key: cord-0261637-ow8kg831
authors: Banerjee, Arinjay; Lew, Jocelyne; Kroeker, Andrea; Baid, Kaushal; Aftanas, Patryk; Nirmalarajah, Kuganya; Maguire, Finlay; Kozak, Robert; McDonald, Ryan; Lang, Amanda; Gerdts, Volker; Straus, Sharon E.; Gilbert, Lois; Li, Angel Xinliu; Mozafarihasjin, Mohammad; Walmsley, Sharon; Gingras, Anne-Claude; Wrana, Jeffrey L.; Mazzulli, Tony; Colwill, Karen; McGeer, Allison J.; Mubareka, Samira; Falzarano, Darryl
title: Immunogenicity of convalescent and vaccinated sera against clinical isolates of ancestral SARS-CoV-2, beta, delta, and omicron variants
date: 2022-01-13
journal: bioRxiv
DOI: 10.1101/2022.01.13.475409
sha: a5e24ce70ac7d0f58355730bf3b4a9487a34b173
doc_id: 261637
cord_uid: ow8kg831

The omicron variant of concern (VOC) of SARS-CoV-2 was first reported in November 2021 in Botswana and South Africa. Omicron has evolved multiple mutations within the spike protein and the receptor binding domain (RBD), raising concerns of increased antibody evasion. Here, we isolated infectious omicron from a clinical specimen obtained in Canada. The neutralizing activity of sera from 65 coronavirus disease (COVID-19) vaccine recipients and convalescent individuals against clinical isolates of ancestral SARS-CoV-2, beta, delta, and omicron VOCs was assessed. Convalescent sera from unvaccinated individuals infected by the ancestral virus during the first wave of COVID-19 in Canada (July, 2020) demonstrated reduced neutralization against beta and omicron VOCs. Convalescent sera from unvaccinated individuals infected by the delta variant (May-June, 2021) neutralized omicron to significantly lower levels compared to the delta variant. Sera from individuals that received three doses of the Pfizer or Moderna vaccines demonstrated reduced neutralization of the omicron variant relative to ancestral SARS-CoV-2. Sera from individuals that were naturally infected with ancestral SARS-CoV-2 and subsequently received two doses of the Pfizer vaccine induced significantly higher neutralizing antibody levels against ancestral virus and all VOCs. Importantly, infection alone, either with ancestral SARS-CoV-2 or the delta variant was not sufficient to induce high neutralizing antibody titers against omicron. This data will inform current booster vaccination strategies, and we highlight the need for additional studies to identify longevity of immunity against SARS-CoV-2 and optimal neutralizing antibody levels that are necessary to prevent infection and/or severe COVID-19.

CoV-2. Sera from individuals that were naturally infected with ancestral SARS-CoV-2 and 53 subsequently received two doses of the Pfizer vaccine induced significantly higher neutralizing 54 antibody levels against ancestral virus and all VOCs. Importantly, infection alone, either with 55 ancestral SARS-CoV-2 or the delta variant was not sufficient to induce high neutralizing 56 antibody titers against omicron. This data will inform current booster vaccination strategies, and 57 we highlight the need for additional studies to identify longevity of immunity against SARS-58

CoV-2 and optimal neutralizing antibody levels that are necessary to prevent infection and/or 59 severe INTRODUCTION 62 SARS-CoV-2 has continued to evolve since its emergence in December 2019 1,2 . Variants of 63 SARS-CoV-2 that demonstrate potential for interference with diagnostics, therapies, and vaccine 64 efficacy, along with evidence for increased transmissibility or disease severity are termed 65 variants of concern (VOCs). The most recent VOC, omicron was first reported in November 66 2021 in Botswana and South Africa 3,4 . The omicron variant has evolved multiple mutations 67 within the spike protein and the receptor binding domain (RBD) that raise concerns regarding a 68 possible increased ability to evade pre-existing antibodies, both from prior infection and from 69 vaccination 5 . The omicron variant has demonstrated increased transmission and a higher level of 70 resistance to antibody-mediated neutralization 4,5 . However, little is known about its 71 pathogenicity and whether disease severity is altered in convalescent, vaccinated or unvaccinated 72 individuals. In Canada, long-term care (LTC) residents were prioritized for third vaccine doses 73 against SARS-CoV-2 based on the observation that antibody titers in older adults waned within 74 six months of their second vaccine dose 6,7 . The neutralizing potential of antibodies generated in 75 LTC residents against VOCs, such as delta and omicron after three doses of mRNA vaccines 76 remain unknown. Thus, to better assess the efficacy of antibody-mediated neutralization against 77 ancestral SARS-CoV-2 and VOCs (beta, delta, and omicron) in naturally infected and vaccinated 78 individuals, we collected sera from multiple cohorts and tested their neutralization ability against 79 clinical isolates of ancestral SARS-CoV-2 and VOCs. 80

Isolates of VOCs used in this study were derived from clinical specimens. Nasopharyngeal 84 swabs were collected from PCR positive patients, and virus isolation was performed on African 85

Green monkey kidney cells (Vero'76) as previously described 8 . We confirmed the whole 86 genome sequence of the isolates and determined their phylogenetic relationship with other 87 SARS-CoV-2 isolates (Figure 1A and see supplementary Table S1 ). Beta, delta and omicron 88 isolates used in this study aligned with their expected lineages ( Figure 1A ). SARS-CoV-2 can 89 rapidly adapt in cell culture and evolve adaptive mutations. We confirmed mutations across the 90 full-length viral genome, including the spike protein for all variants prior to using the viruses in a 91 micro-neutralization assay ( Figure 1B and see supplementary Table S2 ). 92 93

To determine the neutralizing titer of sera from individuals that were naturally infected with 96 SARS-CoV-2, we tested convalescent sera from individuals that were infected with ancestral 97 SARS-CoV-2 during the first wave of coronavirus disease in Canada (July 2020; 98 see supplementary Table S3 ). Serum samples were collected 1-5 months after the onset of 99 COVID-19 (see supplementary Table S3 ). Convalescent sera (n=15) from individuals infected 100 with ancestral SARS-CoV-2 during the first wave of COVID-19 in Canada contained 101 significantly lower neutralizing antibodies against both beta (p=0.0058) and omicron (p=0.0019) 102 variants, relative to the ancestral virus ( Figure 2A ). However, neutralizing antibody titers in these 103 serum samples were not significantly different between ancestral SARS-CoV-2 and the delta 104 variant (p=0.1691; Figure 2A ). Next, the neutralizing antibody titers in convalescent sera (n=10) 105 from individuals who were infected with the delta variant in Canada between May and June, 106 2021 were determined (see supplementary Table S3 ). Serum samples were collected 1-2 months 107 after the date of onset of COVID-19 (see supplementary Table S3 ). Convalescent delta sera 108 contained lower levels of neutralizing antibodies against both beta (p=0.0468) and omicron 109 variants (p=0.049) relative to delta variant ( Figure 2B ), while titers against ancestral SARS-CoV-110 2 and delta were comparable (p=0.6034; Figure 2B ). These data suggest that infection with either 111 ancestral SARS-CoV-2 or the delta variant induces cross-neutralizing antibody with comparable 112 titers against both viruses. However, natural infection with either ancestral SARS-CoV-2 or the 113 delta variant induces significantly lower levels of neutralizing antibodies against both beta and 114 omicron variants. 115 116

Our data suggest that natural infection with ancestral SARS-CoV-2 is not sufficient to induce 119 comparable levels of neutralizing antibodies against beta and omicron variants ( Figure 2A ). 120

Next, we determined levels of neutralizing antibodies in sera (n=10) from individuals who had 121 received two doses of the Pfizer BNT162b2 vaccine after being naturally infected with ancestral 122 SARS-CoV-2 (see supplementary Table S3 ). Two doses of the Pfizer BNT162b2 vaccine after 123 natural infection led to higher levels of neutralizing antibodies against ancestral SARS-CoV-2 124 that were significantly higher than levels against the beta (p=0.0318) and omicron (p=0.0001) 125 variants ( Figure 2C ). Infection and subsequent two dose vaccination with Pfizer BNT162b2 126 induced higher levels of neutralizing antibodies against ancestral virus and all VOCs, including 127 omicron than infected only individuals ( Figure 2D ). 128 129

BNT162b2 mRNA vaccine 131

To determine neutralizing antibody titers in sera from individuals that received one dose of the 132 Pfizer BNT162b2 mRNA vaccine, we collected sera (n=10) one month after the first dose of the 133 vaccine and tested neutralizing antibody titers against ancestral SARS-CoV-2, beta, delta and 134 omicron variants (see supplementary Table S3 ). Low neutralizing antibody titers against the 135 ancestral virus were detected in some samples; however, no neutralization of the three VOCs 136 was observed, with the exception of one serum sample that had detectable levels of neutralizing 137 antibodies against all VOCs ( Figure 3A ). 138 139

Additional booster vaccinations have been deemed critical in protecting us from VOCs in part by 141 inducing higher levels of neutralizing antibodies. Thus, we tested the levels of neutralizing 142 antibodies in sera collected from long-term care residents that received three doses of the Pfizer 143 BNT162b2 (n=10) 9 or the Moderna mRNA-1273 (n=10) 10 vaccines (see supplementary Table  144 S3). For both vaccine recipients, doses 1 and 2 were received 3-4 weeks apart. The third vaccine 145 dose was received ~7 months after dose 2, and serum samples were collected 1 month after the 146 third dose. Sera from individuals that received three doses of the Pfizer BNT162b2 vaccine 147 induced high neutralizing titers against ancestral SARS-CoV-2, but levels of neutralizing 148 antibodies were significantly lower against beta (p=0.0007), delta (p=0.0045) and omicron 149 (p<0.0001) variants, compared to ancestral SARS-CoV-2 ( Figure 3B ). Sera from individuals that 150 received three doses of the Moderna mRNA-1273 vaccine induced high neutralizing titers 151 against ancestral SARS-CoV-2, but levels of neutralizing antibodies were significantly lower 152 against the omicron variant, relative to ancestral SARS-CoV-2 (p=0.0012; Figure 3C ). 153

Neutralizing antibody titers were not significantly different against ancestral SARS-CoV-2, beta 154 and delta variants. Serum samples from individuals that received 3x doses of the Pfizer 155

BNT162b2 vaccine contained 2.86x, 2.25x and 10.3x lower mean neutralizing antibody titers 156 against beta, delta and omicron VOCs, respectively, relative to ancestral SARS-CoV-2 ( Figure  157 4). Serum samples from individuals that received 3x doses of the Moderna mRNA-1273 vaccine 158 contained 1.7x, 1.26x and 3.48x lower mean neutralizing antibody titers against beta, delta and 159 omicron VOCs, respectively, relative to ancestral SARS-CoV-2 ( Figure 4) . 160

The emergence of an yet another SARS-CoV-2 VOC, omicron has led to increasing speculation 163 about the ability of this variant to escape vaccine and natural infection-mediated immunity. The 164 current generation of COVID-19 mRNA vaccines are designed using the spike gene sequence of 165 ancestral SARS-CoV-2 9,10 . The omicron variant has accumulated 29 amino acid substitutions, 3 166 amino-acid deletions and a 3-residue insertion within the spike protein compared to the ancestral 167 SARS-CoV-2 Wuhan isolate 5 . Accumulating data suggest that the omicron variant is at least 168 partially resistant to neutralization by antibodies in vaccinated individuals, along with partial or 169 complete resistance to neutralization by therapeutic monoclonal antibodies 5 . Emerging data 170 demonstrate that T-cell-mediated immunity generated upon infection or vaccination likely 171 remain effective against the omicron variant 11 , and an additional booster vaccine dose results in 172 higher levels of antibodies against the omicron variant when tested using pseudotyped viruses 12 . 173

Despite these recent advances, considerable gaps currently exist in our knowledge regarding the 174 ability of omicron to cause severe COVID-19 and whether partial or complete escape of vaccine 175 or natural infection-mediated immunity occurs and if escape is age dependent. In addition, it is 176 not known if omicron has altered host range, or if transmissibility is increased and whether there 177 are changes in cellular tropism. Furthermore, data on neutralizing antibody titers against clinical 178 isolates of omicron are limited. Thus, as part of this study, we determined the levels of 179 neutralizing antibodies in individuals that were naturally infected, infected and subsequently 180 vaccinated, or vaccinated with three doses of mRNA vaccines using clinical isolates of ancestral 181 SARS-CoV-2, beta, delta and omicron variants. 182

When omicron was first detected, multiple laboratories reported difficulties in isolating 183 and generating laboratory stocks of this variant. In this study, we used Vero'76 cells to isolate 184 the omicron variant from a clinical specimen (nasopharyngeal swab) that was collected from a 185

Canadian patient. We also confirmed the whole genome sequences of the omicron variant, along 186 with beta and delta variants ( Figure 1 and Table 1 ). Thus, we report that Vero'76 cells are 187 sufficient to facilitate the isolation and propagation of the omicron variant. 188

Next, we tested the levels of neutralizing antibody titers in convalescent sera against the 189 ancestral virus, beta, delta and omicron variants ( were lower compared to both the ancestral virus and the delta variant (Figures 2A and 2B) . Thus, 193 our data suggest that infection alone, either with the ancestral virus or the delta variant may not 194 be sufficient to induce high levels of neutralizing antibodies against the omicron variant. Indeed, 195 our data demonstrate that two doses of the Pfizer BNT162b2 mRNA vaccine following infection 196 with ancestral SARS-CoV-2 induced significantly higher levels of neutralizing antibodies against 197 ancestral SARS-CoV-2 and beta, delta and omicron variants ( Figure 2D ). 198

Our data highlight that one dose of the Pfizer vaccine is not sufficient to induce high 199 levels of neutralizing antibodies against ancestral virus or variants ( Figure 3A) , thus the second 200 vaccine doses appears to be critically required to induce neutralizing antibodies. Three doses of 201 either the Pfizer BNT162b2 or Moderna mRNA-1273 vaccine induced comparable neutralizing 202 antibodies against the beta and omicron variants ( Figures 3B and 3C) . However, levels of 203 neutralizing antibodies against omicron were significantly lower compared to ancestral SARS-204

CoV-2 in serum samples from individuals vaccinated with 3x doses of either of the mRNA 205 vaccines ( Figures 3B, 3C and 4) . 206

In summary, our data demonstrate that infection alone, either with ancestral SARS-CoV-207 2 or the delta variant is not sufficient to induce high levels of neutralizing antibodies against 208 omicron. However, two doses of the Pfizer vaccine in previously infected individuals induces 209 higher levels of neutralizing antibodies. While we did not test the effect of two doses of the 210

Moderna mRNA-1273 vaccine in previously infected and recovered individuals, we speculate 211 that the results will be comparable to the Pfizer BNT162b2 vaccine. Our data also show that 212 while 3x doses of both mRNA vaccines induce neutralizing titers against omicron variant in 213 long-term care residents, the levels of neutralizing antibodies remain significantly lower 214 compared to ancestral SARS-CoV-2. Thus, our data support the ongoing third vaccine dose 215 booster strategy for long-term care residents in Canada. Indeed, there is a need for studies to 216 assess the optimal levels of neutralizing antibodies that are required for protection against 217 infection and/or severe COVID-19, which will inform policies around vaccine boosters and 218 enable equitable distribution of vaccines to end the ongoing pandemic. figure generated using ETE v3.1.2 27 . Contexual sequences were incorporated into the 412 phylogenetic analysis by using Nexstrain's ingested GISAID metadata and pandas to randomly 413 sample a representative subset of sequences (jointly deposited in NCBI and GISAID) that 414 belonged to lineages observed in Canada (see also Supplemental Tables S1 and S2 for metadata). 415 416 Micro-neutralization assay. Serum samples were heat-inactivated at 56°C for 30 minutes and 417 then serially diluted 1:2 in DMEM supplemented with 2% FBS and 1% P/S (vDMEM) in 96-418 well blocks. The final volume of diluted serum per well was 180 μL. Each virus was diluted to 419

A pneumonia outbreak associated with a new 433 coronavirus of probable bat origin

Evolutionary trajectory of 435 SARS-CoV-2 and emerging variants

Omicron SARS-CoV-2 variant: a new chapter in the COVID-19 437 pandemic

Rapid epidemic expansion of the SARS-CoV-2 439 Omicron variant in southern Africa

Considerable escape of SARS-CoV-2 Omicron to 441 antibody neutralization

Neutralizing antibody responses to SARS-443

CoV-2 variants in vaccinated Ontario long-term care home residents and workers. medRxiv 2021

After a Third Dose of the SARS-CoV-2 BNT162b2 Vaccine in Adults Aged >/=60 Years

Replication Kinetics of Severe Acute Respiratory Syndrome Coronavirus 2

Safety and Efficacy of the BNT162b2 mRNA 451 Covid-19 Vaccine

Efficacy and Safety of the mRNA

CoV-2 Vaccine

SARS-CoV-2 spike T cell responses induced 455 upon vaccination or infection remain robust against Omicron. medRxiv 2021

mRNA-based COVID-19 vaccine 457 boosters induce neutralizing immunity against SARS-CoV-2 Omicron variant. medRxiv 2021

A Comparison of Whole Genome Sequencing of 459 SARS-CoV-2 Using Amplicon-Based Sequencing, Random Hexamers, and Bait Capture

Persistence of serum and saliva antibody responses to 462 SARS-CoV-2 spike antigens in COVID-19 patients

Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM

Fast and accurate short read alignment with Burrows-Wheeler transform

The Sequence Alignment/Map format and 470 SAMtools

An amplicon-based sequencing framework 472 for accurately measuring intrahost virus diversity using PrimalSeq and iVar

Haplotype-based variant detection from short-read sequencing

A program for annotating and predicting the 477 effects of single nucleotide polymorphisms

Assignment of Epidemiological Lineages in an 479 Emerging Pandemic Using the Pangolin Tool

A dynamic nomenclature proposal for SARS-481

CoV-2 lineages to assist genomic epidemiology

Data Structures for Statistical Computing in Python

Augur: a bioinformatics toolkit for phylogenetic 485 analyses of human pathogens

New Models and 487

Efficient Methods for Phylogenetic Inference in the Genomic Era

ETE 3: Reconstruction, Analysis, and Visualization of 490 Phylogenomic Data

supplementary Table S3 ). Cohort participants provided informed consent for sharing of serum, 375 and studies were approved by the Sunnybrook Research Institute (REB# 149-1994) and/or the 376 Mount Sinai Hospital (REB# 02-0118-U, 20-0339-E, and 21-0069-E) Research Ethics Board 13 . 377For samples from each cohort, samples were selected to have a representative range of anti-spike 378 trimer and anti-RBD antibodies as measured by enzyme-linked immunosorbent assay 14 . 379 380

Sequencing and bioinformatic analyses. cDNA was synthesized from extracted RNA. In brief, 382 4 µL LunaScript RT SuperMix 5X (New England Biolabs, NEB, USA) and 8 µL nuclease free 383 water, were added to 8 µL extracted RNA. cDNA synthesis was performed using the following 384 conditions: 25 °C for 2 min, 55 °C for 20 min, 95 °C for 1 min, and holding at 4 °C. 385Amplicons were generated from cDNA using ARTIC V4 primer pools 386 (https://github.com/artic-network/artic-ncov2019). Two multiplex PCR tiling reactions were 387 prepared by combining 2.5 µL cDNA with 12.5 µL Q5 High-Fidelity 2X Master Mix (NEB, 388 USA), 6µL nuclease free water, and 4 µL of respective 10 µM ARTIC v4 primer pool 389 (Integrated DNA Technologies). PCR cycling was then performed in the following manner: 98 390 °C for 30 s followed by 35 cycles of 98 °C for 15 s and 63 °C for 5 min.