Development of a Multiplexed Fluorescent Pseudovirus Neutralization Test for Simultaneous Assessment of Immunity to Three SARS-CoV-2 Variants

Abstract

In the post-pandemic era, monitoring adaptive immunity of the population to emerging SARS-CoV-2 variants remains an important public health priority. To address this need, we developed a test that can simultaneously assess the neutralization ability of three SARS-CoV-2 variants. A panel of lentiviral pseudoviruses, each bearing the S-protein of different SARS-CoV-2 variants (Wuhan-Hu-1, BA.1, and XBB.1.5) and expressing a unique fluorescent protein (Clover, mRhubarb713, or mRuby3) was generated and used to transduce hACE2-overexpressing cells. The percentage of infected target cells for each variant was quantified via flow cytometry. Co-infection led to a minor reduction in the percentage of infected cells compared to mono-infection controls, confirming the robustness of the assay. We then applied the test to the analysis of human sera samples, which were collected in the Sirius Federal Territory (Russian Federation) and revealed the following: (1) sera collected in 2021 neutralized the Wuhan-Hu-1 variant and demonstrated cross-specificity to the BA.1 variant, but not to the XBB.1.5 variant; (2) sera collected after the Omicron emergence point neutralized Wuhan-Hu-1 and BA.1, and possessed a weak ability to neutralize the XBB.1.5. This assay provides a valuable tool for efficient profiling of humoral immunity and monitoring its development in response to ongoing viral diversity.

1. Introduction

Following the emergence of the Omicron variant (November of 2021), global SARS-CoV-2 seroprevalence demonstrated significant variability depending on the region, with the reported rates ranging from 48% to 99% by the end of 2022 [1,2,3,4]. This period was remarkable for a notable increase in seropositivity, driven by two key factors: the continuation of mass vaccination programs initiated in 2021 (including booster doses), and the high transmissibility of the Omicron lineage, which led to a rise in breakthrough infections [5,6]. For instance, a 2023 study in Iran revealed that over half of the population possessed anti-nucleocapsid protein IgG antibodies, a marker of natural infection, regardless of whether they received whole-virion or vector-based vaccines [7]. This finding underscores the high incidence of breakthrough infections despite the widespread vaccination of the population.

The continuous evolution of SARS-CoV-2 and the emergence of new variants with increased immune evasion potential necessitate constant monitoring of cross-protective immunity. Pseudovirus-based neutralization assays are a key tool for this purpose. These tests utilize replication-incompetent viral particles pseudotyped with the spike (S) proteins of different variants, and can be performed in BSL-2 laboratories to evaluate the neutralizing capacity of antibodies against diverse viruses [8]. However, these assays are often labor-intensive and only have low throughput.

To enhance efficiency, multiplex systems that enable simultaneous evaluation of neutralization activity against several variants in a single sample are employed [9,10]. This approach is particularly relevant given that since 2022 until November 2025 the World Health Organization (WHO) has tracked multiple “Variants of Interest” (VOIs) rather than designating new “Variants of Concern” (VOCs), reflecting a new landscape defined by the co-circulation of multiple sublineages. Therefore, multiplex tools are useful in the characterization of the emerging variants and can be applied in their subsequent classification now shifted towards the identification of distinct SARS-CoV-2 serotypes [11,12,13].

To address this need, we developed and characterized a multiplex, cell-based pseudovirus system. Pseudoviral particles contain fluorescent reporters, which allow for their detection using flow cytometry, confocal microscopy, or cell imaging readers. Human serum neutralizing antibodies are able to bind to pseudoviruses, inhibiting their ability to transduce cells. Thus, the level of fluorescent signal decreases and we can measure the level of neutralization and NT50 in the serial dilutions of human sera.

In this study we showed that the multiplex system enables simultaneous evaluation of neutralizing antibody (nAb) titers against three SARS-CoV-2 variants: the ancestral Wuhan-Hu-1, BA.1 (Omicron), and XBB.1.5, and can be employed to track changes in antiviral immunity in the population.

2. Materials and Methods

Cell lines. Pseudovirus particles (PVPs) were produced by a classic PEI-mediated transfection of the HEK293T cell line. The HEK293T cell line overexpressing hACE2 (HEK-hACE2) was used as the target cell line for the lentiviral-based transduction [10]. All cell lines were cultured in high glucose Dulbecco’s modified Eagle’s medium (DMEM, G4514-500ML, Servicebio, Wuhan, China) supplied with 10% fetal bovine serum (FBS, SV30160.3, Cytiva, Wilmington, DE, USA), 2 mM L-glutamine (25030-024, Gibco, Waltham, MA, USA), 100 IU/mL penicillin and 100 μg/mL streptomycin (G4003-100ML, Servicebio, Wuhan, China) in standard cell culture conditions (37 °C, 5% CO₂).

Plasmids. For the PVP production further plasmids were used: pCG1-SARS-2S coding a codon-optimized S-protein (Wuhan-Hu-1, BA.1, or XBB.1.5), pUCHR-IR-FluorProtein encoding a codon-optimized fluorescent protein (Clover, mRhubarb713, or mRuby3), and the packaging plasmid pCMV-dR8-2. Reporter plasmids were generated based on the pUCHR-IR-GFP plasmid [10]. Codon-optimized expression sequences of mRuby3, Clover, and mRhubarb713 were ligated into pUCHR-IR-GFP replacing the eGFP coding sequence (Appendix A) [14,15,16].

Human samples. Human serum samples from 5 donors were collected during the period of 2021–2025 (

Table 1. Donors’ characteristics.

Donors	Gender	Vaccination Status	Positive PCR Test	Sample ID (Date of Collection)
1	F	Gam-COVID-Vac date 1—12 February 2021, date 2—15 March 2021, date 3—26 January 2022	4 February 2022	1.1 (31 March 2021) 1.2 (20 April 2022) 1.3 (24 November 2022) 1.4 (20 June 2024) 1.5 (5 November 2025)
2	M	Gam-COVID-Vac date 1—29 January 2021, date 2—20 February 2021, date 3—19 October 2021	5 February 2022	2.1 (23 March 2021) 2.2 (16 February 2022) 2.3 (3 August 2022) 2.4 (6 November 2025)
3	M	ChAdOx1 nCoV-19 date 1—16 March 2021, date 2—28 May 2021, Gam-COVID-Vac date 1—17 December 2021, date 2—4 February 2022	No data	3.1 (1 July 2021) 3.2 (20 April 2022) 3.3 (5 October 2023) 3.4 (5 November 2025)
4	F	None	27 December 2020	4.1 (31 March 2021)
5	M	Gam-COVID-Vac date 1—1 June 2021, date 2—22 June 2021	5 September 2020	5.1 (3 February 2021) 5.2 (27 July 2021) 5.3 (3 August 2022) 5.4 (7 November 2025)

) at Sirius Federal Territory, Russian Federation. Informed consent was obtained from each participant. The aliquoted serum was stored at −80 °C. Before the neutralization assay samples were heat-inactivated at 56 °C to avoid the impact of the complement system on the results.

Production of SARS-CoV-2 PVPs. For PVP production three plasmids were co-transfected into the HEK293T using a polyethyleneimine (PEI) solution (408727-100ML, Sigma Aldrich, St. Louis, MO, USA or G1802-1ML, Servicebio, Wuhan, China). The optimal quantity of PEI was determined by titration (Appendix B,

Table A4. Transfection mean depending on PEI:pDNA ratio and PEI solution origin.

PEI:pDNA Ratio	PEI Origin	Transfection Mean, %
1:1	Sigma Aldrich, St. Louis, MO, USA	62.7
	Servicebio, Wuhan, China	53.9
2:1	Sigma Aldrich, St. Louis, MO, USA	48.0
	Servicebio, Wuhan, China	63.1
3:1	Sigma Aldrich, St. Louis, MO, USA	19.8
	Servicebio, Wuhan, China	61.3

). For the transfection, we mixed a reporter plasmid bearing a distinct fluorescent protein (Clover, mRhubarb713 or mRuby3), S-protein coding plasmid and packaging plasmid (

Table 2. Plasmids’ quantity for the transfection.

Plasmid	Mass Per 10 cm Dish, μg
pCMV-dR8-2	5.00
pCG1-SARS-2S	3.33
pUCHR-IR-FluorProtein	6.67

) with PEI solution in the ratio of pDNA:PEI = 1:2 in Opti-MEM medium (11058-021, Gibco, Waltham, MA, USA). In the outcome we got lentiviral particles carrying S-protein on the surface and expressing a fluorescent protein inside the packaging envelope.

DMEM with 5% FBS, 2 mM L-glutamine, 100 IU/mL penicillin and 100 μg/mL streptomycin, 1 mM sodium pyruvate, 20 mM HEPES and MEM-essential amino acids (M7145-100ML, Sigma-Aldrich, USA) was utilized as the PVP packaging medium. Cells were incubated for approximately 48 h, then the PVPs containing the supernatant were collected, purified with a sterile 0.45 μm PES syringe filter (SSF250-45-PES, Biocomma limited, Shenzhen, China) and concentrated by centrifugation (23,000× g, 4 °C, 2.5 h). The supernatant was removed except for 1/20 of its volume, and the pellet was resuspended in this medium, aliquoted and stored at −80 °C until use (Figure 1B).

After the transfection, the cells were collected in complete DMEM with 10% DMSO (67-68-5, Sisco Research Laboratories Pvt. Ltd., Mumbai, India) and stored at −80 °C. Frozen cells were further used to evaluate transfection efficiency with a BD LSRFortessa (BD Biosciences, San Jose, CA, USA) flow cytometer.

Quantification of SARS-CoV-2 PVPs. To quantify the PVPs expressing different fluorochromes, p24 concentrations were measured with ELISA. All samples were diluted at a ratio of 1:400 and assessed in duplicate across different production timepoints with the HIV-1 p24 ELISA Kit (Vector-Best, Novosibirsk, Russia). The amount of p24 was calculated using the 1st WHO International Standard for HIV-1 p24 Antigen (NIBSC code: 22/230). Statistical significance was assessed using a one-way ANOVA (GraphPad Prism v. 8.0.1).

SARS-CoV-2 PVPs transduction assay. HEK-hACE2 cells were seeded in a 96-well plate at 1.2 × 10⁴ cells per well in DMEM/F12 cell growth medium (G4610-500ML, Servicebio, Wuhan, China) with 10% FBS, 2 mM L-glutamine, 100 IU/mL penicillin and 100 μg/mL streptomycin and incubated under standard conditions for 16–18 h. Preliminary transduction mix of the PVPs was prepared in a separate 96-well plate using DMEM/F12 medium supplemented with 2% FBS, 2 mM L-glutamine, 100 IU/mL penicillin and 100 μg/mL streptomycin and 50 μg/mL protamine sulfate. For pseudovirus stabilization and enhanced transduction, each well was supplemented with protamine sulfate at a final concentration of 50 μg/mL. The final volume in each well equaled 200 μL. Diluted PVPs were then incubated for one hour at room temperature to simulate the neutralization test workflow. After that, the cell medium in the HEK-hACE2 plate was replaced with the transduction mix and then the plate was centrifuged (200× g, 1 h, RT) [19]. The plate was incubated for approximately 48 h and then assessed by flow cytometry (Figure 1C).

For the quantification of PVP stocks, we conducted a transduction assay as described earlier. Titration started at 100 μL of PVPs in the first well and gradually decreased at 20–30 μL steps until 10 μL. The remaining volume was replenished with DMEM/F12 medium (supplements are listed in the previous paragraph). The optimal concentration of PVPs corresponded to the volume of PVPs, resulting in a minimum transduction efficiency of 10–15% for all variants. Transduction efficiency was assessed via flow cytometry and described in detail in the Flow cytometry analysis Section (see below).

Neutralization assay for multiple variants of SARS-CoV-2. To assess the serum neutralization activity, we prepared 2-fold serial dilutions of human serum samples from 1:20 to 1:320 in DMEM/F12 with 2% FBS, 2 mM L-glutamine, 100 IU/mL penicillin and 100 μg/mL streptomycin and 50 μg/mL protamine sulfate. Then we added PVPs at the optimal concentration, incubated the plate at room temperature for one hour to allow nAbs in sera to bind PVPs, and followed the PVP transduction protocol as described above. The total transduction level in the multiplexed samples usually averaged in the range of 30–50% (Appendix C, Figure A1B). Neutralization percentage was calculated with the following formula:

$$\mathrm{N}\mathrm{e}\mathrm{u}\mathrm{t}\mathrm{r}\mathrm{a}\mathrm{l}\mathrm{i}\mathrm{z}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}=(1-{\displaystyle \raisebox{1ex}{${\%}_{serum}$}\!\left/ \!\raisebox{-1ex}{${\%}_{noserum}$}\right.})\times 100\%$$

(1)

where ${\%}_{serum}$ stands for the percentage of fluorescent cells in the sample with human sera, and ${\%}_{noserum}$ stands for the percentage of fluorescent cells in the sample with no human sera added. The 50% neutralization titer (NT50) was evaluated via non-linear regression analysis of GraphPad Prism Software ver. 8.0.1. We employed log(inhibitor) vs. normalized response equation with a variable slope. Pearson correlation between transfection and transduction efficiency was obtained with GraphPad Prism, as well as the correlation between the transducing unit (TU)/mL parameter and transfection rate. TU/mL was calculated with the formula:

$$\mathrm{T}\mathrm{U}/\mathrm{m}\mathrm{L}=-\ln \left(1-{\displaystyle \frac{transductionlevel\%}{100\%}}\right)\times \mathrm{N}\mathrm{u}\mathrm{m}\mathrm{b}\mathrm{e}\mathrm{r}\mathrm{o}\mathrm{f}\mathrm{p}\mathrm{l}\mathrm{a}\mathrm{t}\mathrm{e}\mathrm{d}\mathrm{c}\mathrm{e}\mathrm{l}\mathrm{l}\mathrm{s}\times {\displaystyle \frac{1}{V_{PVPs}}}$$

(2)

Multiplicity of infection (MOI) was calculated with the standard formula [20]:

$$\mathrm{M}\mathrm{O}\mathrm{I}={\displaystyle \frac{TU}{Numberofplatedcells}}$$

(3)

To assess the difference between the singleplex neutralization percentage (monofluorescent control) and the multiplex neutralization percentage, we have also calculated the Pearson correlation in GraphPad Prism.

Flow cytometry analysis.

The samples were assessed with a BD LSRFortessa flow cytometer equipped with the necessary lasers and filters (

Table 3. Fluorochromes, the flow cytometer’s laser lines and bandpass filters used for the detection.

Fluorochrome	Excitation Max., nm	Emission Max., nm	Laser Line, nm	Bandpass Filter
DAPI	354	456	405	450/50
Clover	505	515	488	530/30
mRuby3	558	592	561	610/20
mRhubarb713	690	713	640	730/45

).

The flow cytometry data were analyzed with FlowJo ver. 10.10.0 software (BD Biosciences). At the first step, the data were cleaned from abnormalities with the FlowAI plugin (ver. 2.3.1, basic settings) [21]. Compensation was calculated using the Autospill algorithm [22]. Transduction or transfection efficiency was determined as a fluorescent protein (FP) positive event percentage of live single cells (Figure 2).

3. Results

3.1. Production Conditions for PVPs

Lentiviral particles pseudotyped with SARS-CoV-2 S-protein variants were engineered to express distinct fluorescent protein transgenes (Clover, mRhubarb713, and mRuby3), enabling multiplexed detection (Figure 1A). To assess whether the choice of fluorescent reporter influenced particle production, we generated pseudoviruses bearing the Wuhan-Hu-1 spike protein encoding different fluorescent proteins. This result demonstrates that the incorporation of the different fluorescent reporters did not impact pseudovirus production efficiency (Figure 3A). We next assessed whether different plasmid combinations used for PVP production affect pseudovirus yield. To this end, PVPs incorporating different S proteins and fluorescent reporters were produced in multiple independent batches. The yield was quantified by measuring the HIV p24 antigen via ELISA. As shown in Figure 3B, we observed no significant differences in production efficiency between the various PVP variants, indicating that these variations did not impair PVP assembly or release.

All pseudotyped lentiviral particles were produced in HEK293T cells using PEI-mediated co-transfection. We optimized the DNA-to-PEI ratio (PEI:DNA = 2:1) and established a minimal transfection level (35–40%) required to generate PVPs for subsequent assays (Appendix B, Figure 3). Based on our previous experience, pseudoviral particles, independently of transfection efficiency, required at least a 20-fold concentration before being utilized in neutralization experiments. All PVPsbatches produced without concentration demonstrated low transduction efficiency relative to the concentrated ones (Appendix C, Figure A1A). To standardize the singleplex assay, we titrated each concentrated pseudovirus batch on HEK293-hACE2 target cells to determine the MOI required to achieve approximately 15–20% transduction efficiency. The MOI values for all pseudoviral particles were similar, averaging approximately 0.01. The volume of PVPs mandatory for that MOI differs based on the batch and the type of the S protein. For example, the Wuhan-Hu-1 variant typically requires a smaller volume of pseudovirus since the transduction efficiency is generally better. It is important to note that no volume of each PVP type exceeded 80 μL for the neutralization assay and the final mix of all three types of PVPs was no more than 190 μL per well (the final volume of each well was 250 μL).

Among the three variants of SARS-CoV-2 utilized in this study, BA.1 appeared to be the most unstable one, with a substantial transduction efficiency difference between batches. Batches of PVPs produced on different dates tended to vary in transfection level. Notably, a direct correlation between transfection efficiency and the resulting TU titer was observed for all tested variants except the BA.1 variant (Figure 4). Hence, transfection efficiency around 50–60% was considered preferable for subsequent successful lentiviral transduction.

3.2. Efficiency of Neutralization Did Not Differ in Single- and Multiplex Setup

Omicron variants have heightened transmissibility and infectivity rates; there are also various studies demonstrating elicited transduction efficiency of HEK-hACE2 cells with the Omicron-pseudotyped PVPs, so we expected to record an intense signal from the PVPs pseudotyped with BA.1 and XBB.1.5 [23,24,25,26]. Unexpectedly, the Omicron-pseudotyped PVPs appeared to be less effective in lentiviral transduction, typically conceding to the Wuhan-Hu-1 variant in the multiplex system (Figure 5). Hence, the Omicron-pseudotyped PVPs required a bigger volume to successfully conduct a multiplexed neutralization assay, which makes them laborious to produce. Most likely, our deviating observations are caused by modifications in the furin-cleavage site of the S protein plasmid utilized in this study: while significantly increasing the infectivity of Wuhan-Hu-1-pseudotyped PVPs, it diminishes one of the Omicron-pseudotyped PVPs [10]. It is also plausible that the target cell line—HEK-hACE2—lacks the proteases necessary for the viral entry pathway, which nullifies Omicron advantages [27]. Consequently, the transduction level was optimized to approximately 10% for each pseudovirus, achieving an overall transduction rate of 25–35%.

The core advantage of the developed system is the opportunity to simultaneously evaluate neutralization against three variants (Appendix D,

Table A5. Mean of the neutralization percentage for each sample.

	Dilution	Wuhan-Hu-1		BA.1		XBB.1.5
		Singleplex	Multiplex	Singleplex	Multiplex	Singleplex	Multiplex
1.1	1:20	10.78%	39.87%	0.00%	0.00% *	0.00%	0.00%
	1:40		15.76%		7.93%		15.29%
	1:80		22.36%		12.35%		0.00%
	1:160		0.00%		0.00%		0.00%
	1:320		0.00%		0.00%		0.00%
1.2	1:20	94.12%	72.79%	95.45%	93.24%	23.79%	47.21%
	1:40		94.77%		91.15%		35.13%
	1:80		91.83%		89.03%		17.40%
	1:160		86.17%		78.76%		32.36%
	1:320		63.21%		60.09%		3.71%
1.3	1:20	91.50%	89.63%	80.92%	87.72%	75.98%	79.56%
	1:40		89.56%		87.79%		56.95%
	1:80		90.40%		80.59%		29.50%
	1:160		94.27%		82.11%		12.94%
	1:320		82.47%		74.60%		19.47%
1.4	1:20	97.46%	84.25%	96.51%	92.84%	0.00%	28.13%
	1:40		72.42%		72.89%		5.23%
	1:80		81.19%		68.74%		18.38%
	1:160		53.67%		61.47%		27.40%
	1:320		50.01%		23.77%		20.55%
1.5	1:20	No data	92.96%	No data	95.79%	No data	93.20%
	1:40		91.94%		96.41%		64.80%
	1:80		90.91%		89.62%		46.06%
	1:160		86.13%		72.54%		20.44%
	1:320		46.80%		46.49%		17.27%
2.1	1:20	87.72%	70.62%	0.00%	3.07%	0.00%	0.00%
	1:40		79.64%		3.75%		2.72%
	1:80		54.96%		2.39%		0.00%
	1:160		32.60%		10.20%		8.65%
	1:320		30.93%		10.46%		16.69%
2.2	1:20	97.24%	97.13%	95.00%	88.55%	52.02%	33.28%
	1:40		97.86%		90.35%		54.18%
	1:80		88.64%		78.33%		63.27%
	1:160		91.44%		64.34%		36.19%
	1:320		49.92%		52.34%
2.3	1:20	98.22%	76.37%	95.86%	90.70%	29.94%	22.27%
	1:40		92.65%		82.53%		27.40%
	1:80		81.93%		64.91%		14.65%
	1:160		69.75%		37.83%		12.87%
	1:320		49.65%		26.32%		12.12%
2.4	1:20	No data	95.53%	No data	92.46%	No data	36.17%
	1:40		90.67%		90.26%		66.07%
	1:80		73.28%		44.10%		13.45%
	1:160		69.05%		44.82%		33.85%
	1:320		56.13%		48.41%		40.42%
3.1	1:20	0.00%	17.82%	0.00%	0.00%	0.00%	0.00%
	1:40		6.57%		13.14%		0.00%
	1:80		0.00%		0.00%		0.00%
	1:160		0.00%		0.00%		0.00%
	1:320		0.00%		0.00%		0.00%
3.2	1:20	95.69%	94.52%	96.18%	92.59%	37.58%	13.28%
	1:40		91.74%		90.48%		12.41%
	1:80		91.20%		79.63%		0.00%
	1:160		60.34%		26.19%		0.00%
	1:320		9.94%		6.26%		6.35%
3.3	1:20	98.23%	88.30%	85.47%	38.38%	0.00%	0.00%
	1:40		56.12%		17.98%		0.00%
	1:80		29.72%		27.20%		26.01%
	1:160		27.26%		32.35%		14.21%
	1:320		25.66%		20.64%		15.82%
3.4	1:20	No data	94.07%	No data	95.38%	No data	81.41%
	1:40		88.22%		93.00%		60.82%
	1:80		75.57%		85.44%		47.72%
	1:160		25.89%		49.59%		25.50%
	1:320		28.06%		32.56%		17.70%
4.1	1:20	67.24%	44.90%	61.02%	33.44%	39.31%	10.11%
	1:40		34.87%		42.95%		22.31%
	1:80		18.79%		47.69%		16.57%
	1:160		7.48%		29.87%		3.23%
	1:320		8.28%		29.87%		0.00%
5.1	1:20	No data	77.69%	No data	52.82%	No data	41.50%
	1:40		63.71%		44.39%		50.34%
	1:80		37.28%		27.65%		31.02%
	1:160		29.66%		23.86%		21.19%
	1:320		20.12%		28.21%		38.67%
5.2	1:20	No data	93.04%	No data	78.31%	No data	17.99%
	1:40		94.08%		53.95%		13.82%
	1:80		83.44%		22.88%		8.36%
	1:160		71.71%		10.82%		0.00%
	1:320		20.12%		2.35%		0.00%
5.3	1:20	No data	78.62%	No data	46.53%	No data	0.00%
	1:40		58.13%		30.04%		0.00%
	1:80		16.80%		1.43%		0.00%
	1:160		0.64%		0.00%		0.00%
	1:320		0.00%		0.00%		0.00%
5.4	1:20	No data	95.01%	No data	88.65%	No data	45.67%
	1:40		95.67%		74.48%		0.00%
	1:80		74.01%		35.17%		0.00%
	1:160		66.72%		34.80%		0.00%
	1:320		69.24%		36.36%		12.80%

* Outliers.

). We noticed that the reduction in PVP transduction efficiency in the multiplexed assay compared to the singleplex format potentially constrains the dynamic range of the assay. A key validation step was that the neutralization titer measured in the multiplex assay showed no significant deviation from that obtained in the traditional singleplex assay for any of the variants (Figure 5).

The congruence of the singleplex and multiplex assays was validated by testing all serum samples at a uniform 1:20 dilution in both formats. For each sample, we conducted singleplex and multiplex assays on the same 96-well plate and at the same timepoint to obtain analyzable data. Neutralization efficiency, expressed as the percent reduction in infection, demonstrated a high degree of correlation between the two assay formats (Figure 5).

High correlation between neutralization in the singleplex assay and the multiplexed assay does not imply that multiplexed transduction is equivalent to singleplexed one; however, it provides an opportunity to interpret data obtained from the multiplexed pseudoviral neutralization assay as valid.

3.3. Analysis of Neutralization Effectiveness Dynamics Against Three Variants of SARS-CoV-2 Using a Multiplex Assay

We calculated NT50 for each of the three SARS-CoV-2 variants (Wuhan-Hu-1, BA.1, and XBB.1.5) using the developed multiplex assay and analyzed its dynamics in donors’ sera obtained during the 2021–2025 period (Figure 6).

The vaccination with either Gam-COVID-Vac or ChAdOx1 nCoV-19 elicited a robust neutralizing antibody response. This is evident for the first samples collected from Donors 1, 2, and 3 in the first half of 2021, which showed high NT50 titers against the Wuhan-Hu-1 variant (Figure 6A, Table 1). Donor 4, who had a confirmed COVID-19 infection in late 2020, also presented a similar neutralization capacity against the Wuhan-Hu-1 variant at the first time point (Appendix D, Table A5).

At the early stage (2021), the NT50 against the BA.1 variant was lower than against Wuhan-Hu-1 (Figure 6B), though sera still demonstrated neutralization activity. In contrast, almost no reactivity against the XBB.1.5 variant was detected in any donor at this time (Figure 6C).

Following the spread of the BA.1 variant in early 2022, a rapid increase in NT50 against both BA.1 and Wuhan-Hu-1 was observed in Donors 1 and 2, both of whom had breakthrough infections despite revaccination. A similar boost in cross-reactive antibodies was observed in Donor 3, whose sample collection coincided with the peak of BA.1 circulation, suggesting a likely undiagnosed infection [23].

Despite the broadening immunity, neutralization of the XBB.1.5 variant remained low or undetectable in most samples until the period following the Omicron emergence and XBB.1.5 documented circulation in the Russian Federation beginning in early 2023 [28]. A sharp, transient increase in XBB.1.5 neutralization was detected shortly after BA.1 widespread worldwide, which evidently demonstrates the cross-protection of BA.1 immunity against the XBB.1.5 variant. It is imperative to note that the highest NT50 of the mentioned period belonged to Donors 1 and 2, both of whom had confirmed PCR-positive infection in February 2022. This suggests that the early cross-reactivity was likely induced by a high viral load during a specific convalescent period.

It should be noted that samples collected in late 2025 demonstrate increased neutralization of the XBB.1.5 variant, alongside a broadening of neutralizing efficacy against earlier strains.

4. Discussion

More than five years after the onset of the COVID-19 pandemic, continuous monitoring of population susceptibility to emerging SARS-CoV-2 variants remains important. Recent reports indicate the evolution of distinct serotypes, underscoring the need for precise serological tools [11,29]. Pseudovirus-based neutralization assays have become a standard tool for this purpose, as they enable quantification of the neutralizing capacity of immune sera against diverse variants. Since the initial SARS-CoV-2 outbreak, these assays have been widely employed to evaluate therapeutics targeting the viral spike protein and to measure antibodies elicited by vaccination or infection [30,31,32].

Building upon this foundation, the field has progressed from the initial singleplex format to exceedingly sophisticated multiplex systems. This evolution began with duo-plex assays and has advanced to more complicated platforms utilizing multiple fluorescent markers to enable high-throughput, simultaneous assessment of neutralization against several variants [9,33,34].

We developed a multiplexed lentiviral pseudovirus-based assay that simultaneously evaluates neutralization against three SARS-CoV-2 variants belonging to different serotypes (Wuhan-Hu-1, BA.1, and XBB.1.5). This assay utilizes pseudoviruses pseudotyped with the respective S-proteins and encoding three distinct fluorescent proteins (Clover, mRhubarb713, and mRuby3). We confirmed that the neutralization titers obtained in the multiplex format did not significantly differ from those generated by traditional singleplex assays (Figure 5).

By applying this assay to a longitudinal panel of serum samples (2021–2025), we successfully tracked the dynamics of the neutralizing antibody response. The assay correctly identified peaks in variant-specific neutralization following initial vaccination and the BA.1 Omicron outbreak. Furthermore, the data revealed distinct patterns of cross-reactivity between Wuhan-Hu-1 and BA.1, and between BA.1 and XBB.1.5 (Figure 6), providing a detailed view of the evolving humoral immune landscape and insight into the immune evasion of Omicron sublineages.

As SARS-CoV-2 continues to evolve, the developed multiplexed setup—utilizing distinct fluorophores alongside DAPI for viability staining—can be rapidly adapted to other emerging spike protein variants. This pilot study establishes and validates a robust instrumental framework, which will now be employed for longitudinal exploration of the SARS-CoV-2 herd immunity evolution from the pandemic period into the present era.

It should be noted that cell-based neutralization assays require a sufficiently long exposure time. Therefore, the ability to assess neutralization against multiple viral variants simultaneously significantly conserves research time, reduces the consumption of reagents, and optimizes the operational time of laboratory equipment. The selection of flow cytometry for the test readout was largely determined by the institute’s laboratory capabilities. Additionally, the implementation of a 96-well autosampler for the cytometer has markedly enhanced the overall productivity of the assay. Nevertheless, the multiplexed format offers significant advantages, as it not only reduces processing time but also enables the analysis of samples with limited volumes.

The findings of this study should be interpreted in light of several limitations. First, with regard to assay multiplexing capacity, the present system enables simultaneous neutralization assessment against three SARS-CoV-2 variants. Although this represents an advancement over conventional singleplex formats, recent studies have demonstrated the feasibility of quadruplex assays [34]. It is important to note that the maximum level of multiplexing is inherently constrained by the specifications of the available instrumentation, which may currently preclude the broader adoption of such tests in routine laboratory settings. Second, methodological considerations merit discussion. Flow cytometry, while offering quantitative precision, remains a relatively labor-intensive technique. Although fluorescence-based imaging platforms present certain operational advantages, the selection of a readout modality is ultimately dictated by the laboratory’s existing infrastructure. Furthermore, while all experiments were conducted in duplicate or triplicate and were replicated using independent batches of PVPs, the absence of external validation cohorts or orthogonal assay formats represents a constraint on the generalizability of the findings. Third, limitations inherent to the experimental system should be acknowledged. The sample size, despite its longitudinal design, remains modest and may limit the extrapolation of the results to broader populations. Additionally, although pseudovirus-based neutralization assays are widely employed due to their favorable biosafety profile and scalability, they do not fully recapitulate the biological complexity of authentic viral infection. Unlike live virus assays, pseudovirus systems lack other viral structural proteins and do not capture post-entry replication events. Nevertheless, such assays remain a well-validated and broadly accepted surrogate for the assessment of neutralizing antibody responses.