A worldwide outbreak of severe acute respiratory syndrome (SARS) has been associated with exposures originating from a single ill health care worker from Guangdong Province, China. We conducted studies to identify the etiologic agent of this outbreak.
We received clinical specimens from patients in seven countries and tested them, using virus-isolation techniques, electron-microscopical and histologic studies, and molecular and serologic assays, in an attempt to identify a wide range of potential pathogens.
None of the previously described respiratory pathogens were consistently identified. However, a novel coronavirus was isolated from patients who met the case definition of SARS. Cytopathological features were noted in Vero E6 cells inoculated with a throat-swab specimen. Electron-microscopical examination revealed ultrastructural features characteristic of coronaviruses. Immunohistochemical and immunofluorescence staining revealed reactivity with group I coronavirus polyclonal antibodies. Consensus coronavirus primers designed to amplify a fragment of the polymerase gene by reverse transcription–polymerase chain reaction (RT-PCR) were used to obtain a sequence that clearly identified the isolate as a unique coronavirus only distantly related to previously sequenced coronaviruses. With specific diagnostic RT-PCR primers we identified several identical nucleotide sequences in 12 patients from several locations, a finding consistent with a point-source outbreak. Indirect fluorescence antibody tests and enzyme-linked immunosorbent assays made with the new isolate have been used to demonstrate a virus-specific serologic response. This virus may never before have circulated in the U.S. population.
A novel coronavirus is associated with this outbreak, and the evidence indicates that this virus has an etiologic role in SARS. Because of the death of Dr. Carlo Urbani, we propose that our first isolate be named the Urbani strain of SARS-associated coronavirus.
In late 2002, cases of life-threatening respiratory disease with no identifiable cause were reported from Guangdong Province, China; they were followed by reports from Vietnam, Canada, and Hong Kong of severe febrile respiratory illness that spread to household members and health care workers. The syndrome was designated “severe acute respiratory syndrome” (SARS) in March 2003,1-5 and global efforts to understand the cause of this illness and prevent its spread were instituted in March 2003. Many cases can be linked through chains of transmission to a health care worker from Guangdong Province, China, who visited Hong Kong, where he was hospitalized with SARS and died. Clinical specimens from patients meeting the case definition of SARS were sent to the Centers for Disease Control and Prevention (CDC) by collaborators in Vietnam, Singapore, Thailand, Hong Kong, Canada, Taiwan, and the United States as part of the etiologic investigation. In this report, we describe the efforts of the CDC to detect a wide range of possible etiologic agents for this disease outbreak, and we describe the identification and initial characterization of a novel coronavirus associated with cases of SARS.
The nonspecific nature of the clinical presentation of patients with SARS and the urgency of finding a cause required that clinical specimens be tested rapidly for a broad range of viral, bacterial, chlamydial, and rickettsial agents (the CDC case definition of SARS is available as Supplementary Appendix 1 with the full text of this article at http://www.nejm.org). Laboratory testing focused foremost on known respiratory pathogens, especially those that might specifically target the lower respiratory tract through the progression of disease. A combination of traditional methods was applied, including virus isolation in suckling mice and cell culture, electron microscopy, histopathological examination, serologic analysis, and general and specialized bacterial culture techniques. The molecular techniques of polymerase chain reaction (PCR), reverse-transcription PCR (RT-PCR), and real-time PCR were used. Priority was given to testing for the following agents: yersinia, mycoplasma, chlamydia, legionella, Coxiella burnetii, spotted fever and typhus group rickettsiae, influenzaviruses A and B, Paramyxovirinae and Pneumovirinae subfamily viruses (specifically, human respiratory syncytial virus and human metapneumovirus), Mastadenoviridae, Herpetoviridae, Picornaviridae, Old and New World hantaviruses, and Old World arenaviruses.
Given the serious nature of SARS and the suggestion of person-to-person transmission, it was decided to handle all clinical specimens in a biosafety level 3 environment. All division into aliquots, pipetting, and culture attempts were performed in laminar-flow safety cabinets in a biosafety level 3 laboratory. Serum specimens that were tested serologically outside the laboratory were exposed to 60Co gamma irradiation at 2×106 rad while frozen on dry ice. Initial division into aliquots, handling, and culturing were undertaken in a biosafety level 3 laboratory area in which no culturing of known viruses is done. A similar environment was used when specimens from which nucleic acid was to be extracted were placed in a solution of chaotropic salts; after this step, the specimens were removed to other areas for completion of the extraction protocols.
Isolation of Virus
To identify viruses associated with SARS, we inoculated a variety of clinical specimens (blood, serum, material from oropharyngeal swabs or washings, material from nasopharyngeal swabs, and tissues of major organs collected at autopsy) onto a number of continuous cell lines, including Vero E6, NCI-H292, MDCK, LLC-MK2, and B95-8 cells, and into suckling ICR mice by the intracranial and intraperitoneal routes. All cultures were observed daily for cytopathic effect. Maintenance medium was replenished at day 7, and cultures were terminated 14 days after inoculation. Any cultures exhibiting identifiable cytopathic effect were subjected to several procedures to identify the cause of the effect. Suckling mice were observed daily for 14 days, and we further tested any sick or dead mice by preparing a brain suspension that was filtered and subcultured. Mice that remained well after 14 days were euthanized, and their test results were recorded as negative. Tissue-culture samples showing cytopathic effect were prepared for electron-microscopical examination. Negative-stain electron-microscopical specimens were prepared by drying culture supernatant, mixed 1:1 with 2.5 percent paraformaldehyde, onto Formvarcarbon-coated grids and staining with 2 percent methylamine tungstate. Thin-section electron-microscopical specimens were prepared by fixing a washed cell pellet with 2.5 percent glutaraldehyde and embedding it in epoxy resin. For RT-PCR assays, cell-culture supernatants were placed in lysis buffer. In addition, a master seed was prepared from the remaining culture supernatant and cells by freeze-thawing the culture flask, clarifying the thawed contents by centrifugation at 1000×g, and dispensing the supernatant into aliquots stored in gas phase over liquid nitrogen. The master seed was subcultured into 850-cm2 roller bottles of Vero E6 cells for the preparation of formalin-fixed positive control cells for immunohistochemical analysis, mixed with normal E6 cells, and gamma-irradiated for preparation of spot slides for indirect fluorescence antibody tests or extracted with detergent and gamma-irradiated for use as an enzyme-linked immunosorbent assay (ELISA) antigen for antibody tests.
Spot slides were prepared by applying 15 μl of the suspension of gamma-irradiated mixed infected and noninfected cells onto 12-well Teflon-coated slides. Slides were allowed to air dry before being fixed in acetone. Slides were then stored at –70°C until used for indirect fluorescence antibody tests.6 An ELISA antigen was prepared by detergent extraction of infected Vero E6 cells and subsequent gamma irradiation.7 The optimal dilution (1:1000) for the use of this antigen was determined by checkerboard titration against serum from a patient with SARS in the convalescent phase; a control antigen, similarly prepared from uninfected Vero E6 cells, was used to control for specific reactivity of tested serum. The conjugates used were goat antihuman IgG, IgA, and IgM conjugated to fluorescein isothiocyanate and horseradish peroxidase (Kirkegaard and Perry), for the indirect fluorescence antibody test and ELISA, respectively. Specificity and cross-reactivity of a variety of serum samples to the newly identified virus were evaluated by using the tests described above. For this evaluation, we used serum from patients with SARS in Singapore, Bangkok, and Hong Kong and serum from healthy blood donors from the CDC serum bank and from persons infected with known human coronaviruses (human coronaviruses OC43 and 229E) (samples provided by E. Walsh and A. Falsey, University of Rochester School of Medicine and Dentistry, Rochester, N.Y.).
Pathological and Immunohistochemical Studies
Formalin-fixed, paraffin-embedded Vero E6 cells infected with the novel coronavirus and tissues obtained from patients with SARS were stained with hematoxylin and eosin and various immunohistochemical stains. Immunohistochemical assays were based on a method described previously for hantavirus.8 In brief, 4-μm sections were deparaffinized, rehydrated, and digested in Proteinase K for 15 minutes. Slides were then incubated for 60 minutes at room temperature with monoclonal antibodies, polyclonal antiserum, or ascitic fluids derived from animal species with reactivities to various known coronaviruses, and with a convalescent-phase serum specimen from a patient with SARS.
Optimal dilutions of the primary antibodies were determined by titration experiments with cells infected with the new coronavirus and with noninfected cells or, when available, with concentrations recommended by the manufacturers. After sequential application of the appropriate biotinylated link antibody, avidin–alkaline phosphatase complex, and naphthol–fast red substrate, sections were counterstained in Mayer's hematoxylin and mounted with aqueous mounting medium. We used the following antibody and tissue controls: serum specimens from noninfected animals, various coronavirus-infected cell cultures and animal tissues, noninfected cell cultures, and normal human and animal tissues. Tissues from patients were also tested by immunohistochemical assays for various other viral and bacterial pulmonary pathogens. In addition, a bronchoalveolar-lavage specimen was available from one patient with SARS for thin-section electron-microscopical evaluation.
RNA extracts were prepared from 100 μl of each specimen (or culture supernatant) with the automated NucliSens extraction system (bioMérieux). Oligonucleotide primers used for amplification and sequencing of the SARS-related coronavirus were designed from alignments of open reading frame 1b of the coronavirus polymerase gene sequences obtained from GenBank, including human coronaviruses 229E and OC43 (accession numbers X69721 and AF124989, respectively), canine coronavirus (AF124986), feline infectious peritonitis virus (AF124987), porcine transmissible gastroenteritis virus (Z34093), porcine epidemic diarrhea virus (NC_003436), bovine coronavirus (NC_003045), porcine hemagglutinating encephalomyelitis virus (AF124988), sialodacryoadenitis virus (AF124990), mouse hepatitis virus (NC_001846), turkey coronavirus (AF124991), and avian infectious bronchitis virus (NC_001451). Primer pair IN-2 (+) 5'GGGTT-GGGACTATCCTAAGTGTGA3' and IN-4 (–) 5'TAACACACAACICCATCATCA3' was previously designed to conserved regions of open reading frame 1b to achieve broad reactivity with the genus coronavirus. These primers were used to amplify DNA from SARS isolates, and the amplicon sequences obtained were used to design SARS-specific primers Cor-p-F2 (+) 5'CTAACATGCTTAGGATAATGG3', Cor-p-F3 (+) 5'GCCTCTCTTGTTCTTGCTCGC3', and Cor-p-R1 (–) 5'CAGGTAAGCGTAAAACTCATC3', which were used in turn to test patient specimens. Primers used for specific amplification of human metapneumovirus have been described previously.9 Well-characterized primer sets for other respiratory virus pathogens (unpublished data), including human respiratory syncytial virus, parainfluenzaviruses 1, 2, and 3, influenzaviruses A and B, adenovirus, and picornavirus (rhinovirus and enterovirus), were also used to test clinical specimens in this study (primers available on request). All specimens were tested for human glyceraldehyde-3-phosphate dehydrogenase to confirm RNA integrity and control for RT-PCR inhibition.
One primer for each set was 5'-end-labeled with fluorescent dye 6-carboxyfluorescein (6-FAM) to facilitate GeneScan analysis. One-step amplification reactions were performed with the Access RT-PCR System (Promega) as previously described.9 Positive and negative RT-PCR controls, containing standardized viral RNA extracts, and nuclease-free water were included in each run. Amplified 6-FAM–labeled products were analyzed by capillary electrophoresis on an ABI 3100 Prism Genetic Analyzer with GeneScan software (version 3.1.2). Specimens were considered positive for SARS-associated coronavirus if the amplification products were within 1 nucleotide of the expected product size (368 nucleotides for Cor-p-F2 or Cor-p-R1 and 348 nucleotides for Cor-p-F3 or Cor-p-R1) for both specific primer sets, as confirmed by a second PCR reaction from another aliquot of RNA extract in a separate laboratory. Where DNA yield was sufficient, the amplified products were also sequenced. The microarray used in this study was an expanded version of an array previously described.10,11
For sequencing, amplicons were purified with ExoSAP-IT (USB). Both strands of unlabeled products (or one strand of the 6-FAM–labeled products) were sequenced on an ABI PRISM 3100 Genetic Analyzer with use of a fluorescent dye-terminator kit (ABI). The nucleotide sequences were edited with Sequencher for Power Macintosh (version 3.1.1, Gene Codes). The partial nucleotide sequences of the polymerase gene were aligned with published coronavirus sequences, using CLUSTAL W for Unix (version 1.7).12 Phylogenetic trees were computed by maximum parsimony, distance, and maximum likelihood–based criteria analysis with PAUP (version 4.0.d10).13
Two cell lines, Vero E6 cells and NCI-H292 cells, inoculated with oropharyngeal specimens from Patient 16 (a 46-year-old male physician with an epidemiologic link to a hospital with multiple patients with SARS) initially showed cytopathic effect (Table 1). Blood, nasopharyngeal, and throat-swab specimens were collected on March 12, day 1 after onset. At that time, the patient's physical examination was normal except for fever and shortness of breath. During the course of the disease, his status worsened, and he died. A rhinovirus was isolated from the inoculated NCI-H292 cells. Further study suggested that this virus was not associated with patients with SARS, so it will not be discussed here.
Cytopathic effect in the Vero E6 cells was first noted on the fifth post-inoculation day. The cytopathic effect was focal, with cell rounding and a refractive appearance in the affected cells (Figure 1) that was soon followed by cell detachment. The cytopathic effect quickly spread to involve the entire cell monolayer within 24 to 48 hours. Subculture of material after preparation of a master seed resulted in the rapid appearance of cytopathic effect, as noted above, and in complete destruction of the monolayer in the inoculated flasks within 48 hours. Similar cytopathic effect has since been noted in four additional cultures: three cultures of respiratory specimens (two oropharyngeal washes and one sputum specimen) and one culture of a suspension of kidney tissue obtained at autopsy. In these specimens, the initial cytopathic effect was observed between day 2 and day 4 and, as noted above, the cytopathic effect rapidly progressed to involve the entire cell monolayer.
Examination of cytopathic-effect–positive Vero E6 cells by thin-section electron microscopy revealed characteristic coronavirus particles within the cisternae of the rough endoplasmic reticulum and in vesicles (Figure 2A).14,15 Extracellular particles were found in large clusters and adhering to the surface of the plasma membrane. Negative-stain electron microscopy identified coronavirus particles, 80 to 140 nm in diameter, with 20-to-40-nm complex surface projections surrounding the periphery (Figure 2B). Hemagglutinin esterase-type glycoprotein projections were not seen.
A 405-nucleotide segment of the coronavirus polymerase gene open reading frame 1b was amplified from the isolation material by RT-PCR with the broadly reactive primer set IN-2–IN-4. In contrast, this primer set produced no specific band against uninfected cells.
When compared with other human and animal coronaviruses, the nucleotide and deduced amino acid sequence from this region had similarity scores ranging from 0.56 to 0.63 and from 0.57 to 0.74, respectively. The highest sequence similarity was obtained with group II coronaviruses. The maximum-parsimony tree obtained from the nucleotide-sequence alignment is shown in Figure 3. Bootstrap analyses of the internal nodes at the internal branches of the tree provided strong evidence that the SARS-associated coronavirus is genetically distinct from other known coronaviruses. The microarray analyses from infected and uninfected cell cultures gave a positive signal for a group of eight oligonucleotides derived from two virus families: Coronaviridae and Astroviridae. All of the astroviruses and two of the coronavirus oligonucleotides share a consensus sequence motif that maps to the extreme 3' end of astroviruses and two members of the coronavirus family: avian infectious bronchitis and turkey coronavirus.16 Results were consistent with the identity of the isolate as a coronavirus.
Immunohistochemical and Histopathological Analysis and Electron-Microscopical Analysis of Bronchoalveolar-Lavage Fluid
Lung tissues were obtained at autopsy from three patients and by open-lung biopsy in one patient, 14 to 19 days after the onset of SARS. Confirmatory laboratory evidence of infection with coronavirus was available for two patients (Patients 6 and 17) and included PCR amplification of coronavirus nucleic acids from tissues, viral isolation from bronchoalveolar-lavage fluid, or detection of serum antibodies reactive with coronavirus (Table 1). For two patients, no samples were available for molecular, cell-culture, or serologic analysis; however, the condition of both patients met the CDC definition of probable SARS, and both had strong epidemiologic links with laboratory-confirmed cases of SARS. Histopathological evaluation of lung tissues from the four patients showed diffuse alveolar damage at various levels of progression and severity. Changes included hyaline-membrane formation, interstitial mononuclear inflammatory infiltrates, and desquamation of pneumocytes in alveolar spaces (Figure 4A). Other findings identified in some patients included focal intraalveolar hemorrhage, necrotic inflammatory debris in small airways, and organizing pneumonia. Multinucleated syncytial cells were identified in the intraalveolar spaces of two patients who died 14 and 17 days after onset of illness. These cells contained abundant vacuolated cytoplasm with cleaved and convoluted nuclei. No obvious intranuclear or intracytoplasmic viral inclusions were identified (Figure 4B), and electron-microscopical examination of a limited number of these syncytial cells revealed no coronavirus particles. No definitive immunostaining was identified in tissues from SARS patients with the use of a battery of immunohistochemical stains reactive with coronaviruses from antigenic groups I, II, and III. In addition, no staining of patient tissues was identified with the use of immunohistochemical stains for influenzaviruses A and B, adenoviruses, Hendra and Nipah viruses, human metapneumovirus, respiratory syncytial virus, measles virus, Mycoplasma pneumoniae, and Chlamydia pneumoniae.
Evaluation of Vero E6 cells infected with coronavirus isolated from a patient with SARS revealed viral cytopathic effect that included occasional multinucleated syncytial cells but no obvious viral inclusions (Figure 4C). Immunohistochemical assays with various antibodies reactive with coronaviruses from antigenic group I, including human coronavirus 229E, feline infectious peritonitis virus 1, and porcine transmissible gastroenteritis virus, and with an immune serum specimen from a patient with SARS demonstrated strong cytoplasmic and membranous staining of infected cells (Figure 4C and Table 2); however, cross-reactivity with the same immune human serum sample and feline infectious peritonitis virus 1 antigen was not observed. No staining was identified with any of several monoclonal or polyclonal antibodies reactive with coronaviruses in antigenic group II (human coronavirus OC43, bovine coronavirus, and mouse hepatitis virus) or group III (turkey coronavirus and avian infectious bronchitis virus). Electron-microscopical examination of bronchoalveolar-lavage fluid from one patient revealed many coronavirus-infected cells (Figure 5).
Spot slides with infected cells reacted with serum from patients with probable SARS in the convalescent phase. Screening of a panel of serum from patients with suspected SARS from Hong Kong, Bangkok, Singapore, and the United States showed a high level of specific reaction with infected cells and conversion from negative to positive reactivity or diagnostic rises in the indirect fluorescence antibody test by a factor of four. Similarly, tests of these same serum samples with the ELISA antigen showed high specific signal in the convalescent-phase samples and conversion from negative to positive antibody reactivity or diagnostic increases in titer (Table 3). Information from the limited number of samples tested thus far suggests that antibody is first detectable in these two tests between one and two weeks after the onset of symptoms in the patient. Indirect fluorescence antibody testing and ELISA of a panel of 384 randomly selected serum samples (from U.S. blood donors) were negative for antibodies to the new coronavirus, with the exception of 1 specimen that had minimal reactivity on ELISA. A panel of paired human serum samples with diagnostic increases (by a factor of four or more) in antibody (with very high titers to the homologous viral antigen in the convalescent-phase serum) to the two known human coronaviruses, OC43 (13 pairs) and 229E (14 pairs), showed no reactivity in either acute- or convalescent-phase serum with the newly isolated coronavirus by either the indirect fluorescence antibody test or the ELISA.
Nineteen patients with SARS have been identified as infected with the new coronavirus by virus isolation, RT-PCR, or serologic tests; all have direct or indirect links to the SARS outbreak in Hong Kong or Guangdong Province, China (Table 1). We were able to amplify by RT-PCR and obtain the virus sequence from clinical specimens or virus isolates from 12 of these patients. All 12 sequences were identical to those of the first isolate as noted above. For four convalescent patients, infection was detected serologically alone; for nine patients it was detected by RT-PCR alone; for three by virus isolation and RT-PCR; for two by virus isolation, RT-PCR, and serologic analysis; and for one by RT-PCR and serologic analysis. We found none of the coronavirus-infected patients to be infected with human metapneumovirus. In only one patient was both SARS coronavirus and another respiratory virus detected; Patient 16 had both SARS coronavirus and a rhinovirus. A variety of respiratory pathogens were also identified by RT-PCR in other patients whose samples were submitted for SARS testing, including 5 with human metapneumoviruses (sequencing showed that each was distinct) and 12 with rhinoviruses (sequencing showed that each was distinct). None of the patients who were positive for human metapneumovirus had pneumonia.
The isolation of a novel coronavirus from the respiratory secretions of a patient with SARS and the subsequent demonstration of this virus or a serologic response to this virus in others with SARS demonstrate an etiologic association between this virus and SARS. The discovery of this new virus occurred through a broad-based and multidisciplinary effort by clinical, epidemiologic, and laboratory investigators and speaks to the power of a global collaborative effort to address the ever-present threat of emerging infectious diseases.17
The three known groups of coronavirus are associated with a variety of diseases of humans and domestic animals, including gastroenteritis and upper and lower respiratory tract disease. Although the known human coronaviruses are associated with a mild disease (the common cold), the ability of coronavirus of animals to cause severe disease raises the possibility that coronavirus could also cause more severe disease in humans. Other than rare instances in children or immunocompromised patients, it appears that the SARS-related coronavirus may be the first example of a coronavirus that causes severe disease in humans. The novel human coronavirus identified in this study shares antigenic features with various group I coronaviruses, but genetic comparisons suggest it is distinct from group I coronaviruses and from coronaviruses in groups II and III. The factor or factors responsible for this apparent dichotomy remain to be elucidated; however, correlation between antigenic and genetic characteristics of these viruses is occasionally unclear, and the placement of some other human coronaviruses within specific antigenic groups has not always been well defined.18-20
The identification of this novel coronavirus relied on classic tissue-culture isolation to amplify the pathogen and then on electron-microscopical studies to identify the type of virus, a member of the family Coronaviridae, and molecular studies to confirm the identity of the virus, characterize its unique nature, and help link it to the disease. The discovery of this new virus underscores the importance of versatile techniques such as virus isolation and electron microscopy in identifying etiologic pathogens. As with previous outbreak investigations, electron microscopy proved to be a rapid technique that did not require specific reagents for or prior knowledge of a particular agent but that could nevertheless categorize a pathogen on the basis of its appearance and morphogenesis.21-24
In this report, we describe infection in 19 SARS patients with well-defined direct or indirect epidemiologic links either to the outbreak in Hong Kong or to Guangdong Province, China, the origin of the index patient in Hong Kong. As expected with a point-source outbreak, the sequences from a limited region of the polymerase gene are identical. Coronaviruses with identical sequences have also been detected in patients with SARS in Canada.5 The virus was found in multiple specimens, including lung and kidney tissue extracts by virus isolation or RT-PCR, bronchoalveolar-lavage specimens by electron microscopy and PCR, and sputum or upper respiratory tract swab, aspirate, or wash specimens by RT-PCR or virus isolation. Although we tested specimens from the coronavirus-positive patients for a variety of other respiratory pathogens, including human metapneumovirus, by RT-PCR, none were detected in these coronavirus-positive patients except for a rhinovirus in Patient 16. The relation between this novel coronavirus and disease is further evidenced by detection of virus in lung tissue and a bronchoalveolar-lavage specimen, thus placing the virus at the site of diseased tissue. We were not, however, able to demonstrate coronavirus antigens in patient tissues by histologic and immunohistochemical methods or to demonstrate a direct involvement in the pathologic process. Neither were we able to demonstrate SARS-associated coronavirus infection in all suspected patients with SARS.
Possible reasons for the inability to demonstrate infection in some patients with suspected SARS include the lack of sufficient sensitivity of the assays to detect the pathogen and the immune response and the timing and type of specimens tested. For example, we have not yet received convalescent-phase serum specimens from many patients with suspected SARS and have not serologically ruled out infection in many such patients. In addition, we are just beginning to study the type and timing of clinical specimens most likely to support a diagnosis of infection with this new virus. We have made rapid progress in developing our diagnostic assays and are continuing to improve them for the detection of this virus or an immune response to it. In addition, the case definition of SARS is very broad and most likely includes other infectious diseases. We are also continuing to test for other infectious agents that might be associated with SARS, including those that might contribute to the severity of disease or increase the efficiency of viral transmission. Further clinical analysis of patients with SARS in whom there is laboratory confirmation of infection with the new coronavirus might help refine the case definition further.
The apparent lack of antibody in all serum specimens except those from patients with SARS suggests that this virus has not previously circulated in humans. Certainly, it has not circulated widely in humans, which is further evidence of an association between infection with this novel coronavirus and SARS. Presumably, this virus originated in animals and mutated or recombined in a fashion that permitted it to infect, cause disease, and pass from person to person. The available sequence data for this novel coronavirus suggest that it is distinct from those previously reported in animals and humans and that the parent virus has not yet been discovered. The isolation and growth of a human-derived coronavirus in Vero E6 cells were unexpected. The only human or animal coronavirus that has been shown to grow in Vero cells is porcine epidemic diarrhea virus (isolated in China from pigs), and it requires the addition of trypsin to culture medium for growth in Vero E6 cells.25 However, like the sequences of the other known coronaviruses, the sequences for porcine epidemic diarrhea virus are distinct from those of SARS-associated coronavirus, indicating that porcine epidemic diarrhea virus is not the parent virus to this new coronavirus. Because of the death of Dr. Carlo Urbani during the investigation, we propose that our first isolate be named the Urbani strain of SARS-associated coronavirus.
Pathological studies in patients who died with SARS show diffuse alveolar damage as the most notable feature, a finding consistent with the severe respiratory illness seen in some patients with SARS. The primary histopathological lesions seen in the lungs of four patients we studied are consistent with a nonspecific response to acute lung injury that can be caused by infectious agents, trauma, drugs, or toxic chemicals. However, the multinucleated syncytial cells without viral inclusions seen in the lungs of two patients, including one patient positive by PCR, serologic, and virus-isolation methods, are suggestive of a number of viral infections, including measles and parainfluenzavirus, respiratory syncytial virus, and Nipah virus infection. Syncytia have occasionally been observed in culture cells inoculated with other group I and II coronaviruses that infect humans,26-28 but they are more often described in culture cells infected with animal coronaviruses.29-31 To our knowledge, syncytial cells have not been previously described in human tissues infected with coronaviruses.
We did not detect antigens of viruses associated with syncytial formation or SARS-associated coronavirus in these patient tissues, despite the severe pulmonary pathological processes. To detect this novel coronavirus antigen, we used an extensive panel of antibodies against coronaviruses representative of the three antigenic groups. The failure of these antiserum specimens to react with coronavirus antigens in the lung tissues of these patients could be attributed to clearance of viral antigens by the host immune response during the late stage of disease. The tissues of the four patients evaluated by immunohistochemical assays were obtained approximately two to three weeks into the course of the illness. For many viral infections, viral antigens and nucleic acids are cleared within two weeks of disease onset.8,32 It is also possible that the pulmonary damage associated with SARS is not caused directly by the virus but represents a secondary effect of cytokines or other factors induced by viral infection proximal to but not within the lung tissue. In influenzavirus infections, viral antigens are seen predominantly in respiratory epithelial cells of large airways and are only rarely identified in pulmonary parenchyma, despite concomitant and occasionally severe interstitial pneumonitis.33
The clinical virology of this new disease is obviously in its infancy, and there is much to be learned about the behavior of the pathogen in the human host. For example, patterns of viral shedding, including timing and site, and the timing and characteristics of the host immune response, when defined, will help us understand disease transmission, infection-control practices, and strategies for developing vaccines. The CDC, in conjunction with other groups, is planning epidemiologic studies to address these and a wide range of other issues including risk factors for infection and severe disease and the efficacy of infection-control measures. These types of studies should help us refine and focus our efforts to control and treat SARS.
The investigation of the SARS outbreak serves as a positive template for laboratory and epidemiologic response to possible future infectious-disease pandemics. The rapid isolation and characterization of the novel coronavirus associated with SARS have allowed for the timely development of diagnostic tests that should aid our ability to understand the epidemiology of SARS and its prevention. Early recognition of this novel coronavirus has also made it possible to investigate antiviral compounds promptly and to begin developing vaccines. The speed with which this novel coronavirus was detected, characterized, and linked to SARS is a tribute to the power of the prompt communication and exchange of information among the World Health Organization collaborating laboratories about virus-isolation systems, PCR primers and virus sequences, and other diagnostic methods. This collaborative approach can be invaluable in our efforts to understand and control emerging public health threats.
Funding and Disclosures
This article was published at www.nejm.org on April 10, 2003.
We are indebted to Ann R. Falsey and Edward E. Walsh, Rochester General Hospital, University of Rochester School of Medicine and Dentistry; Linda Saif, Ohio Agricultural Research and Development Center, Ohio State University, and the staff of the Hanoi French Hospital, Hanoi, Vietnam; to John Black, James Guy, Miladin Kostovic, Julia Ridpath, Joan Beck, Edward Dubovi, Sanjay Kapil, Chris Grant, David Swayne, Julian Leibowitz, and S.A. Naqi for the provision of reference serum, monoclonal antibodies, and coronavirus control tissues; to the many private and public health physicians and laboratorians who treated patients with SARS or provided clinical materials for evaluation for their assistance and cooperation; and to Claudia Chesley for editorial review of the manuscript. This article is dedicated to the memory of Carlo Urbani.
Members of the SARS Working Group include the following (asterisks indicate group members temporarily assigned from other CDC sections): A.D.L. Cannon, M. Curtis,* B. Farrar, L. Morgan, L. Pezzanite,* A.J. Sanchez, K.A. Slaughter, T.L. Stevens, P.C. Stockton, K.D. Wagoner, A. Sanchez, S. Nichol, M. Vincent, J. Osborne, J. Honig, B.R. Erickson (Special Pathogens Branch, Division of Viral and Rickettsial Diseases, National Center for Infectious Diseases, CDC); B. Holloway, K. McCaustland (DNA Chemistry Section, Scientific Resources Program, National Center for Infectious Diseases); J. Lingappa, L. Lowe, S. Scott, X. Lu, Y. Villamarzo, B. Cook, Q. Chen, C. Birge, B. Shu, M. Pallansch (Respiratory and Enteric Virus Branch, Division of Viral and Rickettsial Diseases); K.M. Tatti, T. Morken, C. Smith, P. Greer, E. White, T. McGlothen, J. Bhatnagar, M. Patel, J. Bartlett, J. Montague, W. Lee, M. Packard (Infectious Diseases Pathology Activity, Division of Viral and Rickettsial Diseases); H.A. Thompson (Viral and Rickettsial Zoonoses Branch); A. Moen, K. Fukuda, T. Uyeki, S. Harper, A. Klimov, S. Lindstrom (Influenza Branch, Division of Viral and Rickettsial Diseases); R. Benson, G. Carlone, R. Facklam, P. Fields, P. Levett, L. Mayer, D. Talkington, W.L. Thacker, M.L.C. Tondella, C. Whitney (Division of Bacterial and Mycotic Diseases, National Center for Infectious Diseases); B. Robertson, D. Warnock (SARS Laboratory Team); J.T. Brooks, S. Schrag, N. Rosenstein (SARS Epidemiology Team); R. Arthur (SARS International Team); D. Ganem (University of California, San Francisco); S.M. Poutanen (Department of Laboratory Medicine and Pathobiology, University of Toronto); T.-J. Chen (Center for Disease Control, Department of Health, Taiwan); C.-H. Hsiao (Department of Pathology, National Taiwan University Hospital, Taipei, Taiwan); N.G. Wai-Fu (Yan Chai Hospital, Hong Kong); M. Ho (K. Wah Hospital, Hong Kong); T.-K. Keung (Princess Margaret Hospital, Hong Kong); K.H. Nghiem, H.K.L. Nguyen, M.Q. Le (Department of Virology, National Institute of Hygiene and Epidemiology, Hanoi, Vietnam); H.H.T. Nguyen, L.T. Hoang (National Institute of Hygiene and Epidemiology, Hanoi, Vietnam); T.H. Vu, H.Q. Vu (Hanoi French Hospital, Hanoi, Vietnam); WHO SARS Investigation Team, Vietnam; S. Chunsuttiwat (Ministry of Public Health, Thailand); WHO Vietnam; and the WHO SARS Collaborating Laboratory Network: Centers for Disease Control and Prevention, Atlanta; Public Health Laboratory Service, Central Public Health Laboratory, London; Government Virus Unit, Hong Kong; Prince of Wales Hospital, the Chinese University of Hong Kong, Hong Kong; Queen Mary Hospital, Hong Kong; Centers for Disease Control, Beijing, China; Erasmus Universiteit, National Influenza Center, Rotterdam, the Netherlands; Federal Laboratories for Health Canada, Winnipeg, Man., Canada; Bernhard Nocht Institute, Hamburg, Germany; Institut für Virologie, Marburg, Germany; National Institute of Infectious Disease, Tokyo, Japan.
1. Acute respiratory syndrome in China — update 3: disease outbreak reported. Geneva: World Health Organization, February 2003.
2. Update: outbreak of severe acute respiratory syndrome -- worldwide, 2003. MMWR Morb Mortal Wkly Rep 2003;52:241-248
3. Tsang KW, Ho PL, Ooi GC, et al. A cluster of cases of severe acute respiratory syndrome in Hong Kong. N Engl J Med 2003;348:1975-1983
4. Lee N, Hui D, Wu A, et al. A major outbreak of severe acute respiratory syndrome in Hong Kong. N Engl J Med 2003;348:1984-1992
5. Poutanen SM, Low DE, Henry B, et al. Identification of severe acute respiratory syndrome in Canada. N Engl J Med 2003;348:1993-2003
6. Wulff H, Lange JV. Indirect immunofluorescence for the diagnosis of Lassa fever infection. Bull World Health Organ 1975;52:429-436
7. Ksiazek TG, West CP, Rollin PE, Jahrling PB, Peters CJ. ELISA for the detection of antibodies to Ebola viruses. J Infect Dis 1999;179:Suppl 1:S192-S198
8. Zaki S, Greer PW, Coffield LM, et al. Hantavirus pulmonary syndrome: pathogenesis of an emerging infectious disease. Am J Pathol 1995;146:552-579
9. Falsey AR, Erdman D, Anderson LJ, Walsh EE. Human metapneumovirus infections in young and elderly adults. J Infect Dis 2003;87:785-790
10. Wang D, Coscoy L, Zylberberg M, et al. Microarray-based detection and genotyping of viral pathogens. Proc Natl Acad Sci U S A 2002;99:15687-15692
11. Bohlander SK, Espinosa R III, Le Beau MM, Rowley JD, Diaz MO. A method for the rapid sequence-independent amplification of microdissected chromosomal material. Genomics 1992;13:1322-1324
12. Thompson JD, Higgins DG, Gibson TJ. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 1994;22:4673-4680
13. Swofford DL. PAUP 4.0: phylogenetic analysis using parsimony (and other methods). Sunderland, Mass.: Sinauer, 1999.
14. Becker WB, McIntosh K, Dees JH, Chanock RM. Morphogenesis of avian infectious bronchitis virus and a related human virus (strain 229E). J Virol 1967;1:1019-1027
15. Oshiro LS, Schieble JH, Lennette EH. Electron microscopic studies of coronavirus. J Gen Virol 1971;12:161-168
16. Jonassen CM, Jonassen TO, Grinde B. A common RNA motif in the 3' end of the genomes of astroviruses, avian infectious bronchitis virus and an equine rhinovirus. J Gen Virol 1998;79:715-718
17. Smolinski MS, Hamburg MA, Lederberg J, eds. Microbial threats to health: emergence, detection, and response. Washington, D.C.: National Academy Press, 2003.
18. McIntosh K, Kapikian AZ, Hardison KA, Hartley JW, Chanock RM. Antigenic relationships among the coronaviruses of man and between human and animal coronaviruses. J Immunol 1969;102:1109-1118
19. Bradburne AF. Antigenic relationships amongst coronaviruses. Arch Gesamte Virusforsch 1970;31:352-364
20. Bradburne AF, Somerset BA. Coronavirus antibody titres in sera of healthy adults and experimentally infected volunteers. J Hyg (Lond) 1972;70:235-244
21. Hazelton PR, Gelderblom HR. Electron microscopy for rapid diagnosis of infectious agents in emergent situations. Emerg Infect Dis 2003;9:294-303
22. Johnson KM, Lange JV, Webb PA, Murphy FA. Isolation and partial characterisation of a new virus causing acute haemorrhagic fever in Zaire. Lancet 1977;1:569-571
23. Chua KB, Bellini WJ, Rota PA, et al. Nipah virus: a recently emergent deadly paramyxovirus. Science 2000;288:1432-1435
24. Murray K, Selleck P, Hooper P, et al. A morbillivirus that caused fatal disease in horses and humans. Science 1995;268:94-97
25. Hofmann M, Wyler R. Propagation of the virus of porcine epidemic diarrhea in cell culture. J Clin Microbiol 1988;26:2235-2239
26. Bruckova M, McIntosh K, Kapikian AZ, Chanock RM. The adaption of two human coronavirus strains (OC38 and OC43) to growth in cell monolayers. Proc Soc Exp Biol Med 1970;135:431-435
27. Patterson S, Macnaughton MR. Replication of human respiratory coronavirus strain 229E in human macrophages. J Gen Virol 1982;60:307-314
28. Luby JP, Clinton R, Kurtz S. Adaptation of human enteric coronavirus to growth in cell lines. J Clin Virol 1999;12:43-51
29. Lavi E, Wang Q, Weiss SR, Gonatas NK. Syncytia formation induced by coronavirus infection is associated with fragmentation and rearrangement of the Golgi apparatus. Virology 1996;221:325-334
30. Kusanagi K, Kuwahara H, Katoh T, et al. Isolation and serial propagation of porcine epidemic diarrhea virus in cell cultures and partial characterization of the isolate. J Vet Med Sci 1992;54:313-318
31. Jacobse-Geels HE, Horzinek MC. Expression of feline infectious peritonitis coronavirus antigens on the surface of feline macrophage-like cells. J Gen Virol 1983;64:1859-1866
32. Wong KT, Shieh W-J, Kumar S, et al. Nipah virus infection: pathology and pathogenesis of an emerging paramyxoviral zoonosis. Am J Pathol 2002;161:2153-2167
33. Guarner J, Shieh WJ, Dawson J, et al. Immunohistochemical and in situ hybridization studies of influenza A virus infection in human lungs. Am J Clin Pathol 2000;114:227-233
Citing Articles (2767)
Only the 1000 most recent citing articles are listed here.
- Table 1. Specimens from Patients with SARS That Were Positive for SARS-Associated Coronavirus by One or More Methods.
- Figure 1. Vero E6 cells Inoculated with Oropharyngeal Specimens from Patients with SARS.
- Figure 2. Ultrastructural Characteristics of SARS-Associated Coronavirus Grown in Vero E6 Cells.
- Figure 3. Estimated Maximum-Parsimony Tree Based on the Sequence Alignment of 405 Nucleotides of the Coronavirus Polymerase Gene Open Reading Frame 1b (Nucleotide Numbers 15173 to 15578 Based on Bovine Coronavirus Complete Genome Accession Number NC_003045) Comparing SARS Coronavirus with Other Human and Animal Coronaviruses.
- Figure 4. Histopathological Evaluation of Lung Tissue from a Patient with SARS and Immunohistochemical Staining of Vero E6 Culture Cells Infected with SARS-Associated Coronavirus.
- Table 2. Immunohistochemical Reactivities of Various Polyclonal Group I Anti-Coronavirus Reference Antiserum Samples with a Coronavirus Isolated from a Patient with SARS and with Selected Antigenic Group I Coronaviruses.
- Figure 5. Ultrastructural Characteristics of a Coronavirus-Infected Cell in Bronchoalveolar-Lavage Fluid from a Patient with SARS, with Numerous Intracellular and Extracellular Particles.
- Table 3. Results of Serologic Testing with Both Indirect Fluorescence Antibody (IFA) Test and Indirect Enzyme-Linked Immunosorbent Assay (ELISA) in Patients with SARS Tested against a Newly Isolated Coronavirus (200300592).
Original ArticleMay 15 Original ArticleMay 15 Original ArticleMay 12