Influenza neutralizing antibodies correlate with protection of influenza infections. Microneutralization assays measure neutralizing antibodies in human sera and are often used for influenza human serology. We describe a microneutralization assay using MDCK-SIAT1 cells to measure neutralizing antibody titers to contemporary 3C.2a and 3C.3a A(H3N2) viruses following influenza vaccination or infection.
Neutralizing antibodies against hemagglutinin (HA) of influenza viruses are considered the main immune mechanism that correlates with protection for influenza infections. Microneutralization (MN) assays are often used to measure neutralizing antibody responses in human sera after influenza vaccination or infection. Madine Darby Canine Kidney (MDCK) cells are the commonly used cell substrate for MN assays. However, currently circulating 3C.2a and 3C.3a A(H3N2) influenza viruses have acquired altered receptor binding specificity. The MDCK-SIAT1 cell line with increased α-2,6 sialic galactose moieties on the surface has proven to provide improved infectivity and more faithful replications than conventional MDCK cells for these contemporary A(H3N2) viruses. Here, we describe a MN assay using MDCK-SIAT1 cells that has been optimized to quantify neutralizing antibody titers to these contemporary A(H3N2) viruses. In this protocol, heat inactivated sera containing neutralizing antibodies are first serially diluted, then incubated with 100 TCID50/well of influenza A(H3N2) viruses to allow antibodies in the sera to bind to the viruses. MDCK-SIAT1 cells are then added to the virus-antibody mixture, and incubated for 18 – 20 h at 37 °C, 5% CO2 to allow A(H3N2) viruses to infect MDCK-SIAT1 cells. After overnight incubation, plates are fixed and the amount of virus in each well is quantified by an enzyme-linked immunosorbent assay (ELISA) using anti-influenza A nucleoprotein (NP) monoclonal antibodies. Neutralizing antibody titer is defined as the reciprocal of the highest serum dilution that provides ≥50% inhibition of virus infectivity.
Influenza viruses continue to cause morbidity and mortality in the human population each year. HA is the major surface glycoprotein of influenza viruses. Neutralizing antibodies targeting HA are the main immune mechanism that correlates with protection of influenza infection1,2. Hemagglutination inhibition (HI) assays and MN assays are two methods widely used to measure antibody responses in human sera after influenza infection or vaccination3. The HI assay measures antibody inhibition of virus hemagglutination of red blood cells and is considered a surrogate assay. Unlike HI, the MN assay can directly measure the levels of antibodies in human sera that neutralize influenza infection in cell cultures. MDCK cells are often used in influenza virus isolation and MN assays4.
Influenza viruses constantly undergo antigenic drift and shift, acquiring mutations on HA proteins that can alter the receptor binding specificity of viruses. Since 2014, new clusters of A(H3N2) viruses emerged and continued to circulate until the current season. The majority of these viruses belong to the genetic group 3C.2a and 3C.3a based on phylogenetic analysis of the HA proteins. Many of the circulating 3C.2a viruses had reduced ability to hemagglutinate red blood cells and therefore cannot be characterized by HI assays5. Neutralization assays must be used to measure the antibody responses to these viruses that do not hemagglutinate6. Furthermore, studies have shown that these contemporary A(H3N2) viruses have altered receptor binding properties compared with earlier A(H3N2) viruses and tend to accumulate culture adapted mutations and polymorphism when passaged in in vitro cell cultures7,8,9. Compared with conventional MDCK cells, MDCK-SIAT1 is a cell line developed by Matrosovich et al. through stable transfection of MDCK cells with cDNA of human α2,6-sialtransferase (SIAT1). This cell line expresses increased amounts of α2,6-sialic galactose moieties and decreased amounts of α2,3-sialic acid moieties than the parent MDCK cells10. MDCK-SIAT1 cells have shown to improve isolation rates for A(H3N2) viruses compared to MDCK cells11. Recently, Lin et al. reported that for newly emerged 3C.2a and 3C.3a human A(H3N2) influenza viruses, more faithful virus replications and better virus infectivity were achieved when viruses were cultured in MDCK-SIAT1 cell lines compared with MDCK cells7. Thus, the MDCK-SIAT1 cells are better suited in MN assays to characterize antibody responses to the recent clusters of A(H3N2) viruses.
Here we describe a MN assay using MDCK-SIAT1 cells to measure antibody responses to contemporary 3C.2a and 3C.3a A(H3N2) viruses in human sera. Viruses grown in either eggs or cells can be used in this assay. Heat inactivated sera containing neutralizing antibodies are first serially diluted, then incubated with 100 TCID50/well of influenza A(H3N2) virus to allow antibodies in the sera to bind to the virus. MDCK-SIAT1 cells are then added to the virus-antibody mixture, and incubated for 18 – 20 h at 37 °C, 5% CO2 to allow the A(H3N2) virus to infect MDCK-SIAT1 cells and replicate. After overnight incubation, the plates are fixed and the amount of virus in each well is quantified by an ELISA using anti-influenza A NP monoclonal antibodies. The detection of NP indicates the presence of virus infection and the absence of neutralizing antibodies. Neutralizing antibody titer is defined as the reciprocal of the highest serum dilution that provides ≥ 50% inhibition of virus infectivity.
All influenza viruses should be handled according to appropriate biosafety level requirements (BSL-2 or higher) as defined in the Biosafety on Microbiological and Biomedical Laboratories (BMBL)12.
1. Preparation of Reagents and Starting Material
2. Passage of MDCK-SIAT1 Cell Culture
Note: All cell cultures must be performed in a biological safety cabinet to prevent contamination.
3. Propagation of A(H3N2) Viruses in MDCK-SIAT1 Cells
NOTE: A(H3N2) virus can be propagated either in 10 – 11 day old embryonated hen's eggs according to standard protocol3, or MDCK-SIAT1 cell cultures. MDCK-SIAT1 cells should reach over 100% confluency for virus inoculation. Virus seed stocks can be inoculated with multiple dilutions of inoculum. The inoculum dilutions with the best harvest HA and infectivity can be used for further MN assays.
4. Determination of TCID of the Virus
5. MN Assay Using MDCK-SIAT1 Cells
Determination of the infectivity of the virus stocks is the first step in the MN assay. Figure 2 illustrates the plate layout to determine the TCID50 of the virus stocks. For virus stocks with unknown infectivity, the virus can be titrated from multiple pre-dilutions, for example both 10-2 and 10-3, in order to capture the best titration curve to calculate infectivity of the virus. The virus amount used in the MN assay should be standardized to 100 TCID50/well (50 µL). The minimum dilution of virus stocks to achieve a 100 TCID50/well is 1:100.
Figure 3 describes the critical steps of a MN assay using MDCK-SIAT1 cells. Heat-inactivated sera are serial-diluted in 96-well plates as illustrated in Figure 4. 100 TCID50/50 µL virus is then added to each well. After 1 h incubation to allow the antibody in the sera bind to the virus, 100 µL of 1.5 x 105 cells/mL of MDCK-SIAT1 cells are added to each well. For optimal results, MDCK-SIAT1 cells should be at 75 – 95% confluency (Figure 1) for both the TCID50 and MN assays. The plates are then incubated for 18 – 20 h at 37 °C, 5% CO2. After overnight incubation, the amount of virus in each well is detected and quantified by an anti-influenza A NP ELISA (Figure 3).
The OD in each well represents the amount of virus infection and replication in MDCK-SIAT1 cells in the presence of serially diluted sera containing neutralizing antibodies. It can be plotted against the sera dilutions (Figure 5). The reciprocal of the highest serum dilution that achieved ≥50% neutralization is considered the antibody titer for the serum sample. Figure 5 contains examples of results from 2 patients with paired sera tested against A/HongKong/4801/2014 A(H3N2) 3C.2a virus pre- and post-influenza vaccination. With patient 1 pre-vaccination sera, none of the sera dilutions inhibited the virus infection (Figure 5, light blue curve) and therefore is considered negative (<10). In contrast, post-vaccination sera from this patient inhibited virus replication, indicating vaccination induced neutralizing antibodies. The third dilution of this sera is the highest dilution below the 50% cut-off, thus this patient has a post-vaccination titer of 40 (Figure 5, dark blue curve). In comparison, the serum from the second patient pre-vaccination contains pre-existing neutralizing antibodies at a titer of 320. Following vaccination, the titer increased to >1,280. In this case, at a 1:10 pre-dilution of serum, no antibody end titer was achieved. The serum should be re-tested at a higher pre-dilution in order to achieve end titer.
Figure 1. MDCK-SIAT1 cell culture. MDCK-SIAT1 cells. (A) MDCK-SIAT1 confluent monolayer (>100% confluency) for virus inoculation; (B) MDCK-SIAT1 cells at log phase with 75 – 95% confluency for TCID50 and MN assays.
Figure 2. TCID plate layout. Virus infectivity is determined by TCID50. Virus stock is pre-diluted to 10-2, and then serially diluted at a ½ log per dilution to 10-7 (column 1 – 11), at 8 replicates per dilution (row A -H). Column 12 is used as the cell only control. TCID50 of the virus stock can be calculated using the Reed-Muench method.
Figure 3. Schematic of the MN assay using MDCK-SIAT1 cells. 50 µL of heat-inactivated sera are serially 2-fold diluted into 96-wells plates. 50 µL of 100 TCID50 of A(H3N2) virus are then added to each well. Plates are incubated at 37 °C for 1 h to allow the binding of virus and antibody. Then 1.5 x 104 MDCK-SIAT1 cells are added to each well. Plates are incubated for 18 – 20 h at 37 °C, 5% CO2. After overnight incubation, plates are washed and fixed, the amount of virus in each well is quantified by an ELISA using anti-influenza A NP monoclonal antibodies. Sera antibody titers are calculated based on cut-offs defined by virus controls and cell controls. Please click here to view a larger version of this figure.
Figure 4. The MN assay plate layout. Heat-inactivated sera are pre-diluted at 1:10, then serially 2-fold diluted in 96-well plates in column 1 – 10. Virus controls (virus and cells only, no sera) and cell controls (cells only, no virus and no sera) are in column 12. Column 11 is used for back titration or control sera. Please click here to view a larger version of this figure.
Figure 5. The MN assay results of human sera tested against A/HongKong/4801/2014 A(H3N2) virus using MDCK-SIAT1 cells. Sera from two patients pre- and post-2016-17 seasonal influenza vaccination were tested against A/HongKong/4801/2014 A(H3N2) virus using MDCK-SIAT1 cells. Neutralization titers are the reciprocal of the highest serum dilution that achieves ≥50% virus neutralization, as indicated by arrows on each curve. Patient 1: pre-vaccination titer <10, post-vaccination titer 40; patient 2:pre-vaccination titer 320, post-vaccination titer >1,280.
The MN assay is one of the main assays used for influenza serology to detect antibody responses following influenza infection or vaccination. Titers generated from MN assays are often used as the primary outcome of many influenza seroepidemiology studies. MN assays are also widely used for sero-diagnosis, and the evaluation of vaccine immunogenicity. International inter-lab studies have been conducted to compare MN assays performed in multiple laboratories14.
In contrast to HI, the MN assay is designed to directly measure functional antibodies that can neutralize virus infection in cell culture. There are various forms of MN assays used in field laboratories around the world. The read-out of the assays (i.e., reduction of virus infectivity in the presence of neutralizing antibodies) can be based on ELISA quantification of virus NP antibodies, HA quantification of virus, or immuno-staining of virus plaques in each well following virus infection in the presence of antibodies and incubation in cell cultures14,15. Various forms of fluorescent-based plaque reduction neutralization assays (e.g., focus reduction assays and ViroSpot assays) are often used for antigenic characterization of large numbers of influenza virus field isolates, partly due to the low virus amount required in these assays15,16. For the characterization of antibody responses to influenza in human serology, the 2-day ELISA based MN assay is recommended by World Health Organization (WHO) global influenza surveillance network for serological diagnosis of influenza3. This method is also widely used in sero-epidemiology studies and the evaluation of antibody responses following influenza vaccination. It primarily detects antibodies to influenza HA surface protein and therefore can detect functional strain-specific antibodies.
Here, we describe a 2-day ELISA-based MN assay that utilizes MDCK-SIAT1 cells. This assay is optimized to detect neutralizing antibodies to contemporary A(H3N2) viruses in human sera. Several critical steps should be considered when performing MN assays using MDCK-SIAT1 cells. First, MDCK-SIAT1 cells should be passaged in culture media containing G418 sulphate to maintain the stability of the human αSIAT1 cDNA in the cell line. G418 sulphate (e.g., geneticin) is not required in the media during virus propagation and MN assays using MDCK-SIAT1 cells. Cells used in MN assays should be at log growth phase with 75 – 95% confluency. Second, good virus stocks are essential in order to carry out successful MN assays. Virus stocks should have high infectivity as measured by TCID50 titration and a minimum amount of defective particles. The minimum dilution of virus stocks to achieve 100 TCID50/well virus should be equal or greater than 1:100. Although both egg and cell passaged viruses can be used in MN assays, MDCK-SIAT1 cell line maintains better genetic stability for propagating 3C.2a and 3C.3a A(H3N2) viruses7. After virus propagation, HA and NA genes of the virus stocks should be sequenced to ensure that no egg or cell culture adapted mutations are introduced at key antigenic sites that may cause changes in the antigenicity of the virus used in the assay. Third, the amount of infectious virus used in each assay should be carefully titrated, and verified with the inclusion of a back titration for each virus in each assay. Too much virus used in the assay may lower the antibody titers detected and reduce the sensitivity of the assay. Likewise, too little virus used in the assay will cause weak VC signals and high backgrounds (CC), and false positive results. Thus, it is important to include appropriate positive and negative control sera in the assays to monitor performance. When comparing antibody responses to multiple viruses using the same sera, it is critical to ensure that back titrations of all viruses are within an acceptable range.
The MN assay using MDCK-SIAT1 cells is sensitive and specific in detecting antibody responses to contemporary 3C.2a and 3C.3a A(H3N2) influenza viruses in human sera, including antigenically drifted A(H3N2) viruses6. This assay has been used to evaluate vaccinated human sera panels to 3C.2a and 3C.3a A(H3N2) viruses following inactivated influenza vaccination5,17,18. However, as HAs of influenza viruses continue to acquire mutations that can alter antigenicity and receptor binding properties of viruses and cause antigenic drift, continuous efforts are needed to optimize existing influenza serology assays in order to maintain the sensitivity and specificity required to characterize new emerging influenza viruses.
The authors have nothing to disclose.
We thank Dr. Xiuhua Lu, Dr. Feng Liu, and Ms. Ashley Burroughs from the Influenza Division of the CDC for their critical review and assistance in preparation of this manuscript. We thank Dr. Adrian Reber from Influenza Division of the CDC for his assistance in preparing the graphics of Figure 3. Lastly, we thank Dr. M. Matrosovich, Marburg, Germany for providing the MDCK-SIAT1 cells.
Dulbecco’s Modified Eagle Medium (DMEM) with high Glucose | Life Science | 11965 | A critical component of Sterile Cell Culture, Virus Propagation and Virus Diluent Media |
Fetal bovine serum (FBS) | Hyclone | SH30070.03 | |
Bovine Serum Albumin (BSA) Fraction V, Protase Free | Sigma-Aldrich | 3117332001 | |
L-Glutamine | Life Science | 25030-081 | |
Sodium pyruvate | Life Science | 11360-070 | |
Geneticin G-418 disulfate salt | Sigma-Aldrich | A1720-5G | |
HEPES | Life Science | 15630-080 | |
Penicillin/Streptomycin | Life Science | 15140-122 | |
Acetone | VWR | 67-64-1 | Used at an 80% concentration |
Phosphate-Citrate Buffer with Sodium Perborate | Sigma-Aldrich | SLBF2806V | |
O-Phenylenediamine Dihydrochloride tablet | Sigma-Aldrich | SLBQ1086V | 1 tablet per 100ml of cell culture grade water |
Sulfuric Acid | Fisher Scientific | A510-P500 | Used 0.5M final concentration |
Ethanol, Denatured, 4L | VWR | EM-AX0422-3 | Used at an 70% concentration |
Trypsin-EDTA | Life Science | 1748048 | |
RDE II "Seiken" | Denka Seiken | 370013 | |
Tween 20 | Sigma-Aldrich | P1379-500ml | |
Anti-NP mouse monoclonal Ab | Millipore pool | MAB 8257 MAB 8258 | |
Anti-mouse IgG HRP | KPL | 074-1802 | |
96-well flat-bottom plates | Thermo Scientific | 3455 | |
Plate reader | Molecular Device | Spectromax 384 plus | |
Cell Culture Flask 162 cm2/Vent Cap | Corning/VWR | 3151 |