Understanding the Spike Protein in COVID-19 Vaccine in Recombinant Vesicular Stomatitis Virus (rVSV) Using Automated Capillary Western Blots

Abstract

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the viral agent that is responsible for the coronavirus disease-2019 (COVID-19) pandemic. One of the live virus vaccine candidates Merck and Co., Inc. was developing to help combat the pandemic was V590. V590 was a live-attenuated, replication-competent, recombinant vesicular stomatitis virus (rVSV) in which the envelope VSV glycoprotein (G protein) gene was replaced with the gene for the SARS-CoV-2 spike protein (S protein), the protein responsible for viral binding and fusion to the cell membrane. To assist with product and process development, a quantitative Simple Western (SW) assay was successfully developed and phase-appropriately qualified to quantitate the concentration of S protein expressed in V590 samples. A strong correlation was established between potency and S-protein concentration, which suggested that the S-protein SW assay could be used as a proxy for virus productivity optimization with faster data turnaround time (3 h vs 3 days). In addition, unlike potency, the SW assay was able to provide a qualitative profile assessment of the forms of S protein (S protein, S1 subunit, and S multimer) to ensure appropriate levels of S protein were maintained throughout process and product development. Finally, V590 stressed stability studies suggested that time and temperature contributed to the instability of S protein demonstrated by cleavage into its subunits, S1 and S2, and aggregation into S multimer. Both of which could potentially have a deleterious effect on the vaccine immunogenicity.

Figure 1

V590 vaccine HVF material was assayed with anti-S1 and/or anti-S2 using the Simple Western technology to confirm the presence of S protein and its subunits, S1 and S2, in rVSV; (A) virtual blot-like image and (B) electropherograms of S protein in V590 samples. Peak assignments were confirmed using recombinant forms of S protein, S1 subunit, and S2 subunit (Figure S3). CLU: chemiluminescence units.

Figure 2

V590 DS material was treated with 500 U/mL PNGase-F treatment to remove the N-linked glycans. Electropherograms of control and PNGase-treated material assayed with anti-S1 and anti-S2 antibodies are displayed. Similar experiment was conducted for glycosylated recombinant S protein (Figure S4). CLU: chemiluminescence units.

Figure 3

Stabilized recombinant S protein was used as the reference standard for the S-protein mass assay to quantify S protein in V590 samples. A standard curve was established from 6.25 to 100 ng/mL CLU: chemiluminescence units.

Figure 4

S-protein mass assay, μplaque assay, and/or rVSV protein mass assay were used to analyze both upstream and downstream V590 samples to understand the S-protein concentration, infectious particle concentration, and rVSV protein concentration throughout the downstream purification steps, respectively. (A) Semi-log plot of S-protein concentration and potency for HVF material from 0 to 3 dpi. (B) Plot of S protein and potency at each process step normalized to HVF. (C) Percent cumulative mass yield plot of S protein, potency, and rVSV protein across all downstream purification steps.

Figure 5

Electropherograms of S-protein profile across all downstream purification steps (HVF through SFP). Inset: Zoomed-in view of overlayed CB and CCP S-protein profiles. CLU: chemiluminescence units.

Figure 6

Stability of S protein in V590 BRFP samples generated in two different processes was monitored at 25 °C for 15 days. (A) S-protein profile of V590 using process 1 at t = 0, 3, 7, and 15 days; (B) S-protein profile of V590 using process 2 at t = 0, 3, 7, and 15 days; and (C) plot of S protein and potency normalized to time zero (t = 0) time point result in correlating degradation rates of S-protein levels and potency. CLU: chemiluminescence units.