Finger stick blood test to assess postvaccination SARS‐CoV‐2 neutralizing antibody response against variants

Abstract There is clinical need for a quantifiable point‐of‐care (PoC) SARS‐CoV‐2 neutralizing antibody (nAb) test that is adaptable with the pandemic's changing landscape. Here, we present a rapid and semi‐quantitative nAb test that uses finger stick or venous blood to assess the nAb response of vaccinated population against wild‐type (WT), alpha, beta, gamma, and delta variant RBDs. It captures a clinically relevant range of nAb levels, and effectively differentiates prevaccination, post first dose, and post second dose vaccination samples within 10 min. The data observed against alpha, beta, gamma, and delta variants agrees with published results evaluated in established serology tests. Finally, our test revealed a substantial reduction in nAb level for beta, gamma, and delta variants between early BNT162b2 vaccination group (within 3 months) and later vaccination group (post 3 months). This test is highly suited for PoC settings and provides an insightful nAb response in a postvaccinated population.

facilities that are difficult to integrate into PoC testing. 6,7 PoC lateral flow tests are currently limited, as they either detect total immunoglobulin level which is not a reliable indicator for immune protection or only provide qualitative assessment. 8,9 The availability of a quick and accurate PoC nAb test to track vaccination-induced immune responses especially against variants at both the population as well as individual level would be a valuable tool in enabling public health authorities to manage breakthrough infections and to develop an effective booster vaccination strategy for more vulnerable individuals.
We previously developed a rapid paper-based SARS-CoV-2 neutralization assay known as cellulose pulldown virus neutralization test (cpVNT) that detects SARS-CoV-2 neutralizing antibody (nAb) in plasma or serum within 10 min. 10

| Blood sample processing and storage
Blood samples were kept at 4 C for delivery, venous blood storage in heparin tubes (BD Vacutainer; #367874) while finger stick blood were stored in either heparin (Xinle Medical MP0540) or EDTA (Xinle Medical MP0581) microtainer tubes. A portion of the sample volume was separated into plasma content by centrifugation at 4000g for 5 min in 4 C. Plasma were stored in À20 C. Both WHO International Standard (20/136) and Reference Panel for anti-SARS-CoV-2 immunoglobulin (20/268) plasma were purchased from National Institute for Biological Standards and Control and were stored in À20 C upon receipt.

| Protein production and purification
The expression and purification of soluble extracellular fragment of human ACE2 (residues 19-615; GenBank: AB046569.1) and WT SARS-CoV2-Spike (EMBL: QHD43416.1 with silent mutations c. A1452>G and c.T1470>C) RBD fused to CBD followed the same protocol as described in Kongsuphol et al. 10 Similarly, alpha c. Finally, the purified RBD-CBD variants were concentrated and stored in 20 HEPES pH 7.5, 300 mM NaCl, 10% glycerol and 0.5 mM TCEP at À80 C.

| Fluorescence conjugation of monoFc-ACE2
Alexa Fluor 594 conjugation of monoFc-ACE2 was carried out by using Alexa Fluor 594 Conjugation Kit (Fast)-Lightning-Link (abcam; ab269822). For each labeling reaction, 100 μl of 1 mg/ml of monoFc-ACE2 in phosphate buffer saline (PBS) pH 7.6 was mixed with 10 μl of Modifier reagent. The 110 μl of mixture was transferred to Alexa Fluor 594 Conjugation Mix followed by 30-min incubation at room temperature in the dark. Then, the reaction was stopped by adding 10 μl of Quencher reagent and for 15-min incubation in the dark.
Finally, the labeled protein was stored in aliquots of 5 μl at À80 C freezer before use. The final 80 μl reaction was applied equally onto the test and control spot with 40 μl for each. Once sample was fully absorbed, both test and control spots were washed once with 40 μl of PBS pH 7.6. The cassette was then placed in an Atto Testbed for fluorescence measurement. All steps described above were performed at room temperature.

| Surrogate virus neutralization assay cPass (Genscript)
The assay was performed as per manufacturer's protocol by first diluting the selected plasma samples 1:10 in the sample dilution buffer provided by the kit, and incubated with horseradish peroxidase (HRP)conjugated RBD for 30 min at 37 C. Then, the sample-RBD mixtures were transferred to an ACE2 coated ELISA plate for 15-min incubation at 37 C before washing with the kit's washing solution. The sample read-out was performed by adding 100 μl 3,3 0 ,5,5 0 -tetramethylbenzidine (TMB) solution per reaction well for 15 min, followed by 50 μl of stop solution. Absorbance was measured at 450 nm using Infinite 200 PRO multimode TECAN plate reader and the percent of inhibition were calculated according to manufacturer's recommendation.

| SARS-CoV-2 pseudovirus neutralization assay
We applied the same protocol for production of SARS-CoV-2 pseudotyped lentiviral particles and pseudovirus neutralization assay as previously reported. 10 (1): To evaluate this new test format, we made a series of contrived blood samples by spiking 1, 5, 10, 25, 50, and 100 nM of mouse monoclonal SARS-CoV-2 nAb into a blood prepared with pre-SARS-CoV-2 pandemic plasma and washed red blood cells. The assay demonstrated an IC 50 of 3.38 nM nAb using blood as sample matrix (Figure 1c,d).  (Figure 3a). This showed that the test can produce a dose dependent response that captures the clinical range of nAb activity in plasma in under 10 min assay time. Since plasma represents approximately 55% of whole blood, the percent blocking test results in plasma samples was expected to be higher than that of whole-blood due to the lack of erythrocytes. To correlate the percent blocking in the WHO standard and reference panel plasma to corresponding whole blood, we analyzed 30 matching samples of blood and plasma in the modified cpVNT. We found that the percent blocking in blood samples is approximately 0.87 times of that in plasma samples assuming the relationship between the two sample types are linear (Figure 3b). We observed that the overall median percent blocking in Pre-Vac samples was found below 30% blocking in modified cpVNT using blood as matrix (Figure 2a (Figure 3c), which was validated to be comparable to gold standard plaque reduction neutralization test with live virus. 6,13,14 Meanwhile, as compared to the lab based pVNT test, the modified cpVNT showed 100% sensitivity (CI: 54.1%-99.9%) and 71.4% specificity (CI: 41.9%-91.6%) (Figure 3d). For reference, the WHO plasma of nAb activity at 1000, 210, and 44 IU/ml when performed with sVNT cPass yielded 94%, 78%, and 19% inhibition, respectively (Table S3). The lower specificity and sensitivity relative to ELISA and pVNT can be attributed to the difference in sample type (whole blood vs. plasma/serum) and different assay procedures.   Figures S3 and S4A). These variants contain mutations in the RBD region, which may reduce the binding affinities of antibodies generated against the WT protein and/or increase ACE2 receptor binding. 15 We found that the binding affinity of alpha, beta, gamma, and delta are higher than that of WT, especially gamma that showed a threefold increase (4.3 nM) in binding affinity comparing to WT (12.7 nM) consistent with previous reports 15,16 (Table 1). Furthermore, our result supports published data that the N501Y mutation in the alpha and gamma variants of RBD contributes to the slow off-rate of the complex 17 (Table 1). Meanwhile T478K appears to promote fast complex formation based on comparison among delta, kappa, epsilon, and a delta plus variant that shared the L452R mutation (Table 1). We engineered a hypothetical RBD variant containing N501Y, T478K mutation and annotated it as "AD" (alpha-delta) variant that is speculated to have fast on-rate and slow off-rate with ACE2. This hypothetical variant confirmed our hypothesis where it binds ACE2 with the highest binding affinity (K D of 3 nM) among the 10 variants (Table 1). Next, we assess the activity of these RBD-CBD variants on the modified cpVNT with Pre-Vac blood.

| Evaluation of postvaccination nAb responses using modified cpVNT
Although variants with high affinity to ACE2 showed increased fluorescence intensity than WT RBD-CBD in the modified cpVNT assay at the same reagent concentration, that is, alpha (1.8-fold), beta (1.6-fold), gamma (2.1-fold), delta (1.3-fold), the correlation is not direct. As we observed variant RBD-CBDs epsilon and lambda still generate comparable signal as WT despite the lower ACE2 binding affinity, while AD variant showed merely 1.6-fold increase in signal despite binding ACE2 strongly ( Figure S4A). Besides the binding kinetics, the capture rate of RBD-CBD on the cellulose paper and possible avidity of the different RBD-variants on ACE2 could contribute to the effect.
We then tested the four VOCs: RBD-CBD alpha, beta, gamma, and delta with 33 blood samples from participants within 3 months of completing vaccination. There were considerable variations in the nAb responses to the different variants. The nAb percent blocking against beta and gamma variants being reduced significantly to 72.4% and 70.1%, while the percent blocking reduced only minimally to 87.2% and 91.9% for alpha and delta, respectively, as compared to WT (95.6%) ( Figure 4a). These were in line with previous reports using pVNT and VNT, whereby neutralization of beta and gamma variants had considerable reduction for both mRNA vaccines. [18][19][20] About 91.8% nAb blocking was observed against the engineered AD variant even though the RBD-CBD variant binds strongly to ACE2, suggesting that vaccine induced nAb can outcompete stronger interaction ( Figure S4B). While this data indicated a heterogenous response it was important to evaluate whether our test was able to stratify response in relation to the different vaccines used. The median percent blocking for BNT162b2 recipients against alpha was 78% (p < 0.01) and delta was 89.2% (p = n. s.) as compared to WT (94.8%). (Figure 4b). The most substantial reductions of nAb response were observed with beta and gamma variants reaching 55.7% and 49.6% blocking, respectively, among BNT162b2 recipients (Figure 4b). In the cohort of mRNA-1273 recipients, we observed reduction to 87.5% with beta variant (p < 0.0001) and 80.5% with gamma variant (p < 0.0001).
Next, we also examined and compared the percent blocking of nAb in whole-blood samples from participants within 3 months or greater than 3 months after completion of vaccination against WT and four VOCs RBD-CBD. Only samples from BNT162b2 recipients were available to us for the greater than 3 months cohort as it was the first vaccine rolled-out in the Singapore national vaccination program. There was a modest drop of nAb percent blocking from 96% to 68% (28%, p < 0.0001) observed in WT RBD-CBD and 77.6%-40.2% (37.4%, p < 0.001) in alpha RBD-CBD between the two groups of samples (Figure 4d). Meanwhile a more substantial reduction was seen    23 We tested the influence of RBD/ACE2 interaction on nAb blocking with the hypothetical AD variant that carries N501Y and T478K mutations. It was found unable to evade vaccine induced nAb inhibition where it shows 91.8% nAb blocking in the modified cpVNT similar to WT RBD-CBD despite its high affinity to ACE2 ( Figure S4B). Since a single T478K mutation did not present compromising effect on the binding of potent neutralizing mAbs previously, 24 we observed that the additional N501Y mutation in AD variant does not affect nAb binding within the modified cpVNT's reaction time ( Figure S4B). In contrast, the beta variant despite showing modest increase in affinity toward ACE2 (K D 9.6 nM) than WT (K D 12.7 nM), exhibited significantly lower nAb percent blocking than WT ( Figure 4a).
As the K417 and E484 sites are known to escape both Class 1 and Class 2 anti-RBD antibodies, 25 the combined effect of RBD/ACE2 binding and poor nAb recognition generate more pronounced immune escape response. These examples indicate that the modified cpVNT can be used to systematically assess the RBD mutations and improve our understanding of its underlying molecular mechanism versus nAb response.
With the emergence of highly transmissible SARS-CoV-2 variants, the durability and persistence of vaccine effectiveness is of major concern. Although nAb response strongly correlates with immune protection, 12 cellular immunity is essential in providing sustained immune protection upon exposure, particularly against severe illness. Therefore, both humoral and cellular immune response are required for a complete assessment of SARS-CoV-2 immunity. While standardized methods for rapid assessment of cellular immunity responses are underway, 26  Colombia that was reported to escape vaccine induced immunity. 30 Our data also shows the strength of the modified cpVNT as a PoC

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.