key: cord-0314215-nl1ud0rz authors: Schmidt, Fabian; Weisblum, Yiska; Rutkowska, Magdalena; Poston, Daniel; Silva, Justin Da; Zhang, Fengwen; Bednarski, Eva; Cho, Alice; Schaefer-Babajew, Dennis J.; Gaebler, Christian; Caskey, Marina; Nussenzweig, Michel C.; Hatziioannou, Theodora; Bieniasz, Paul D. title: High genetic barrier to escape from human polyclonal SARS-CoV-2 neutralizing antibodies date: 2021-08-08 journal: bioRxiv DOI: 10.1101/2021.08.06.455491 sha: 40e50f0d03df9234da6a520811ce162ad1faef2f doc_id: 314215 cord_uid: nl1ud0rz The number and variability of the neutralizing epitopes targeted by polyclonal antibodies in SARS-CoV-2 convalescent and vaccinated individuals are key determinants of neutralization breadth and, consequently, the genetic barrier to viral escape. Using chimeric viruses and antibody-selected viral mutants, we show that multiple neutralizing epitopes, within and outside the viral receptor binding domain (RBD), are variably targeted by polyclonal plasma antibodies and coincide with sequences that are enriched for diversity in natural SARS-CoV-2 populations. By combining plasma-selected spike substitutions, we generated synthetic ‘polymutant’ spike proteins that resisted polyclonal antibody neutralization to a similar degree as currently circulating variants of concern (VOC). Importantly, by aggregating VOC-associated and plasma-selected spike substitutions into a single polymutant spike protein, we show that 20 naturally occurring mutations in SARS-CoV-2 spike are sufficient to confer near-complete resistance to the polyclonal neutralizing antibodies generated by convalescents and mRNA vaccine recipients. Strikingly however, plasma from individuals who had been infected and subsequently received mRNA vaccination, neutralized this highly resistant SARS-CoV-2 polymutant, and also neutralized diverse sarbecoviruses. Thus, optimally elicited human polyclonal antibodies against SARS-CoV-2 should be resilient to substantial future SARS-CoV-2 variation and may confer protection against future sarbecovirus pandemics. were distinct from the majority of mutations arising after rVSV/SARS-CoV-2 passage in plasma 87 (Extended Data Fig. 2 , Table S1 ). Cumulatively, the plasma-selected mutations were enriched 88 in specific elements within NTD, RBD and other spike domains (Fig. 2a , Table S1 ). From the 27 89 plasma-passaged virus populations, 38 individual mutant viruses were isolated by plaque 90 purification, each of which encoded one, two or three spike substitutions (Fig. 2b ) that generally 91 occurred at high frequency in the passaged viral populations (Table S1 ). We compared the distribution of mutations selected by the RU27 plasma panel in cell Mutations conferring resistance to class I RBD antibodies were not selected by plasma passage 99 perhaps reflecting a lower than expected abundance of class I antibodies in this plasma panel 100 . Other sites, including spike amino acids ~680-700 and ~930 exhibited variation in 101 both plasma-passaged and natural variant datasets, but have not yet been demonstrated to be 102 targeted by neutralizing antibodies. Nevertheless, the similarity in the distribution of natural and 103 plasma-selected sequence variation within spike suggests that selection by neutralizing 104 antibodies drives divergence in naturally circulating SARS-CoV-2 populations. Of the 38 plaque-purified rVSV/SARS-CoV-2 mutants recovered following passage in 106 RU27 plasmas, 34 exhibited varying degrees of reduced sensitivity to neutralization by the 107 plasma that was used for its selection (median =3.1 fold reduced NT50, range 0.8 to 39.3 fold, 108 Extended Data Fig. 3 ). Nevertheless, for 37/38 of the selected rVSV/SARS-CoV-2 mutants, the 109 selecting plasma exhibited residual neutralizing activity. We aggregated 13 mutations from the 110 plasma selected viruses based on their effects on plasma neutralization sensitivity (Extended 111 data Fig. 3 ) and distribution throughout the spike protein, generating a single synthetic 112 'polymutant' spike (PMS) protein sequence, termed PMS1-1 (Extended Data Fig. 4a) . Notably, 113 an rVSV/SARS-CoV-2 derivative encoding these spike mutations (rVSV/SARS-CoV-2PMS1-1) 114 exhibited resistance to neutralization by the RU27 plasma panel that was significantly greater in 115 magnitude (p<0.0001) and consistency than the individual plasma selected mutants (median 8.0 -fold, range 2.7 to 52.9 fold, Fig. 2e, Extended Data Fig. 4b) . Nevertheless, 26/27 of the RU27 117 plasmas retained residual neutralizing activity against rVSV/SARS-CoV-2PMS1-1 (Fig. 2e, 118 Extended Data Fig. 4b ). We conclude that some neutralizing epitopes are shared among the 119 convalescent antibodies in high-titer plasmas, but neutralizing activity against SARS-CoV-2 is 120 clearly polyclonal and heterogeneous among individuals with respect to epitope targets. 174 pseudotypes bearing the PMS20 spike (Fig 4a, b) . Indeed, the PMS20 mutations that reduced 175 Ran21 and Vac14 plasma NT50 by a median of 50-fold (range 5.9 to 225 fold) and 81-fold 176 (range 8.4 to 229-fold), respectively caused a median NT50 reduction of only 18.6-fold (range 177 3.9 to 100-fold) for the vaccinated convalescent (VC1-15) plasma panel (Fig 4b) . Analysis of 178 chimeric SARS-CoV-2/PMS20 spike proteins in which the respective RBDs were exchanged 179 [PMS20(2-RBD) and SARS-CoV-2(PMS-RBD)] indicated that the relative resistance of the 180 PMS20 to both Ran and VC plasmas was conferred by multiple spike determinants and that the 181 neutralization breadth in the VC plasmas was due to antibodies directed at both RBD and non-182 RBD determinants (Fig. 4a) Plasma from the vaccinated/convalescent group also had substantial neutralizing activity 189 against heterologous sarbecovirus HIV-1 pseudotypes, including those that were poorly 190 neutralized by Ran21, Vac14 and RU27 plasma panels and whose RBD and/or NTD sequences 191 are extensively divergent from SARS-CoV-2 (Fig 4c, d) (range 621-2705) SARS-CoV (Fig. 4d, Extended Data Fig. 7) . Notably, the neutralizing activity 195 of the VC plasmas against the highly divergent sarbecoviruses bCoV-WIV16 and SARS-CoV 196 ( Fig. 4c ) was similar to that found in the random convalescent plasmas against SARS-CoV-2 197 Wuhan-hu-1 (Extended Data Fig. 7) , Thus, the neutralization potency and breadth of polyclonal 198 plasma following mRNA vaccination of previously SARS-CoV-2 infected individuals appears 199 greater than previously appreciated. suggests that affinity maturation, over months, of SARS-CoV-2 neutralizing antibodies in 208 convalescents confers additional flexibility and affinity 24,25,27 that may not be afforded by 209 standard mRNA vaccine regimens 28 . Indeed, individual affinity-matured antibodies can impose a 210 requirement for multiple viral substitutions for antibody escape and enable substantial activity 211 against VOC. Some human monoclonal antibodies also have activity against divergent 212 sarbecoviruses 24,29 . Thus, affinity maturation, the availability of numerous epitope targets and 213 the generation of high levels of circulating antibodies may explain why polyclonal plasma from 214 individuals who have been both infected and subsequently vaccinated could effectively 215 neutralize the otherwise highly neutralization resistant PMS20 spike, as well as sarbecoviruses 216 whose NTD and/or RBD domains are divergent from SARS-CoV-2. It remains to be seen 217 whether similar neutralization potency and breadth can be achieved using appropriately timed 218 boosting with existing SARS-CoV-2 vaccines. If so, existing immunogens may be sufficient to 219 provide robust protection against SARS-CoV-2 variants that may arise in future, and a degree of Variant Frequency RU3 RU5 RU17 RU4 RU21 RU9 RU7 RU22 RU24 RU12 RU10 RU11 RU25 RU19 RU1 RU18 RU23 RU16 RU8 RU26 RU14 RU15 RU27 RU20 RU6 RU2 100 1000 10000 <50 Plasma donor NT 50 SARS-CoV-2 SARS-CoV-2(1-RBD) RU3 RU5 RU17 RU4 RU21 RU9 RU7 RU22 RU24 RU12 RU10 RU11 RU25 RU19 RU1 RU18 RU23 RU16 RU8 RU26 RU14 RU15 RU27 RU20 RU6 RU2 E154K N450D K444T E484K T604I K378R V47M G142E V445E/M Q493R F759S P792H I931M V1061M P1162S Recovered mutants A623T P681S D1118G W258R V1128A V143G Q1010R N801D L244P S247P G446R S12P Q14H Y145C K147N/E S254Y C15G P25L Y248H A243D A260V/T F140S S813I W64R NTD RBD FP HR1d e RU3 RU4 RU9 RU12 RU13 RU5 RU25 RU22 RU7 RU21 RU24 RU15 RU10 RU11 RU19 RU18 RU17 RU23 RU14 RU1 RU27 RU20 RU16 RU26 RU2 RU8 RU6a b Figure 3 N440K E484K L244P K378R S12P F140S N801D N450D W258R P792H PMSD4 Y144C V445E G446R E484K T604I K378R V47M G142E V445E N450D Q493R F759S P792H I931M V1061M P1162S PMS1-1 Δ145 N501Y K417N L18F D80A PMS20 D215G Δ242-4 A701V D614G V47M Δ69-70 W258R R346S N440K V445E L455R E484K A475V N801D P792H Vaccine recipient Ran8 For analysis of the Illumina sequencing data, adapter sequences were removed from the raw 472 reads and low-quality reads (Phred quality score <20) using BBDuk. Filtered reads were 473 mapped to the codon-optimized SARS-CoV-2 S sequence in rVSV/SARS-CoV-2/GFP and 474 mutations were annotated using using Geneious Prime (Version 2020.1.2), using a P-value 475 cutoff of 10 -6 . RBD-specific variant frequencies, P-values, and read depth were compiled using Inoculum (µl /well) Spike codon Spike codon Comparative neutralization potency (NT50 values) of high titer convalescent (RU27) 575 plasma against WT (grey symbols) and indicated PMS (a), SARS-CoV-2 variant sarbecovirus (c) (red symbol) HIV-1 pseudotypes. Median and range of two independent 577 experiments, each with two technical replicates