Longitudinally profiling neutralizing antibody response to SARS coronavirus with pseudotypes

Abstract

The severe acute respiratory syndrome-associated coronavirus (SARS-CoV) spike protein (S) is a major target for neutralizing antibodies. Retroviral SARS-CoV S pseudotypes have been constructed and used to develop an in vitro microneutralization assay that is both sensitive and specific for SARS-CoV neutralizing antibodies. Neutralization titers measured by this assay are highly correlated to those measured by an assay using replication-competent SARS-CoV. No cross-neutralization occurred with human sera known to contain antibodies to coronavirus strains OC43 and 229E. The pseudotype assay was used to profile neutralizing antibody responses against SARS-CoV S in sequential serum samples taken from 41 confirmed SARS patients during the 2003 outbreak in Hong Kong and shows long-lasting immunity in most recovered patients. The pseudotype assay does not require handling live SARS virus; it is a useful tool to determine neutralizing titers during natural infection and the preclinical evaluation of candidate vaccines.

Figure 1

Infectivity of retroviral severe acute respiratory syndrome–associated coronavirus (SARS-CoV) spike protein (S) pseudotypes on target cells. SARS-CoV S-mediated infection of human 293T, TE671, and Quail QT6/ACE2 was assessed. Murine leukemia virus (MLV) or HIV pseudotypes bearing either the pantropic vesicular stomatitis virus envelope protein (VSV-G) as a positive control, or the SARS-CoV S, were added to target cells. After 72 h, green fluorescent protein (GFP)–positive cells were counted by fluorescence-activated cell sorter analysis. Infection titers are given as infectious units per milliliter (IU/mL). Arrow indicates that infection titer was less than the detection limit, 10² IU/mL.

Figure 2

Correlation of neutralizing antibody titers measured by plaque reduction assay with titers measured with pseudotype assay. LV, neutralizing antibody titer by using replication-competent severe acute respiratory syndrome–associated coronavirus (SARS-CoV) (live virus); PV, neutralizing antibody titer by using pseudotype virus; PV₉₀ (filled black diamonds), 90% neutralizing antibody titer by using murine leukemia virus (MLV) (SARS) pseudotype virus; PV₅₀ (open squares), 50% neutralizing antibody titer. Logarithmic trendlines were fitted to the data by using Microsoft Excel 2003 (Microsoft Corp., Redmond, WA, USA). Correlation coefficients for LV versus PV₉₀ and LV versus PV₅₀ are 0.69 and 0.78, respectively.

Figure 3

Severe acute respiratory syndrome–associated coronavirus (SARS-CoV) neutralizing antibody–positive rate by time of blood sample collection (days after onset of fever). Black bars represent the number of patients tested for neutralizing antibodies (Nab). White bars represent the number of patients whose assayed samples were positive for neutralizing antibodies (Nab+). Samples are considered positive for Nab if the 90% neutralizing antibody titer determined by using murine leukemia virus (MLV) (SARS) pseudotypes is >10. Line plot with open white circles shows the geometric mean (GM) Nab titer within each time frame. IC₉₀, 90% inhibitory concentration.

Figure 4

Neutralizing antibodies to severe acute respiratory syndrome–associated coronavirus spike protein in sequential blood samples from 4 representative patients. Lines represent profiles of individual patients. Filled black symbols represent geometric mean titers at individual time points. IC₉₀, 90% inhibitory concentration.