- Email: [email protected]
Comparison of hemagglutination inhibition assay, an ELISA-based micro-neutralization assay and colorimetric microneutralization assay to detect antibody responses to vaccination against influenza A H1N1 2009 virus
Journal of Virological Methods 171 (2011) 369–373
Contents lists available at ScienceDirect
Journal of Virological Methods journal homepage: www.elsevier.com/locate/jviromet
Protocols
Comparison of hemagglutination inhibition assay, an ELISA-based micro-neutralization assay and colorimetric microneutralization assay to detect antibody responses to vaccination against influenza A H1N1 2009 virus S. Grund a,∗,1 , O. Adams a,1 , S. Wählisch b , B. Schweiger b a b
Institute of Virology of the University of Düsseldorf, Heinrich-Heine-University, D-40225 Düsseldorf, Germany National Reference Centre for Influenza, Robert-Koch-Institute, Berlin, Germany
a b s t r a c t Article history: Received 25 June 2010 Received in revised form 23 November 2010 Accepted 30 November 2010 Available online 10 December 2010 Keywords: Influenza (H1N1) 2009 virus Vaccination Antibody responses Hemagglutination inhibition assay Neutralization assay
The hemagglutination inhibition (HI) assay has been the main method used to investigate immune responses to vaccination against influenza H1N1 (2009) virus. However microneutralization tests (MNT) have been shown to be more sensitive and more specific. In this study, the three methods of choice: (i) the HI assay, (ii) an ELISA-based conventional MNT and (iii) a colorimetric MNT in terms of their ability to detect antibody responses in serum pairs collected from 43 healthy individuals before and 21 days after vaccination were compared. The colorimetric MNT was established yielding intra- and inter-run imprecisions of 7.5% and 12.4%, respectively. Testing of antisera to seasonal influenza viruses demonstrated the assay to be specific for antibodies to influenza H1N1 (2009) virus. A good correlation between the three methods was found, being highest for the ELISA-MNT and the colorimetric MNT (r = 0.714 for geometric mean titers (GMT) and r = 0.695 for titer increases). Similar rates of fourfold titer increases were detected: 95.3% in the ELISA-MNT vs. 93.0% in colorimetric MNT and 95.3% in HI assay. The ELISA-based MNT demonstrated the highest titer range leading to the highest postvaccination GMT and the highest titer increase (>50-fold). The lowest GMTs were measured with the HI assay, while the colormetric MNT detected the highest GMT in prevaccination sera. Taken together, similar seroconversion rates were obtained with the three assays. The ELISA-MNT appeared to be the best method to compare absolute pre- and postvaccination GMTs. The colorimetric MNT, being less labour-intensive than the ELISA-MNT, seems to be a suitable tool in vaccination studies. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Clinical trials have investigated immune responses to the monovalent split-virus inactivated vaccine against influenza H1N1 (2009) virus (Liang et al., 2010; Plennevaux et al., 2010; Zhu et al., 2009). In these studies antibody responses were measured mainly by the hemagglutination inhibition (HI) assay because this method can be easily adapted to new influenza viruses. However, it has been shown for seasonal and for avian influenza that microneutralization tests (MNT) are more sensitive and more specific (Frank et al., 1980; Gitelman et al., 1986; Rowe et al., 1999). A previous study for seasonal influenza A indicated that strain-specific IgG antibodies with neutralizing capacity may be present, but these antibodies were not generally detected in homologous HI assays (Remarque et al.,
∗ Corresponding author at: Institut für Virologie, Geb. 22.21, Universitätsstr. 1, D-40225 Düsseldorf, Germany. Tel.: +49 211 811 0551; fax: +49 211 811 2227. E-mail address: [email protected] (S. Grund). 1 These authors contributed equally to this work. 0166-0934/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jviromet.2010.11.024
1998). Conventional microneutralization tests are performed in different ways with regard to titer determination, cytopathic effect formation in cell cultures, red blood cells for detection of released virus or usage of the ELISA technique to identify virus infected cells. Especially the ELISA method is time-consuming, difficult to perform under biosafety-level 3 (BSL-3) conditions and requires the need for monoclonal antibodies that are not widely available for new viruses. Recently, a colorimetric MNT has been introduced for seasonal influenza (Lehtoranta et al., 2009). This assay is an adaptation of micro-cytotoxicity tests (van de Water et al., 1993) and is based on the cleavage of a tetrazolium salt by mitochondrial dehydrogenases reflecting the amount of viable and metabolically active cells. In the present study, the three assays of choice: (i) HI assay, (ii) ELISA-MNT and (iii) colorimetric MNT were compared to detect immune responses in serum pairs collected from 43 healthy individuals before and 21 days after administration of a monovalent vaccine against pandemic influenza H1N1 (2009) virus. The colorimetric MNT was adapted and validated to detect antibodies to pandemic influenza and the method was modified to include less hands-on manipulations.
370
S. Grund et al. / Journal of Virological Methods 171 (2011) 369–373
2. Materials and methods
2.5. ELISA-based conventional microneutralization (ELISA-MNT) assay
2.1. Cells MDCK cells (obtained from BioWhittaker Europe, Verviers, Belgium) were grown in Dulbecco’s modified essential medium (DMEM) supplemented with penicillin and streptomycin (500 U/ml each) and 2% fetal calf serum and were maintained in a 5% CO2 humified atmosphere at 37 ◦ C. Two to three times a week cells were subcultured at a split-ratio of 1:2–1:5. 2.2. Viruses For the HI assay and the ELISA-MNT the pandemic influenza virus reference strain A/California/7/2009 H1N1 was used. The colorimetric MNT was performed with a pandemic influenza A H1N1 virus isolated from an infected patient in September 2009. This virus reacted well with a specific antiserum to the influenza virus A/California/7/2009 in HI assay. According to this close relatedness in the antigenic profile the virus was declared to be an influenza A/California/7/2009-like virus by the German national reference institute for influenza at the Robert-Koch-Institute in Berlin. Sequencing of the HA gene (984 nucleotides) yielded seven nucleotide exchanges compared to the reference strain A/California/7/2009 without any exchange in amino acid sequence. In comparison to the reference virus A/California/7/2009 a concentration of 500–1000 TCID50 of this virus demonstrated a stronger cytopathic effect in MDCK cells leading to a cell destruction of roughly 90% after 72 h of incubation. Strong destruction of target cells has been described to be essential for a colorimetric MNT (Lehtoranta et al., 2009). 2.3. Human sera In November 2009, 43 serum pairs were collected from healthy staff members of the Institute of Virology of the University of Düsseldorf before and 21 days after vaccination with an adjuvant monovalent split-virus inactivated vaccine against influenza H1N1 (2009) virus (Pandemrix® , GlaxoSmithKline, London, England). Informed consent was obtained from all individuals participating in the study. The mean age of subjects was 36.2 years (SD 10.4 years). Individuals with a history of infection with influenza H1N1 (2009) virus were excluded from the study.
The method of the ELISA-based microneutralization test was modified from an assay described previously (Rowe et al., 1999). All sera used were heat inactivated at 56 ◦ C for 30 min and twofold serial dilutions were prepared (diluent: Eagle’s minimal essential medium supplemented with 1% BSA and gentamycin 0.05 g/ml). 50 l of each dilution were added to a microtiter plate (flat-bottom, Corning Incorporated, Corning, USA) and mixed with 50 l virus suspension (2 × 103 TCID50 /ml in diluent). The plates included four wells containing 50 l of diluted virus and 50 l diluent as positive controls, four wells containing 100 l diluent as cell control and two wells each for back titration (a serial twofold dilution in diluent, to 1:16). The plates were incubated at 37 ◦ C for 2 h and 5% CO2 . Afterwards 100 l of MDCK-2 cells (1.5 × 105 /ml) were added to each well. After overnight incubation (18–22 h) at 37 ◦ C and 5% CO2 , the medium was removed from the wells and the monolayers were fixed with cold fixative (acetone 80% in PBS 1:5) for 10 min. The titer was calculated by performing an ELISA to detect virus-infected cells. The fixed plates were washed three times with the wash buffer (PBS, 0.1% Tween 20), the anti-influenza A NP mouse monoclonal antibody (WHO Influenza Reagent Kit, Centers for Disease Control and Prevention (CDC), Atlanta, USA) was diluted 1:4000 in blocking buffer (PBS, 1% bovine serum albumin and 0.1% Tween 20) and 100 l were added to each well. After 1 h incubation the plates were washed four times with wash buffer and 100 l of horseradish peroxidase-conjugated goat anti-mouse IgG (diluted 1:1000 in blocking buffer) was added to each well. The plates were incubated at room temperature for 1 h and then washed six times with wash buffer. 100 l of freshly prepared substrate solution (hydrogen peroxide, 1 g/L, in sodium acetate buffer solution, 25 mmol/L [pH 4.1] with tetramethylbenzidine dihydrochloride, 5 g/L) were added to each well and the reaction was stopped after 5 min with 100 l stop solution (0.5 N sulfuric acid) per well. The absorbance was measured at 450 nm (A450 ) with Titertek Multiscan PLUS (Labsystems) and the virus neutralization endpoint titer of each serum was determined with the following equation: X = [(average A450 of virus control wells) − (average A450 of cell control wells)]/2 + (average A450 of cell control wells). X = 50% of specific signal and all values below this value are positive for neutralization activity. All sera were tested twice on different days and the final titer value was the average of two separate runs. 2.6. Colorimetric microneutralization (MNT) assay
2.4. Hemagglutination inhibition (HI) assay The hemagglutination inhibition assay was performed as previously described (Lenette and Schmidt, 1979). Prior to testing, each serum was treated with receptor destroying enzyme (Cholera filtrate, Sigma, Germany) to inactivate non-specific inhibitors achieving a final serum dilution of 1:10. The sera were then diluted serially twofold in V-bottom microtiter plates. The virus was adjusted to 4 HA units/25 l, which was verified by back titration, and 25 l of this virus suspension was added to each of the 96 wells. After incubation at room temperature (RT) for 30 min freshly prepared 0.5% turkey red blood cells (RBCs) were added (obtained from the Bundesinstitut für Risikobewertung, Berlin, Germany). The plates were agitated briefly followed by a further incubation at RT for 30 min. Human sera serving as positive controls and negative controls were included on each plate. The determination of the HI titer was performed by calculation of the reciprocal of the last serum dilution which contained non-agglutinated RBCs. The sera were tested only once in HI assay because there was not sufficient sample material available for further assay runs for the majority of vaccines.
To detect specific antibodies to influenza H1N1 (2009) virus a colorimetric MNT was established on the basis of different methods that had been introduced for seasonal influenza (Lehtoranta et al., 2009; van de Water et al., 1993). Sera were heat inactivated at 56 ◦ C for 30 min, and twofold serial dilutions in 50 l were prepared in flat-bottom 96-well tissue plates (TPP, Trasadingen, Switzerland) with Dulbecco’s modified essential medium (DMEM) supplemented with penicillin and streptomycin (500 U/ml each), 2% fetal calf serum (FCS) and 1% trypsin. An equal volume of the same medium containing 500–1000 TCID50 of the challenge virus was added to each serum dilution, and plates were incubated at 37 ◦ C for 1 h. All sera were tested in duplicate and the average absorption values were used for the calculation of neutralizing titers. Wells containing virus dilutions without serum, MDCK cells only and DMEM only were included on each plate serving as controls and system background, respectively. In each run four control sera yielding: (i) negative, (ii) low, (iii) middle and (iv) high reactive results in the HI assay were included to assess the between-run imprecision of the assay. After incubation 100 l of a suspension containing 3 × 104 /ml freshly trypsinized MDCK cells in DMEM
S. Grund et al. / Journal of Virological Methods 171 (2011) 369–373
371
Table 1 Intra- and inter-run reproducibility of the colorimetric microneutralization assay established to detect antibodies to influenza H1N1 (2009) virus. Control serum
Low reactive Middle reactive asssaareactive High reactive
Intra-assay (8-fold)
Inter-assay (16-fold)
GMT
5 × log2 GMT
5 × log2 SD (mean)
CV (%)
GMT
5 × log2 GMT
5 × log2 SD (mean)
CV (%)
160.0 586.9 1810.2
5.00 6.88 8.50
0.54 0.36 0.54
10.8 5.2 6.4
140.5 472.7 1522.2
4.81 6.56 8.25
0.75 0.73 0.86
15.6 11.1 10.4
Geometric mean titers (GMT), standard deviations (SD) and coefficients of variation (CVs) were calculated from 5 × log2 -transformed data.
2.7. Statistical analysis Statistical analysis was performed using SigmaPlot version 10.0 and Mircosoft Excel 2002 software. Sera negative at a dilution of 1/10 were assigned a titer of 5 and sera positive at a dilution of 1/5120 were assigned a titer of 5120 in each method. A fourfold titer increase between pre- and postvaccination sera was considered significant for both MNTs and the HI assay. Titers obtained for the three methods were analysed as 5 × log2 -transformed data and titer increases as log2 -transformed data. Correlation between methods was calculated by linear regression analysis and either the paired t-test or the Wilcoxon signed rank test was used to investigate the significance of titer increases between pre- and postvaccination sera.
3. Results
3.2. Correlation between the results of the conventional ELISA-based MNT, the colorimetric MNT and the hemagglutination inhibition assay Geometric mean titers (GMT) and titer increases of the three assays are given in Table 2. The ELISA-MNT demonstrated the highest titer increase (>50-fold) leading to the highest postvaccination GMT and the highest titer range between pre- and postvaccination sera. With the HI assay the lowest pre- and postvaccination GMTs were measured. The colorimetric MNT detected the lowest titer increase of the assays but yielded the highest GMT in prevaccination sera whereas the postvaccination GMT was between the values measured by the two other assays. No significant difference was found between the assays concerning the rates of fourfold titer increases that were roughly 95%. Linear regression analysis of the 5 × log2 -transformed postvaccination titers obtained from ELISA-MNT and colorimetric MNT demonstrated these assays to have the best correlation of the three assays (r = 0.714). The regression equation varied from the line of identity: colorimetric MNT = 0.53 × ELISA-MNT + 2.56 (Fig. 2A). The mean absolute difference between titers was 13.86 (95% CI: 12.06–15.91). Linear regression analysis of the log2 transformed titer increases also demonstrated a good correlation
2,5
2,0
1,5
A480-A680
(2% FCS, 1% trypsin, penicillin and streptomycin 500 U/ml each) were added to each well except the system background wells. After incubation for 72 h at 37 ◦ C in a 5% CO2 atmosphere, 50 l of XTT reagent (Cell Proliferation Kit II (XTT), Roche diagnostics, no. 11465015001, Basel, Switzerland) were added to each well. Following incubation of 120 min at 37 ◦ C optical density was measured with an ELISA reader (Tecan, sunrise Absorbance Reader with magellan software version 1.1, Grödig/Salzburg, Austria) at 480 nm using 680 nm as the reference wavelength. The average A480 was calculated from six infected (VC) and eight uninfected (CC) control wells and the neutralizing endpoint was calculated by a 50% signal calculation modified from another method (Rowe et al., 1999): the endpoint titer was expressed as the reciprocal of the highest dilution of serum with A480 more than X, where X = (average A480 of CC wells) − 0.5 × (average A480 of CC wells − average A480 of VC wells).
1,0
Cut-Off: 0.96
0,5
3.1. Validation of the colorimetric microneutralization assay Specificity was tested with sheep antisera to the purified hemagglutinin of the viruses included in the trivalent split-virus vaccine of the influenza season 2009/2010 (anti-A/Brisbane/59/2007 (H1N1), anti-A/Brisbane/10/2007 (H3N2) and anti-B/Brisbane/60/2008, obtained from NIBSC, London, England). The assay appeared to be specific for antibodies to influenza H1N1 (2009) virus, yielding negative results for all three antisera. A dose response curve of sheep antiserum to the hemagglutinin of the influenza A virus strain A/California/7/2009 demonstrated a high protective capacity (Fig. 1). Intra-run reproducibility was assessed by eightfold testing of two post-vaccination sera (low and high reactive) and one convalescent serum (middle reactive). Coefficients of variation (CV) were between 5.2 and 10.8% and the mean intra-run reproducibility was 7.5% (Table 1). The CV for between-run reproducibility were determined by 16-fold testing of the same sera and ranged from 10.4 to 15.6% (mean CV of 12.4%).
0,0
Anti-A/California/7/2009 (H1N1) Anti-A/Brisbane/59/2007(H1N1) Anti-A/Brisbane/10/2007(H3N2) Anti-B/Brisbane/60/2008
Fig. 1. Absorption values (A480 –A680 ) in colorimetric MNT of twofold serial diluted antisera from sheep to the purified hemagglutinin of influenza A virus/California/7/2009 (the vaccination virus of the monovalent pandemic vaccine) and the influenza viruses included in the vaccine of the winter season 2009/2010: filled circles, anti-influenza A virus/California/7/2009 H1N1; open circles, anti-influenza A virus/Brisbane/59/2007 H1N1; filled triangles, anti-influenza A virus/Brisbane/10/2007 H3N2; open triangles, anti-influenza B virus/Brisbane/60/2008. The cut-off value was associated with an absorption higher than 0.96 as indicated by the horizontal bar.
372
S. Grund et al. / Journal of Virological Methods 171 (2011) 369–373
Table 2 Antibody responses of 43 paired sera to vaccination against influenza H1N1 (2009) measured by hemagglutination inhibition (HI) assay, ELISA-based microneutralisation assay (MNT) and colorimetric MNT.
GMT on day 0 (95% CI) GMT on day 21 (95% CI) Fold titer increase (95% CI) % of sera with significant antibody increase (95% CI)
ELISA-based MNT
Colorimetric MNT
HI assay
11.3 (8.0–15.9) 571.6 (348.2–938.3) 50.6 (31.8–80.8) 95.3 (88.7–100)
26.7 (21.6–33.1) 358.3 (248.3–517.0) 13.4 (9.7–18.5) 93.0 (85.1–100)
5.3 (0–5.8) 94.0 (64.6–136.8) 17.6 (12.1–25.7) 95.3 (88.7–100)
GMT: geometric mean titer, calculations were performed from 5 × log2 -transformed data. A fourfold titer increase between pre- and postvaccination sera was considered significant for the three assays.
A
1:2560
1:2560 1:1280
Colorimetric MNT titers
Colorimetric MNT titers
1:1280 1:640 1:320 1:160 1:80 1:40 1:20
y = 0.53 *x+2.56 r = 0.714
1:10 0
B
0
1:10
1:20
1:40
1:640 1:320 1:160 1:80 1:40 1:20
y = 0.63 *x+3.51 r = 0.642
1:10 0
1:80 1:160 1:320 1:640 1:1280 1:2560 1:5120 1:10240
0
1:10
1:20
1:40
ELISA-MNT titers
1:80 1:160 1:320 1:640 1:1280 1:2560 1:5120 1:10240
HI titers
1:20480
C
ELISA-MNT titers
1:5120
1:1280
1:320
1:80
y = 0,86 *x+3.2 r = 0.648
1:20
0 0
1:20
1:80
HI titers
1:320
1:1280
1:5120
Fig. 2. Correlation and relationship of the titers of 43 healthy individuals 21 days after vaccination with a monovalent vaccine against pandemic influenza H1N1 (2009) virus measured by (A) ELISA-MNT and colorimetric MNT, (B) HI assay and colorimetric MNT and (C) HI assay and ELISA-MNT. Conversion equations and correlation coefficients given in the inserts were calculated by linear regression analysis of 5 × log2 -transformed data.
between the assays (r = 0.695; regression equation: colorimetric MNT = 0.48 × ELISA-MNT + 1.02; data not shown). Postvaccination titers obtained from HI assay and colorimetric MNT were also correlated (r = 0.642; regression equation: colorimetric MNT = 0.63 × HI + 3.51 (Fig. 2B) as were titer increases (r = 0.646, regression equation: colorimetric MNT = 0.55 × HI + 1.46, data not shown). A similar correlation was found between HI assay and ELISA-MNT: r = 0.648 for postvaccination titers (regression equation: ELISA-MNT = 0.86 × HI + 3.2, Fig. 2C) and r = 0.582 for titer increases (regression equation: ELISA-MNT = 0.72 × HI + 2.68, data not shown). The mean absolute titer difference was 23.31 (95% CI: 19.86–27.32) between HI and colorimetric MNT and 18.53 (95% CI: 14.31–24.0) between HI and ELISA-MNT. In all assays the GMT rose significantly between pre- and postvaccination sera (p < 0.001 for colorimetric MNT and HI assay in Wilcoxon signed rank test and p < 0.001 for ELISA-MNT in paired t-test).
4. Discussion The intention of this study was to compare the GMTs and the titer increases of healthy individuals after vaccination against pandemic influenza H1N1 (2009) virus measured by: (i) HI assay, (ii) conventional ELISA-based MNT and (iii) colorimetric MNT. One of the main findings was that comparable results for fourfold titer increases were obtained with the three assays: 95.3% in the ELISAMNT (95% CI: 88.7–100) vs. 93.0% (95% CI: 85.1–100) in colorimetric MNT and 95.3% (88.7–100) in HI assay. Therefore, each of the three assays appeared to be suitable for the investigation of immune responses after vaccination. The rate of a fourfold or higher titer increase of roughly 95% is comparable with the results that have been obtained in clinical vaccination trials for subjects 18–60 years of age (Plennevaux et al., 2010; Roman et al., 2010; Zhu et al., 2009).
S. Grund et al. / Journal of Virological Methods 171 (2011) 369–373
The fact that higher titers and titer increases were detected by the ELISA-MNT in comparison to the HI assay is in concordance with studies from seasonal and avian influenza (Benne et al., 1994; Rowe et al., 1999). The HI assay has been shown to be less sensitive than the ELISA-MNT and not to detect all neutralizing antibodies (Remarque et al., 1998). A study comparing the HI assay with the colorimetric MNT found higher pre- and postvaccination titers in the colorimetric MNT than in the HI assay (Lehtoranta et al., 2009). However, this study investigated antibodies to seasonal and not to pandemic influenza and no comparison of the colorimetric MNT with the ELISA-MNT was performed. The present study demonstrated that postvaccination titers, titer increases and titer range were higher in the ELISA-MNT than in the colorimetric MNT and the HI assay. This suggests the ELISAbased MNT to be the method of choice in studies comparing absolute pre- and postvaccination titers and titer increases. It can also be concluded that, at least for antibodies to influenza H1N1 (2009) virus, the highest prevaccination titers are measured by the colorimetric MNT. It would be interesting to investigate the crossreactivity of sera to swine origin and seasonal influenza viruses with the colorimetric MNT. Recent studies in this field have been performed merely with the HI assay (Ikonen et al., 2010) or with the ELISA-MNT (Hancock et al., 2009). Further it should be analysed how useful the colorimetric MNT is to detect antibodies after natural infection with influenza H1N1 (2009) virus. Convalescent sera over a time course should be tested to find out if this assay can detect seroconversions earlier than the ELISA-based MNT or the HI assay. To be used in vaccination trials the costs produced by an assay is an important point to be addressed. Comparing the three assays used in this study, the HI assay is the least expensive test whereas both MNT assays are more cost intensive. This is due to the use of antibodies and substrate in the ELISA-MNT and to the use of tetrazolium salts in the colorimetric MNT. For this reason the HI assay might be preferred in studies investigating large cohorts of vaccines. To measure antibodies to influenza H1N1 (2009) virus the colorimetric MNT was adapted from earlier studies including a further simplification of the underlying method resulting in less hands-on manipulations (Lehtoranta et al., 2009; van de Water et al., 1993). By using the novel tetrazolium salt XTT transfer steps between plates and centrifugation can be omitted. Validation data indicate a high specificity and a good reproducibility of the assay. Antisera to seasonal influenza did not have any neutralizing capacity in the colorimetric MNT. Interestingly, the antiserum to the influenza strain A/Brisbane/59/2007 (H1N1) was found negative (<1:10) but showed the highest background absorption value of the three antisera to seasonal influenza (Fig. 1). This might be explained by the higher phylogenetical relatedness of A/Brisbane/59/2007 to the influenza virus A/California/7/2009 strain (Pascua et al., 2009). The fact that antisera to the influenza strains A/Brisbane/ 10/2007 (H3N2) and B/Brisbane/60/2008 showed only low background absorption values suggests that in the present colorimetric assay virus replication is not inhibited non-specifically by high serum concentrations as found in earlier studies (Lehtoranta et al., 2009). The dose–response dependency of the antiserum to the strain A/California/7/2009 can be attributed to the high neutralization titer (1:10,240) of this antiserum. With regard to seasonal influenza, an HI titer of 1:40 is considered commonly to indicate immune protection as described in vaccination trials (Liang et al., 2010; Plennevaux et al., 2010; Zhu et al., 2009). Studies investigating the susceptibility of volunteers to influenza challenge infection have found different titers in the HI and in the ELISA-MNT assay to be protective (Frank et al., 1980). From the data of the present study it can be concluded that a titer of 1:40 in the HI assay corresponds to a titer of 1:211 in the col-
373
orimetric MNT and a titer of 1:275 in the ELISA-MNT. Limited data are available at present with respect to different methods as far as seroprotection is concerned. Seroprevalence studies are now underway in several countries and will continue to shed more light on comparison of HI and MNT results as well as seroprotection data. Taken together, the present study demonstrated that after vaccination against influenza H1N1 (2009) virus similar rates of fourfold antibody increases were measured by the HI assay, the ELISA-MNT and the colorimetric MNT. The ELISA-MNT demonstrated the highest titer increase leading to the highest postvaccination titers and titer range and appears to be the assay of choice to investigate absolute GMTs. With the colorimetric MNT the highest prevaccination titers are measured. Being less labour-intensive than the ELISAMNT, this assay seems to be a suitable tool in vaccination studies and to be interesting for the investigation of cross-reactivities between sera to swine origin and seasonal influenza viruses. Acknowledgements We wish to thank S. Höveler, H. Lehmann and the other medical technicians of our laboratories for their excellent assistance and Prof. Dr. C.R. MacKenzie for the grammatical review of the manuscript. References Benne, C.A., Harmsen, M., De Jong, J.C., Kraaijeveld, C.A., 1994. Neutralization enzyme immunoassay for influenza virus. J. Clin. Microbiol. 32, 987–990. Frank, A.L., Puck, J., Hughes, B.J., Cate, T.R., 1980. Microneutralization test for influenza A and B and parainfluenza 1 and 2 viruses that uses continuous cell lines and fresh serum enhancement. J. Clin. Microbiol. 12, 426–432. Gitelman, A.K., Kaverin, N.V., Kharitonenkov, I.G., Rudneva, I.A., Sklyanskaya, E.L., Zhdanov, V.M., 1986. Dissociation of the haemagglutination inhibition and the infectivity neutralization in the reactions of influenza A/USSR/90/77 (H1N1) virus variants with monoclonal antibodies. J. Gen. Virol. 67 (Pt. 10), 2247–2251. Hancock, K., Veguilla, V., Lu, X., Zhong, W., Butler, E.N., Sun, H., Liu, F., Dong, L., DeVos, J.R., Gargiullo, P.M., Brammer, T.L., Cox, N.J., Tumpey, T.M., Katz, J.M., 2009. Crossreactive antibody responses to the 2009 pandemic H1N1 influenza virus. N. Engl. J. Med. 361, 1945–1952. Ikonen, N., Strengell, M., Kinnunen, L., Osterlund, P., Pirhonen, J., Broman, M., Davidkin, I., Ziegler, T., Julkunen, I., 2010. High frequency of cross-reacting antibodies against 2009 pandemic influenza A (H1N1) virus among the elderly in Finland. Euro Surveill., 15. Lehtoranta, L., Villberg, A., Santanen, R., Ziegler, T., 2009. A novel, colorimetric neutralization assay for measuring antibodies to influenza viruses. J. Virol. Methods 159, 271–276. Lenette, H.L., Schmidt, N.J., 1979. Influenza viruses. In: Diagnostic Procedures for Viral Rickettsial and Chlamydial Infections, pp. 585–609. Liang, X.F., Wang, H.Q., Wang, J.Z., Fang, H.H., Wu, J., Zhu, F.C., Li, R.C., Xia, S.L., Zhao, Y.L., Li, F.J., Yan, S.H., Yin, W.D., An, K., Feng, D.J., Cui, X.L., Qi, F.C., Ju, C.J., Zhang, Y.H., Guo, Z.J., Chen, P.Y., Chen, Z., Yan, K.M., Wang, Y., 2010. Safety and immunogenicity of 2009 pandemic influenza A H1N1 vaccines in China: a multicentre, double-blind, randomised, placebo-controlled trial. Lancet 375, 56–66. Pascua, P.N., Song, M.S., Lee, J.H., Park, K.J., Kwon, H.I., Baek, Y.H., Hong, S.P., Rho, J.B., Kim, C.J., Poo, H., Ryoo, T.S., Sung, M.H., Choi, Y.K., 2009. Evaluation of the efficacy and cross-protectivity of recent human and swine vaccines against the pandemic (H1N1) 2009 virus infection. PLoS One 4, e8431. Plennevaux, E., Sheldon, E., Blatter, M., Reeves-Hoche, M.K., Denis, M., 2010. Immune response after a single vaccination against 2009 influenza A H1N1 in USA: a preliminary report of two randomised controlled phase 2 trials. Lancet 375, 41–48. Remarque, E.J., de Bruijn, I.A., Boersma, W.J., Masurel, N., Ligthart, G.J., 1998. Altered antibody response to influenza H1N1 vaccine in healthy elderly people as determined by HI ELISA, and neutralization assay. J. Med. Virol. 55, 82–87. Roman, F., Vaman, T., Gerlach, B., Markendorf, A., Gillard, P., Devaster, J.M., 2010. Immunogenicity and safety in adults of one dose of influenza A H1N1 v 2009 vaccine formulated with and without AS03A-adjuvant: preliminary report of an observer-blind, randomised trial. Vaccine 28, 1740–1745. Rowe, T., Abernathy, R.A., Hu-Primmer, J., Thompson, W.W., Lu, X., Lim, W., Fukuda, K., Cox, N.J., Katz, J.M., 1999. Detection of antibody to avian influenza A (H5N1) virus in human serum by using a combination of serologic assays. J. Clin. Microbiol. 37, 937–943. van de Water, C., van Dura, E.A., van der Stap, J.G., Brands, R., Boersma, W.J., 1993. Rapid in vitro micro-cytotoxicity tests for the detection and quantitation of neutralizing antibodies to both viruses and toxins. J. Immunol. Methods 166, 157–164. Zhu, F.C., Wang, H., Fang, H.H., Yang, J.G., Lin, X.J., Liang, X.F., Zhang, X.F., Pan, H.X., Meng, F.Y., Hu, Y.M., Liu, W.D., Li, C.G., Li, W., Zhang, X., Hu, J.M., Peng, W.B., Yang, B.P., Xi, P., Wang, H.Q., Zheng, J.S., 2009. A novel influenza A (H1N1) vaccine in various age groups. N. Engl. J. Med. 361, 2414–2423.
Contents lists available at ScienceDirect
Journal of Virological Methods journal homepage: www.elsevier.com/locate/jviromet
Protocols
Comparison of hemagglutination inhibition assay, an ELISA-based micro-neutralization assay and colorimetric microneutralization assay to detect antibody responses to vaccination against influenza A H1N1 2009 virus S. Grund a,∗,1 , O. Adams a,1 , S. Wählisch b , B. Schweiger b a b
Institute of Virology of the University of Düsseldorf, Heinrich-Heine-University, D-40225 Düsseldorf, Germany National Reference Centre for Influenza, Robert-Koch-Institute, Berlin, Germany
a b s t r a c t Article history: Received 25 June 2010 Received in revised form 23 November 2010 Accepted 30 November 2010 Available online 10 December 2010 Keywords: Influenza (H1N1) 2009 virus Vaccination Antibody responses Hemagglutination inhibition assay Neutralization assay
The hemagglutination inhibition (HI) assay has been the main method used to investigate immune responses to vaccination against influenza H1N1 (2009) virus. However microneutralization tests (MNT) have been shown to be more sensitive and more specific. In this study, the three methods of choice: (i) the HI assay, (ii) an ELISA-based conventional MNT and (iii) a colorimetric MNT in terms of their ability to detect antibody responses in serum pairs collected from 43 healthy individuals before and 21 days after vaccination were compared. The colorimetric MNT was established yielding intra- and inter-run imprecisions of 7.5% and 12.4%, respectively. Testing of antisera to seasonal influenza viruses demonstrated the assay to be specific for antibodies to influenza H1N1 (2009) virus. A good correlation between the three methods was found, being highest for the ELISA-MNT and the colorimetric MNT (r = 0.714 for geometric mean titers (GMT) and r = 0.695 for titer increases). Similar rates of fourfold titer increases were detected: 95.3% in the ELISA-MNT vs. 93.0% in colorimetric MNT and 95.3% in HI assay. The ELISA-based MNT demonstrated the highest titer range leading to the highest postvaccination GMT and the highest titer increase (>50-fold). The lowest GMTs were measured with the HI assay, while the colormetric MNT detected the highest GMT in prevaccination sera. Taken together, similar seroconversion rates were obtained with the three assays. The ELISA-MNT appeared to be the best method to compare absolute pre- and postvaccination GMTs. The colorimetric MNT, being less labour-intensive than the ELISA-MNT, seems to be a suitable tool in vaccination studies. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Clinical trials have investigated immune responses to the monovalent split-virus inactivated vaccine against influenza H1N1 (2009) virus (Liang et al., 2010; Plennevaux et al., 2010; Zhu et al., 2009). In these studies antibody responses were measured mainly by the hemagglutination inhibition (HI) assay because this method can be easily adapted to new influenza viruses. However, it has been shown for seasonal and for avian influenza that microneutralization tests (MNT) are more sensitive and more specific (Frank et al., 1980; Gitelman et al., 1986; Rowe et al., 1999). A previous study for seasonal influenza A indicated that strain-specific IgG antibodies with neutralizing capacity may be present, but these antibodies were not generally detected in homologous HI assays (Remarque et al.,
∗ Corresponding author at: Institut für Virologie, Geb. 22.21, Universitätsstr. 1, D-40225 Düsseldorf, Germany. Tel.: +49 211 811 0551; fax: +49 211 811 2227. E-mail address: [email protected] (S. Grund). 1 These authors contributed equally to this work. 0166-0934/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jviromet.2010.11.024
1998). Conventional microneutralization tests are performed in different ways with regard to titer determination, cytopathic effect formation in cell cultures, red blood cells for detection of released virus or usage of the ELISA technique to identify virus infected cells. Especially the ELISA method is time-consuming, difficult to perform under biosafety-level 3 (BSL-3) conditions and requires the need for monoclonal antibodies that are not widely available for new viruses. Recently, a colorimetric MNT has been introduced for seasonal influenza (Lehtoranta et al., 2009). This assay is an adaptation of micro-cytotoxicity tests (van de Water et al., 1993) and is based on the cleavage of a tetrazolium salt by mitochondrial dehydrogenases reflecting the amount of viable and metabolically active cells. In the present study, the three assays of choice: (i) HI assay, (ii) ELISA-MNT and (iii) colorimetric MNT were compared to detect immune responses in serum pairs collected from 43 healthy individuals before and 21 days after administration of a monovalent vaccine against pandemic influenza H1N1 (2009) virus. The colorimetric MNT was adapted and validated to detect antibodies to pandemic influenza and the method was modified to include less hands-on manipulations.
370
S. Grund et al. / Journal of Virological Methods 171 (2011) 369–373
2. Materials and methods
2.5. ELISA-based conventional microneutralization (ELISA-MNT) assay
2.1. Cells MDCK cells (obtained from BioWhittaker Europe, Verviers, Belgium) were grown in Dulbecco’s modified essential medium (DMEM) supplemented with penicillin and streptomycin (500 U/ml each) and 2% fetal calf serum and were maintained in a 5% CO2 humified atmosphere at 37 ◦ C. Two to three times a week cells were subcultured at a split-ratio of 1:2–1:5. 2.2. Viruses For the HI assay and the ELISA-MNT the pandemic influenza virus reference strain A/California/7/2009 H1N1 was used. The colorimetric MNT was performed with a pandemic influenza A H1N1 virus isolated from an infected patient in September 2009. This virus reacted well with a specific antiserum to the influenza virus A/California/7/2009 in HI assay. According to this close relatedness in the antigenic profile the virus was declared to be an influenza A/California/7/2009-like virus by the German national reference institute for influenza at the Robert-Koch-Institute in Berlin. Sequencing of the HA gene (984 nucleotides) yielded seven nucleotide exchanges compared to the reference strain A/California/7/2009 without any exchange in amino acid sequence. In comparison to the reference virus A/California/7/2009 a concentration of 500–1000 TCID50 of this virus demonstrated a stronger cytopathic effect in MDCK cells leading to a cell destruction of roughly 90% after 72 h of incubation. Strong destruction of target cells has been described to be essential for a colorimetric MNT (Lehtoranta et al., 2009). 2.3. Human sera In November 2009, 43 serum pairs were collected from healthy staff members of the Institute of Virology of the University of Düsseldorf before and 21 days after vaccination with an adjuvant monovalent split-virus inactivated vaccine against influenza H1N1 (2009) virus (Pandemrix® , GlaxoSmithKline, London, England). Informed consent was obtained from all individuals participating in the study. The mean age of subjects was 36.2 years (SD 10.4 years). Individuals with a history of infection with influenza H1N1 (2009) virus were excluded from the study.
The method of the ELISA-based microneutralization test was modified from an assay described previously (Rowe et al., 1999). All sera used were heat inactivated at 56 ◦ C for 30 min and twofold serial dilutions were prepared (diluent: Eagle’s minimal essential medium supplemented with 1% BSA and gentamycin 0.05 g/ml). 50 l of each dilution were added to a microtiter plate (flat-bottom, Corning Incorporated, Corning, USA) and mixed with 50 l virus suspension (2 × 103 TCID50 /ml in diluent). The plates included four wells containing 50 l of diluted virus and 50 l diluent as positive controls, four wells containing 100 l diluent as cell control and two wells each for back titration (a serial twofold dilution in diluent, to 1:16). The plates were incubated at 37 ◦ C for 2 h and 5% CO2 . Afterwards 100 l of MDCK-2 cells (1.5 × 105 /ml) were added to each well. After overnight incubation (18–22 h) at 37 ◦ C and 5% CO2 , the medium was removed from the wells and the monolayers were fixed with cold fixative (acetone 80% in PBS 1:5) for 10 min. The titer was calculated by performing an ELISA to detect virus-infected cells. The fixed plates were washed three times with the wash buffer (PBS, 0.1% Tween 20), the anti-influenza A NP mouse monoclonal antibody (WHO Influenza Reagent Kit, Centers for Disease Control and Prevention (CDC), Atlanta, USA) was diluted 1:4000 in blocking buffer (PBS, 1% bovine serum albumin and 0.1% Tween 20) and 100 l were added to each well. After 1 h incubation the plates were washed four times with wash buffer and 100 l of horseradish peroxidase-conjugated goat anti-mouse IgG (diluted 1:1000 in blocking buffer) was added to each well. The plates were incubated at room temperature for 1 h and then washed six times with wash buffer. 100 l of freshly prepared substrate solution (hydrogen peroxide, 1 g/L, in sodium acetate buffer solution, 25 mmol/L [pH 4.1] with tetramethylbenzidine dihydrochloride, 5 g/L) were added to each well and the reaction was stopped after 5 min with 100 l stop solution (0.5 N sulfuric acid) per well. The absorbance was measured at 450 nm (A450 ) with Titertek Multiscan PLUS (Labsystems) and the virus neutralization endpoint titer of each serum was determined with the following equation: X = [(average A450 of virus control wells) − (average A450 of cell control wells)]/2 + (average A450 of cell control wells). X = 50% of specific signal and all values below this value are positive for neutralization activity. All sera were tested twice on different days and the final titer value was the average of two separate runs. 2.6. Colorimetric microneutralization (MNT) assay
2.4. Hemagglutination inhibition (HI) assay The hemagglutination inhibition assay was performed as previously described (Lenette and Schmidt, 1979). Prior to testing, each serum was treated with receptor destroying enzyme (Cholera filtrate, Sigma, Germany) to inactivate non-specific inhibitors achieving a final serum dilution of 1:10. The sera were then diluted serially twofold in V-bottom microtiter plates. The virus was adjusted to 4 HA units/25 l, which was verified by back titration, and 25 l of this virus suspension was added to each of the 96 wells. After incubation at room temperature (RT) for 30 min freshly prepared 0.5% turkey red blood cells (RBCs) were added (obtained from the Bundesinstitut für Risikobewertung, Berlin, Germany). The plates were agitated briefly followed by a further incubation at RT for 30 min. Human sera serving as positive controls and negative controls were included on each plate. The determination of the HI titer was performed by calculation of the reciprocal of the last serum dilution which contained non-agglutinated RBCs. The sera were tested only once in HI assay because there was not sufficient sample material available for further assay runs for the majority of vaccines.
To detect specific antibodies to influenza H1N1 (2009) virus a colorimetric MNT was established on the basis of different methods that had been introduced for seasonal influenza (Lehtoranta et al., 2009; van de Water et al., 1993). Sera were heat inactivated at 56 ◦ C for 30 min, and twofold serial dilutions in 50 l were prepared in flat-bottom 96-well tissue plates (TPP, Trasadingen, Switzerland) with Dulbecco’s modified essential medium (DMEM) supplemented with penicillin and streptomycin (500 U/ml each), 2% fetal calf serum (FCS) and 1% trypsin. An equal volume of the same medium containing 500–1000 TCID50 of the challenge virus was added to each serum dilution, and plates were incubated at 37 ◦ C for 1 h. All sera were tested in duplicate and the average absorption values were used for the calculation of neutralizing titers. Wells containing virus dilutions without serum, MDCK cells only and DMEM only were included on each plate serving as controls and system background, respectively. In each run four control sera yielding: (i) negative, (ii) low, (iii) middle and (iv) high reactive results in the HI assay were included to assess the between-run imprecision of the assay. After incubation 100 l of a suspension containing 3 × 104 /ml freshly trypsinized MDCK cells in DMEM
S. Grund et al. / Journal of Virological Methods 171 (2011) 369–373
371
Table 1 Intra- and inter-run reproducibility of the colorimetric microneutralization assay established to detect antibodies to influenza H1N1 (2009) virus. Control serum
Low reactive Middle reactive asssaareactive High reactive
Intra-assay (8-fold)
Inter-assay (16-fold)
GMT
5 × log2 GMT
5 × log2 SD (mean)
CV (%)
GMT
5 × log2 GMT
5 × log2 SD (mean)
CV (%)
160.0 586.9 1810.2
5.00 6.88 8.50
0.54 0.36 0.54
10.8 5.2 6.4
140.5 472.7 1522.2
4.81 6.56 8.25
0.75 0.73 0.86
15.6 11.1 10.4
Geometric mean titers (GMT), standard deviations (SD) and coefficients of variation (CVs) were calculated from 5 × log2 -transformed data.
2.7. Statistical analysis Statistical analysis was performed using SigmaPlot version 10.0 and Mircosoft Excel 2002 software. Sera negative at a dilution of 1/10 were assigned a titer of 5 and sera positive at a dilution of 1/5120 were assigned a titer of 5120 in each method. A fourfold titer increase between pre- and postvaccination sera was considered significant for both MNTs and the HI assay. Titers obtained for the three methods were analysed as 5 × log2 -transformed data and titer increases as log2 -transformed data. Correlation between methods was calculated by linear regression analysis and either the paired t-test or the Wilcoxon signed rank test was used to investigate the significance of titer increases between pre- and postvaccination sera.
3. Results
3.2. Correlation between the results of the conventional ELISA-based MNT, the colorimetric MNT and the hemagglutination inhibition assay Geometric mean titers (GMT) and titer increases of the three assays are given in Table 2. The ELISA-MNT demonstrated the highest titer increase (>50-fold) leading to the highest postvaccination GMT and the highest titer range between pre- and postvaccination sera. With the HI assay the lowest pre- and postvaccination GMTs were measured. The colorimetric MNT detected the lowest titer increase of the assays but yielded the highest GMT in prevaccination sera whereas the postvaccination GMT was between the values measured by the two other assays. No significant difference was found between the assays concerning the rates of fourfold titer increases that were roughly 95%. Linear regression analysis of the 5 × log2 -transformed postvaccination titers obtained from ELISA-MNT and colorimetric MNT demonstrated these assays to have the best correlation of the three assays (r = 0.714). The regression equation varied from the line of identity: colorimetric MNT = 0.53 × ELISA-MNT + 2.56 (Fig. 2A). The mean absolute difference between titers was 13.86 (95% CI: 12.06–15.91). Linear regression analysis of the log2 transformed titer increases also demonstrated a good correlation
2,5
2,0
1,5
A480-A680
(2% FCS, 1% trypsin, penicillin and streptomycin 500 U/ml each) were added to each well except the system background wells. After incubation for 72 h at 37 ◦ C in a 5% CO2 atmosphere, 50 l of XTT reagent (Cell Proliferation Kit II (XTT), Roche diagnostics, no. 11465015001, Basel, Switzerland) were added to each well. Following incubation of 120 min at 37 ◦ C optical density was measured with an ELISA reader (Tecan, sunrise Absorbance Reader with magellan software version 1.1, Grödig/Salzburg, Austria) at 480 nm using 680 nm as the reference wavelength. The average A480 was calculated from six infected (VC) and eight uninfected (CC) control wells and the neutralizing endpoint was calculated by a 50% signal calculation modified from another method (Rowe et al., 1999): the endpoint titer was expressed as the reciprocal of the highest dilution of serum with A480 more than X, where X = (average A480 of CC wells) − 0.5 × (average A480 of CC wells − average A480 of VC wells).
1,0
Cut-Off: 0.96
0,5
3.1. Validation of the colorimetric microneutralization assay Specificity was tested with sheep antisera to the purified hemagglutinin of the viruses included in the trivalent split-virus vaccine of the influenza season 2009/2010 (anti-A/Brisbane/59/2007 (H1N1), anti-A/Brisbane/10/2007 (H3N2) and anti-B/Brisbane/60/2008, obtained from NIBSC, London, England). The assay appeared to be specific for antibodies to influenza H1N1 (2009) virus, yielding negative results for all three antisera. A dose response curve of sheep antiserum to the hemagglutinin of the influenza A virus strain A/California/7/2009 demonstrated a high protective capacity (Fig. 1). Intra-run reproducibility was assessed by eightfold testing of two post-vaccination sera (low and high reactive) and one convalescent serum (middle reactive). Coefficients of variation (CV) were between 5.2 and 10.8% and the mean intra-run reproducibility was 7.5% (Table 1). The CV for between-run reproducibility were determined by 16-fold testing of the same sera and ranged from 10.4 to 15.6% (mean CV of 12.4%).
0,0
Anti-A/California/7/2009 (H1N1) Anti-A/Brisbane/59/2007(H1N1) Anti-A/Brisbane/10/2007(H3N2) Anti-B/Brisbane/60/2008
Fig. 1. Absorption values (A480 –A680 ) in colorimetric MNT of twofold serial diluted antisera from sheep to the purified hemagglutinin of influenza A virus/California/7/2009 (the vaccination virus of the monovalent pandemic vaccine) and the influenza viruses included in the vaccine of the winter season 2009/2010: filled circles, anti-influenza A virus/California/7/2009 H1N1; open circles, anti-influenza A virus/Brisbane/59/2007 H1N1; filled triangles, anti-influenza A virus/Brisbane/10/2007 H3N2; open triangles, anti-influenza B virus/Brisbane/60/2008. The cut-off value was associated with an absorption higher than 0.96 as indicated by the horizontal bar.
372
S. Grund et al. / Journal of Virological Methods 171 (2011) 369–373
Table 2 Antibody responses of 43 paired sera to vaccination against influenza H1N1 (2009) measured by hemagglutination inhibition (HI) assay, ELISA-based microneutralisation assay (MNT) and colorimetric MNT.
GMT on day 0 (95% CI) GMT on day 21 (95% CI) Fold titer increase (95% CI) % of sera with significant antibody increase (95% CI)
ELISA-based MNT
Colorimetric MNT
HI assay
11.3 (8.0–15.9) 571.6 (348.2–938.3) 50.6 (31.8–80.8) 95.3 (88.7–100)
26.7 (21.6–33.1) 358.3 (248.3–517.0) 13.4 (9.7–18.5) 93.0 (85.1–100)
5.3 (0–5.8) 94.0 (64.6–136.8) 17.6 (12.1–25.7) 95.3 (88.7–100)
GMT: geometric mean titer, calculations were performed from 5 × log2 -transformed data. A fourfold titer increase between pre- and postvaccination sera was considered significant for the three assays.
A
1:2560
1:2560 1:1280
Colorimetric MNT titers
Colorimetric MNT titers
1:1280 1:640 1:320 1:160 1:80 1:40 1:20
y = 0.53 *x+2.56 r = 0.714
1:10 0
B
0
1:10
1:20
1:40
1:640 1:320 1:160 1:80 1:40 1:20
y = 0.63 *x+3.51 r = 0.642
1:10 0
1:80 1:160 1:320 1:640 1:1280 1:2560 1:5120 1:10240
0
1:10
1:20
1:40
ELISA-MNT titers
1:80 1:160 1:320 1:640 1:1280 1:2560 1:5120 1:10240
HI titers
1:20480
C
ELISA-MNT titers
1:5120
1:1280
1:320
1:80
y = 0,86 *x+3.2 r = 0.648
1:20
0 0
1:20
1:80
HI titers
1:320
1:1280
1:5120
Fig. 2. Correlation and relationship of the titers of 43 healthy individuals 21 days after vaccination with a monovalent vaccine against pandemic influenza H1N1 (2009) virus measured by (A) ELISA-MNT and colorimetric MNT, (B) HI assay and colorimetric MNT and (C) HI assay and ELISA-MNT. Conversion equations and correlation coefficients given in the inserts were calculated by linear regression analysis of 5 × log2 -transformed data.
between the assays (r = 0.695; regression equation: colorimetric MNT = 0.48 × ELISA-MNT + 1.02; data not shown). Postvaccination titers obtained from HI assay and colorimetric MNT were also correlated (r = 0.642; regression equation: colorimetric MNT = 0.63 × HI + 3.51 (Fig. 2B) as were titer increases (r = 0.646, regression equation: colorimetric MNT = 0.55 × HI + 1.46, data not shown). A similar correlation was found between HI assay and ELISA-MNT: r = 0.648 for postvaccination titers (regression equation: ELISA-MNT = 0.86 × HI + 3.2, Fig. 2C) and r = 0.582 for titer increases (regression equation: ELISA-MNT = 0.72 × HI + 2.68, data not shown). The mean absolute titer difference was 23.31 (95% CI: 19.86–27.32) between HI and colorimetric MNT and 18.53 (95% CI: 14.31–24.0) between HI and ELISA-MNT. In all assays the GMT rose significantly between pre- and postvaccination sera (p < 0.001 for colorimetric MNT and HI assay in Wilcoxon signed rank test and p < 0.001 for ELISA-MNT in paired t-test).
4. Discussion The intention of this study was to compare the GMTs and the titer increases of healthy individuals after vaccination against pandemic influenza H1N1 (2009) virus measured by: (i) HI assay, (ii) conventional ELISA-based MNT and (iii) colorimetric MNT. One of the main findings was that comparable results for fourfold titer increases were obtained with the three assays: 95.3% in the ELISAMNT (95% CI: 88.7–100) vs. 93.0% (95% CI: 85.1–100) in colorimetric MNT and 95.3% (88.7–100) in HI assay. Therefore, each of the three assays appeared to be suitable for the investigation of immune responses after vaccination. The rate of a fourfold or higher titer increase of roughly 95% is comparable with the results that have been obtained in clinical vaccination trials for subjects 18–60 years of age (Plennevaux et al., 2010; Roman et al., 2010; Zhu et al., 2009).
S. Grund et al. / Journal of Virological Methods 171 (2011) 369–373
The fact that higher titers and titer increases were detected by the ELISA-MNT in comparison to the HI assay is in concordance with studies from seasonal and avian influenza (Benne et al., 1994; Rowe et al., 1999). The HI assay has been shown to be less sensitive than the ELISA-MNT and not to detect all neutralizing antibodies (Remarque et al., 1998). A study comparing the HI assay with the colorimetric MNT found higher pre- and postvaccination titers in the colorimetric MNT than in the HI assay (Lehtoranta et al., 2009). However, this study investigated antibodies to seasonal and not to pandemic influenza and no comparison of the colorimetric MNT with the ELISA-MNT was performed. The present study demonstrated that postvaccination titers, titer increases and titer range were higher in the ELISA-MNT than in the colorimetric MNT and the HI assay. This suggests the ELISAbased MNT to be the method of choice in studies comparing absolute pre- and postvaccination titers and titer increases. It can also be concluded that, at least for antibodies to influenza H1N1 (2009) virus, the highest prevaccination titers are measured by the colorimetric MNT. It would be interesting to investigate the crossreactivity of sera to swine origin and seasonal influenza viruses with the colorimetric MNT. Recent studies in this field have been performed merely with the HI assay (Ikonen et al., 2010) or with the ELISA-MNT (Hancock et al., 2009). Further it should be analysed how useful the colorimetric MNT is to detect antibodies after natural infection with influenza H1N1 (2009) virus. Convalescent sera over a time course should be tested to find out if this assay can detect seroconversions earlier than the ELISA-based MNT or the HI assay. To be used in vaccination trials the costs produced by an assay is an important point to be addressed. Comparing the three assays used in this study, the HI assay is the least expensive test whereas both MNT assays are more cost intensive. This is due to the use of antibodies and substrate in the ELISA-MNT and to the use of tetrazolium salts in the colorimetric MNT. For this reason the HI assay might be preferred in studies investigating large cohorts of vaccines. To measure antibodies to influenza H1N1 (2009) virus the colorimetric MNT was adapted from earlier studies including a further simplification of the underlying method resulting in less hands-on manipulations (Lehtoranta et al., 2009; van de Water et al., 1993). By using the novel tetrazolium salt XTT transfer steps between plates and centrifugation can be omitted. Validation data indicate a high specificity and a good reproducibility of the assay. Antisera to seasonal influenza did not have any neutralizing capacity in the colorimetric MNT. Interestingly, the antiserum to the influenza strain A/Brisbane/59/2007 (H1N1) was found negative (<1:10) but showed the highest background absorption value of the three antisera to seasonal influenza (Fig. 1). This might be explained by the higher phylogenetical relatedness of A/Brisbane/59/2007 to the influenza virus A/California/7/2009 strain (Pascua et al., 2009). The fact that antisera to the influenza strains A/Brisbane/ 10/2007 (H3N2) and B/Brisbane/60/2008 showed only low background absorption values suggests that in the present colorimetric assay virus replication is not inhibited non-specifically by high serum concentrations as found in earlier studies (Lehtoranta et al., 2009). The dose–response dependency of the antiserum to the strain A/California/7/2009 can be attributed to the high neutralization titer (1:10,240) of this antiserum. With regard to seasonal influenza, an HI titer of 1:40 is considered commonly to indicate immune protection as described in vaccination trials (Liang et al., 2010; Plennevaux et al., 2010; Zhu et al., 2009). Studies investigating the susceptibility of volunteers to influenza challenge infection have found different titers in the HI and in the ELISA-MNT assay to be protective (Frank et al., 1980). From the data of the present study it can be concluded that a titer of 1:40 in the HI assay corresponds to a titer of 1:211 in the col-
373
orimetric MNT and a titer of 1:275 in the ELISA-MNT. Limited data are available at present with respect to different methods as far as seroprotection is concerned. Seroprevalence studies are now underway in several countries and will continue to shed more light on comparison of HI and MNT results as well as seroprotection data. Taken together, the present study demonstrated that after vaccination against influenza H1N1 (2009) virus similar rates of fourfold antibody increases were measured by the HI assay, the ELISA-MNT and the colorimetric MNT. The ELISA-MNT demonstrated the highest titer increase leading to the highest postvaccination titers and titer range and appears to be the assay of choice to investigate absolute GMTs. With the colorimetric MNT the highest prevaccination titers are measured. Being less labour-intensive than the ELISAMNT, this assay seems to be a suitable tool in vaccination studies and to be interesting for the investigation of cross-reactivities between sera to swine origin and seasonal influenza viruses. Acknowledgements We wish to thank S. Höveler, H. Lehmann and the other medical technicians of our laboratories for their excellent assistance and Prof. Dr. C.R. MacKenzie for the grammatical review of the manuscript. References Benne, C.A., Harmsen, M., De Jong, J.C., Kraaijeveld, C.A., 1994. Neutralization enzyme immunoassay for influenza virus. J. Clin. Microbiol. 32, 987–990. Frank, A.L., Puck, J., Hughes, B.J., Cate, T.R., 1980. Microneutralization test for influenza A and B and parainfluenza 1 and 2 viruses that uses continuous cell lines and fresh serum enhancement. J. Clin. Microbiol. 12, 426–432. Gitelman, A.K., Kaverin, N.V., Kharitonenkov, I.G., Rudneva, I.A., Sklyanskaya, E.L., Zhdanov, V.M., 1986. Dissociation of the haemagglutination inhibition and the infectivity neutralization in the reactions of influenza A/USSR/90/77 (H1N1) virus variants with monoclonal antibodies. J. Gen. Virol. 67 (Pt. 10), 2247–2251. Hancock, K., Veguilla, V., Lu, X., Zhong, W., Butler, E.N., Sun, H., Liu, F., Dong, L., DeVos, J.R., Gargiullo, P.M., Brammer, T.L., Cox, N.J., Tumpey, T.M., Katz, J.M., 2009. Crossreactive antibody responses to the 2009 pandemic H1N1 influenza virus. N. Engl. J. Med. 361, 1945–1952. Ikonen, N., Strengell, M., Kinnunen, L., Osterlund, P., Pirhonen, J., Broman, M., Davidkin, I., Ziegler, T., Julkunen, I., 2010. High frequency of cross-reacting antibodies against 2009 pandemic influenza A (H1N1) virus among the elderly in Finland. Euro Surveill., 15. Lehtoranta, L., Villberg, A., Santanen, R., Ziegler, T., 2009. A novel, colorimetric neutralization assay for measuring antibodies to influenza viruses. J. Virol. Methods 159, 271–276. Lenette, H.L., Schmidt, N.J., 1979. Influenza viruses. In: Diagnostic Procedures for Viral Rickettsial and Chlamydial Infections, pp. 585–609. Liang, X.F., Wang, H.Q., Wang, J.Z., Fang, H.H., Wu, J., Zhu, F.C., Li, R.C., Xia, S.L., Zhao, Y.L., Li, F.J., Yan, S.H., Yin, W.D., An, K., Feng, D.J., Cui, X.L., Qi, F.C., Ju, C.J., Zhang, Y.H., Guo, Z.J., Chen, P.Y., Chen, Z., Yan, K.M., Wang, Y., 2010. Safety and immunogenicity of 2009 pandemic influenza A H1N1 vaccines in China: a multicentre, double-blind, randomised, placebo-controlled trial. Lancet 375, 56–66. Pascua, P.N., Song, M.S., Lee, J.H., Park, K.J., Kwon, H.I., Baek, Y.H., Hong, S.P., Rho, J.B., Kim, C.J., Poo, H., Ryoo, T.S., Sung, M.H., Choi, Y.K., 2009. Evaluation of the efficacy and cross-protectivity of recent human and swine vaccines against the pandemic (H1N1) 2009 virus infection. PLoS One 4, e8431. Plennevaux, E., Sheldon, E., Blatter, M., Reeves-Hoche, M.K., Denis, M., 2010. Immune response after a single vaccination against 2009 influenza A H1N1 in USA: a preliminary report of two randomised controlled phase 2 trials. Lancet 375, 41–48. Remarque, E.J., de Bruijn, I.A., Boersma, W.J., Masurel, N., Ligthart, G.J., 1998. Altered antibody response to influenza H1N1 vaccine in healthy elderly people as determined by HI ELISA, and neutralization assay. J. Med. Virol. 55, 82–87. Roman, F., Vaman, T., Gerlach, B., Markendorf, A., Gillard, P., Devaster, J.M., 2010. Immunogenicity and safety in adults of one dose of influenza A H1N1 v 2009 vaccine formulated with and without AS03A-adjuvant: preliminary report of an observer-blind, randomised trial. Vaccine 28, 1740–1745. Rowe, T., Abernathy, R.A., Hu-Primmer, J., Thompson, W.W., Lu, X., Lim, W., Fukuda, K., Cox, N.J., Katz, J.M., 1999. Detection of antibody to avian influenza A (H5N1) virus in human serum by using a combination of serologic assays. J. Clin. Microbiol. 37, 937–943. van de Water, C., van Dura, E.A., van der Stap, J.G., Brands, R., Boersma, W.J., 1993. Rapid in vitro micro-cytotoxicity tests for the detection and quantitation of neutralizing antibodies to both viruses and toxins. J. Immunol. Methods 166, 157–164. Zhu, F.C., Wang, H., Fang, H.H., Yang, J.G., Lin, X.J., Liang, X.F., Zhang, X.F., Pan, H.X., Meng, F.Y., Hu, Y.M., Liu, W.D., Li, C.G., Li, W., Zhang, X., Hu, J.M., Peng, W.B., Yang, B.P., Xi, P., Wang, H.Q., Zheng, J.S., 2009. A novel influenza A (H1N1) vaccine in various age groups. N. Engl. J. Med. 361, 2414–2423.