Human coronavirus: cause and effect?

April 2005

Editor’s note: Before I start this editorial on human coronavirus (HCoV), I must acknowledge a wonderful editorial commentary that I recommend to everyone that is interested in recent considerations for the impact of HCoV in human disease. Kenneth McIntosh, MD, from the division of infectious diseases, Children’s Hospital Boston, published this editorial in The Journal of Infectious Diseases (2005. 191:489-491). Most of this editorial used Dr. McIntosh’s review of HCoV and the references provided in his editorial for this commentary.

Tyrrell and Byneo first described HCoV in 1965 (Br Med J. 1965;5448:1467-1470). The subsequent evolution of animal diseases associated with coronavirus infections have included progressive peritonitis, nephritis, acute and chronic hepatitis and subacute encephalitis. More traditional respiratory and gastrointestinal syndromes have been identified through studies of viral cytopathogenicity, immunologic damage and genetic susceptibility.

In humans, respiratory pathogenesis for HCoV was elucidated in volunteer studies (Br Med J. 1967;3:767-769). Seroepidemiologic studies identified the two most easily studied strains, HCoV-229E and HCoV-OC43. Continued studies since the 1960s of potentially novel HCoV strains that may be associated with upper respiratory tract infections in humans were hindered by the fact that some of these viruses would not grow in tissue culture and could not be adapted to grow in animal models. They were not further characterized.

Three more recent events have focused the spotlight back on the HCoV as a human pathogen. The first and most obvious was the appearance of severe acute respiratory syndrome (SARS) in 2003. This disease had a unique clinical spectrum and resulted from an animal CoV moving across species lines. The identification of the SARS agent was the result of international collaboration and genomic sequence data that had been accumulated from the study of multiple animal CoVs. This technology and resurgence of interest in HCoV as a possible human pathogen led to the identification of a new coronavirus closely related to HCoV-229E, called NL-63 in two independent laboratories in the Netherlands in 2004 (Nat Med. 2004;10:368-373 and Proc Natl Acad Sci USA. 2004;101:6212-6216). These isolates were from human respiratory tracts.

Subsequently, Esper and colleagues used polymerase chain reaction (PCR) to screen specimens from the respiratory tracts of symptomatic children. Using these probes, they identified genomic sequences of a HCoV that they designated (HCoV-NH) (J Infect Dis. 2005;191:492-498). This novel isolate is similar to the HCoV identified previously in the Netherlands and likely represent the same species (HCoV-NL63). The authors speculated that this HCoV may have worldwide distribution and could account for a significant proportion of respiratory tract disease in infants and young children. As Dr. McIntosh questioned, since the seasonal distribution, the percentage of positive samples, the associated respiratory syndromes and the number of infected children at various ages were heavily influenced by both the particular population and the clinical setting, it is essentially impossible to draw conclusions on the epidemiology, pathogenicity, and the relative importance of this new HCoV in relation to other respiratory viruses at this time. Although both The Netherlands and the U.S. group of investigators are to be congratulated on their discoveries, there is still a lot of work left to do before these issues can move from speculation to understanding.

The second paper by Esper and colleagues (J Infect Dis. 2005;191:499-502) reported a temporal association with statistical significance between infection with HCoV-NH (HCoV-NL63) and Kawasaki Disease (KD) using a case-controlled study of hospitalized children. The editorial by Dr. McIntosh provides several tantalizing considerations for why this might have validity and equally strong skepticism (J Infect Dis. 2005;191:489-491).

I asked a colleague and friend to help me understand the recent description of HCoV and a possible link to KD. Anne Rowley, MD, is a professor of pediatrics and microbiology and immunology, at Northwestern University Feinberg School of Medicine. Anne’s expertise in KD and her research into the immunology and pathogenesis of this disease made my choice easy. Her assessment of this possible link is included in the following four paragraphs:

“A recent article from a group of investigators at Yale University (Esper F, et al. Association between a Novel Human Coronavirus and Kawasaki Disease. J Infect Dis. 2005;191:499-502) proposed an association between NL-63 (a new human coronavirus identified by van der Hoek and associates and reported in Nature Medicine. 2004;10:368-373) and acute KD. The Yale investigators used RT-PCR to amplify NL-63 RNA from archival respiratory tract specimens that tested negative for respiratory syncytial virus, influenza A and B, parainfluenza viruses 1-3, and adenovirus by direct fluorescent assay. They identified samples in their archival bank from 11 children who were ultimately diagnosed with acute KD, and eight of these samples were positive, compared with samples from one of 22 controls that were not diagnosed with KD.”

“To determine whether acute KD is in fact associated with the presence of NL-63 in respiratory tract samples, a multicenter group of KD investigators from the United States and The Netherlands tested respiratory samples from acute KD patients for NL-63 RNA by RT-PCR using multiple sets of primer pairs including those used by Esper and colleagues. Six laboratories participated in this study, the results of which were reported at the 8th International Kawasaki Disease Symposium held February 17-20 in San Diego. NL-63 RNA was identified in respiratory samples from only one of 45 acute KD patients. Therefore, the multicenter group concluded that human coronavirus NL-63 in the respiratory tract is not associated with acute KD.”

“Potential problems with the Yale study include possible misdiagnosis of KD, because most patients with a well-established diagnosis of KD may not have respiratory samples sent to the clinical virology laboratory, incomplete KD patients were included in the study and a nonstandardized method of reporting coronary artery dilatation was used. Another potential problem is the ever-present concern regarding PCR contamination in any study that is entirely PCR-based.”

“On a purely historical note, NL-63 causes a cytopathic effect in standard tissue culture lines, and KD investigators worldwide have sought evidence of CPE in tissue culture lines inoculated with clinical samples from acute KD patients for more than 40 years. It seems highly improbable that all these investigators, who have included many excellent virologists, would have missed the CPE if NL-63 were associated with acute KD.”

A fifth HCoV, HKU1, was recently identified by investigators in Hong Kong from a 71-year-old man with respiratory illness (J Virology. 2005;79:884-895). This virus does not grow in cell culture and was identified by RT-PCR of a nasopharyngeal aspirate sample, using conserved coronavirus polymerase primers. It is more closely related to HCoV-OC43 than to the other known human coronaviruses.

These studies and discussion confirm that with modern molecular technology, the age of discovery in virology is alive and well. Whether attempting to link HCoV to necrotizing enterocolitis or KD, trying to evaluate the significance of HCoV as a respiratory pathogen compared to RSV and human metapneumovirus, or the potential association of these respiratory pathogens as triggers of immunologic based disease is a fertile field for future research in infectious diseases.