A Novel Flow Cytometry Array for High Throughput Detection of SARS-CoV-2 Antibodies

TL;DR

A new flow cytometry method detects SARS-CoV-2 antibodies quickly and reliably, making it useful for large-scale studies.

Contribution

A novel multiplex bead-based assay for high-throughput detection of SARS-CoV-2 antibodies with high reproducibility.

Findings

The assay measured three antibody types in 624 samples within 2 hours.

Intra- and inter-plate coefficients of variation were low, indicating high reproducibility.

The platform effectively quantified background noise for easier data interpretation.

Abstract

Background/Objectives: Although the U.S. Food and Drug Administration (FDA) has approved one antiviral treatment and authorized others for emergency use, there is no fully effective antiviral therapy for coronavirus disease 2019 (COVID-19), which is caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Assays detecting virus-specific immunoglobulins (Ig) or nucleic acids in large-scale epidemiological, vaccine, and drug development studies remain limited due to high costs, reagent accessibility, and cumbersome protocols. Methods: A multiplex bead-based assay was developed to simultaneously detect human IgM, IgG, and IgA antibodies against the SARS-CoV-2 spike receptor binding domain (RBD) in serum using flow cytometry. Assay performance was evaluated for sensitivity, specificity, reproducibility, and cross-reactivity and compared to another immunoassay platform. Results: The assay enabled simultaneous measurement of three antibody isotypes across 624 samples within 2 h. Intra-plate coefficients of variation (CVs) ranged from 3.16 to 6.71%, and inter-plate CVs ranged from 3.33 to 5.49%, demonstrating high reproducibility. The platform also quantified background noise from nonspecific binding, facilitating straightforward data interpretation. Conclusions: This novel, flexible multiplex bead-based assay utilizing a well-established platform provides a rapid and reproducible approach for detecting SARS-CoV-2-specific antibodies. Its high throughput capacity and low variability make it well suited for large-scale epidemiological, vaccine, and therapeutic studies. The platform’s adaptability further supports application to other infectious diseases, offering an ideal tool for broad immunological surveillance.

Figures

Full text

1. Introduction

Coronaviruses represent a group of enveloped viruses with positive-sense single-stranded RNA with one of the largest genomes among RNA viruses, with sizes ranging from approximately 26 to 32 kilobases [1,2]. One of the characteristic features of coronaviruses is their crown-shaped spikes on the viral surface, which play a critical role in viral entry into host cells of mammals and birds [1,3,4]. In humans, three distinct coronaviruses that cause respiratory tract infections include the severe acute respiratory syndrome coronavirus (SARS-CoV), Middle East respiratory syndrome coronavirus (MERS-CoV), and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which were first reported in 2002, 2012, and 2019, respectively [2,5,6,7].

Coronavirus disease 2019 (COVID-19), which is caused by infection with SARS-CoV-2, was declared a worldwide pandemic by the World Health Organization in March 2020. By November 2021, over 245 million infections and 5 million deaths associated with COVID-19 were reported worldwide. Approximately 20% of COVID-19 patients are asymptomatic, while others experience a wide range of symptoms, such as fever, muscle pain, headache, sudden loss of taste, and sore throat [5,8]. Importantly, asymptomatic and pre-symptomatic individuals can still infect others [8], while advanced age and underlying conditions are associated with increased morbidity and mortality [9,10,11,12]. For example, patients who are >70 years old account for 70% of all COVID-19-related deaths in the United States. Extensive production of cytokines, which is also known as a “cytokine storm,” can perpetuate acute respiratory distress syndrome (ARDS) and organ failure in COVID-19 patients [13,14,15].

Detection of humoral antibodies to SARS-CoV-2 antigens is critical in slowing the spread of COVID-19. Routine antibody testing will help determine who has been infected with SARS-CoV-2 and which vaccinated individuals may require additional booster shots. Clinical assessment of antibodies has been used in trials that assess interventions in SARS-CoV-2 infection to improve understanding of clinical outcomes [16]. Assays that quantitatively or qualitatively determine the presence of antibodies have been used since the beginning of the pandemic in both a research-use-only setting and a diagnostic setting [17]. Currently, the antibody titers of a specific immunoglobulin (Ig) isotype for hundreds of patients can be determined with a plate-based indirect enzyme-linked immunosorbent assay (iELISA) in ~3 h. However, one major disadvantage of this platform is that its throughput capacity is limited to the following: each antibody isotype must be detected separately. Multiplex approaches can increase the throughput of antibody detection and can be used in blood [18,19,20,21] and saliva [22]. Additionally, detection of multiple Ig subtypes (IgM, IgG, and IgA) is important as the antibody kinetics vary among isotypes after infection [23] or vaccination [24].

To overcome these limitations, we developed a high-throughput, bead-based assay for flow cytometry to detect three human Ig isotypes (IgM, IgG, and IgA) simultaneously to the SARS-CoV-2 Spike (S) protein’s receptor binding domain (RBD) within 2 h (Figure 1). To enable high-throughput analysis, 13 fluorescent beads of varying sizes (i.e., 5 and 8 µm in diameter) are barcoded with different fluorescent intensities in the allophycocyanin (APC) channel (Figure 2). Anti-human IgM, IgG, and IgA were also conjugated with three different fluorochromes (Figure 1). Using these different size-fluorescent combinations, three antibody isotypes in 624 serum samples can be analyzed simultaneously in a standard 96-well V-bottom microplate using a flow cytometer’s forward scatter/side scatter plot (FSC/SSC) and APC channel. Furthermore, antibody signal is measured accurately with control beads that account for nonspecific binding, thus making data interpretation easy. This high-throughput bead-based assay could be used for population-wide screening for COVID-19 and other infectious diseases.

2. Materials and Methods

Serum Samples. One hundred twenty-one (121) serum samples from patients confirmed to have COVID-19 via quantitative reverse transcription polymerase chain reaction (qRT-PCR) were purchased from Vitrologic, Inc. (2) (Vitrologic, Inc., Charleston, SC, USA), Cantor BioConnect Inc. (45) (Cantor BioConnect Inc., Santee, CA, USA), and Texas Direct Diagnostics (74) (Texas Direct Diagnostic, San Marcos, TX, USA). Samples were collected between March 2020 and May 2020. Of these 121 samples, the age at collection ranged from 16 to 92, with a median age of 61. 64 identified as female, 53 identified as male, and 4 did not share this information. The samples were collected between 7 days and 60 days post symptom onset and at an average of 26 days post date of PCR test (with a range of 2 days to 53 days). Two hundred ninety-nine (299) serum samples from “normal” donors were purchased from Vitrologic, Inc., before the first case of COVID-19 was reported in December 2019; thus, these donors did not have COVID-19. The age at collection ranged from 29 to 64, with a median age of 45. Among these 299 normal donors, 100 identified as female, 100 identified as male, and 99 did not share this information. Serum from patients who tested positive for antinuclear antibodies (ANA), hepatitis C, or respiratory syncytial virus (RSV), was purchased from Cantor BioConnect (Cantor BioConnect Inc., Santee, CA, USA).

SARS-CoV-2 Antigen. Recombinant SARS-CoV-2 spike receptor binding domain (S-RBD) with a C-terminal 6x histidine (His) tag was expressed and purified. Briefly, S-RBD (Arg319-Phe541; Accession number QHD43416) was expressed in human embryonic kidney 293 (HEK293) suspension cells via transient transfection. The protein was then purified by immobilized metal affinity chromatography (IMAC) and buffer exchanged into 1x phosphate-buffered saline (PBS), pH 7.4, with a Thermo Scientific^TM^ Slide -A-Lyzer^TM^ Dialysis Cassette, 3.5 K MWCO (Thermo Fisher Scientific; Waltham, MA, USA). Protein purity was >95% as determined via an SDS-PAGE gel stained with Coomassie blue.

Fluorochrome-conjugated Human IgM, IgG, and IgA Detection Antibody Cocktail. Antibodies to human IgM Fc5μ, IgG Fcγ, or IgA ɑ-chain were purchased from Jackson ImmunoResearch (Jackson ImmunoResearch Labs, West Grove, PA, USA) and conjugated with R-phycoerythrin (R-PE), RayBright^®^ Violet 450 (V450), or RayBright^®^ Blue 488 (B488), respectively. RayBright^®^ V450 and B488 were purchased from RayBiotech Life (RayBiotech Life Inc., Peachtree Corners, GA, USA), whereas the R-PE was purchased from Columbia Biosciences (Columbia Biosciences, Frederick, MD, USA). The detection antibodies were mixed to form the “Detection Antibody Cocktail.”

Immobilization of SARS-CoV-2 S-RBD or BSA to RayPlex^®^ Multiplex Beads. Carboxyl-coated RayPlex^®^ Multiplex Beads (RayBiotech Life, Inc., Peachtree Corners, GA, USA) were barcoded using different fluorescent intensities specific to the allophycocyanin (APC) channel and bead sizes (5 and 8 µm) (Figure 2). SARS-CoV-2 S-RBD was covalently immobilized onto each of the 13 RayPlex multiplex beads using 1-Ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC)/sulfo-N-hydroxysuccinimide (sulfo-NHS) chemistry. Briefly, 1 mg of carboxyl-conjugated RayPlex^®^ beads in 500 µL activation buffer [50 mM 2-(N-morpholino)ethanesulfonic acid (MES) buffer pH 5.0 and 0.002% Tween^®^ 20] were activated with 36 mM EDC/7.2 mM sulfo-NHS (CovaChem, LLC; Loves Park, IL, USA) for 1 hr. Activated beads were washed twice with 1 mL of coupling buffer (PBS, 0.002% Tween^®^ 20) and then incubated with 15 µg of purified SARS-CoV-2 S-RBD or bovine serum albumin (BSA) in 50 µL of coupling buffer with gentle shaking for 3 hrs. Immobilization was quenched with 50 mM Tris (pH 8.0). The beads coated with SARS-CoV-2 S-RBD are hereafter referred to as “Antigen Beads” (R1-RBD to R13-RBD), while the BSA-coated beads are referred to as “Control Beads” (R1-CTL to R13-CTL). The beads were stored in storage buffer (PBS, 0.2% w/v BSA, 0.05% w/v NaN3) at 4 °C until further use. Unless otherwise specified, all reagents were purchased from Sigma-Aldrich (Sigma-Aldrich Inc., St. Louis, MO, USA).

Flow Cytometry Analysis of SARS-CoV-2-specific IgM, IgG, and IgA Antibodies. The bead-based assay was carried out in 96-well Costar^®^ 3897 V-bottom microplates (Corning, Inc., Corning, NY, USA), with antigen beads (25 µL) in the first 4 rows (48 wells) and controls beads (25 µL) in the last 4 rows (48 wells).

Serum samples were first diluted 4000-fold in 1X Assay Diluent [PBS, 0.05% Tween^®^ 20, 1% BSA, 1% polyvinylpyrrolidone (PVP) + 1% casein] in a 2.2 mL deep storage plate (NEST Scientific USA, Inc., Beltsville, MD, USA). Twenty-five microliters (25 µL) of each diluted sample were added to one well of the V-bottom microplate with one of the 13 antigen beads and one well of the V-bottom microplate with the corresponding control beads, then mixed for 90 min at 1000 rpm at room temperature using an orbital shaker. Thus, a total of 50 µL per sample was employed, with 48 unique samples per plate. The beads were then washed with 200 µL of wash buffer (PBS, 0.05% Tween^®^ 20) and spun down at 1000 g for 5 min at room temperature.

With a multichannel pipette, beads in corresponding wells across 13 microplates were combined in 200 µL of wash buffer to form 1 single microplate (e.g., well A1 from 13 plates pooled together into well A1 of the final 96-well microplate), such that each final pooled well contained up to 13 different RayPlex^®^ bead combinations (Supplemental Figure S1). The plate was spun down at 1000 g for 5 min at room temperature, and the supernatant was removed.

To detect human IgM, IgG, and IgA bound to the antigen beads, the beads were incubated with 50 µL of a pre-titrated fluorochrome-conjugated detection antibody cocktail (R-PE goat anti-human IgM, RayBright^®^ V450 goat anti-human IgG, and RayBright^®^ B488 goat anti-human IgA) on an orbital shaker for 30 min at 1000 rpm and room temperature. After washing twice with 200 µL Wash Buffer, the beads were resuspended in 200 µL Wash Buffer and analyzed with a BD FACSCelesta flow cytometer (Becton, Dickinson and Company, Franklin Lakes, NJ, USA). Data were analyzed with FlowJo software, version 10 (Becton, Dickinson and Company, Ashland, OR, USA).

Data Validation, Cutoff Value, and Positivity Definition. The control beads were used to account for the background signal arising from non-specific binding of serological antibodies to proteins. The presence of antibodies to S-RBD, or “positivity,” was determined based on whether any of the antibody isotype data met two conditions: (1) Antigen Bead mean fluorescent intensity (MFI) to Control Bead MFI ratio > 1.5, and (2) Antigen Bead MFI—Control Bead MFI > 200. A serum sample is diagnosed as COVID-19 positive when any one of IgM, IgG, or IgA is positive.

Data Analysis. Using the MFI values for both antigen and control beads, the following statistical analysis was performed:

1. Assay Sensitivity, Specificity, Positive Predictive Value (PPV), and Negative Predictive Value (NPV). We tested 121 patient samples and 299 normal serum samples.
2. Assay Performance. Intra- and inter-plate CVs were used to determine the assay precision and reproducibility.
3. Class Specificity. Dithiothreitol (DTT) inactivates IgM antibodies, but not IgG antibodies. Here, IgM-, IgG-, and IgA-positive serum samples were either treated or not treated with 5 mM DTT at 37 °C for 15 min prior to running the bead-based assay to determine whether the detection of IgM antibodies was specific.
4. Cross-Reactivity. Serum from donors immunized against other infectious agents was employed to determine the cross-reactivity of antibodies to S-RBD.

3. Results

3.1. Establishment of the Bead-Based Assay

To develop the bead-based assay to detect serological IgM, IgG, and IgA antibodies to SARS-CoV-2 S-RBD, we immobilized recombinant S-RBD on 13 RayPlex^®^ multiplex beads (“Antigen Beads”). In parallel, BSA was immobilized to another set of beads (“Control Beads”). The control beads enabled the measurement of background noise arising from nonspecific antibody binding. Serum samples were incubated with antigen or control beads. To determine the optimal conditions for the assay, we compared different serum dilutions (Table 1), detection antibody dilutions, and incubation times. Optimal conditions were defined as those resulting in the highest MFI signal-to-noise (S/N) ratio with minimal background, or the antigen bead MFI to the control bead MFI following incubation with the serum sample. The highest MFI S/N ratio with minimal background was achieved when using a 4000-fold dilution of serum samples, a 90 min incubation with the antigen or control beads, 50 µL of the fluorochrome-conjugated detection antibody cocktail (1:100 dilution from 0.5 mg/mL of each detection antibody), and a 30 min incubation with the detection antibody cocktail at room temperature.

3.2. Determination of the Assay Sensitivity and Specificity

Following statistical analyses of 121 COVID-19-positive samples and 299 normal samples, the presence of antibodies to S-RBD, or “positivity,” was determined based on whether any of the antibody isotype data met two conditions: (1) Antigen Bead MFI to Control Bead MFI ratio > 1.5, and (2) Antigen Bead MFI—Control Bead MFI > 200. 119 COVID-19 positive samples and 297 negative samples were accurately classified (Supplemental Table S1), resulting in a sensitivity of 98.3% and specificity of 99.3%. A subset of these data—13 COVID-19 patient serum samples, incubated with antigen or control beads are shown in Figure 3 and Table 2 (Figure 3, Table 2). The IgM, IgG, and IgA MFI S/N ratios demonstrated that the bead-based assay accurately classified the positive and negative samples. These results suggest the bead-based high-throughput assay has very high sensitivity and specificity in identifying individuals who have been exposed to SARS-CoV-2.

3.3. Assay Intra- and Inter-Plate Coefficient of Variation (CV)

To evaluate the stability and reproducibility of this assay, the intra- and inter-assay CVs were determined based on the MFI S/N ratios of 13 positive samples. Intra-plate CV analysis was determined using four replicates per sample within each plate. Inter-plate CV analysis employed data across four plates. The intra- and inter-assay CVs were 3.16–6.71% and 3.33–5.49%, respectively (Table 3 and Table 4). These data demonstrate that the bead-based assay is stable and reproducible.

3.4. Assay Positive Predictive Value (PPV) and Negative Predictive Value (NPV) Determination

The PPV and NPV are defined as the proportion of individuals with positive or negative test results who are correctly diagnosed, respectively. Using 121 COVID-19-positive samples and 299 normal samples, the PPV was 98.3% and the NPV was 99.3%.

3.5. Comparison of Antibody Measurement Using ELISA

We compared the antibody detection of the multiplex bead-based assay with commercial ELISA tests that measure IgM, IgG, and IgA antibodies to SARS-CoV-2 S-RBD independently. 116 of the 121 serum samples from patients confirmed to have COVID-19 via PCR and 259 of the 299 serum samples from patients who did not have COVID-19 collected pre-2019 were analyzed. We found a 100% concordance in terms of the expected positive and negative results in the single-target ELISA compared to the multiplex array for the COVID-19 positive serum samples (98% vs. 98%, respectively). The concordance in terms of the expected positive and negative results in the single-target ELISA compared to the multiplex array for the control population was only 30% (30% vs. 99%, respectively). The IgM and IgA ELISA had a very high percentage of false positives due to high background, which was significantly improved in the multiplex array, which included a BSA control. The concordance in terms of the expected positive and negative results in the single-target ELISA compared to the multiplex array for IgG alone was the highest of the Ig isotypes (93% vs. 99%, respectively, in the control population and 89% vs. 98%, respectively, in the COVID-19-positive population). The bead-based array had increased sensitivity, specificity, and accuracy compared to the ELISA (Table 5).

3.6. Class Specificity

To ensure that the antibody isotype detection was specific, two patient serum samples were either treated or not treated with 5 mM DTT, since DTT is known to inactivate IgM but not IgG or IgA antibodies. Samples were then tested for IgM, IgG, and IgA antibodies to S-RBD with the bead-based assay. Both untreated samples were IgM, IgG, and IgA positive, but only IgG and IgA were positive in DTT-treated samples (Supplemental Table S1). These data indicate that antibody isotype detection with the bead-based assay is specific, particularly for IgM.

3.7. Crossreactivity Investigation with Samples with Non-COVID-19 Infections

Antibodies produced by the immune system in response to vaccination or previous exposure to viruses other than SARS-CoV-2 may cross-react with the S-RBD immobilized on the antigen beads, thus leading to false positives. To evaluate this possibility, 15 serum samples from patients who were clinically diagnosed with having antinuclear antibodies (ANA), hepatitis C virus (HCV), or Respiratory syncytial virus (RSV) were tested. All samples were classified as being negative for IgM, IgG, and IgA antibodies to S-RBD (Table 6).

4. Discussion

Large-scale screening of humoral antibodies to SARS-CoV-2 antigens is critical in epidemiological, vaccine, and drug development studies as well as slowing the spread of COVID-19 [25]. Assays that quantitatively or qualitatively determine the presence of antibodies have been used since the beginning of the pandemic in both a research-use-only as well as a diagnostic setting [16,17]. The time–course for antibody generation and detection varies, but generally IgM will become detectable 3–5 days post infection, IgA 5–7 days post infection, and IgG 7–14 days post infection. Studies have shown increased sensitivity and specificity when simultaneously detecting the different Ig subtypes [26]. Additionally, antibody testing in combination with PCR has shown improved accuracy in results [27], as together they can detect the presence of the virus or different antibody subtypes over the varying course of disease kinetics. However, the ability to perform high-throughput testing for antibodies has been limited.

The assay described here has the capability to measure 3 antibody isotypes for 624 serum samples within 2 h. In this proof-of-concept study, we demonstrate three antibody isotypes (IgM, IgG, and IgA) can be measured in 420 serum samples with our bead-based assay. We further show that the assay has low inter- and intra-CVs as well as high sensitivity and specificity. One of the key advantages of this assay compared to other bead-based assays is the use of BSA-conjugated control beads. Previous bead-based assays have relied on using the absolute MFI values [28]. However, some individuals produce high levels of antibodies that bind nonspecifically to the beads and proteins, resulting in false positives. By using control beads, we could employ an MFI S/N ratio that can account for high background caused by non-specific antibody binding.

We focused on using the assay to detect antibodies to the SARS-CoV-2 S-RBD because viral entry is facilitated by the interaction between the S-RBD and the host cell’s angiotensin-converting enzyme 2 (ACE2) receptor. As such, the S-protein and the S-RBD are major foci of vaccine and drug development efforts to prevent and treat COVID-19. However, this platform could be easily employed to study antibodies to other SARS-CoV-2 antigens.

Other non-invasive sample types include dried blood spots (DBS) and saliva. When these samples were tested with the assay, the data were inconsistent. However, further investigation using these sample types is warranted since DBS is a convenient sample type in resource-poor settings compared to serum or plasma. Furthermore, DBS can also be collected conveniently in the patient’s home and sent, via mail at room temperature, to a laboratory for testing.

High-throughput multiplex assays like the one developed here offer substantial advantages in both cost-effectiveness and labor efficiency compared to traditional single-target methods by reducing reagent consumption, sample volume requirements, and the number of assay runs needed to generate comprehensive datasets. This consolidation translates into lower per-target costs and minimizes waste. In addition, the streamlined workflow reduces hands-on time for laboratory personnel, alleviates bottlenecks associated with sample preparation and processing, and accelerates data acquisition. Collectively, these efficiencies support not only economic benefits but also faster turnaround times, making multiplex assays particularly valuable for large-scale studies, biomarker discovery, and translational research applications.

Notably, the criteria for identifying a sample as positive or negative for COVID-19 with this assay were determined using a large patient cohort but not validated using an independent cohort. Also, only a small cohort of non-COVID infectious samples were tested to determine the cross-reactivity of the assay to non-COVID conditions. Other types of infectious samples and larger cohorts should be employed in the future to validate the data. It is worth noting that while we used the assay to detect antibodies to S-RBD, other studies have shown that the immune system also responds to other SARS-CoV-2 antigens [29,30,31]. Thus, the use of other SARS-CoV-2 antigens with the assay should be explored.

5. Conclusions

Our bead-based multiplex assay is reproducible and high throughput, enabling simultaneous detection of IgM, IgG, and IgA antibodies to SARS-CoV-2 using standard flow cytometry. It is flexible across a wide range of applications and is compatible with standard flow cytometry, requiring only red and blue lasers. In the context of COVID-19, the assay will help slow the spread of COVID-19 by enabling routine population-wide screening of SARS-CoV-2 antibodies to help identify those who should self-quarantine and those who may require additional booster shots following COVID-19 vaccination. It offers significant advantages in efficiency, scalability, and interpretability, making it suitable for population-wide antibody screening, vaccine monitoring, and epidemiological studies. Beyond COVID-19, the platform’s adaptability supports its application to other infectious diseases, highlighting its potential as a broadly useful immunological tool.