Naturally-acquired immunity in Syrian Golden Hamsters provides protection from re-exposure to emerging heterosubtypic SARS-CoV-2 variants B.1.1.7 and B.1.351

The ability of acquired immune responses against SARS-CoV-2 to protect after subsequent exposure to emerging variants of concern (VOC) such as B1.1.7 and B1.351 is currently of high significance. Here, we use a hamster model of COVID-19 to show that prior infection with a strain representative of the original circulating lineage B of SARS-CoV-2 induces protection from clinical signs upon subsequent challenge with either B1.1.7 or B1.351 viruses, which recently emerged in the UK and South Africa, respectively. The results indicate that these emergent VOC may be unlikely to cause disease in individuals that are already immune due to prior infection, and this has positive implications for overall levels of infection and COVID-19 disease.


Introduction
Towards the end of 2020, variants representing new lineages of SARS-CoV-2 emerged separately on three continents. These so-called variants of concern (VOC) rapidly spread to become dominant lineages. This rapid spread was to the detriment of other lineages and has obvious implications for the effectiveness of pre-existing immunity, either from vaccines or previous infection with SARS-CoV-2. Thus, there is an urgent and unmet need to study the underlying biological consequences of the mutations in the VOC with respect to re-infection as well as disease severity and transmissibility.
The recent VOCs all contain a N501Y substitution in the receptor-binding domain (RBD) of the spike glycoprotein (S), which increases the binding affinity of S for the ACE2 receptor on cells 1 The B.1.1.7 lineage was identified during September 2020 in the UK 2 . It has a deletion of residues 69 & 70, as well as six other mutations in S, a premature stop codon in ORF8, three aa. substitutions and a deletion in ORF1 and two substitutions in nucleoprotein (N). Lineage B.1.351 3

was identified in South
Africa during November 2020 and has two additional aa. substitutions in the RBD, namely, K417N and E484K. The first may disrupt a salt bridge with D30 of ACE2. The second may disrupt the interaction of RBD with K31 of hACE2 and enhance ACE2 binding 1,4 . Recent data have shown that VOC, especially those like B.1.351 with aa.
substitutions at residues 484 and 517, escape neutralisation by antibodies against the ACE2-binding Class 1 and the adjacent Class 2 epitopes, but are susceptible to neutralization by the generally less potent antibodies directed to Class 3 and 4 epitopes on the flanks of the RBD 5 .
The Immunological mechanisms involved in protection against SARS-CoV-2 infection and resulting disease are not fully understood (reviewed by 6 ). Antibodies to the S glycoprotein (whether neutralising or not) have a major role in protection, as well as CD4 and CD8 T cells that recognise a wide range of virus proteins [7][8][9] . By whatever mechanism, prior infection with SARS-CoV-2 does generate protective immunity as symptomatic re-infection within six months after the first wave in the UK was very rare in the presence of anti-S or anti-N IgG antibodies 10,11 . However, like other human coronaviruses, immunity following SARS-CoV-2 infection does not necessarily prevent re-infection, and this may be linked to the emergence of novel strains that partially evade immunity.
The analysis of SARS-CoV-2 in humans is naturally restricted to analysis of clinical samples (e.g. blood, nasopharyngeal swabs and bronchial alveolar lavages) after diagnosis of infection. Therefore, animal models of COVID-19 present critical tools to fill knowledge gaps for the disease in humans and for screening therapeutic or prophylactic interventions. Different animal species can be infected with wild-type SARS-CoV-2 to serve as models of COVID-19 and these include mice, hamsters, ferrets, rhesus macaques and cynomolgus macaques 12 . Of these, the hamster has emerged as the small-animal gold-standard for pathogenesis studies, as well as preclinical vaccine and therapeutics development 13,14 . Hamsters are readily infectable, display both upper and lower respiratory tract replication, clinical signs and also pathology that are similar to humans. In addition, hamsters shed and can transmit from animal to animal making transmission studies also possible.
With the evolution of SARS-CoV-2 in the human population over the past year, new strains have evolved, some of which, so-called variants of concern (VOC), have properties that can evade specific mechanisms of preexisting immunity such as neutralizing antibodies 5,15 . The aim of this work was to directly assess the question of whether immunity, in an in vivo model which includes both humoral and cellular, generated to a strain circulating early in the pandemic in spring 2020 would protect against both B.1.1.7 and B.1.135 VOCs currently circulating worldwide.

Virus infection.
Animals were randomly assigned into multiple cohorts. For SARS-CoV-2 infection, hamsters were anaesthetised lightly with isoflurane and inoculated intra-nasally with

Experimental design.
To address the degree of protection conferred by prior infection with a prototypic B lineage virus against exposure to emerging VOC, one cohort of hamsters (n = 24) was inoculated with 10 4 SARS-CoV-2 strain hCoV-2/human/Liverpool/REMRQ0001/2020 (herein called LIV) and one cohort with PBS (for schematic see Figure 1)  (Table 1).
Serum taken before -re-challenge and at the end of the experiment was assessed for neutralisation activity and swabs taken at multiple time-points were assessed for viral RNA by PCR. Finally, tissues were taken at necropsy and analysed by histopathology and immunohistology for virus antigen. These supplementary evaluations are currently ongoing, and this preprint will be updated as data become available.   Recent work has shown that emerging strains of SARS-CoV-2 such as B.1.1.
7 and, in particular B.1.351 can partially or completely escape neutralisation and binding by antibodies in convalescent sera from patients exposed to prototype strains of SARS-CoV-2, and to vaccine-elicited responses 5,15,19,20 . While escape from neutralisation in vitro is an important line of evidence as to whether emerging strains will escape pre-existing immunity or not, other factors such as non-neutralising antigen-specific antibodies, T cells and innate lymphocytes clearly have the potential to contribute protection in vivo [21][22][23][24] . Indeed, recent work has shown that T cell epitopes that dominate human SARS-CoV-2 responses are not subject to major substitution in the three variants of concern 5 . Further, it is known that previous infection with  25,26 . Accordingly, more detailed studies will be required to define the difference in pathogenesis between these strains.
There are caveats with the presented data. The study was conducted in a hamster model where responses will be similar but likely not identical to humans. Also, the time from exposure to re-challenge is relatively short at only 3 weeks. However, these data do provide a very good, direct indication that exposure to or vaccination against SARS-CoV-2 will protect against exposure to the variants that emerged in late 2020 in the UK and South Africa. Our data suggest that protection will be better than suggested by in vitro neutralising antibody studies and that a degree of herd immunity will be achievable. Ongoing reconnaissance as SARS-CoV-2 continues to evolve is warranted.