In Vitro ELISA Test to Evaluate Rabies Vaccine Potency

Corinne Jallet; No&#235;l Tordo

doi:10.3791/59641

JoVE Journal > Immunology and Infection

İmmünoloji ve Enfeksiyon

In Vitro ELISA Test to Evaluate Rabies Vaccine Potency

Published: May 11, 2020

doi:

10.3791/59641

Corinne Jallet¹, Noël Tordo^1,2

¹Unit Antiviral Strategies, OIE Reference Laboratories for Rift Valley Fever Virus & Crimean Congo Hemorrhagic Fever Virus,Pasteur Enstitüsü, ²Institut Pasteur de Guinée, Conakry

Özet

Here we describe an indirect ELISA sandwich immunocapture to determine the immunogenic glycoprotein contents in rabies vaccines. This test uses a neutralizing Monoclonal Antibody (mAb-D1) recognizing glycoprotein trimers. It is an alternative to the in vivo NIH test to follow the consistency of vaccine potency during production.

Abstract

The growing global concern for the animal welfare is encouraging manufacturers and the National Control Laboratories (OMCLs) to follow the 3Rs strategy for the Replacement, Reduction, and Refinement of the laboratory animal testing. The development of in vitro approaches is recommended at the WHO and European levels as alternatives to the NIH test for evaluating the rabies vaccine potency. At the surface of the rabies virus (RABV) particle, trimers of glycoprotein constitute the major immunogen to induce Viral Neutralizing Antibodies (VNAbs). An ELISA test, where Neutralizing Monoclonal Antibodies (mAb-D1) recognize the trimeric form of the glycoprotein, has been developed to determine the contents of the native folded trimeric glycoprotein along with the production of the vaccine batches. This in vitro potency test demonstrated a good concordance with the NIH test and has been found suitable in collaborative trials by RABV vaccine manufacturers and OMCLs. Avoidance of animal use is an achievable objective in the near future.

The method presented is based on an indirect ELISA sandwich immunocapture using the mAb-D1 which recognizes the antigenic sites III (aa 330 to 338) of the trimeric RABV glycoprotein, i.e., the immunogenic RABV antigen. mAb-D1 is used for both coating and detection of glycoprotein trimers present in the vaccine batch. Since the epitope is recognized because of its conformational properties, the potentially denatured glycoprotein (less immunogenic) cannot be captured and detected by the mAb-D1. The vaccine to be tested is incubated in a plate sensitized with the mAb-D1. Bound trimeric RABV glycoproteins are identified by adding the mAb-D1 again, labeled with peroxidase and then revealed in the presence of substrate and chromogen. Comparison of the absorbance measured for the tested vaccine and the reference vaccine allows for the determination of the immunogenic glycoprotein content.

Introduction

Since more than 50 years, the NIH test¹ is used as a gold standard method to evaluate the rabies vaccine potency before the batch release. This test consists of an intraperitoneal immunization of groups of mice with the vaccine to be tested followed by an intra-cerebral (IC) challenge 14 days later with the Challenge Virus Standard (CVS) strain of rabies virus (RABV). The potency is evaluated from the proportion of mice surviving the IC challenge. Although WHO² and European Pharmacopeia³ still require the NIH test for assessing the vaccine potency, this test suffers several hurdles: results are highly variable⁴; infectious RABV is used during the challenge and this requires both technical skill and strict biosafety measures; large numbers of animals are used, and the severity of the challenge raises serious ethical concerns⁵. A less severe variation of this test has been developed: two weeks after the intra-peritoneal immunization, mice are not challenged by IC but bled and tested for the presence of specific RABV neutralizing antibodies (VNAbs) in their serum using an in vitro neutralization test. However, this test still requires sacrificing a large number of laboratory mice although it is already in use for the veterinary vaccines⁶^,⁷ and has been considered for human vaccines⁸.

As of now, both International⁹ and European¹⁰ recommendations encourage manufacturers and National Control Laboratories (Official Medicine Control Laboratories – OMCLs) to implement the Replacement, Reduction, and Refinement of laboratory animal testing, referred as the 3Rs strategy. European Directive 2010/63/EU (in force since 2013/01/01) related to the protection and welfare of animals has also reinforced the constraints for vaccine manufacturers and laboratories involved in the Quality Control of rabies vaccines as well as in rabies research¹¹. As a result, the development, validation, and use of alternative in vitro approaches have now become a priority. These are not only ethically sound but can also reduce the batch testing costs and shorten the time for results to hours instead of weeks³.

At the surface of the RABV particle, the glycoprotein adopts a trimeric form¹²^,¹³^,¹⁴^,¹⁵^,¹⁶. In rabies vaccine, this native trimeric form constitutes the major immunogen inducing VNAbs¹⁷ while the monomeric, soluble or denatured glycoproteins are poorly immunogenic¹⁸^,¹⁹. Thus, the preservation of trimers of the glycoprotein along the vaccine production process is a good indicator for the preservation of an optimal immunogenic potential. Several immunochemical methods, such as the antibody-binding-test²⁰^,²¹, the single radial immunodiffusion (SRD) test²² and the ELISA test²³^,²⁴^,²⁵^,²⁶^,²⁷ are recommended by the WHO Technical Report Series² and the European monograph³ to quantify the antigen content in rabies vaccines. These are used by manufacturers to monitor the consistency of vaccine production and by the OMCL to assess the consistent formulation of batches of human vaccines²⁸, even if the NIH test is still considered for the potency.

However, all these immunochemical methods are not equivalent. The SRD test requires a pre-treatment which may alter the membrane-anchored trimers and result in a soluble or denatured form of the glycoprotein²²^,²⁹. Hence, SRD is not much efficient in discriminating between immunogenic and non-immunogenic glycoproteins resulting in an imperfect appraisal of the immunogenicity of a vaccine lot. By contrast, the ELISA test is more sensitive²², preserves the native structure of the glycoprotein, and is more appropriate to determine the content of the natively folded trimers of glycoprotein. The ELISA test can use either rabbit polyclonal or mouse monoclonal anti-glycoprotein antibodies purified or concentrated with ammonium sulfate. Studies have demonstrated good concordance between the NIH test and the antigen content evaluated by ELISA in vaccines and concluded that ELISA methods are suitable for the in vitro potency test. This advocates that ELISA tests might at least supplement or even replace the NIH test⁴^,²⁶^,²⁷^,³⁰^,³¹^,³²^,³³. Today, the European Pharmacopoeia recommends the use of validated serological or immunochemical assays as alternatives to the NIH test³. The complete avoidance of animal use for vaccine potency has become a realistic perspective.

The method presented below is based on an indirect ELISA sandwich immunocapture using a mouse monoclonal antibody clone (mAb-D1) which recognizes the antigenic sites III (aa 330 to 338) of the trimeric RABV glycoprotein¹⁵^,³⁴. This method was developed initially at the Institut Pasteur²⁶^,³⁰ then optimized and validated by the Agence Nationale de Sécurité du Médicament et des produits de santé (ANSM) laboratory, i.e., the French OMCL⁴^,³³. The mAb-D1 is used both for sensitizing the plate and subsequently for detecting the captured antigen. This allows for the specific quantification of the glycoprotein trimers, i.e., the immunogenic RABV antigen. The mAb-D1 used for the detection is labeled with peroxidase, which is revealed in the presence of the substrate and chromogen. Comparison of the absorbance measured for the tested vaccine and the reference vaccine allows for the determination of the immunogenic glycoprotein content. It is of note that the same type of assay can be applied for different mAbs recognizing different antigenic sites of the RABV glycoprotein³⁵. The method to obtain and purify or concentrate with ammonium sulfate anti-glycoprotein polyclonal rabbit immunoglobulins G (IgG) or monoclonal mouse globulins have been extensively described previously³⁶ along with the method to conjugate antibodies with peroxidase³⁷.

Protocol

1. Security precautions

NOTE: This method is applicable to both live RABV and inactivated vaccine.

Use good Laboratory Practice and Safety procedures.
Wear adequate Personal Protection Equipment (PPE) including disposable coat, gloves, mask, glasses, etc.
When the live virus is titrated, use a class II biological safety cabinet.
1. Consider any material in contact with the samples (reagents, washing solutions, etc.) as infectious material.
2. Treat the contaminated material by immersing in the bleach solution (2,5% of sodium hypochlorite) for 30 min for decontamination.
Handle chemicals in accordance with the Good Laboratory Practice.

2. Preparation

Use analytical grade reagents where ever possible.
Prepare fresh solutions of the coating buffer/carbonate buffer, passivation buffer, diluent and citrate buffer (Table 1), filter through 0.45 or 0.22 µm filters and store at 4 °C for one day prior to the use to preserve their analytical purity.
Allow reagents to reach to the room temperature (+18 °C to +25 °C) 30 min before the use and homogenize by gentle mixing prior to the use.

3. Microplate sensitization

NOTE: Use 96 well adsorption immunoassay plates which are optimized to bind high amounts of Immunoglobulins (e.g., see Table of Materials).

To each well, add 200 µL of the monoclonal antibody (mAb-D1) diluted in the carbonate buffer.
NOTE: An optimal concentration of about 1 µg/mL has been experimentally determined and corresponds to an approximate 1/2000 dilution of the purified mAb-D1. This recommended concentration is indicated for each mAb-D1 batch and must be periodically verified with the positive control.
Cover the plate with an adhesive film and incubate the microplate for 3 h at 37 °C in a humidified atmosphere.
Carefully aspirate and transfer the well content into a recipient containing 2,5% sodium hypochlorite solution.
Invert the microplate and let it dry on an adsorbent paper at room temperature for 5 min.

4. Microplate passivation

To each well, add 300 µL of the passivation buffer.
Cover the plate with an adhesive film and incubate for 30 min at 37 °C.
Aspirate carefully and transfer the well content into a recipient containing 2,5% sodium hypochlorite solution.
Invert the microplate and let it dry on an adsorbent paper at room temperature for 1 min.
NOTE: The microplate can be immediately used or stored sealed at -20 °C for up to 3 months until use.

5. ELISA assay

NOTE: For establishing the control curve of the reference vaccine, Step 5.3 is not required; to titrate the tested vaccine all Steps 5.1 to 5.6 are necessary.

Washing of the sensitized microplate
1. To each well, add 300 µL of the washing buffer.
2. Aspirate carefully and transfer the well content into a recipient containing 2,5% sodium hypochlorite solution.
3. Repeat steps 5.1.1 and 5.1.2 five more times to extensively wash the sensitized plate.
4. Invert the microplate and let it dry on an adsorbent paper at room temperature for 1 min.
Dilutions of the reference vaccine for the control curve
1. Reconstitute the reference vaccine (validation antigen Lot 09) in 1 mL of distilled water corresponding to a concentration of 10 µg/mL of rabies virus glycoprotein.
2. Prepare a ten-fold dilution of the reconstituted reference vaccine in the diluent to reach 1 µg/mL of rabies virus glycoprotein.
3. Prepare 6 serial two-fold dilutions of this reference vaccine in the diluent as indicated in Table 1.
4. Distribute 200 µL of the diluent in duplicate (wells 1H/2H) to serve as a blank control.
5. Distribute 200 µL per well of each reference vaccine dilution in duplicate (wells G1/G2 to A1/A2).
Dilutions of the tested vaccine for its titration
1. Prepare a ten-fold dilution of the tested vaccine in the diluent.
2. Prepare 7 two-fold serial dilutions of the tested vaccine in diluent as indicated in Table 2.
3. Distribute 200 µL per well of each tested vaccine dilution in duplicate (wells H3/H4 to A3/A4).
Incubation/Washing of the ELISA plate
1. Cover the microplate with an adhesive film and incubate for 1 h at 37 °C.
2. Remove the film, aspirate carefully and transfer the content of each well into a recipient containing 2,5% sodium hypochlorite solution.
3. To each well add 300 µL of washing buffer.
4. Aspirate carefully and transfer the content of each well into a recipient containing 2,5% sodium hypochlorite solution.
5. Repeat steps 5.4.3 and 5.4.4 five times to remove the unbound antigen and conserve the G protein trimers bound to the coated antibody (mAb D1).
6. Invert the microplate and let it dry on an adsorbent paper at room temperature for 1 min.
Binding of the peroxidase conjugated mAb-D1
1. Distribute 200 µL per well of the recommended dilution (1/2000) of peroxidase-labeled mAb-D1 in diluent (approximate concentration of 1µg/mL). A recommended concentration is indicated for each mAb-D1 batch and has to be periodically verified with the positive control.
2. Cover the microplate with an adhesive film and incubate for 1 h at 37 °C.
3. Remove the film, aspirate carefully and transfer the content of each well into a recipient containing 2,5% sodium hypochlorite solution.
4. Add to each well, 300 µL of the washing buffer.
5. Aspirate carefully and transfer the content of each well into a recipient containing 2,5% sodium hypochlorite solution.
6. Repeat steps 5.5.4 and 5.5.5 five times to remove unbound peroxidase-labeled antibody (mAb D1).
7. Invert the microplate and let it dry on an adsorbent paper at room temperature for 1 min.
Revelation using a substrate-chromogen
1. Distribute 200 µL per well of substrate-chromogen solution.
2. Seal the microplate with a film and incubate in the dark at room temperature for 30 min. A yellow-orange color develops the intensity of which is proportional to the amount of bound peroxidase-labeled antibody (mAb D1).
3. Stop the reaction by adding 50 µL of stopping solution per well.
4. Carefully wipe the bottom of the microplate and place it in a spectrophotometer to determine the optical density (OD) at 492 nm of all used wells: negative control (blank), reference vaccine and tested vaccine.
5. Collect OD data as .xls or .xlsx file format for analysis.
Draw the reference vaccine curve, the trimeric glycoprotein content, as a function of optical density (492 nm)
1. Calculate the mean OD at 492 nm for each duplicate at the different dilutions of the reference vaccine (wells G1/G2 to A1/A2 in Table 2)
2. Subtract the mean OD of the Blank (wells, H1/H2 in Table 2) from each calculated mean OD.
3. Plot the resulting OD values on the vertical axis (linear scale) and the corresponding concentration in glycoprotein trimers (ng/mL) on the horizontal axis (logarithmic scale).
4. Draw the reference curve by joining points (Figure 1).

Representative Results

In the following example the reference vaccine Lot 09, consisting of purified inactivated rabies virus particles (PV vaccine strain), is used. The glycoprotein trimers (10 µg/mL) content in this has been established after the determination of the total amount of viral proteins (BCA test) and evaluation of the percentage of the glycoprotein by SDS-polyacrylamide gel electrophoresis. Alternatively, a calibrated reference vaccine, e.g., the WHO 6^th International Standard (IS) for Rabies Vaccine (NIBSC code: 07/162), can be used.

Table 3 shows the OD values (492 nm) for a typical experiment. Using these values, the reference vaccine curve was drawn by plotting (1) the mean OD (minus the mean OD of the blank) at the different dilutions of the reference vaccine on the vertical axis (linear scale); (2) the concentration of the glycoprotein trimers (ng/mL) on the horizontal axis (logarithmic scale) (Figure 1).

The glycoprotein content of the tested vaccine is estimated by comparison to this reference vaccine curve. The evaluation is precise for the dilution of the tested vaccine giving a mean OD value in the linear part of the reference vaccine curve. In the presented experiment (Figure 1), the linear part is from about 1 to 2 OD (Table 3).

The dilution 1/40 of the tested vaccine, with a mean OD for duplicate samples of 1.534, is then appropriate for further evaluation. When this OD is plotted horizontally up to the point of intersection with the reference vaccine curve, the vertical projection on the x-axis corresponds to 500 ng/mL of glycoprotein. Considering the dilution, the tested vaccine contains

40 x 500 ng/mL = 20 µg/mL of trimeric glycoprotein.

Currently, there is no ELISA international unit for RABV glycoprotein content. However, the in vivo potency of the reference vaccine, in international units (IU/mL), has been established using the NIH test in comparison to the 6^th WHO International Standard (NIBSC code: 07/162)⁴. Consequently, the comparison of the mean ODs between the reference and the tested vaccine not only allows the measurement of the trimeric glycoprotein amount of the tested vaccine but also evaluates the in vitro potency estimated in Equivalent International Unit (EIU/mL).

Using the ELISA method described here, the French OMCL (ANSM) has monitored the glycoprotein content of more than 1000 batches of human rabies vaccine to be released in the market and has compared to the NIH test performed at the manufacturer's site⁴. The high variability of the NIH test, due to the heterogeneity in mice strain and challenge procedure³⁸, prevented a statistical correlation between the two tests. However, a concordance in the profile of results and the same pass/fail conclusions were obtained using in vitro and in vivo assays⁴. This concordance confirms that the native trimers of the glycoprotein recognized by the mAb-D1³⁴^,³⁹ constitute the main rabies virus immunogen inducing VNAbs during vaccination¹⁷. These VNAbs are able to protect mice from the intra-cerebral challenge of the NIH test. In conclusion, the in vitro appraisal of the rabies glycoprotein content is an attractive alternative to the NIH test evaluate rabies vaccine potency.

Buffers and reagents	Preparation
Coating buffer (Carbonate buffer 50mM pH=9.6)	Add Sodium carbonate 50 mM (Na₂CO₃-10H₂O) to Sodium bicarbonate 50 mM (NaHCO₃) until the desired pH (about 1/10 volume of sodium bicarbonate)
Passivation buffer	0.3% Bovine Serum Albumin (BSA, fraction V), 5% sucrose in carbonate buffer 50 mM pH 9.6
10x Phosphate buffered saline pH=7 (PBS 10x)	NaCl 80 g, KCl 2 g, Na₂PO₄-12H₂O 11.33 g, KH₂PO₄ 2g in 1L of distilled water. Adjust pH=7 with 4N NaOH
Washing buffer	0.05% Tween in 1x PBS
Diluent	0.5% Bovine Serum Albumin (Fraction V), 0.05% Tween in 1x PBS (adjust pH to 7 because of acidification by BSA)
Citrate buffer pH-5.6 (for peroxidase substrate)	11.67 g Tri-sodium citrate-2H₂O (Na₃C₆H₅O₇-2H₂0), 2.17 g Citric acid-1H20 in 1L of distilled water
Substrate-chromogen solution	50 mg Ortho-phenylene diamine tablet, 0.1% Hydrogen peroxide 30% (110 vol) in 25 ml Citrate buffer pH 5.6
Stopping solution (4N sulfuric acid)	10 ml H2SO4 36N in 80 ml cooled distilled water. Dilution must be carried out in an ice bath

Table 1. Buffers used in the assay.

1

2

3

4

5

6

7

8

9

10

11

12

A

Ref. Vaccine 1/640

Tested Vaccine 1/1280

B

Ref. Vaccine 1/320

Tested Vaccine 1/640

C

Ref. Vaccine 1/160

Tested Vaccine 1/320

D

Ref. Vaccine 1/10

Tested Vaccine 1/160

E

Ref. Vaccine 1/40

Tested Vaccine 1/80

F

Ref. Vaccine 1/20

Tested Vaccine 1/40

G

Ref. Vaccine 1/10

Tested Vaccine 1/20

H

Blank

Tested Vaccine 1/10

Reference vaccine Lot 09 dilution

Concentration (ng/mL)

1/640

15.625

1/320

31.25

1/160

62.5

1/80

125

1/40

250

1/20

500

1/10

1000

Table 2: Microplate plan for rabies glycoprotein titration assay and dilutions for the reference and tested vaccines used in the assay.

Reference vaccine Lot 09 dilution	Concentration (ng/mL)	Tested vaccine dilution	OD column 1	OD column 2	OD Mean	Reference OD mean- Blank OD mean
1/640	15.625	1/1280	0.17	0.16	0.165	0.0825
1/320	31.25	1/640	0.233	0.238	0.2355	0.153
1/160	62.5	1/320	0.378	0.387	0.3825	0.3
1/80	125	1/160	0.619	0.644	0.6315	0.549
1/40	250	1/80	1.006	1.077	1.0415	0.959
1/20	500	1/40	1.559	1.674	1.6165	1.534
1/10	1000	1/20	2.245	2.307	2.276	2.1935
Blank	Blank	1/10	0.078	0.087	0.0825

Table 3. Results obtained at OD₄₉₂ nm for plotting the reference curve

Figure 1: Reference curve showing the trimeric glycoprotein content as the function of the optical density (492 nm) for the reference vaccine.

Discussion

The epitope recognized by the mAb-D1 is located in the antigenic site III of the RABV glycoprotein which is not only immuno-dominant for the induction of VNAbs but also involved in neurovirulence, pathogenicity⁴⁰^,⁴¹ and receptor recognition⁴². There is another important antigenic site along the glycoprotein, site II⁴³, against which several MAbs have been isolated such as mAb-WI-1112³⁵. These can also be used in the similar type of experiments.

One limitation of this in vitro method by ELISA resides in the necessary conservation of the epitope recognized by the used mAb on the rabies virus strain to be tested. Up to now, all the classical strains used for human rabies vaccines are recognized by the mAb-D1. The greater diversity of rabies strains used in animal vaccines may constitute a problem in the future. However, as mentioned earlier, the same assay can be applied using different mAbs recognizing different antigenic sites of the RABV glycoprotein for coating and detection. This will allow circumventing the problem³⁵.

Another sensitive point of this method, quantifying the trimeric form of the RABV glycoprotein, is the possible effect of pH and temperature on the reversible conformation conversion¹⁴. These parameters are taken into account in the proposed protocol.

In summary, quantification of a highly immunogenic epitope of correctly folded glycoprotein trimers by in vitro ELISA method appears as effective as the NIH test to measure the capacity of a vaccine batch to induce humoral immunity against rabies virus infection. In addition, the ELISA method can discriminate sub-potent vaccine lots, in quality or in quantity, from the potent ones⁴^,³⁵. The last step before proposing such an in vitro ELISA assay to replace the NIH test is the organization of an international collaborative study for its improvement and standardization.

A workshop of theInteragency Coordinating Committee on the Validation of Alternative Methods (ICCVAM) entitled International Workshop on Alternative Methods to Reduce, Refine, and Replace the Use of Animals in Vaccine Potency and Safety Testing (Ames, September 2010)⁴⁴, concluded that the NIH test should be replaced by an alternative in vitro test evaluating the vaccine immunogenicity and able to discriminate between potent and sub-potent batches⁴. Another workshop of the European Partnership for Alternatives to Animal Testing (EPAA) in 2012⁴⁵ decided that a standardized sandwich ELISA calibrated against the current international rabies reference standard would be an ideal alternative for the rabies vaccine potency testing. Following, an international collaborative pre-validation study, which included both manufacturers and regulatory bodies, compared various ELISA designs used by the manufacturers and their National Control Laboratories for batch release for their ability to discriminate sub-potent from potent batches from different vaccine brands³⁵. An ELISA design combining mAb-D1 (antigenic site III) and a different mAb-WI-1112 (antigenic site II) was proposed by the European Directorate for the Quality of Medicines & HealthCare (EDQM) for a forthcoming international collaborative study under the umbrella of the Biological Standardization Program (BSP). This shows (1) that several combinations of mAbs for microplate coating and detection can be used as an in vitro alternative to the in vivo NIH test and (2) that these combinations can be dependent on the RABV vaccine strain to be tested.

Açıklamalar

The authors have nothing to disclose.

Acknowledgements

Dr. Sylvie Morgeaux and Dr. Jean-Michel Chapsal must be acknowledged for their key participation to establish the ELISA assay on vaccine batches and to organize international Workshops and collaborative studies. We thank Sabrina Kali for critical reading of the manuscript. Dr. Pierre Perrin was responsible for the isolation and characterization of mAb-D1. This work has been mainly supported by Institut Pasteur funding.

Materials

Class II Biological Safety Cabinet	ThermoFisher Scientific	10445753	if titrating live virus
Clear Flat-Bottom Immuno Nonsterile 96-Well Plates, 400 µL, MAXISORP	ThermoFisher Scientific	439454	good for binding to the loaded antibody
Equip Labo Polypropylene Laboratory Fume Hood	ThermoFisher Scientific	12576606	for the preparation of sulfuric acid
Immunology Plate Strong Adsorption MAXISORP Flat Bottom Well F96	Dutscher	55303	good for binding to the loaded antibody
Microplate Sealing Tape(100 sheets)	ThermoFisher Scientific	15036
Microplate single mode reader Sunrise	TECAN
Microplate shaker-incubator	Dutscher	441504
Microplate washer Wellwash	ThermoFisher Scientific	5165000
Multichannel pipette (30-300 µL) 12 channels	ThermoFisher Scientific	4661180N
Single Channel pipettes (Kit 2 : Finnpipettes F2 0.2-2 μL micro, 2-20 μL, 20-200 μL & 100-1000 μL)	ThermoFisher Scientific	4700880