SARS-CoV-2 vaccine ChAdOx1 nCoV-19 infection of human cell lines reveals low levels of viral backbone gene transcription alongside very high levels of SARS-CoV-2 S glycoprotein gene transcription

Abstract

Background: ChAdOx1 nCoV-19 is a recombinant adenovirus vaccine against SARS-CoV-2 that has passed phase III clinical trials and is now in use across the globe. Although replication-defective in normal cells, 28 kbp of adenovirus genes is delivered to the cell nucleus alongside the SARS-CoV-2 S glycoprotein gene.

Methods: We used direct RNA sequencing to analyse transcript expression from the ChAdOx1 nCoV-19 genome in human MRC-5 and A549 cell lines that are non-permissive for vector replication alongside the replication permissive cell line, HEK293. In addition, we used quantitative proteomics to study over time the proteome and phosphoproteome of A549 and MRC5 cells infected with the ChAdOx1 nCoV-19 vaccine.

Results: The expected SARS-CoV-2 S coding transcript dominated in all cell lines. We also detected rare S transcripts with aberrant splice patterns or polyadenylation site usage. Adenovirus vector transcripts were almost absent in MRC-5 cells, but in A549 cells, there was a broader repertoire of adenoviral gene expression at very low levels. Proteomically, in addition to S glycoprotein, we detected multiple adenovirus proteins in A549 cells compared to just one in MRC5 cells.

Conclusions: Overall, the ChAdOx1 nCoV-19 vaccine's transcriptomic and proteomic repertoire in cell culture is as expected. The combined transcriptomic and proteomics approaches provide a detailed insight into the behaviour of this important class of vaccine using state-of-the-art techniques and illustrate the potential of this technique to inform future viral vaccine vector design.

Fig. 1

Transcription map of detected transcripts in non-permissive and permissive human cells. a A simplified schematic of the S glycoprotein expression cassette as inserted into the ChAdOx1 nCoV-19 vector backbone in place of the E1 region of ChAdOx1. The nucleotide positions of major features in the expression cassette are indicated. Black transcripts originate from the top strand of the dsDNA viral genome (illustrated by two black lines with circles at the 5′ ends) and blue transcripts originate from the complementary strand. Each arrow depicts the expected transcription start and polyadenylation site but omits splicing events within each transcript that give rise to mature mRNA. Most ChAdOx1 structural proteins are transcribed starting at the same promoter (known as the major late promoter) ending at one of 5 different polyadenylation sites (known as L1–L5). However, the 22K gene can also be expressed from its own promoter as shown. Typically, every member of this late transcript group has three obligate exons at the beginning known as the tripartite leader. In addition, an occasional 4th exon is included containing an ORF known as the i-leader protein. In each subsequent part (b to e), the drawing from IGV viewer illustrates the structure of the dominant transcripts expressed from the 35,539 bp ChAdOx1 nCoV-19 genome that codes for each indicated ORF as detected by dRNAseq in MRC5 cells (b), A549 cells (c) and 293 cells (d). For the E4 region, we have marked with an asterisk those E4 ORFs that are in fact derived from HuAd5. The rectangles indicate exons joined by lines indicating introns. In each case, the dominant transcript for the S glycoprotein of SARS-CoV-2 is indicated in red. Transcripts that map to the top strand of the virus genome are in black and those that map to the bottom strand (and are reversed in orientation) are in blue. e The structure of dominant transcripts derived from the 4323 bp of the HuAd5 E1 region integrated into the genome of HEK293 cells

Fig. 2

Transcript heterogeneity. This figure illustrates the structure of various transcripts observed at very low levels (less than 10 transcripts per dataset) all of which initiate at the transcription start site that drives expression of the S glycoprotein. For orientation, the structure of the dominant S glycoprotein transcript is shown in red in the examples selected from the MRC5 dataset. Three sets of example transcripts are shown from MRC5, A549 and HEK293 cells. The transcripts shown arise from splicing events that utilise only the canonical GU-AG splice donor/acceptor pair present in over 90% of eukaryotic splicing events. As can be seen, some of the transcripts arise from unintended splice site usage as well as from missing the intended BGH polyadenylation signal and utlilising other polyadenylation signals further down the vector genome

Fig. 3

Novel polyadenylation site usage for preVI. a The transcript structure of transcripts coding for three proteins classically considered to be L3 proteins viz preVI, hexon and 23K proteinase. The three vertical boxes on the left represent the tripartite exons, labelled TPL1, TPL2 and TPL3, and normally included in all major late transcripts generated by mammalian adenoviruses. The L3 transcripts would classically share the same L3 polyadenylation site (indicated on the diagram). In this diagram, the dominant transcripts seen in ChAdOx1 nCoV-19-infected 293 cells for each ORF are coloured red. This figure includes, in black, the transcript structure for observed transcripts that would both code for preVI and fit the canonical transcript structure for an L3 transcript, utilising the L3 polyadenylation site. In the case of 293 cells, however, there are only 192 copies of this transcript compared to 403 copies of the novel preVI transcript indicated in red at the bottom of a. b The sequences at the proposed polyadenylation site on ChAdOx1 showing the location of the polyadenylation signal and the GU-rich region that is often present downstream of a polyadenylation signal. c The equivalent regions in the genome of human adenoviruses 5 (a species C adenovirus) and human adenovirus type 4 which is, like ChAdOx1, a species E adenovirus. Taken together, b, c and d illustrate the similarities between the two species E adenovirus genomes and their distinction from a species C adenovirus genome