key: cord-0254479-9v4chfos
authors: Yuan, Shuofeng; Ye, Zi-Wei; Liang, Ronghui; Tang, Kaiming; Zhang, Anna Jinxia; Lu, Gang; Ong, Chon Phin; Poon, Vincent Kwok-Man; Chan, Chris Chung-Sing; Mok, Bobo Wing-Yee; Qin, Zhenzhi; Xie, Yubin; Sun, Haoran; Tsang, Jessica Oi-Ling; Yuen, Terrence Tsz-Tai; Chik, Kenn Ka-Heng; Chan, Chris Chun-Yiu; Cai, Jian-Piao; Luo, Cuiting; Lu, Lu; Yip, Cyril Chik-Yan; Chu, Hin; To, Kelvin Kai-Wang; Chen, Honglin; Jin, Dong-Yan; Yuen, Kwok-Yung; Chan, Jasper Fuk-Woo
title: The SARS-CoV-2 Omicron (B.1.1.529) variant exhibits altered pathogenicity, transmissibility, and fitness in the golden Syrian hamster model
date: 2022-01-13
journal: bioRxiv
DOI: 10.1101/2022.01.12.476031
sha: 144213a254180e1d3ccb4dff42eae17804339291
doc_id: 254479
cord_uid: 9v4chfos

The newly emerging SARS-CoV-2 Omicron (B.1.1.529) variant first identified in South Africa in November 2021 is characterized by an unusual number of amino acid mutations in its spike that renders existing vaccines and therapeutic monoclonal antibodies dramatically less effective. The in vivo pathogenicity, transmissibility, and fitness of this new Variant of Concerns are unknown. We investigated these virological attributes of the Omicron variant in comparison with those of the currently dominant Delta (B.1.617.2) variant in the golden Syrian hamster COVID-19 model. Omicron-infected hamsters developed significantly less body weight losses, clinical scores, respiratory tract viral burdens, cytokine/chemokine dysregulation, and tissue damages than Delta-infected hamsters. The Omicron and Delta variant were both highly transmissible (100% vs 100%) via contact transmission. Importantly, the Omicron variant consistently demonstrated about 10-20% higher transmissibility than the already-highly transmissible Delta variant in repeated non-contact transmission studies (overall: 30/36 vs 24/36, 83.3% vs 66.7%). The Delta variant displayed higher fitness advantage than the Omicron variant without selection pressure in both in vitro and in vivo competition models. However, this scenario drastically changed once immune selection pressure with neutralizing antibodies active against the Delta variant but poorly active against the Omicron variant were introduced, with the Omicron variant significantly outcompeting the Delta variant. Taken together, our findings demonstrated that while the Omicron variant is less pathogenic than the Delta variant, it is highly transmissible and can outcompete the Delta variant under immune selection pressure. Next-generation vaccines and antivirals effective against this new VOC are urgently needed. One Sentence Summary The novel SARS-CoV-2 Omicron variant, though less pathogenic, is highly transmissible and outcompetes the Delta variant under immune selection pressure in the golden Syrian hamster COVID-19 model.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was first identified in late 2019 69 and quickly developed into the most important global health challenge in recent decades (1) (2) (3) . (16). What is more apparent from early epidemiological data is that the 85 Omicron variant is spreading rapidly even in populations with high two-dose COVID-19 86 vaccination uptake rates (17, 18) . However, whether this is due to the intrinsic transmissibility of , which closely simulates non-lethal human disease (19) .

The Omicron variant is less pathogenic than the Delta variant in vivo 95 We first compared the clinical signs, viral burden, and cytokine/chemokine profiles of the 96 Omicron and Delta variants in the golden Syrian hamster model (Fig. 1A) . The body weight 97 losses (<5%) (P<0.01 to P<0.0001) (Fig. 1B) and clinical scores (P<0.05 to P<0.01) (Fig. 1C ) of 98 the Omicron-infected hamsters were limited, and consistently and significantly milder than those 99 of the Delta-infected hamsters from 2 days post-infection (dpi) to 7 dpi. The Omicron-infected 100 hamsters started to regain their body weights at about 4-5 dpi, which was earlier than the Delta-101 infected hamsters (Fig. 1B) . Early after infection (2 dpi), the viral loads and infectious virus 102 titers of the two variants in the nasal turbinate and trachea were similar, but the lung viral loads 103 (P<0.05) and virus titers were about 2 log 10 units (P<0.0001 ) lower in the Omicron-infected than 104 Delta-infected hamsters (Fig. 1D) . During the acute (4 dpi) (Fig. 1E ) and regenerative (7 dpi) 105 (Fig. 1F ) phases of infection, the viral burden of the Omicron variant became consistently lower 106 than that of the Delta variant throughout the upper and lower respiratory tract. At 7 dpi, the viral 107 titers in the trachea and lung were already below the detection limit (<100 PFU/mg) in the 108 Omicron-infected hamsters (Fig. 1F) . Virus shedding of the Omicron variant in the oral swabs 109 (Fig. S1A) and feces (Fig. S1B ) became undetectable by 12 dpi and 14 dpi, which were both 110 earlier than the Delta variant. Corroborative to the viral burden findings, the Omicron-infected 111 hamsters had generally lower tissue cytokines/chemokines gene expression levels between 2 dpi 112 and 7 dpi (Fig. 1G) . At 7 dpi, the dysregulated inflammatory cytokine/chemokine response has 113 6 almost completely normalized. The antibody response against variant-specific spike receptor-114 binding domain (RBD) of the Omicron-infected hamsters was also significantly lower than that 115 of the Delta-infected hamsters (Fig. S2) . 116 In the lung sections collected at 2 dpi, the Omicron-infected hamsters showed alveolar 117 wall congestion, while the Delta-infected hamsters demonstrated more severe and diffuse 118 alveolar wall infiltration and congestion (Fig. S3) . At 4 dpi, both groups of hamsters showed 119 bronchiolar epithelial destruction, and peribroncheolar and perivascular inflammatory 120 infiltration, but the alveolitis in the Delta-infected hamsters was more much more diffuse than 121 the Omicron-infected hamsters. At 7 dpi, the lung sections of the Omicron-infected hamsters 122 appeared mostly normal, while those of the Delta-infected hamsters still showed blood vessel 123 congestion and alveolar wall inflammatory infiltration, indicating that the Omicron-infected 124 hamsters had earlier resolution of tissue damage. Viral nucleocapsid protein expression was 125 significantly more abundantly seen in the lung sections of the Delta-infected than the Omicron-126 infected hamsters throughout 2 dpi to 7 dpi (Fig. S4) . Similarly, the Omicron-infected hamsters 127 showed less severe histopathological changes (Fig. S5 ) and less abundant viral nucleocapsid 128 protein expression (Fig. S6) in their nasal turbinate than the Delta-infected hamsters from 4 dpi 129 to 7 dpi. 132 The other key question was the comparative transmissibility of the Omicron and Delta variants in 133 vivo. To this end, we first co-housed 6 index SARS-CoV-2-challenged hamsters (n=3 each for 134 the Omicron and Delta variants) with 6 naïve hamsters for 4 hours in a 1:1 ratio ( Fig. 2A) . The 135 experiment was repeated twice. All the index hamsters had similar viral loads in the nasal 136 7 turbinate at sacrifice at 2 dpi, indicating that they were successfully infected (Fig. 2B) . All 12 137 naïve hamsters were found to be infected at sacrifice at 2 days after exposure (Fig. 2C) , 138 indicating that both variants are highly transmissible through close contact. Next, we randomly 139 grouped 42 hamsters into 6 groups of index and naïve hamsters (1:6 ratio) in our established non-140 contact transmission system, and repeated the experiment twice (total n=84) ( Fig. 2D) (20) . The 141 hamsters were sacrificed at 2 dpi (index) or 2 days post-exposure to index (naïve). All the index 142 hamsters were successfully infected with similarly high nasal turbinate virus titers (Fig. 2E) Delta variant, respectively, through non-contact transmission (Fig. 2F) . Notably, although not 152 To investigate whether this highly transmissible Omicron variant is likely to take over as the 153 dominant circulating SARS-CoV-2 variant, we performed a competition assay to compare its 154 fitness with that of the presently circulating Delta variant. We co-infected human lung-derived Omicron and Delta variants in mixed specimens (Fig. S7) . Consistent with our recent 160 preliminary findings at an early time-point, the Delta variant consistently exhibits significant 161 fitness advantage over the Omicron variant for up to 72hpi in vitro ( Fig. 3A and 3B ) (21).

However, this scenario drastically changed when selection pressure by vaccinated sera 163 containing antibodies with reduced anti-Omicron but preserved anti-Delta neutralizing activity 164 was present (Fig. 3C) , with the Omicron variant significantly (P<0.0001) outcompeting the Delta 165 variant (Fig. 3D) . Then, we validated our in vitro findings with in vivo competition models in in those that were exposed to non-vaccinated index hamsters (Fig. 3J) . Whereas among the naïve 176 hamsters exposed to vaccinated index hamsters, the Omicron variant significantly outcompeted 177 the Delta variant (Fig. 3J) . Overall, our findings demonstrated that the Delta variant exhibits of COVID-19 in human, and has been widely applied for studying these aspects of SARS-CoV-2 190 (19, 20, (22) (23) (24) (25) (26) .

In our pathogenicity study, we showed that while the viral load and infectious virus titer United Kingdom also showed that the hospitalization rate of Omicron-infected patients may be 214 40-45% lower than that of Delta-infected patients (29). Importantly, we also showed that virus 215 shedding in the oral swabs and feces were both significantly lower and shorter in the Omicron- 

Study design 276 The emerging Omicron variant has an alarmingly high number of mutations in its spike, which 277 may not only affect its susceptibility to existing vaccines and monoclonal antibodies, but also its 278 pathogenicity, transmission, and fitness. This study was designed to characterize these important we described previously (19) . Briefly, male and female hamsters, aged 8-10 weeks old, were The body weight and clinical signs of disease of the hamsters were monitored daily. A score of 1 321 was given to each of the following clinical signs: lethargy, ruffled fur, hunchback posture, and 322 rapid breathing as previously described (36). The hamsters were sacrificed at 2 dpi, 4 dpi, and 7 323 dpi (n=5 per variant per time-point) for viral load quantitation by qRT-PCR, virus titer 324 quantitation by plaque assay, histopathological studies and immunofluorescent staining as 325 described previously (19, 37) . Additional hamsters (n=5 per variant) were kept beyond 7 dpi for 326 serial viral load detection in oral swabs and feces. Contact transmission study. The contact transmission study was performed as we described 341 previously (19) . Briefly, Index hamsters were intranasally challenged with either the Omicron or 342 Delta variant at 0 dpi. Twenty-four hours later, each virus-challenged index hamster was 343 transferred to a new cage, with each cage containing one naïve hamster as a close contact. The 344 index and naïve hamsters were co-housed for 4 hours before transferral to new cages. The index 345 and contact hamsters were sacrificed at 2 dpi and 2 days post-exposure, respectively. Hamsters 346 with cycle threshold (Ct) value <40 in either nasal turbinate or lung were considered as infected.

The experiment was repeated twice, each time with 6 index (n=3 for Omicron and n=3 for Delta hamsters were transferred to cages adjacent to the cages housing the index hamsters (1:6 ratio).

The index and exposed naïve hamsters were removed from the non-contact transmission system 359 and transferred to separate new cages 6 hours later. The hamsters were then sacrificed at 2 dpi 360 (index) or 2 days post-exposure (naïve) for organ tissue collection. Hamsters with Ct value <40 361 in either nasal turbinate or lung were considered as infected. The experiment was repeated twice, 362 with a total of 12 index and 72 naïve hamsters. Cell cytopathic effects (CPE) were monitored to validate complete inactivation using Vero cells.

Hamster sera were collected at day 0, day 14, day 28, and day 100 for neutralization antibody 388 titration using enzyme-linked immunosorbent assay and micro-neutralization assay. At day 100 389 post-vaccination, the vaccinated index hamsters were infected (ie: 0dpi) and the competition 390 experiments were performed as described for the non-vaccinated hamster model.

Validation of competition assay. The experiments were performed as previously described 393 (26). To validate the consistency and accuracy of the competition assay in both system, the 394 Omicron and Delta variants were mixed at ratios of 10:1, 5:1, 3:1, 1:1, 1:3, 1:5, and 1:10 based 395 on their PFU titres (total 10 5 PFU viruses), or mixed with 10 6 , 10 5 , 10 4 , 10 3 , and 10 2 PFU of the 396 two variants at a ratio of 1:1. The total RNA of these mixed variants was isolated and amplified 397 by RT-PCR followed by Sanger sequencing, or directly analysed by one-step RT-qPCR using 398 strain-specific primers and probes as mentioned below. The Omicron to Delta ratio was 399 calculated by the peak heights of Sanger sequencing, i.e. Spike_R493 for Omicron (CGA) and 

New SARS-like virus in China triggers alarm

A pneumonia 462 outbreak associated with a new coronavirus of probable bat origin

Yuen 465 KY. A familial cluster of pneumonia associated with the 2019 novel coronavirus indicating 466 person-to-person transmission: a study of a family cluster

Lessons learned 1 year after SARS-CoV-2 emergence leading to 469 COVID-19 pandemic

Working Group; COVID-19

Estimated transmissibility and impact of SARS-CoV-2 lineage B.1.1.7 in England

Enhanced fitness of SARS-CoV-2 variant of concern Alpha but not Beta

World Health Organization. Classification of Omicron (B.1.1.529): SARS-CoV-2 Variant of 484

Concern

Neutralization of SARS-CoV-2 Omicron variant by sera from BNT162b2 or Coronavac 491 vaccine recipients

Striking antibody evasion manifested by the Omicron variant of 495 SARS-CoV-2. Nature

499 Network for Genomic Surveillance in South Africa

Omicron 501

Comparative tropism, 605 replication kinetics, and cell damage profiling of SARS-CoV-2 and SARS-CoV with 606 implications for clinical manifestations, transmissibility, and laboratory studies of COVID-19: an 607 observational study

Clofazimine broadly inhibits coronaviruses including SARS-613

CoV-2

Establishes Subclinical Respiratory Infection with Virus Shedding in Golden Syrian Hamsters

Metallodrug ranitidine bismuth citrate suppresses SARS-CoV-2 620 replication and relieves virus-associated pneumonia in Syrian hamsters

Real-Time Reverse Transcription-PCR Assay Validated In Vitro and with Clinical Specimens

Funding: This study was partly supported by funding to The University of Hong Kong: the 629

The Government of the Hong Kong Special Administrative Region

Innovation and Technology Commission, the Government of the 632

); the General Research Fund

) and Theme-Based Research Scheme of the Research Grants Council

); the Consultancy Service for Enhancing Laboratory Surveillance of Emerging 635

Infectious Diseases and Research Capability on Antimicrobial Resistance for Department of 636 Health of the Hong Kong Special Administrative Region Government

C.); the High Level-Hospital Program, Health Commission of Guangdong Province

Major Science and Technology Program of Hainan Province (ZDKJ202003)

641 the research project of Hainan Academician Innovation Platform (YSPTZX202004)

); the University of Hong Kong Outstanding Young Researcher Award

National Key Research and Development Programme on Public Security Risk 645 Prevention and Control Emergency Project (K

); and donations from the Shaw 647

Lee Wan Keung Charity Foundation Limited, Providence Foundation Limited (in 649 memory of the late Lui Hac Minh

Chan Yin Chuen Memorial Charitable Foundation, Marina 651 Man-Wai Lee, the Hong Kong Hainan Commercial Association South China Microbiology 652 Research Fund, the Jessie & George Ho Charitable Foundation

The funding sources had no role in the 655 study design, data collection

665 Project administration

report 673 31 collaboration with Sinovac Biotech Ltd. and China National Pharmaceutical Group Co., Ltd

Data availability: Complete sequences of the SARS-CoV-2 Omicron (hCoV-19/Hong

GISAID accession number EPI_ISL_7357684) and Delta

GISAID: EPI_ISL_3221329) variants available through

Other supporting raw data are available from the corresponding author upon reasonable

Pathogenicity of the Omicron and Delta variants in the golden Syrian hamster 683 model. (A) Scheme of the pathogenicity study comparing infections caused by the Omicron and 684

The hamsters were sacrificed at 2dpi, 4dpi, and 7dpi (n=5 per variant per time-687 point) for viral load quantitation by qRT-PCR, virus titer quantitation by plaque assay, and 688 histopathological studies. (B) Body weight changes and (C) clinical scores of the hamsters after 689 infection with either variant. A score of 1 was given to each of the following clinical signs: 690 lethargy, ruffled fur, hunchback posture, and rapid breathing

Data are mean ± standard deviations. n = 5 694 biological replicates per variant time-point

Contact and non-contact transmissions of the Omicron and Delta variants among 698 golden Syrian hamsters. (A) Scheme of the contact transmission study. (B) The nasal turbinate 699 infectious virus titres in the virus-challenged index hamsters at 2 dpi were determined by plaque 700 assay to ensure successful infection of both groups of hamsters. n.s. indicates non-significant 701 (Student's t-test). (C) Positive rates of infection among the exposed naïve hamsters after 702 exposure to either the Omicron or Delta variant

Scheme of the non-contact transmission study. (E) The nasal turbinate infectious virus titres 704 in the virus-challenged index hamsters at 2 dpi were determined by plaque assay to ensure 705 successful infection of both groups of hamsters. n.s. indicates non-significant

Positive rate of infection after exposure to Omicron or Delta virus. P-values were determined

Omicron and Delta 710 variants. (A) In vitro virus competition assay using a mixture of the Omicron and Delta variants 711 with an initial ratio of 1:1 was inoculated onto Calu-3 cell cultures (final MOI of 0.10 for each 712 variant). The ratios after competition in the cell culture (A) supernatant and (B) lysates were 713 measured by qRT-PCR. (C) Serum samples from 4 donors at least 14 days after receiving the 714 second dose of mRNA

Omicron and Delta variants, and the resulting 50% inhibitory dilution (ID 50 ) were displayed

In vitro virus competition with or without addition of the vaccinees' serum samples in cell 717 culture supernatants. Scheme of the in vivo competition models using (E) non-vaccinated and (F) 718 vaccinated index hamsters. (G) The hamster sera (n=6) were collected at the indicated days after

vaccination for detection of antibody response against spike RBD. (H) The neutralizing activity 720 of the hamster serum samples collected at day 100 after vaccination against live Omicron and 721

The Omicron:Delta ratio in the nasal turbinate, trachea, and lung of the (I) non-722 vaccinated and vaccinated index hamsters, and (J) naïve hamsters exposed to the non-vaccinated 723 or vaccinated index hamsters. Data are mean ± standard deviations

01, and *P<0.05 by two-way ANOVA

Virus shedding in the oral swabs and feces of the Omicron-infected and Delta-727 infected hamsters. (A) Oral swabs and (B) feces were serially collected from the Omicron-728 infected and Delta-infected hamsters at the indicated time-points for viral load detection by 729 quantitative reverse transcription-polymerase chain reaction. The dotted line represented the 730 limit of detection

01 and *P<0.05 by two-way ANOVA

Variant-specific anti-spike receptor-binding domain antibody response of the 734

Omicron-infected and Delta-infected hamsters. Microwell plates were coated with 100

SinoBiological 736 #Cat:40592-V08H121 and Delta RBD, Sinobiological Cat# Cat

After removal of the 738 blocking solution, aliquots of 100 μl/well of serially diluted hamster sera from the dilution of 739 1:100 were added to microwell plates coated with RBD protein and incubated at 37°C for 1 h

The plates were then washed 6 times, rabbit anti-hamster horseradish peroxidase antibody

μL/well) at the dilution of 1:2000 was added and incubated for 30 min at 37°C. After the plates 742 were washed 6 times, tetramethylbenzidine substrate (100 μL/well) was added. The reaction was 743 stopped after 10 min by the addition of 0.3N sulfuric acid, and the plates were then examined in 744 an ELISA plate at 450 nm and 620 nm

**P<0.01, and *P<0.05 by two-way ANOVA

Representative histopathological findings of the lung of Omicron-infected and 748

Delta-infected hamsters. Representative haematoxylin and eosin-stained lung sections of the 749

At 2 dpi and 4 dpi, the Delta-infected hamsters showed more diffuse 751 peribronchiolar, perivascular, and alveolar inflammatory infiltrates, bronchiolar epithelial 752 destruction, and alveolar congestion than the Omicron-infected hamsters

Representative immunofluorescence staining findings of the lung of Omicron-758 infected and Delta-infected hamsters. (A) Immunofluorescence staining of the lung sections of 759

Omicron-infected and Delta-infected hamsters at 2 days-post infection (dpi), 4 dpi

These representative 761 images were selected from a pool of 15 images captured in 5 hamsters per variant. (B) NP-762 positive cells per 50× field per lung section of each hamster. Scale bars = 200 μm. Data are mean 763 ± standard deviations. n = 5 biological replicates per variant

At 4 dpi, the Delta-infected hamsters 771 showed severe destructive changes of the respiratory and olfactory epithelia, while the Omicron-772 infected hamsters showed intact respiratory epithelium with mild intra-epithelial inflammatory 773 infiltrates

while the Delta-infected hamsters still demonstrated some residual inflammation. Scale 775 bars = 100µM or

Representative immunofluorescence staining findings of the nasal turbinate of 778

Omicron-infected and Delta-infected hamsters. (A) Immunofluorescence staining of the nasal 779 turbinate sections of Omicron-infected and Delta-infected hamsters at 2 days-post infection (dpi), 780 4 dpi, and 7 dpi. SARS-CoV-2 nucleocapsid (NP, green) and cell nuclei (blue) were stained

These representative images were selected from a pool of 15 images captured in 5 hamsters per 782 variant. (B) NP-positive cells per 50× field per nasal turbinate section of each hamster

01 and *P<0.05 by two-way ANOVA

Validation of the competition assay. (A) The correlation between input PFU ratios and 787 output RT-PCR amplicon ratios determined by Sanger sequencing

The Omicron and Delta variants were mixed at PFU ratios of 10:1

The R493/Q493 ratio was calculated by the peak heights of Sanger sequencing. Data were 791 analyzed by linear regression with correlation coefficients (r) and significance (p). The symbols 792 represent individual replicates and the solid line represents the fitted line. Data were derived 793 from an experiment conducted in duplicate. (B) Assay range evaluation

The two variants were mixed at 1:1 ratio. The total titers of the mixed virus 796 stock were 10 2 , 10 3 , 10 4 , 10 5 , and 10 6 PFU. The total RNA of the virus mixture was isolated and 797 amplified by RT-PCR. The R493/Q493 ratio was calculated by the peak heights from Sanger 798 sequencing. The symbols represent individual replicates, bar heights represent the means, and 799 error bars represent standard deviations. Data are derived from an experiment conducted in 800 triplicate. (C) Validation of the competition assay by spike-targeting qRT