Coronavirus Pseudotypes for All Circulating Human Coronaviruses for Quantification of Cross-Neutralizing Antibody Responses

Abstract

The novel coronavirus SARS-CoV-2 is the seventh identified human coronavirus. Understanding the extent of pre-existing immunity induced by seropositivity to endemic seasonal coronaviruses and the impact of cross-reactivity on COVID-19 disease progression remains a key research question in immunity to SARS-CoV-2 and the immunopathology of COVID-2019 disease. This paper describes a panel of lentiviral pseudotypes bearing the spike (S) proteins for each of the seven human coronaviruses (HCoVs), generated under similar conditions optimized for high titre production allowing a high-throughput investigation of antibody neutralization breadth. Optimal production conditions and most readily available permissive target cell lines were determined for spike-mediated entry by each HCoV pseudotype: SARS-CoV-1, SARS-CoV-2 and HCoV-NL63 best transduced HEK293T/17 cells transfected with ACE2 and TMPRSS2, HCoV-229E and MERS-CoV preferentially entered HUH7 cells, and CHO cells were most permissive for the seasonal betacoronavirus HCoV-HKU1. Entry of ACE2 using pseudotypes was enhanced by ACE2 and TMPRSS2 expression in target cells, whilst TMPRSS2 transfection rendered HEK293T/17 cells permissive for HCoV-HKU1 and HCoV-OC43 entry. Additionally, pseudotype viruses were produced bearing additional coronavirus surface proteins, including the SARS-CoV-2 Envelope (E) and Membrane (M) proteins and HCoV-OC43/HCoV-HKU1 Haemagglutinin-Esterase (HE) proteins. This panel of lentiviral pseudotypes provides a safe, rapidly quantifiable and high-throughput tool for serological comparison of pan-coronavirus neutralizing responses; this can be used to elucidate antibody dynamics against individual coronaviruses and the effects of antibody cross-reactivity on clinical outcome following natural infection or vaccination.

Keywords: COVID-19; SARS-CoV-2; coronavirus; neutralization; pseudotyped virus.

Figure 1

The seven human coronaviruses belong to either the alphacoronavirus or betacoronavirus genus of the Coronavirinae sub-family of the Coronaviridae. The alphacoronaviruses include seasonal coronaviruses HCoV-229E and HCoV-NL63, whilst the betacoronaviruses include high pathogenicity coronaviruses such as SARS-CoV-1, SARS-CoV-2 and MERS-CoV in addition to seasonal coronaviruses HCoV-HKU1 and HCoV-OC43. The Coronavirinae are ubiquitous pathogens of mammalian and avian species; each of the seven human coronaviruses emerged from independent zoonotic spillover events.

Figure 2

The genomic organization of the seven human coronaviruses: all coronaviruses encode two major Open Reading Frames (ORFs) known as ORF1a and ORF1b in addition to Spike (S), Envelope (E), Membrane (M) and Nucleocapsid (N) proteins. S proteins form trimeric envelope glycoproteins utilized for entry after receptor binding, whilst other surface proteins (E and M) have roles in viral assembly. All HCoVs also encode a variable set of non-structural proteins (shown in grey), and the two seasonal betacoronaviruses HCoV-HKU1 and HCoV-OC43 additionally encode surface Haemagglutinin-Esterase (HE) proteins as receptor-destroying enzymes.

Figure 3

Optimising production of pseudotyped viruses: (a) Various amounts of spike expression plasmid were transfected for each human coronavirus, shown here with SARS-CoV-1, SARS-CoV-2 and HCoV-NL63. High titres were achieved for the entire panel using 50 ng of the pEVAC expression plasmid alongside 250 ng of p8.91 and 375 ng of pCSFLW. (b) SARS-CoV-2 PVs were harvested at either 48 h or 72 h after transfection; collection of PV containing supernatant at 72 h led to increased titres. 2-step transfection, with initial transfection of the S gene followed 24 h later by the lentiviral gag-pol and luciferase reporter plasmids and collection 72 h after the first transfection, led to a reduction in titre. Significant differences in titre shown by asterisks: * represents p ≤ 0.05, ** p ≤ 0.005 and *** p ≤ 0.0005, respectively.

Figure 4

Schematic representing pseudotyped virus (PV) production and use in PV neutralization assays. Producer cells were co-transfected with plasmids encoding a coronavirus spike protein, a lentiviral gag-pol construct and the pCSFLW firefly luciferase reporter. The lentiviral capsid assembled into a PV core and packaged the pCSFLW reporter construct (consisting of the firefly luciferase coding sequence flanked by lentiviral LTRs), before budding from the plasma membrane and acquiring an outer membrane studded with coronavirus spike proteins. After spike-mediated entry into target cells, the pCSFLW reporter was integrated and expressed, leading to measurable luminescence from successfully transduced target cells. Neutralization of PVs by anti-spike antibodies following incubation with human sera can be measured as a reduction in luminescence output after target cell transduction.

Figure 5

Enhancing titres of HCoV-OC43 and HCoV-HKU1: (a) entry of HKU1 PVs into CHO cells was enhanced after incubation with exogenous NA relative to CHO-K1 cell only control; (b) HCoV-OC43 or HCoV-HKU1 PVs were generated incorporating their corresponding Haemagglutinin Esterase (HE) protein in addition to the spike protein. Co-transfection of 12.5 ng of HE led to comparable titres to spike-only PVs for both viruses when used with TMPRSS2-transfected HEK293T/17 cells (see below). Transfection of higher inputs of HE led to reduced titres for both viruses. (c) Production of HCoV-OC43 and HCoV-HKU1 was enhanced by co-transfection of producer cells with TMPRSS2; this enhanced entry of HCoV-OC43 over 3000-fold compared to un-transfected HEK293T/17 cells. Significant differences in titre shown by asterisks: * represents p ≤ 0.05, ** p ≤ 0.005 and *** p ≤ 0.0005, respectively.

Figure 6

PVs were pseudotyped with the SARS-CoV-2 S protein in addition to combinations of the Envelope (Env) and Membrane (Memb) proteins; PVs produced with 100 ng Envelope and 400 ng Membrane in addition to 50 ng spike had the second highest titres relative to spike only.

Figure 7

Entry of PVs was assessed by titration into various cell lines. (a) Entry of SARS-CoV-1 was highest in HEK293T/17 cells, whilst entry of SARS-CoV-2 and HCoV-NL63 was comparable in both HEK293T/17 and HUH7 cells. (b) MERS-CoV and HCoV-229E preferentially entered HUH7 cells compared to HEK293T/17 cells. Significant differences in titre shown by asterisks: * represents p ≤ 0.05, ** p ≤ 0.005 and *** p ≤ 0.0005, respectively.

Figure 8

Enhancement of PV entry by target cell transfection: (a) Transfection of either HEK293T/17 cells or HUH7 cells with TMPRSS2 and ACE2 considerably boosted entry of the ACE2-using SARS-CoV-1, SARS-CoV-2 and HCoV-NL63 PVs. The optimal combination of ACE2 and TMPRSS2 for transfection into HEK293T/17 cells was determined for each PV—75 ng of TMPRSS2 expression plasmid and 500 ng of ACE2 expression plasmid supported the highest entry. (b) Entry of HCoV-HKU1 and HCoV-OC43 was enhanced by transfection of HEK293T/17 cells with TMPRSS2.

Figure 9

Neutralization of SARS-CoV-2, HCoV-NL63, HCoV-229E and HCoV-HKU1 by serum samples from 5 SARS-CoV-2 seropositive individuals (SP), 5 SARS-CoV-2 seronegative individuals (SN), 5 COVID-19 patients (patients) and 2 SARS-CoV-2 seropositive NIBSC reference sera (NISBC ref); (a) neutralization of SARS-CoV-2, HCoV-HKU1, HCoV-229E and HCoV-NL63 by 5 COVID-19 patient serum samples. Further neutralization curves by additional serum samples are shown in Supplementary Figure S1. Curves fit by non-linear regression using GraphPad Prism 9; (b) IC50 values compared between each group of serum samples for each PV. IC50 values were calculated using GraphPad Prism 9 as the dilution of sera needed for 50% neutralization of luciferase output/PV entry.

Figure 10

Example data showing log IC50 values for neutralization of HCoV-NL63 and SARS-CoV-2 at sequential time points in a cohort of patients hospitalized with COVID-19; samples were collected through the HICC consortium (https://www.hicc-consortium.com/, accessed on 10 May 2021). Serum samples were collected weekly with timepoint 1 representing a serum sample taken at admission.