We describe the use of a phenotypic fluorescence-based neuraminidase inhibition assay to assess the susceptibility of influenza A and B viruses to the neuraminidase inhibitor class of antivirals.
The neuraminidase (NA) inhibitors are the only class of antivirals approved for the treatment and prophylaxis of influenza that are effective against currently circulating strains. In addition to their use in treating seasonal influenza, the NA inhibitors have been stockpiled by a number of countries for use in the event of a pandemic. It is therefore important to monitor the susceptibility of circulating influenza viruses to this class of antivirals. There are different types of assays that can be used to assess the susceptibility of influenza viruses to the NA inhibitors, but the enzyme inhibition assays using either a fluorescent substrate or a chemiluminescent substrate are the most widely used and recommended. This protocol describes the use of a fluorescence-based assay to assess influenza virus susceptibility to NA inhibitors. The assay is based on the NA enzyme cleaving the 2′-(4-Methylumbelliferyl)-α-D-N-acetylneuraminic acid (MUNANA) substrate to release the fluorescent product 4-methylumbelliferone (4-MU). Therefore, the inhibitory effect of an NA inhibitor on the influenza virus NA is determined based on the concentration of the NA inhibitor that is required to reduce 50% of the NA activity, given as an IC50 value.
Haemagglutinin (HA) and neuraminidase (NA) are the two major surface glycoproteins of influenza A and B viruses. HA binds to the sialic acid-galactose of cell surface glycoproteins or glycolipids, while the NA releases virus by cleaving the sialic acid from the galactose on the cell surface1. The NA inhibitors are a class of influenza antivirals that were rationally designed to bind tightly to the NA enzymatic active site, thereby preventing the release and spread of virus progeny. Oseltamivir and zanamivir are two NA inhibitors that have been approved in many countries worldwide for the treatment and prophylaxis of influenza. In recent years, two additional NA inhibitors, peramivir and laninamivir, have been approved for use in a limited number of countries. Screening of influenza viruses for susceptibility to NA inhibitors and the identification of mutations that confer resistance are important in determining and monitoring the effectiveness of this class of antivirals.
In the last 16 years, the fluorescence-based NA inhibition assay has been performed routinely at the WHO Collaborating Centre for Reference and Research on Influenza, Melbourne (Melbourne WHOCCRRI) to monitor the changing trend of antiviral susceptibility among circulating influenza viruses. Annually, more than 2,000 influenza viruses are tested for antiviral susceptibility. In most influenza seasons, >98% of the viruses are susceptible to all four NA inhibitors2,3,4, although during the 2007-2008 northern hemisphere influenza season, there was a surge in the number of former seasonal A(H1N1) viruses that had reduced susceptibility to oseltamivir5. This group of viruses, which contained the NA amino acid substitution H275Y, spread to the rest of the world by the end of 2008, making oseltamivir inappropriate globally for the treatment of this virus. The vast majority of currently circulating influenza B, influenza A(H3N2), and influenza A(H1N1)pdm09 strains are susceptible to oseltamivir, although community clusters of A(H1N1)pdm09 variants containing the NA amino acid substitution H275Y that confers reduced oseltamivir and peramivir susceptibility, have been reported in various parts of the world6,7.
Because of the need for a sufficiently high virus titer, clinical specimens (including animal nasal washes) must be passaged in either cell culture or embryonated chicken eggs prior to antiviral susceptibility testing. The NA inhibition assay described in this article can be divided into three sections:
Determining the linear range for the fluorescent product 4-methylumbelliferone (4-MU) on a particular fluorometer
Due to the inherent differences between fluorometers, the linear range for the fluorescent end-product, 4-MU, and the relative fluorescence unit (RFU), need to be established. Once the linear range for 4-MU is established, the optimal target signal is selected, to which the concentration of the influenza viruses is adjusted in the NA activity assay. Once completed for a particular fluorometer, this should not need to be repeated.
Determining the NA activity of the viruses
The NA activity assay is a simple assay that involves the addition of the 2′-(4-Methylumbelliferyl)-α-D-N-acetylneuraminic acid (MUNANA) substrate to serially diluted viruses. The amount of fluorescent end-product 4-MU generated from the cleavage of MUNANA by the NA is measured using a fluorometer. The appropriate virus dilution to use in the NA inhibition assay is selected by plotting fluorescence units against virus dilution. From the sigmoidal curve produced, the mid-point of the linear section should correspond to the 4-MU linear range of the fluorometer determined in section 1 and will inform the appropriate concentration of viruses to be used in section 3.
Assessing virus susceptibility to NA inhibitors using the NA inhibition assay
To assess the susceptibility of viruses to a particular NA inhibitor, viruses at the dilution determined in section 2 are incubated with a range of NA inhibitor concentrations. Following a subsequent incubation with MUNANA, the 4-MU generated by uninhibited viruses is measured in RFU by the fluorometer. The inhibitory effect of the NA inhibitor on the NA enzyme activity of a virus is calculated according to the NA inhibitor concentration required to reduce 50% of the NA activity, given as an IC50 value.
1. Determining the Linear Range of Fluorescent Product 4-MU on a Fluorometer
2. Determining the NA Activity of the Viruses
NOTE: Influenza viruses are cultured to sufficient titers in Madin-Darby Canine Kidney (MDCK) cells or embryonated chicken eggs8.
3. Assessing Virus Susceptibility to NA Inhibitors Using the NA Inhibition Assay
4. Calculation of IC50 Values
NOTE: The JASPR v1.2 is curve-fitting software that enables the calculation of IC50 values. The software was developed by the Influenza Division at the CDC, Atlanta, USA. The software utilizes the equation: V=Vmax x(1-([I]/(Ki +[I]))), where Vmax is the maximum rate of metabolism, [I] is the inhibitor concentration, V is the response being inhibited, and Ki is the IC50 for the inhibition curve.
Using standardized reporting guidelines from the WHO Working Group on Surveillance of Influenza Antiviral Susceptibility9, the susceptibility of influenza viruses to the NA inhibitors are reported using the terms normal inhibition (NI), reduced inhibition (RI), and highly reduced inhibition (HRI). NI viruses are those with IC50 values less than 10-fold compared to the reference median IC50 for influenza A viruses (or less than 5-fold for influenza B viruses). RI viruses are those with IC50 values between 10- and 100-fold above the reference median IC50 for influenza A viruses (or 5- and 50-fold for influenza B). HRI viruses are those with IC50 values of 100-fold above the reference median IC50 for influenza A viruses (or above 50-fold for influenza B viruses); see Table 2.
The reference median IC50 values for A(H1N1)pdm09, A(H3N2), and B Yamagata/B Victoria viruses are calculated and updated annually at the Melbourne WHOCCRRI to reflect minor changes in the IC50 values of circulating influenza strains to the NA inhibitors (Table 3). The median IC50 values in influenza A(H1N1)pdm09 viruses are almost the same across the four NA inhibitors, but the median zanamivir and laninamivir IC50 values for A(H3N2) viruses are 2- to 4-fold higher compared to oseltamivir and peramivir IC50 values (Table 3). The median oseltamivir IC50 value for influenza B viruses is generally 5- to 10-fold higher than the zanamivir, peramivir, and laninamivir IC50 values (Table 3).
The NA inhibition assay is a phenotypic assay that does not provide information on the genetic changes associated with RI or HRI. Therefore, it is important that genetic analysis is performed following the identification of viruses with RI or HRI. At the Melbourne WHOCCRRI, the NA gene of variants is analyzed using Sanger sequencing and pyro-sequencing. A representative list of amino acid substitutions that can be found in the NA gene of viruses with RI and HRI variants is presented in Table 4. A more extensive list of amino acid substitutions that can alter NAI susceptibility is also available on the WHO website10.
Figure 1: RFU against 4-MU concentration. (a) Standard curve of RFU against 4-MU concentration (µM). The dotted box shows the linear range of 4-MU for the fluorometer. The fluorescence signals above the linear range may be saturated, and therefore, any small changes in fluorescence may not be detected by the fluorometer. (b) Close-up linear section of the standard curve of Figure 1a for the identification of the "optimal target signal." The fluorometer at the Melbourne WHOCCRRI has a linear range of 2.5-40 µM 4-MU and an optimal target signal of ~30 µM 4-MU, which corresponds to ~1,500 RFU. Please click here to view a larger version of this figure.
Figure 2: Example of the NA activity curves of influenza viruses. The average background value of 50.61 RFU has been subtracted from every dilution point on the NA activity curves. The arrows indicate the appropriate virus dilution to use in the NA inhibition assay for each virus. For easier preparation of virus dilutions, one may choose to perform a 1/100 dilution for VIRUS 3 instead of a 1/96 dilution. Please click here to view a larger version of this figure.
Figure 3: Plate layout for the setup of the NA inhibition assay. Each plate includes the last column, which acts as a negative control that contains no virus but only 1x assay buffer (AB), NA inhibitor, MUNANA, and stop solution. NOTE: JASPR uses readings from column 12 of each plate to determine the average blank signal used in the calculation of the IC50 values. Please click here to view a larger version of this figure.
Figure 4: Example of an inhibition curve and IC50 value of an A(H1N1)pdm09 virus, A/Perth/82/2015. The JASPR software presents the inhibition curve as fluorescence (RFU) against the increasing concentration (nM) of NA inhibitor, with every point fit within the curve. Based on the inhibition curve, the IC50 value is determined as the concentration of NA inhibitor to reduce 50% of the virus NA activity. Please click here to view a larger version of this figure.
Virus | Virus dilution required | 1x assay buffer volume (µL) | Surfactant-Amps-NP-40 (10%) (mL) | Virus volume (mL) |
1 | 1/20 | 940 | 10 | 50 |
2 | 1/40 | 965 | 10 | 25 |
3 | 1/100 | 890 | 10 | 10 |
Table 1: Preparation of virus dilutions for VIRUS 1, 2, and 3 in the NA inhibition assay.
Virus type/subtype/lineage | Normal inhibition | Reduced inhibition | Highly reduced inhibition |
(NI) | (RI) | (HRI) | |
A(H1N1)pdm09 | <10 fold | 10-100 fold | >100 fold |
A(H3N2) | <10 fold | 10-100 fold | >100 fold |
B Yamagata and B Victoria | <5 fold | 5-50 fold | >50 fold |
Table 2: The WHO Antiviral Working Group recommended guidelines for the classification of influenza virus susceptibility to NA inhibitors.
Virus type/subtype/lineage | N | Zanamivir | Oseltamivir | Peramivir | Laninamivir |
Median (range) IC50 nM | Median (range) IC50 nM | Median (range) IC50 nM | Median (range) IC50 nM | ||
A(H1N1)pdm09 | 1,326 | 0.42 (0.1-3.43) | 0.36 (0.01-3.48) | 0.19 (0.07-1.60) | 0.55 (0.05-2.29) |
A(H3N2) | 1,654 | 0.9 (0.11-4.0) | 0.38 (0.01-3.65) | 0.33 (0.12-3.06) | 1.38 (0.01-9.38) |
B Yamagata and B Victoria | 1,115 | 2.2 (1.24-10.72) | 15.12 (2.39-70.75) | 1.36 (0.57-6.67) | 2.89 (1.62-9.15) |
Table 3: Median IC50 and IC50 range of normal inhibition (NI) viruses from 2015 derived in the WHO CCRRI, Melbourne.
Amino acid substitution | Type/subtype/lineage | IC50 fold-change compared to reference median IC50 values. | |||
Zanamivir | Oseltamivir | Peramivir | Laninamivir | ||
H275Y | A(H1N1)pdm09 | 1 | 557 (HRI) | 123 (HRI) | 2 |
E119V | A(H3N2) | 1 | 63 (RI) | 1 | 1 |
H134Y | B Victoria | 1 | 4 | 76 (HRI) | 2 |
N151T | B Victoria | 4 | 4 | 42 (HRI) | 1 |
G104E | B Victoria | 1,220 (HRI) | 87 (HRI) | 17,724 (HRI) | 701 (HRI) |
E105K | B Victoria | 3 | 5 (RI) | 59 (HRI) | 2 |
I222T | B Victoria | 2 | 7 (RI) | 8 (RI) | 3 |
H273Y | B Yamagata | 1 | 230 (HRI) | 377 (HRI) | 2 |
D197N | B Yamagata | 4 | 7 (RI) | 32 (RI) | 3 |
Table 4: Representative list of amino acid substitutions linked to reduced inhibition (RI) or highly reduced inhibition (HRI) to NA inhibitors.
Problem | Possible reason(s) | Solution(s) |
No or low NA activity | No virus was present or low virus yield. | Clinical specimen must be cultured in cell lines (i.e. Madin-Darby Canine Kidney cells) or in embryonated chicken eggs to a higher virus load for use in the NA inhibition assay. |
Some mutant viruses have extremely low NA activity despite at high virus load. | Use neat virus concentration for testing. Lower pH assay buffer (e.g. pH 5.3) may be used. However, caution must be taken when comparing data. | |
No or low NA activity in NA inhibition assay | No virus was added. | Re-dilute the virus. Ensure the virus is directly added into the 1x assay buffer. |
Wrong virus dilution was used. | Repeat the NA activity assay. | |
Insufficient incubation time. | Ensure the incubation time is followed. | |
Data points fall outside the IC50 curve | Cross contamination of NA inhibitor of higher concentration. | Ensure that the tips are not in contact with the NA inhibitor when dispensing diluted viruses into the 96 well plate. |
If an 8 deep well reservoir was used, discard and re-dispense the NA inhibitor concentrations into a fresh 8 deep well reservoir. | ||
The volume of the NA inhibitor or MUNANA or diluted virus was not added equally into each well. | Repeat the assay with a calibrated multi-channel pipette. Ensure equal volume of each reagent is dispensed into each well. | |
Unusually high IC50 values | Too high concentration of virus was added. | Repeat the NA activity assay and NA inhibition assay. |
Test sample contained mixtures of influenza A and influenza B. | Perform real-time PCR to identify the presence of virus mixtures. | |
Bacterial contamination in the sample | Culture virus in the sterile condition with the presence of antibiotic. | |
High background fluorescence signal | MUNANA substrate may degrade over time. | Use a new batch of MUNANA subtrate. |
Detection of fluorescence from neighboring wells. | Use black 96-well flat-bottom plates |
Table 5: Troubleshooting for potential problems in the NA inhibition assay.
The global monitoring of influenza virus susceptibility to NA inhibitors is currently being conducted by a number of laboratories using either fluorescent or chemiluminescent NA inhibition assays11,12. The fluorescent assay is more commonly used than the chemiluminescent assay. Although both assays are robust and reproducible, the IC50 values obtained from the fluorescence-based assay are often higher than the chemiluminescence-based assay, making a direct comparison of the data from the two assays difficult13. Even with the use of the same protocol, data generated from one laboratory may vary from another. Because of these variations between laboratories, the WHO Working Group on Surveillance of Influenza Antiviral Susceptibility produced a guideline to assist in inter-laboratory comparisons. Rather than comparing the absolute IC50 values, this guideline uses a comparison based on the IC50 fold difference to the median IC50 of the NI influenza viruses tested in each particular laboratory. The ability to compare data from the five collaborating centers has resulted in the annual publication of global influenza antiviral susceptibility data2,3,4. The availability of the large amount of influenza susceptibility data in the public domain allows researchers to compare IC50 data from those studies with that generated in their own laboratories.
Other NA inhibition assays that adopt a similar concept are also commercially available. These commercial kits that contain ready-to-use reagents (NA inhibitors not included) are equally reproducible. However, the in-house NA inhibition assay is substantially cheaper than the commercial kits, because the majority of the reagents can be made in-house in larger quantities and the MUNANA substrate, which previously made up the major cost of the assay, can now be purchased from various sources at competitive prices. The cost of testing one influenza isolate per drug is approximately $1 (USD). At the Melbourne WHOCCRRI, improvements have been made to the in-house NA inhibition assay after the incorporation of a robotic platform for the liquid handling components of the assay. Apart from the manual preparation of virus dilutions, the majority of the procedures are performed using the liquid-handling robot. Not only does this minimize manual handling, but it also increases the numbers of assays that can be run in a day.
Although the NA inhibition assay is highly robust, there are a number of critical steps that need to be completed with additional care. First, any irregularity in the NA inhibitor concentrations can shift the inhibition curves and the IC50 values; therefore, careful attention should be paid when preparing the NA inhibitor concentrations. Second, accurate pipetting and precise incubation periods are crucial to maintaining consistent results across assays; this can be achieved by using calibrated pipettes and timers. The inclusion of control viruses in every assay also enables the monitoring of assay performance from assay to assay and over long periods of time. Third, because the NA enzyme activity of seasonal influenza viruses is optimal at pH 6.5, the correct pH of the assay buffer is important. Some reports have found that the use of lower pH conditions may improves the identification of influenza variants, such as the A(H7N9) variant containing the R292K mutation14,15. However, the modification to the pH of the assay buffer will shift the IC50 values, and this may complicate the comparison of data within laboratories and between laboratories. Other modifications and troubleshooting that can be performed are listed in Table 5.
The NA inhibitors are the only class of approved antivirals that are currently effective against circulating influenza viruses. Until other antiviral classes become available for clinical use, the antiviral susceptibility surveillance of circulating influenza viruses will be focused on NA inhibitors alone. Because of the simplicity and reproducibility of results, the use of the NA inhibition assay to assess influenza virus susceptibility to NA inhibitors will continue.
The authors have nothing to disclose.
The Melbourne WHO Collaborating Centre for Reference and Research on Influenza is supported by the Australian Government Department of Health.
Influenza A and B viruses | Cultured in MDCK cells or 9 day old embryonated specific pathogen free (SPF) eggs | ||
Madin-Darby Canine Kidney (MDCK) cells | ATCC | PTA-6500 | |
2-(4-methylumbelliferyl)-a-D-N-acetylneuraminic acid (MUNANA) | Biosynth AG | M-5507 | |
2-(4-methylumbelliferyl)-a-D-N-acetylneuraminic acid (MUNANA) | Sigma | M8639 | |
4-Methylumbelliferone (4-MU) | Sigma | M1381-25G | |
2-[N-morpholino]ethanesulphonic acid (MES hydrate) (free acid) | Sigma | M8250-250G | |
Calcium Chloride (Ca Cl2) | APS AJAX Finechem | 127-500G | |
Surfactant-Amps-NP-40 (10% solution) | Thermo Fisher Scientific | PIE28324 | |
Sodium Hydroxide (NaOH) | APS AJAX Finechem | 482-2.5KG | |
Absolute Ethanol | APS AJAX Finechem | 214-2.5L GL | |
96-well clear flat-bottom plates | NUNC | 456537 | |
96-well U-bottom plates | Greiner Bio-one | 4650101 | |
8 channel deep well block | Pacific Laboratory Products | RES-MW8-HP | |
96-well deep plates, 2.0mL square wells | Pacific Laboratory Products | P-2ML-SQ-C | |
Plate sealers | Thermo Fisher Scientific | 236366 | |
Bottle-top vacuum filter system (cellulose membrane (nitrate), pore size 0.2 μm, membrane area 33.2 cm2, filter capacity 500 mL) | Sigma-Aldrich | CLS430758-12EA | |
Single-channel pipettes (1 µL – 1000 µL) | Variety of suppliers (eg. Eppendorf, Sartorius) | ||
Multi-channel pipettes | Variety of suppliers (eg. Eppendorf, Sartorius) | 8 or 12 channel electronic and manual pipette (5 – 1250 µL volume) | |
Pipette tips (1 µL – 1250 µL) | Variety of suppliers (eg. Eppendorf, Sartorius) | ||
Disposable pipettes (10 mL and 25 mL) | Greiner Bio-one | P7740-200EA and P7865-200EA | |
Pipette controller | Eppendorf | 4430000018 | |
Centrifuge tubes 50 mL | BD Bioscience | 352070 | |
Racked tubes | Scientific Specialties, Inc. | 1750-00 | |
Fluorometer with excitation wavelength setting of 355 nm and an emission wavelength setting of 460 nm | TermoFisher Scientific | ASCENT FL 374 | |
Ascent software | TermoFisher Scientific | 5185410CD | |
Incubator set at 37°C | Lab Supply | Biocell 1000 | |
Zanamivir | GlaxoSmithKline | Request directly from the company | |
Oseltamivir carboxylate | Roche | ||
Peramivir (BCX-1812) | BioCryst | ||
Laninamivir (R-125489) | Daiichi-Sankyo | ||
JASPR v1.2 | Influenza Division at the CDC Atlanta, USA | freely available upon request (fluantiviral@cdc.gov) |