S protein are involved. Mimicks of these parts would be excellent candidate for blocking the receptor. Unfortunately, most work on the subject has been discontinued in the past few years, because the fashion that plagues support to research did not go in this direction... However we must be vigilant: if the TGEV form segregates continuously mutants (deletions in particular), that result in HRCoV, then the disease, spread extremely rapidly through oro-faecal transmission, might suddenly appear again at places where it was supposed to have been eradicated. We need more research to understand this phenomenon, remembering that the rate of stable, seleceted, mutation per nucleotide in an RNA virus is of the order of one thousandth per position and per year, a very high figure. And nothing is known yet about the rate of deletions... One should note that using analogs of the building blocks of RNA as antibiotics, as one is tempted to do, may in fact be counterproductive, by favoring the apparition of deletions. All protocols should be experimentally tested before being used in humans. Acting too fast, with too limited knowledge, can be very dangerous in the long term. And remember that this virus is only one in the millions of microbes that may suddenly change their lifestyle and become virulent to humans. In the same way as we cannot predict where discovery will happen, we cannot predict from which side microbes will suddently invade us. Work on the virus develops at many places in the world. Data begin to indicate that it may have unusual stability on surfaces for a virus of its type (April 23d).

Two very informative articles: Vet Res 1993;24(2):125-50
Porcine respiratory coronavirus: molecular features and virus-host interactions.
Laude H, Van Reeth K, Pensaert M.
INRA, Unite de Virologie et Immunologie Moleculaires, Jouy-en-Josas, France.

Since 1984, a previously unrecognized respiratory coronavirus, causing a mostly unapparent infection, has rapidly and massively spread within the swine population in Europe, and few years later, a virus with similar characteristics has been identified in the USA. The agent, designated PRCV, appears to be derived from the porcine enteric coronavirus TGEV. The aim of the present article is to review comprehensively the state of the knowledge about this new virus and its infection. The review includes the following topics: epizootiology, molecular characterization and antigenic features of PRCV, pathogenesis and clinical aspects, immunity and laboratory diagnosis.

The authors' views concerning the impact of the emergence of PRCV on both coronavirus research and swine production are presented in the conclusion. J Gen Virol 1990 Nov;71 ( Pt 11):2599-607 Porcine respiratory coronavirus differs from transmissible gastroenteritis virus by a few genomic deletions.
Rasschaert D, Duarte M, Laude H.
Laboratoire de Virologie et d'Immunologie Moleculaires, Centre de Recherches de Jouy-en-Josas, France.

The genome organization of porcine respiratory coronavirus (PRCV), a newly recognized agent which has a close antigenic relationship to the enteropathogenic transmissible gastroenteritis virus (TGEV), was studied. Genomic RNA from cell-cultured PRCV (French isolate RM4) was used to produce cDNA clones covering the genomic 3' end to the start of the spike (S) glycoprotein gene (7,519 nucleotides). Six open reading frames (ORFs) were identified that allowed the translation of three coronavirus structural proteins and three putative non-structural (NS) polypeptides, homologous to TGEV ORFs designated NS3-1, NS4 and NS7. Pairwise alignment of PRCV nucleotide and amino acid sequences with sequence data available for three TGEV strains revealed a 96% overall homology. However, the genome of PRCV exhibited two important distinctive features. The first was that the S gene lacked 672 nucleotides in the 5' region and encoded a truncated form of the S polypeptide, and secondly, the first NS ORF downstream of the S gene was predicted to be non-functional as a consequence of a double deletion. The significance of genomic deletions with respect to tissue tropism and evolution of coronaviruses is discussed. A consequence of the double epidemic hypothesis is that, when there is simultaneous infection by the enteric and the respiratory form, the disease may be much more severe. This may explain the Amoy Gardens event.

Summary of the scenario

In the "double epidemic" scenario an animal virus with gut tropism infects people in Guangdong, and the disease spreads as a gastroenteritis (significant symptoms, but not too severe, mild fever in general, lasts for a couple of days, the first incapacitating somewhat the people, sometimes much more severe). At some point last autumn the virus XECV (for "unknown" Entero CoronaVirus) mutates (probably by deletion of an amino terminal portion of the S protein, downstream from the signal peptide). This mutant, HRCoV, changes tropism, infects the respiratory tract and gives SARS. Now the two epidemics will spread more or less independently of each other. What we can suspect is the following: at the onset of the process the first persons had both viruses together, and this gave very severe symptoms, with a high proportion of deaths (this may correspond to the time of panick in Guangdong). Then, when the gut virus spreads alone, and infects people before they contact HRCoV they are protected, (some two weeks are probably needed to get a strong protection). In a "naive" population, which did not see XECV, HRCoV gives SARS, with a low death toll. In a population that just got XECV, the situation will be similar to the original one, with a high death toll. I suspect that this is what could have happened in Amoy Gardens. A completely different scenario (similar to that of dengue fever) fortunately much less likely, would be that the infection by XECV would make the symptoms of HRCoV infection much more severe. This has to be explored by epidemic models. Of course, an unrelated epidemic, with epitopes similar to those of the SARS virus might also be protective: this might account for the apparent lack of sensitivity of children to the disease, and the huge differences in the extent of the outbreak in different parts of China.

Let us be optimistic: I think that the situation in Vietnam allows us to be. Indeed we see that it is in the cleanest hospital that the symptoms were more severe, and that was the place were a real outbbreak took place. This suggests that the less hygienic conditions might have been ... protective! Perhaps because people were already infected by XECV, or by similar coronaviruses, which would confer some protection. This would also account for the apparent surprising difference between Guangdong and Hong Kong. The consequence is that dealing with hygiene will be extremely difficult: for mild infections lack of hygiene is good but as soon as a disease becomes dangerous stringent hygiene measures must be enforced! This shows that we really need deep reflection and research in that domain. Also, a consequence of this is that, when one deals with patients who have both SARS and diarrhoea one should try to isolate the virus both from respiratory secretion and from stool, and then sequence relevant parts (the obvious candidate is the S gene, but of course it would be best to sequence the whole RNA of the virus). A second optimistic view (to be challenged, of course) is that the double epidemic suggests immediately a route for vaccination. Finally, of course, this is just an hypothetical scenario, and much work remains to be done to see how adequate to reality it is. It must also be remembered (and experiments have to be performed to analyse the situation), that in some cases a first exposure to a disease, makes the second one more serious, rather than protect. If the "double epidemic" hypothesis is right, then this must be tested rapidly. In the best case, people who got first the "gut" form would be protected, and could be used as a sanitary cordon to contain the disease. In contrast segregation of "gut" tropism from the "respiratory" virus might increase virulence.

Return to the Journalists page....

The future: emphasis over hygiene and plead for more academic research

Common cold is extremely contagious. The most efficient mode of contagion is well known: from face to hands, to any object, to hands, to face... Naturally, droplets emitted during sneezing or coughing, when they reach the mouth, nose or eyes are also highly effective to transmit the disease. Fortunately the virus does not survive long outside of the human body. The virus survival is longer in colder places. One should therefore avoid low temperature air conditioning. In contrast, good aeration is essential, since a low concentration of viruses lowers its probability to infect persons in good health. For these reasons, the most important protection is washing hands, and, if infected, wearing a mask covering mouth and nose. Apart from psychological conforting effects, masks are not protecting very well unaffected persons, in particular because they do not cover eyes, and because they are so inconvenient that the persons wearing them have a tendency to touch their face, an act that should be avoided as much as possible. Usual rules of hygiene should be carefully followed (washing hands frequently in particular) and people should limit contacts with large concentrations of people. One should avoid travelling, to limit the spread of the disease. Also, remember that panic is one of the most efficient way to spread diseases. If having fever, one should not travel, but rather, use phones to discuss the situation, and be monitored by staff with appropriate protection. If having been in contact with a person who later on, was known to be infected, a similar behaviour should be followed. Everybody should understand his or her responsability towards the general public. The most surprising feature of the present epidemic is the role of medical doctors and medical staff in general in the extremely fast propagation of the disease. It is typical of these hospital acquired diseases (nosocomial infections) that now plague the majority of hospital in the developed world. It is therefore of the utmost importance to analyze the world-wide situation in this domain by re-implementing appropriate hygiene measures in the medical domain. Departments of Clinical Microbiology should have a leading role in this domain. One should also investigate all possible routes of contamination (including water and food at the work place). Buildings should be constructed according to rules that take hygiene as a major concern. Apart for ongoing work on infectious diseases world-wide we must certainly divert the course of what has become fashionable when circumstances require. Between the late fourties and the late eighties all renowned university in the world had an active Department of Microbiology affiliated to their Faculty of Sciences. And this department was at the forefront of what has become molecular biology and molecular genetics. Most major discoveries that allowed, today, to identify the deadly virus as fast as it was, were derived from techniques invented from the study of microbes and of their multiplication, including, naturally, microbes that have nothing to do with medicine: the role of Archaea, bacteria that as yet have never been found to be pathogenic, and that live in extreme environments, in the discovery of Polymerase Chain Reaction is a case in point in the situation we are witnessing today. This of course also involved Biochemistry, Chemistry and Structural Biology. Since the early nineties the situation changed dramatically: it was no longer fashionable to study microbiology, in particular basic microbiology and these departments disappeared. Also of course, it is indispensible to know how to grow the virus. This type of skill is not shared by many scientists however... Knowledge accumulated during more than one century on bacteria viruses and fungi became obsolete, and now that it will be necessary to invent new drugs and new fighting measures, the scientists who had the knowledge retired and passed away.... We are the victims of our interest for daily events, not for long-term plans. And it may not be by chance that the present outbreak occurs in Hong Kong, a place where value is stressed only on extremely fast returns... But what is wealth worth of, if one is doomed to be infected by the next deadly virus or bacteria? Should not we invest in the future? Any event that appears suddenly can only be dealt with properly if the field is ready. Let us see what happens with the present SARS epidemic. This tells us more about the future, and the reason why only long-term goals will, in time, have much importance, and let us see what areas are important:

Naturally we cannot predict the future, and we will only be able to tackle difficult situations if we have prepared the field by making appropriate discoveries. This, to my view, is what justifies basic academic work, and this must be understood by journalists, so that they can communicate this essential idea to the general public. Failing to do so will lead us to witness much deadlier diseases in the future. A special issue of the Journal of Applied Microbiology provides much information about the spread of new diseases, associated to the new structures of our societies.

See in particular: The traveller and emerging infections: sentinel, courier, transmitter
M.E. Wilson
Journal of Applied Microbiology; Volume 94, Issue 1, Page 1

Remarks on possible treatments

Viral diseases are notoriously difficult to cure. Despite progresses in antivirals — the best known case is the "tritherapy" treatment of AIDS, that required thousands of men/year of work — most viruses are resistant to foreign molecules. More precisely, because viruses are not real living organisms but extract their multiplication power from the very host cells they infect, most molecules that are active against viruses are also active against their hosts. Hence their general toxicity. In some cases (this is the case of AIDS "tritherapy") it has been possible to custom molecules exquisitely adapted to the folds of proteins coded by the virus, with minimal interference with the host cell's machinery. In some other cases one can limit toxicity with local treatment: the molecules are toxic both to the virus and the host, but because they are delivered locally (on the surface of skin, on the outside of the eye...) their adverse effects are not worrying, and one can get rid of the virus. A virus is just a piece of genetic program, made of nucleotides, enclosed in a more or less complex envelope. Fighting the virus means preventing its entry into the host cells where it multiplies, prevent its multiplication inside the cell, or prevent the construction of its envelope once it is developing in the host cell. All curative treatments are based on these processes. The most evident ones are those which poison the replication process by using analogs of the building blocks of the virus genome, or by poisoning the cell machinery that makes these building blocks. In the case of a previously unknown virus such as the coronavirus complex causing SARS, the only immediate possible "textbook" treatment was to follow this reasoning. While this may sound reasonable, in particular in association with anti-inflammatory treatment to alleviate the adverse over-reaction of the host, this was actually very dangerous (and this was much discussed among clinicians and scientists world-wide from the very start of the epidemic). An anti-inflammatory treatment appears to be essential for the relief of the most adverse symptoms. However there are many drawbacks associated to these molecules. Of major concern is the fact that they usually decrease the immune response of the host, and may therefore tend to increase the viral load, and make the virus stay longer in the body. It is well established that an inappropriate use of corticosteroids would appear to relieve symptoms for some time, while, later on, the status of the patients may deteriorate with a rebound of the virus load, due to poor immune response. Things are even much less clear with antivirals. On the one hand this family of antivirals are very toxic for the host (in particular for the liver as well as for the cells that multiply the fastest in the body, small intestine and stem blood cells). On the other hand, because they interfere with viral replication, they favor mutations, as demonstrated in the biases found in genomes and highly depending on availability of building blocks [PDF]. This is most unfortunate in the case of coronaviruses, as demonstrated by the cause of the outbreak of respiratory diseases in pig livestocks.

Lack of understanding the behaviour of a new virulent pathogen may have extremely dangerous consequences if an inappropriate treatment is provided: in the present case, the very fact that coronaviruses can infect both the respiratory tract and the gut (places which are in communication by the pharynx) make them particularly prone to shift tropism (and they may even infect the liver, as seen in rodents), so that any increase in the viral load (for example after indiscriminate use of anti-inflammatory molecules) may result in a catastrophy. This is the most important, as noticed in Chinese forums on the 3rd of May, because the "spike" protein of the virus has significant similarities with the murine virus causing hepatitis in mice. This means that, in the absence of other knowledge, we only have symptomatic treatment available for the moment, aiming at alleviating the most adverse symptoms. And even in this case one must be extremely prudent in the way they are used since they involve respiratory assistance, with usually significant risk of aerosol formation. Other treatments are possible, but they must be thoroughly tested by randomized testing in a large number of patients. The lesson is that we need research at a high level, to test for existing molecules, but also to create new ones which, as in the case of the Human Immunodeficiency Virus (HIV) would interfere with the enzyme that allows the virus to replicate, or would interfere with its permeation into the host cells. As can be seen from the case of HIV, this will need a huge academic effort, world-wide. It is most unfortunate that, for the past twenty years or so, microbiology became out of fashion, leaving us with an immense gap in our knowlegde of microbes. The most efficient treatment, of course, is prevention. Vaccination would be the best approach. But the lesson given by AIDS should sober us. However coronaviruses are not retroviruses and we have much better hopes to be able to construct vaccines, but this also will require time. If the fact that children are apparently much less affected than adults is confirmed, then this may suggest that there exist some kind of immunity (and this may also be reflected in the large differences in infection seen at different places) and we have good hope for the future. It seems difficult to hope that we will have a vaccine in less than one year time, however. At the moment, therefore, the most urgent is to contain the disease by preventing contamination. This will require enforcement of stringent hygiene measures as well as control of travels and quarantine. At the date of May 5th, hope exists that we may be able to construct a vaccine against SARS. Indeed treatment using sera collected from recovering patients apparently relieved the symptoms of patients that were at a very dangerous stage of the disease. This type of serotherapy, which has been widely used in the past, is not always effective. In some cases it can even increase the symptoms. Results of these protocols were therefore eagerly awaited. At a time when the standard textbook treatment that is widely used does not seem as efficient as hoped (the death toll in Hong Kong passed the 12% figure), this is an important hope. One should however remember that blood and blood derived products are dangerous (remember HIV and hepatitis C), but in the absence of other treatments this may be better than nothing, in the worst cases.

Return to the Journalists page....