key: cord-0297291-szj3rpye
authors: Junker, D.; Dulovic, A.; Becker, M.; Wagner, T. R.; Kaiser, P. D.; Traenkle, B.; Rothbauer, U.; Kienzle, K.; Bunk, S.; Strümper, C.; Häberle, H.; Schmauder, K.; Malek, N.; Althaus, K.; Koeppen, M.; Bitzer, M.; Göpel, S.; Schneiderhan-Marra, N.
title: Reduced serum neutralization capacity against SARS-CoV-2 variants in a multiplex ACE2 RBD competition assay
date: 2021-08-22
journal: nan
DOI: 10.1101/2021.08.20.21262328
sha: fb038ec346e83b20ca71270573db12ec08624b38
doc_id: 297291
cord_uid: szj3rpye

As global vaccination campaigns against SARS-CoV-2 proceed, there is emerging interest in the longevity of immune protection, especially with regard to increasingly infectious virus variants. Neutralizing antibodies (Nabs) targeting the receptor binding domain (RBD) of SARS-CoV-2 are promising correlates of protective immunity and have been successfully used for prevention and therapy. To assess neutralizing capacity, we developed a bead-based multiplex ACE2 RBD competition assay as a large scalable, time-, cost-, and material-saving alternative to infectious live-virus neutralization tests. By mimicking the interaction between ACE2 and RBD, this assay detects the presence of Nabs against SARS-CoV2 in serum. Using this multiplex approach allows the simultaneous analysis of Nabs against all SARS-CoV-2 variants of concern and variants of interest in a single well. Following validation, we analyzed 325 serum samples from 186 COVID-19 patients of varying severity. Neutralization capacity was reduced for all variants examined compared to wild-type, especially for those displaying the E484K mutation. The neutralizing immune response itself, while highly individualistic, positively correlates with IgG levels. Neutralization capacity also correlated with disease severity up to WHO grade 7, after which it reduced.

Neutralizing antibodies (Nabs) prevent infection of the cell with pathogens or foreign particles 48 by neutralizing them, eliminating a potential threat and rendering the pathogen or particle 49 harmless (1). The longevity of a Nab response has important implications for immune 50 protection and vaccination strategies. In SARS-CoV-2, Nabs interfere with the cell entry 51 mechanism primarily by blocking the interaction of the receptor binding domain (RBD) with the 52 human cell receptor angiotensin converting enzyme 2 (ACE2) (2, 3). The RBD of SARS-CoV-53 2 is the target of approximately 90 % of the neutralizing activity present in immune sera (4), 54

with a lack of Nabs correlating with risk of fatal outcome (5, 6). Passive transfer of Nabs has 55 been shown to provide protection from infection (7-9), with several Nabs drugs granted 56 emergency use authorization by the U.S. Food and Drug Administration (10-13). 57

Since the first documented infections in Wuhan China (14), SARS-CoV-2 has continually 58 evolved, with the emergence of global variants of concern (VOC) being of particular 59

importance. As of this moment, the WHO currently lists the alpha (B.1.1.7)(15), beta 60 against the original B1 isolate (hereon referred to as "wild-type") spike protein (25). Several 65 studies have already found that both convalescent and post-vaccinated sera have lower 66 neutralization capacities against beta and gamma VOCs (26-28). Of particular concern are 67 mutations on amino acid residue (aa) 484 (e.g. E484K), which seem to confer escape from 68 vaccine control, with an additional mutation on aa 501 (e.g. N501Y) increasing this effect (29) . is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted August 22, 2021. ; https://doi.org/10.1101/2021.08.20.21262328 doi: medRxiv preprint SARS-CoV-2. In this study we developed and applied a multiplex ACE2 RBD competition 74 assay based upon the antibody-mediated inhibition of RBD-ACE2 binding. This automatable 75 assay enables simultaneous screening of serum samples for the presence of Nabs against an 76 unlimited number of VOCs/VOIs in a single well, making it a time-, material-and cost-effective 77 alternative to live VNTs or classical ELISAs. Following in-depth validation of the assay, we 78 analyzed the IgG antibody response and neutralizing capacity of 325 serum samples from 186 79 COVID-19 patients with mild to severe disease progression towards eleven different SARS-80

CoV-2 variant RBDs including the alpha, beta, gamma and delta VOCs. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted August 22, 2021. ; https://doi.org/10.1101/2021.08.20.21262328 doi: medRxiv preprint

To investigate neutralizing capacity of serum from COVID-19 convalescent individuals, we 85 developed and established a high-throughput bead-based multiplex ACE2 RBD competition 86 assay. This assay mimics the ACE2 RBD interaction, detecting the presence of Nabs against 87 SARS-CoV-2. For this, we expressed RBDs of SARS-CoV-2 wild-type and 11 different variants 88 (alpha, beta, gamma, epsilon, eta, theta, kappa, delta, lambda, Cluster 5 and A.23.1) and 89 coupled them on spectrally distinct beads (see methods for details). All beads were then pooled 90 to produce a bead mix, which was incubated with patient serum diluted in assay buffer 91 containing biotinylated ACE2. RBD-bound ACE2 was then detected using phycoerythrin-92 labeled streptavidin and MFI values were normalized to a control consisting of biotinylated 93

Neutralization capacity and IgG antibody responses were positively correlated (all Pearsons 136 correlation coefficients above 0.70, Figure 4 ) with variant-specific differences still present. The 137 percentage of samples considered to be negative (<20% neutralization), followed the same 138 variant-specific order as Figure 3 . For all RBD mutants, the increase in neutralization capacity 139 most commonly occurred once IgG RBD MFI levels exceeded 10,000. Notably, there was 140 individual variation among the samples, with some having high neutralizing capacity but 141 relatively low IgG responses. This was especially true for the RBD wild type and similar 142

performing RBD mutants such as A.23.1, alpha and Cluster 5. For RBD mutants eta, gamma, 143 theta and beta more than 75 % of all samples were considered negative, compared to 42.5 % 144 for wild-type (Figure 4) . 145

We then examined longitudinal changes on neutralizing capacity and IgG responses using 147 samples from 44 study participants, ranging from 1 to 290 days after the first positive SARS-148

CoV-2 PCR. For a subset of six individuals with similar sample collection points, both IgG 149 antibody response and neutralization capacity against wild-type ( Figure 5A , B) remained 150 initially low following infection, with an increase no later than 10 days after the first positive 151 PCR test result, although one individual (Donor3) peaked 6 days after the first positive PCR 152 test and then declined. Neutralization capacity for most individuals (Donors1, 4, 5 and 6) 153 declined no later than 19 days post PCR. Generally, neutralization capacity and IgG levels 154 peaked at similar time points (Donors 1, 2 and 6), although for some individuals (Donors 4 and 155 5) neutralization capacity peaked before IgG titers. Among all longitudinal samples, while 156 neutralization and antibody titers followed a similar pattern (Figure 5C, D) , responses were 157 highly individualistic with large variances at all time points. 158

As delta represents the current dominant global strain, we then examined how neutralization 159 capacity and antibody binding compared for this variant to wild-type. Neutralizing capacity and 160

IgG binding against delta was reduced for all samples (Figure 5E and F) . Overall, neutralizing 161 capacity and IgG response followed the same pattern for all samples as for wild-type. When 162 . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted August 22, 2021. ; https://doi.org/10.1101/2021.08.20.21262328 doi: medRxiv preprint examining all RBD mutants, mean neutralization capacity and IgG antibody responses ( Figure  163 5G, H) clustered as before. Beta, theta and gamma had the largest decrease in neutralizing 164 capacity with eta, kappa and lambda occupying the gap between another cluster of delta, 165 epsilon, Cluster5 and alpha. As expected, all RBD mutants showed reduced neutralization 166 capacity compared to wild-type. Mean neutralization capacity was highly individualistic, 167 peaking for all RBD mutants 22 days post-positive PCR before decreasing below 20 % at 168 approximately 92 days post PCR. 169

We then examined correlations between neutralization capacity and COVID-19 disease 171 severity. Neutralizing capacities for wild-type and RBD delta within 7-49 days post-positive 172 PCR increased steadily with disease severity up to grade 7 (WHO grading scale, hospitalized 173 patients needing intubation and mechanical ventilation), before decreasing for patients of 174 grade 8 (fatal disease course) (Figure 6A and B) . Mean neutralization capacities at WHO 175 grade 7 were 72.0 % for wild-type and 58.7 % for delta. From 49 days post positive PCR, 176 neutralization capacities correlated with disease severity (Figure 6C and D) , however there 177 was an overall reduction for grades 4 to 7 compared to the first timeframe for both wild-type 178 and delta (Figure 6C and D) . Within this timeframe, mean neutralization capacity for both WT 179 (43.56%) and delta (30.08%) decreased. Anti-RBD IgG levels also correlated with disease 180 is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted August 22, 2021. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted August 22, 2021.

With vaccination campaigns beginning to take effect and result in reduced case numbers in 193 certain countries, interest in the serological response to SARS-CoV-2 is increasingly changing 194 from quantitative to qualitative in nature. Whereas in the early-phase of the pandemic SARS-195

CoV-2 antibody assays were helpful in determining seroprevalence and support vaccine 196 development, now a reliable correlate of immune protection is needed to securely lift social 197 restrictions and guide future vaccine developments. 198

While cell-culture based virus neutralization tests (VNTs) (e.g. plaque reduction neutralization 199 test) are the gold standard for neutralization assays, they have many disadvantages over 200 conventional protein-based surrogate assays. Such assays require rapid access to continually 201 changing virus variants and as such special biosafety level 3 laboratories are necessary. 202

Additionally, VNTs are cell-culture based and therefore it takes multiple days to conduct an 203 experiment with reproducibility potentially affected by either the cells or their long culture 204 conditions. Consequently, highly reproducible assays under substantially faster and safer 205 working conditions (e.g. BSL 1) would be highly beneficial. Our newly developed protein-based 206 ACE2 RBD competition assay is finished in under 4 hours and only requires 5 µL of patient 207 sample to measure neutralizing capacities simultaneously against multiple SARS-CoV-2 VOCs 208 and VOIs. As a protein-based assay, it does not require enhanced safety protocols to be 209 followed and can be completed safely in a BSL1 laboratory. Due to the bead-based nature and 210 plate format, it is automatable, suitable for high-throughput, standardized and highly 211

reproducible. The protein-based nature also allows for the rapid inclusion of emerging variants 212 or single mutations. 213

As shown in our technical validation, assay components and as such the results are highly 214 stable. Neutralization percentages correlate with FDA-approved in vitro diagnostic tests such 215

as NeutraLISA (Euroimmun). The lower than 20% reduction cut-off used in the NeutraLISA to 216 identify negative samples is also suitable for the ACE2 RBD competition assay as both 217 measure the steric inhibition of ACE2. Direct comparison between the two assays shows that 218 . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted August 22, 2021. ; https://doi.org/10.1101/2021.08.20.21262328 doi: medRxiv preprint the ACE2 RBD competition assay has a resolution range more suited to highly neutralizing 219 samples, while the NeutraLISA has increased resolution for samples with weak neutralization 220 at the cost of early saturation of neutralization percentage. Samples which had different 221 classifications between the two assays had a median MFI of 9056, which explains the low 222 neutralization capacities in the ACE2 RBD competition assay, as high neutralization capacities 223 are generally achieved at an MFI > 10,000. (death) has not to our knowledge been reported before. This decrease requires further 236 investigation to determine its cause given its likely role in patient mortality. 237

As expected, neutralizing capacity towards VOCs was highly variable. The strongest 238 reductions in neutralizing capacities were all from variants (eta, gamma, theta and beta) with 239 a E484K mutation. This specific mutation has been reported in multiple studies as an escape 240 mutation that enhances the RBD-ACE2 affinity (36). Neutralizing capacity was further reduced 241 among these variants for those which additionally had a N501Y mutation (gamma, theta and 242 beta), which is known to further enhance RBD-ACE2 binding (37). These results are in-line 243

with previous findings that have reported significant reductions in neutralization for gamma and 244 beta (38-40). The gamma and beta RBDs in our assays are only separated by a single K417N 245 . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted August 22, 2021. ; https://doi.org/10.1101/2021.08.20.21262328 doi: medRxiv preprint mutation, which is known to significantly reduce both the RBD-ACE2 binding affinity as well as 246 the binding affinity to monoclonal therapeutic antibodies or other human antibodies (41). This 247 reduction in binding affinity could explain why beta neutralizing capacity was lower than 248 gamma. Among recently emerged variants (delta, kappa and lambda), neutralizing capacity 249 compared to wild-type was reduced for all. The reduction in neutralizing capacity seen for 250 kappa and delta are comparable to recent findings (42), although we could not confirm the 251 reduction seen by other authors for Lambda (43). This is likely due to the 7-amino acid deletion 252 in the N-terminal domain of lambda's spike protein, which is not present in the RBD and is 253 thought to contribute to its immune evading properties (44). Overall the decrease in 254 neutralization capacity against RBDs of all analyzed variants compared to wild-type has 255 important implications for the design of second generation vaccines. 256

Our ACE2 RBD competition assay has limitations similar to other protein-based in vitro 257 neutralization assays, such as only accounting for the Nabs that block the RBD-ACE2 258 interaction site through steric hindrance, and not for Nabs that interfere with cell entry 259 mechanisms as would be analyzed in a VNT. Furthermore, the binding assay is also more 260 prone to non-specific binding events. However, a major advantage of the multiplex ACE2 RBD 261 competition assay over VNTs is the speed of response toward viral evolution such as emerging 262 variants of concern. The bead-based format of the assay is also highly reproducible and not 263 susceptible to changes in experimental conditions, as is the case for cell culture based VNTs. 264

The plate format of the assay also enables automation and high-throughput screening. Our 265 ACE2 RBD competition assay only requires recombinant expressed RBD proteins which can 266 be quickly and easily produced. Additionally, this assay has the possibility of introducing 267 artificial mutants to screen for possible escape variants that could arise in the future. Similar is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted August 22, 2021. ; https://doi.org/10.1101/2021.08.20.21262328 doi: medRxiv preprint time points. However, this large variation is also beneficial as it clearly demonstrates the 273 individual variability in neutralizing capacity. 274

In conclusion, we have developed and validated an ACE2 RBD competition assay that 275 analyzes neutralizing antibodies in serum for all current variants of concern and interest of 276 SARS-CoV-2. Using this assay, we have identified differing responses in serum from infected 277 individuals, indicating that neutralizing antibody production is directly proportional to total 278 antibody production. Neutralization capacity was highly variable among all variants examined, 279

with the 484 residue appearing to be critical in reducing capacity. Overall, the speed and ease is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted August 22, 2021. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint 

To determine the intra-assay precision of the ACE2 RBD competition assay, 12 replicates of is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint 

Freeze-thaw stability of the biotinylated ACE2 stocks was determined by analyzing five serum 421 samples (Vac1 -Vac4 (vaccinated) and pre-pandemic) in triplicates, with ACE2 stocks 422 undergoing 1 to 6 freeze-thaw cycles. In addition to that, every sample was also processed 423 with ACE2 not re-frozen once thawed (fresh). The MFI values of every sample were normalized 424

to the values of the respective ACE2 control. For analysis, the mean, standard deviation and 425 coefficient of variation in percent of all replicates were calculated. 426

Short-term stability was determined by storing ACE2 buffer under six different conditions 428 before proceeding with the assay protocol. The prepared ACE2 buffer was stored 2 h, 4 h and 429 24 h at both 4°C and room temperature and compared to ACE2 buffer without storage (fresh). is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted August 22, 2021. ; https://doi.org/10.1101/2021.08.20.21262328 doi: medRxiv preprint

Regression analysis comparing neutralizing capacity (%) and IgG responses (MFI) for wild-type and all RBD mutants included in the study. Each circle represents one sample (n=186). For longitudinal donors with more than one sample available, the sample closest to 20 days post positive PCR diagnosis was selected. The percentage next to the bracket indicates the proportion of samples with neutralization capacity ≤ 20% (in orange). Pearson's correlation coefficient (r) is specified for every correlation. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted August 22, 2021. ; https://doi.org/10.1101/2021.08.20.21262328 doi: medRxiv preprint limitation of activities, 3 -hospitalized, mild disease / no oxygen therapy, 4 -hospitalized, mild disease / mask or nasal prongs, 6 -hospitalized, severe disease / intubation + mechanical ventilation, 7 -hospitalized, severe disease / ventilation + additional organ support (pressors, RRT, ECMO), 8 -Death. The study did not contain samples of WHO grade 5.

. CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint

The copyright holder for this this version posted August 22, 2021. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint 

Neutralizing antibodies for the treatment of COVID-19

A human neutralizing antibody targets the receptor-binding site of SARS-459

CoV-2

Mapping Neutralizing and Immunodominant Sites on the SARS-CoV-2

Neutralizing antibody responses to SARS-CoV-2 in symptomatic 464

COVID-19 is persistent and critical for survival

COVID-19-neutralizing antibodies predict disease severity 467 and survival

Correlates of protection against SARS-CoV-2 in rhesus macaques

Critical role of neutralizing antibody for SARS-CoV-2 reinfection and 471 transmission. Emerging microbes & infections

Isolation of potent SARS-CoV-2 neutralizing antibodies and 473 protection from disease in a small animal model

REGN-COV2, a Neutralizing Antibody Cocktail, in Outpatients 476 with Covid-19. The New England journal of medicine

Antibodies for Treatment of COVID-19

announcements/coronavirus-covid-19-update-fda-authorizes-monoclonal-antibodies-480 treatment-covid-19

Effect of Bamlanivimab as Monotherapy or in Combination With 482

Etesevimab on Viral Load in Patients With Mild to Moderate

FDA) FaDA. Coronavirus (COVID-19) Update: FDA Authorizes Monoclonal

Antibodies for Treatment of COVID-19

A pneumonia outbreak associated with a new coronavirus of probable 489 bat origin

Changes in symptomatology, reinfection, and transmissibility 491 associated with the SARS-CoV-2 variant B.1.1.7: an ecological study. The Lancet 492 Public health

Detection of a SARS-CoV-2 variant of concern in South Africa

Control ECfDPa. Threat Assessment Brief: Emergence of SARS-CoV-2 B.1.617 498 variants in India