Human Coronavirus 229E Remains Infectious on Common Touch Surface Materials

Abstract

The evolution of new and reemerging historic virulent strains of respiratory viruses from animal reservoirs is a significant threat to human health. Inefficient human-to-human transmission of zoonotic strains may initially limit the spread of transmission, but an infection may be contracted by touching contaminated surfaces. Enveloped viruses are often susceptible to environmental stresses, but the human coronaviruses responsible for severe acute respiratory syndrome (SARS) and Middle East respiratory syndrome (MERS) have recently caused increasing concern of contact transmission during outbreaks. We report here that pathogenic human coronavirus 229E remained infectious in a human lung cell culture model following at least 5 days of persistence on a range of common nonbiocidal surface materials, including polytetrafluoroethylene (Teflon; PTFE), polyvinyl chloride (PVC), ceramic tiles, glass, silicone rubber, and stainless steel. We have shown previously that noroviruses are destroyed on copper alloy surfaces. In this new study, human coronavirus 229E was rapidly inactivated on a range of copper alloys (within a few minutes for simulated fingertip contamination) and Cu/Zn brasses were very effective at lower copper concentration. Exposure to copper destroyed the viral genomes and irreversibly affected virus morphology, including disintegration of envelope and dispersal of surface spikes. Cu(I) and Cu(II) moieties were responsible for the inactivation, which was enhanced by reactive oxygen species generation on alloy surfaces, resulting in even faster inactivation than was seen with nonenveloped viruses on copper. Consequently, copper alloy surfaces could be employed in communal areas and at any mass gatherings to help reduce transmission of respiratory viruses from contaminated surfaces and protect the public health.

Importance: Respiratory viruses are responsible for more deaths globally than any other infectious agent. Animal coronaviruses that "host jump" to humans result in severe infections with high mortality, such as severe acute respiratory syndrome (SARS) and, more recently, Middle East respiratory syndrome (MERS). We show here that a closely related human coronavirus, 229E, which causes upper respiratory tract infection in healthy individuals and serious disease in patients with comorbidities, remained infectious on surface materials common to public and domestic areas for several days. The low infectious dose means that this is a significant infection risk to anyone touching a contaminated surface. However, rapid inactivation, irreversible destruction of viral RNA, and massive structural damage were observed in coronavirus exposed to copper and copper alloy surfaces. Incorporation of copper alloy surfaces in conjunction with effective cleaning regimens and good clinical practice could help to control transmission of respiratory coronaviruses, including MERS and SARS.

FIG 1

Persistence of infectious human coronavirus on common surface materials. Approximately 10³ PFU HuCoV-229E (20 µl infected-cell lysate) was applied to 1-cm² coupons of test surface materials and incubated at ambient conditions (21°C; relative humidity, 30% to 40%). Virus was removed and assayed for infectivity at various time points as described in the text. Although the initial inoculum concentration was quite low, the virus retained infectivity for 5 days on all surfaces, except silicon rubber. Therefore, natural contamination of common surface material with very few coronavirus particles could represent a considerable risk of infection spread if touched and transferred to facial mucosa. Error bars represent ± SEM, and data are from the results of multiple experiments.

FIG 2

Rapid inactivation of human coronavirus occurs on brass and copper nickel surfaces. Approximately 10³ PFU HuCoV-229E (20 µl infected-cell lysate) was applied to 1-cm² coupons of a range of brasses (A and B [early time points only]), copper nickels (C), and control metal surfaces that did not contain copper (stainless steel, zinc, and nickel). Virus was removed at various time points and assayed for infectivity as described in the text. Coronavirus was inactivated in ≤40 min on brasses and 120 min on copper nickels containing less than 70% copper. Analysis of the initial 30 min of contact between virus and brasses (Fig. 2B) reveals an initial lag followed by rapid inactivation. Stainless steel and nickel did not demonstrate any antiviral activity, although mild antiviral activity was observed on zinc (this was significant only at 60 min [P = 0.046]). (D) The same inoculum was applied as 1 µl/cm², was dried immediately to simulate fingertip touch contamination, and was found to have inactivated the virus approximately 8 times faster. Error bars represent ± SEM, and data are from the results of multiple experiments.

FIG 3

Comparison between brasses and copper nickels (containing the same percentage of copper) used to inactivate human coronavirus to determine if zinc content enhances the antiviral effect. Approximately 10³ PFU was inoculated onto alloys containing 90% copper for 0, 5, and 20 min (A) or 70% copper for 0, 30, and 60 min (B) and was then removed and assessed for infectivity as described in the text. Alloys containing 90% copper were very effective at inactivating human coronavirus (A), but variations in efficacy did not appear to be related only to the presence of zinc. The presence of copper nickel C70600 resulted in increased efficacy compared to that of copper nickel C72500; that result may be linked to surface oxide layer or copper ion release from this alloy. However, at a lower percentage of copper (B), synergy with zinc or Cu(I) release may be important because contact with cartridge brass resulted in virus inactivation that was at least 3 times faster than that seen with copper nickel C71500.

FIG 4

Inactivation of coronavirus on copper and cartridge brass surfaces in the presence of chelators EDTA and BCS (A and C) and quenchers d-mannitol and Tiron (B and D) to remove Cu(II) or Cu(I) ionic species and hydroxyl radical or superoxide, respectively. Both chelators protected coronavirus from inactivation on copper and brass surfaces, suggesting that release of Cu(I) and Cu(II) is required for antiviral activity. Tiron was protective for the first hour of contact on copper and brass surfaces, indicating that superoxide is directly or indirectly involved in the inactivation mechanism. However, d-mannitol gave minimum protection on copper but prolonged protection on brass surfaces. Increasing the concentration of d-mannitol did not affect the results (not shown). This suggests that copper ions are the main moieties responsible for inactivation of coronavirus on 100% copper surfaces but that generation of hydroxyl radicals becomes more significant as the concentration of copper in the alloy is reduced. EDTA, BCS, d-mannitol, and Tiron did not significantly affect the infectivity of HuCoV-229E on stainless steel controls or in suspensions (not shown). Error bars represent ± SEM, and data are from the results of multiple experiments.

FIG 5

Destruction of human coronavirus viral genome on copper and copper alloy surfaces. (A) Analysis of a small fragment (136-bp region of the nsp4 gene) of the coronavirus genome revealed a reduction in copy number from virus exposed to copper and cartridge brass surfaces in reverse transcriptase real-time PCR. There was some reduction on stainless steel but none in viral suspension (lightest gray bars), suggesting that this was due to sample drying. (B) Analysis of the entire viral genome is represented in electrophoretic separation of viral RNA extracted from virus exposed to copper (lanes 1, 4, and 7), cartridge brass (lanes 2, 5, and 8), and stainless steel (lanes 3, 6, and 9) for 0 min (lanes 1 to 3), 120 min (lanes 4 to 6), and 240 min (lanes 7 to 9). The genomic RNA from virus exposed to copper and brass degraded with increased contact time. This did not occur on stainless steel; the genomic RNA remained as fragments too large to pass through the gel. However, the total amount of intact RNA was reduced at 4 h, possibly due to drying damage as seen in panel A. Lane 10 represents untreated virus, and the unmarked lane is a Bioline marker (Hyperladder I). The same procedure was used with mock-infected cells, revealing the same pattern of RNA breakdown following application to copper surfaces (not shown).

FIG 6

Exposure to copper surfaces results in morphological changes to human coronavirus. Purified HUCoV-229E was applied to metal surfaces and then removed, and a negatively stained preparation was observed using transmission electron microscopy. (A) Intact virions were visible following exposure to stainless steel for 10 min. (B) However, following exposure to copper for 10 min, many virus particles appeared to be disintegrating (indicated by a star), although some intact virions were still present (arrow). (C) After a 30-min exposure to copper, further damage had occurred and virions appeared shrunken (indicated by a star), with damage to surface spikes (arrow).