Development of a Time and Cost Benefit Antibody Binding Test-Based Method for Determination of Rabies Vaccine Potency

Abstract

This study is an improvement on the antibody binding test, known as ABT method, to develop a simple and fast method in comparison with NIH for determination of rabies vaccine potency. In the current study, several commercial human and veterinary vaccines were tested using both modified ABT and NIH methods. The ED₅₀ was calculated using the probit method and the relative potency of each vaccine was measured based on the reference vaccine. The test was repeated four times to calculate the reproducibility of the method. Statistical analysis indicated that there was no significant difference between the result obtained from NIH and modified ABT method for either human or veterinary vaccines (p > 0.05). In addition, the linearity of the method (R ²) was calculated as 0.94 by serial dilution of a test vaccine. Coefficient variances were determined as less than and more than 10% for the human and veterinary rabies vaccines, respectively. In conclusion, the findings suggest that the modified method could be considered as an alternative approach for rabies vaccine potency determination in in-process quality control tests at industrial scale. It is a time and cost benefit method and accuracy may further be increased by employing monoclonal antibodies against trimeric form of G glycoprotein. However, the use of serum samples may be useful compared with an artificial mix of antibodies because other components from the serum samples could have a positive impact on cell sensitivity and mimic more the complexity of the immune response. Although the modified test has solved a fundamental problem, it is still not sensitive enough for veterinary vaccine assessment and needs further modifications to obtain the acceptability criteria.

Introduction

Rabies is a viral zoonosis disease where veterinary and wild animals are considered as the sources of infection. Rabies is an acute and progressive disease that attacks the central nervous system and with the onset of symptoms, there is no possibility of recovery (28). On average, every 15 min, someone dies of rabies in the world, and in 2016, the World Health Organization (WHO) reported tens of thousands of people die annually due to rabies (29). It should be noted that death often occurs in developing countries (20). Over 2.5 billion people in the world are located in endemic areas and there is a permanent risk for them (14). Fortunately, rabies is a preventable disease and vaccination is one of the main strategies to control the disease. Annually, over 15 million people in the world are receiving multidose postexposure prophylaxis to prevent the disease (19).

Due to the low immunogenicity of some rabies vaccines, which leads to insufficient protective response and subsequent death, the most important quality control (QC) test is determination of potency (24). NIH method is recommended by both pharmacopoeias and WHO as the main method for determination of rabies vaccine potency (3,10). However, the NIH test, like other in vivo methods, has some limitations such as (a) variable results (due to the use of animal, dilution manner of the test, peritoneal and intracerebral injections, and use of challenge virus at different passages), (b) time-consuming (minimum 28 days for every test), (c) large number of mice (usually 200 mice per test), (d) multiple facilities (for the purchase and maintenance of animals), (e) not economical, (f) stress on the animals, (g) unknown variables in the test (e.g., the role of virus-neutralizing antibodies in resistance against intracerebral injection of the virus is not clear), and (h) intracerebral injection in the NIH test is not similar to the natural transmission of virus (4).

Because of mentioned obstacles, many investigations have been done to find a suitable in vitro test for replacement of NIH method (6,9,17,21). Different studies have proven the key role of G glycoprotein in induction of a potent immune response against rabies virus (30). Moreover, it has become clear that there is a strong relationship between the amount of G glycoprotein and the potency of a rabies vaccine (7,25). Hence, measurement of G glycoprotein content is considered as an appropriate method to evaluate the potency of the rabies vaccine (12). Methods that have been proposed in this area include single radial immunodiffusion assay, enzyme-linked immunosorbent assay (ELISA), and antibody binding test (ABT) (26). Among these techniques, the ABT method is one of the most rapid and economic antigen quantification tests and assesses the antigenic content of a vaccine after neutralization with a standard antibody. ABT was introduced for the first time by Arko in 1972 (2). Later, the test was modified to in vitro by replacement of mice with cells, and at the end of the test, a virus titration assay was done based on plaque-forming unit (PFU) in the cells. This method was called plaque reduction method (2,18). Since the rabies virus is not cytopathogenic and the plaques were formed in a very long time, Barth modified the test by replacement of PFU with fluorescent staining of infected cells in 1981 (5). After that the ABT method was not further studied by other scientists. The current study is an effort to reform the ABT method and develop a modified test for determination of either human or veterinary rabies vaccine potency.

Materials and Methods

Reagents, media, and vaccines

Four potent human vaccines (not containing any adjuvant), four potent veterinary vaccines (containing alum adjuvant), and two subpotent vaccines were used as substandard vaccines (Table 1). MEM medium and fetal bovine serum (FBS) were purchased from GIBCO, Thermo Fisher Scientific. Sodium chloride, Na₂HPO₄, and KH₂PO₄ powders were purchased from Sigma-Aldrich. Challenge virus strain (CVS) and BSR cell line were obtained from WHO collaborating center for reference and research on rabies of Pasteur Institute of Iran. Antinucleocapsid polyclonal antibody conjugated with FITC and antirabies serum were purchased from Bio-Rad and Haffkine Bio-Pharmaceutical Corporation, respectively. One hundred percent substandard vaccine was prepared by heating of Rabipur^® vaccine at 90°C for 30 min. Fifty percent substandard vaccine was prepared by 1:1 mix of completely denatured Rabipur vaccine with the original Rabipur vaccine.

Table 1.

Detailed Information of Test Vaccines Used in the Study

Vaccine name	Type	Lot no.	Country of manufacture	Manufacturer
Abhayrab	Human	AYB 43/13	India	Human Biologicals Institute
Rabivax	Human	031L2018	India	Serum Institute of India
Rabivax	Human	031L2019	India	Serum Institute of India
Rabipur	Human	2078	Germany	Novartis Vaccines and Diagnostics GmbH
Rabisin	Veterinary	L401147	France	Merial
CANVAC^® R	Veterinary	121011	Czech Republic	Dyntec
Pasteur Institute of Iran	Veterinary	90-1	Iran	Pasteur Institute of Iran
Pasteur Institute of Iran	Veterinary	90-2	Iran	Pasteur Institute of Iran
Rabipur (100% Substandard)^a	Human	—	—	—
Rabipur (50% Substandard)^b	Human	—	—	—

Completely (100%) denatured Rabipur vaccine by heating at 90°C for 30 min.

1:1 mix of completely denatured Rabipur vaccine with the original (potent) Rabipur vaccine.

NIH test

The NIH test was carried out according to WHO guidelines. Briefly, 4-week-old SW mice (Pasteur Institute of Iran) were used for the assay. The animal experiment protocol was approved by the Review Board and Ethics Committee of Pasteur Institute of Iran and conformed to the European Communities Council Directive of November 1986 (86/609/EEC) in such a way to minimize the number of animals used and their suffering. The mice randomly divided into 10 groups and cages were placed randomly in different places to minimize the effect of the environment (e.g., temperature and light) on immune responses. Serial dilutions (1/5, 1/25, 1/125, and 1/625) of test and reference vaccines using NIH buffer (0.83 g NaCl, 0.11 g Na₂HPO₄, and 0.027 g KH₂PO₄ in 100 mL of distilled water adjusted to pH 7.4) were prepared and 0.5 mL of each dilution was intraperitoneally injected into the mice. Injections were performed in two doses with an interval of 1 week. Seven days after the last injection, 0.03 mL CVS (20 LD₅₀, LD₅₀ = 10^−6.8) was injected intracerebrally into the immunized mice. To determine the actual LD₅₀ of injected virus, serial dilutions (10⁻⁴, 10⁻⁵, 10⁻⁶, and 10⁻⁷) of CVS in NIH buffer were prepared and subsequently injected intracerebrally into the mice and finally the actual LD₅₀ was calculated. For inactivity test, reference and test vaccines were also injected intracerebrally into 10 mice. Five days after the last injection, the mice were monitored and the number of diseased, paralyzed, and dead mice was recorded for 20 days. After the test was finished, the diseased or paralyzed mice were euthanized. To confirm the death by rabies virus, an immunofluorescence staining was performed on dead mouse brains. Finally, mouse survival was compared with the reference vaccine and relative potency was calculated using the probit method (23). The cutoff values of approved vaccines are at least 2.5 and 1 IU for human and veterinary vaccines, respectively.

Immunofluorescence staining of mouse brains

Immunofluorescence staining of mouse brains was carried out according to WHO guideline. After isolation of mouse brains and impression on glass slides, the cells were fixed with cold acetone. Then, 100 μL of antinucleocapsid ployclonal antibody conjugated with FITC (Bio-Rad) was added on the glass slides and the glass slides were incubated for 1 h in wet condition at 37°C. After washing with phosphate-buffered saline (PBS) and drying, the slides were observed with a fluorescence microscope for fluorescence-stained infected cells (8).

ABT procedure on human rabies vaccines

Serial dilutions (1:3, 1:9, 1:27, 1:81, and 1:243) of the test and reference vaccines with a final volume of 50 μL were prepared in NIH buffer. Fifty microliters of rabies antiserum (0.2 IU/mL) was then added to each dilution and incubated for 1 h at 37°C. Again, 50 μL of CVS (5000 TCID50, TCID50 = 10^−6.17) was added to each dilution and incubated for one additional hour at 37°C. Finally, 50 μL of BSR cell suspension (50,000 cells in MEM supplemented with 10% FBS/well) was added to each dilution and the mixtures were transferred into a 96-well plate and incubated for 24 h in a 5% CO₂ incubator. In addition, positive control (without serum), negative control (without virus), serum control, and virus control (without vaccine) were examined as triplicate. After fixation with cold acetone, cells were stained by FITC-conjugated polyclonal antibodies (Bio-Rad) and studied using florescence microscopy. The number of fluorescent foci were counted and compared together with the endpoint 50% method. Finally, relative potency was calculated using the probit method (31). The cutoff value of approved human vaccines is at least 2.5 IU.

ABT procedure on veterinary rabies vaccines

Because veterinary vaccines contain alum adjuvant, which has a potential destructive effect on cells, the method was carried out with some modifications. Briefly, 50,000 BSR cells were added to each well of a 96-well plate and incubated at 37°C for 24 h. The prepared mixtures were then added to each well as the standard ABT method. To avoid toxicity of adjuvant, after an hour, the mixture was removed from the well and cells were washed by sterile PBS and reincubated for 24 h at 37°C. The rest of the procedure was practically the same as the standard ABT method for human rabies vaccines. The cutoff value of approved veterinary vaccines is at least 1 IU.

Statistical analysis

The potency of in vitro and in vivo results was assessed by the probit method using bioassyst software, version 3. The correlation between two tests was calculated using Excel (Microsoft Office 2010). The relationship between each pair of variables was performed by SPSS software, version 16.0. Results are reported as mean ± standard deviation (SD) for quantitative variables. Differences between NIH and ABT results were analyzed using the two-tailed paired Student's t-test where the p-value cutoff or alpha level (α) is 0.05. A value of p > 0.05 was considered statistically significant.

Results

Comparison of calculated potencies

Potencies, calculated by NIH and ABT methods, are shown in Table 2. Statistical analysis indicates there is no significant difference between the results obtained from NIH and ABT methods for either human or veterinary vaccines (all p-values were more than 0.05).

Table 2.

Calculated Relative Potencies of Human and Veterinary Rabies Vaccines by ABT and NIH Methods

Vaccine name	NIH method (IU)	ABT method (IU)	p
Abhayrab	13.87	14.66 ± 0.59	0.1739
Rabivax (031L2018)	9.74	11.30 ± 0.66
Rabivax (031L2018)	7.47	9.02 ± 0.39
Rabipur	12.93	14.22 ± 0.27
Rabisin	3.67	2.70 ± 0.12
CANVAC R	1.42	0.84. ± 0.07
Pasteur Institute of Iran (90-2)	1.11	0.52 ± 0.09
Pasteur Institute of Iran (90-1)	1.24	0.75 ± 0.07
100% Substandard	0.26	0.65 ± 0.09
50% Substandard	4.35	7.64 ± 0.21

The difference between groups was not significant based on two-tailed paired Student's t-test (p = 0.1739). Data are represented as mean ± SD and p-value less than 0.05 is assumed as a statistically significant difference between two methods.

ABT, antibody binding test; SD, standard deviation.

Linearity test

In the present study, for linearity testing, Rabipur vaccine was used. Serial dilutions (1, 1:2, 1:4, 1:8, and 1:16) of the vaccine were prepared in NIH buffer. The potency of each dilution was determined by ABT method. Finally, dilutions versus obtained potencies were plotted and linearity of the method was calculated as 0.94 using the measurement of regression (R ²) between points (Fig. 1).

FIG. 1.

The linearity of the method was investigated by linear regression analysis between ABT-calculated potencies and dilutions of a test vaccine. ABT, antibody binding test.

Reproducibility test

To test the reproducibility (test–retest) of the method, potencies of either human or veterinary vaccines were determined four times a day and then the mean, SD, and coefficient variance were calculated. Coefficient variances were determined less than 10% and more than 10% for human (Table 3) and veterinary rabies vaccines (Table 4), respectively.

Table 3.

ABT Repeatability in Human Rabies Vaccines

Vaccine name (lot no.)	Rabipur (2078)	Rabivax (031L2019)	Rabivax (031L2018)	Abhayrab (AYB 43/13)
First iterance	14.2	8.94	11.16	14.62
Second iterance	13.86	9.27	11.37	14.98
Third iterance	14.51	8.5	10.58	15.2
Fourth iterance	14.3	9.35	12.18	13.86
Mean	14.22	9.02	11.32	14.66
Standard deviation	0.27	0.39	0.66	0.59
Coefficient variance (%)	3.60	6.90	7.20	5.20

Table 4.

ABT Repeatability in Veterinary Rabies Vaccines

Vaccine name (lot no.)	CANVAC R (121011)	Pasteur Institute of Iran (90-2)	Pasteur Institute of Iran (90-1)	Rabisin (L401147)
First iterance	0.91	0.46	0.67	2.63
Second iterance	0.87	0.62	0.82	2.21
Third iterance	0.77	0.43	0.71	2.84
Fourth iterance	0.79	0.55	0.78	3.1
Mean	0.835	0.515	0.745	2.695
Standard deviation	0.07	0.09	0.07	0.38
Coefficient variance (%)	30	57	34	22

Correlation test

The correlation between NIH and modified ABT method was measured using Pearson coefficient. It was found that there is a significant correlation between the two tests (p = 0.00, R = 0.99) (Fig. 2).

FIG. 2.

The correlation between NIH and modified ABT data sets was measured using Pearson coefficient and a significant correlation between the two data sets was observed.

Discussion

In the field of vaccinology, one of the main QC tests is to determine the potency of a vaccine. To determine the potency, parameters such as protective efficacy, immunogenicity, and antigenicity could be checked. To assess the protective efficacy of a vaccine, survival after immunization with the antigen is monitored. In approaches that are designed by measuring immunogenicity, the vaccine is injected into the animals and immune responses are assessed according to which potency of the vaccine is calculated. Evaluation of antigenicity of vaccine is another strategy to determination of vaccine potency. In this approach, the major antigen in the vaccine, which is responsible for immune responses, is measured and the potency of vaccine was then calculated based on the amount of the quantifiable antigen. In such methods, the antigen amount would have a direct and quantitative relationship with the potency value of vaccine. According to 3R guideline (reduce, refine, and replace), in vivo tests need to be replaced by in vitro methods (27). Among the above strategies, only methods that are planned based on antigenicity are consistent with the 3R guideline. However, it should be noted that the potency of any vaccine cannot be determined by evaluating antigenicity, and it is required to take many investigations to prove a strong correlation between antigenicity and potency.

According to WHO, the NIH test is the main approach for determination of potency of veterinary and human rabies vaccines (23). Since the method is an in vivo method, large numbers of mice are used and this prolongs the duration of the test (4). Therefore, according to the 3R guideline, the method needs to be replaced by a suitable in vitro test. It must be emphasized that based on ICH Guideline Q2 (R1), the accuracy, precision, repeatability, reproducibility, sensitivity, specificity, linearity, acceptable ranges, and robustness of the intended test must be analyzed. Different methods have been proposed that measured the amount of G glycoprotein in an inactivated rabies vaccine. Among suggested approaches, the antibody binding test (ABT) was a fast, affordable, and accurate method.

In the current investigation, a new modified ABT method was developed for determination of rabies vaccine potency. It must be emphasized that the change in the virus strains is the most important modification in the current method. Barth used Hep Flury virus, but in the current study, CVS (which is also used in the NIH test) is applied. Other modification is the use of a serum containing neutralizing antibodies obtained from vaccinated people against rabies. For assessment of the test, potency of several human and veterinary rabies vaccines was determined by both modified ABT and NIH test. A fairly good correlation between the results of two methods was observed. However, the data revealed that for vaccines with less potency than 10 IU/mL (determined by NIH test), the difference between the results of ABT and NIH is more. It seems that the soluble monomer G glycoproteins are the reason of this difference (11,22). In rabies vaccine production, one of the most important steps is deactivation of the virus with minimal structural changes. So, antigens responsible for immune responses, especially trimeric G glycoprotein, remain intact and maintain their three-dimensional structures (15). Several studies have shown that spatial structure of trimeric antigens is a very important factor in stimulation of a potent immune response against rabies virus. When the G glycoprotein misses its trimeric structure and comes to the monomeric form, it rapidly loses immunogenicity (11). It can be concluded that one of the fundamental problems in the ABT method for accurate determination of potency is antibodies, which are not able to distinguish between monomeric and trimeric forms of antigen. It must be noted that lack of differentiation between monomeric and trimeric forms of G glycoprotein is not only related to the ABT method but also is a problem for all methods that are designed based on antigenicity assay. One of the feasible strategies to overcome this problem is filtration of the vaccine before carrying out the test to remove soluble antigens. Vaccine filtration is a good strategy, which other researchers have also used, for example, Gamoh carried out vaccine filtration to remove soluble antigens before determination of potency by an ELISA test. Another strategy is usage of monoclonal antibodies, which specifically recognize only the trimeric form of G glycoprotein. Initially, in the ELISA tests, polyclonal antibodies were used and gradually polyclonal antibodies were replaced by monoclonal antibodies. For the first time, Lafon used monoclonal antibodies in 1985 (16). However, the use of these monoclonal antibodies did not seem to be useful because these antibodies still had the ability of binding to monomeric form of G glycoprotein. Thus, the use of monoclonal antibodies just against trimeric form of G glycoprotein was considered by Gibert et al. (13). However, the use of serum samples may be useful compared with an artificial mix of antibodies because other components from the serum samples could have a positive impact on cell sensitivity and mimic more the complexity of the immune response. Moreover, even with the use of antitrimeric form monoclonal antibodies in ELISA, finding a good correlation between ELISA and NIH seems unattainable because of high variation in the latter test (1).

For evaluation of linearity of the test, serial dilutions of a commercial vaccine were prepared and each dilution was separately evaluated by the test. As expected, the results revealed a direct relationship between either the amount of antigen present in the vaccine or the diluted vaccines and their potencies. Other results also indicated that the modified test has high reproducibility. In contrast to the NIH test, the difference between obtained results on a vaccine, either at different times or days, is very negligible.

Preliminary results obtained from the modified ABT on veterinary vaccines demonstrated that alum, which is used as an adjuvant in veterinary vaccines, has a toxic effect on cells in the assay. Therefore, the ABT assay was carried out for veterinary vaccines with some modifications. Comparison of the results of two tests for veterinary vaccines showed a lower value of potency when the ABT method is utilized. This is by reason of the fact that for in vitro procedures based on measurement of antigen content such as the ABT method, the synergic effect of the adjuvant on immune system boosting cannot be measured. Therefore, the developed test is not accurate enough for determination of veterinary rabies vaccine potency and consequently did not fulfill the acceptance criteria for other adjuvant-containing vaccines and more modifications are needed to overcome the mentioned obstacle.

Conclusion

The findings suggest that the modified ABT method, developed in the current study, could be considered as an alternative approach for human rabies vaccine potency determination, especially in in-process quality control tests at industrial scale. It has time and cost benefits and has high reproducibility and its accuracy may be increased with using antibodies just against trimeric form of G glycoprotein. However, the use of serum samples may be useful compared with an artificial mix of antibodies because other components from the serum sample could have a positive impact on cell sensitivity and mimic more the complexity of the immune response. It should be highlighted that although the modified test has solved the toxicity of alum-containing vaccines on cells, it still is not sensitive and accurate enough to detect adjuvanticity effect of adjuvant-containing vaccines in vitro.

Footnotes

Acknowledgments

The authors are grateful to the WHO collaborating center for reference and research on rabies of Pasteur Institute of Iran for technical support and services. Authors note that this work was carried out in Pasteur Institute of Iran, which was not supported by any governmental funding.

Author Disclosure Statement

No competing financial interests exist.

References

Abhinay

, Dessain

, Srikanth

, et al. A novel site-II directed glycoprotein estimation ELISA to aid rabies vaccine manufacture for veterinary and human use. Vaccine, 2014; 32:209–213.

Arko

, Wiktor

, and Sikes

. Laboratory techniques in rabies: the antibody binding test for vaccine potency. Monogr Ser World Health Organ, 1973:292–294.

Asgary

, Shoari

, Baghbani-Arani

, et al. Green synthesis and evaluation of silver nanoparticles as adjuvant in rabies veterinary vaccine. Int J Nanomedicine. 2016; 11:3597–3605.

Barth

, Diderrich

, and Weinmann

. NIH test, a problematic method for testing potency of inactivated rabies vaccine. Vaccine, 1988; 6:369–377.

Barth

, Gross-Albenhausen

, Jaeger

, and Milcke

. The antibody-binding-test, a useful method for quantitative determination of inactivated rabies virus antigen. J Biol Stand, 1981; 9:81–89.

Bruckner

, and Kihm

. Alternatives for potency testing of veterinary vaccines: rabies vaccine as a model. Dev Biol Stand, 1986; 64:93–97.

Cox

, Dietzschold

, and Schneider

. Rabies virus glycoprotein. II. Biological and serological characterization. Infect Immun, 1977; 16:754–759.

Dean

, and Abelseth

. Laboratory techniques in rabies: the fluorescent antibody test. Monogr Ser World Health Organ, 1973:73–84.

Eng

, Fishbein

, Talamante

, et al. Immunogenicity of rabies vaccines used during an urban epizootic of rabies in Mexico. Vaccine, 1994; 12:1259–1264.

10.

Fontana

, Kratje

, Etcheverrigaray

, and Prieto

. Immunogenic virus-like particles continuously expressed in mammalian cells as a veterinary rabies vaccine candidate. Vaccine, 2015; 33:4238–4246.

11.

Gamoh

, Senda

, Itoh

, et al. Use of ELISA for in vitro potency test of rabies vaccines for animal use. Biologicals, 1996; 24:95–101.

12.

Gamoh

, Shimazaki

, Senda

, et al. Establishment of a potency test by ELISA for a rabies vaccine for animal use in Japan. J Vet Med Sci, 2003; 65:685–688.

13.

Gibert

, Alberti

, Poirier

, et al. A relevant in vitro ELISA test in alternative to the in vivo NIH test for human rabies vaccine batch release. Vaccine, 2013; 31:6022–6029.

14.

Hampson

, Coudeville

, Lembo

, et al. Estimating the global burden of endemic canine rabies. PLoS Negl Trop Dis, 2015; 9:e0003709.

15.

Koraka

, Bosch

, Cox

, et al. A recombinant rabies vaccine expressing the trimeric form of the glycoprotein confers enhanced immunogenicity and protection in outbred mice. Vaccine, 2014; 32:4644–4650.

16.

Lafon

, Perrin

, Versmisse

, and Sureau

. Use of a monoclonal antibody for quantitation of rabies vaccine glycoprotein by enzyme immunoassay. J Biol Stand, 1985; 13:295–301.

17.

Lyng

, Bentzon

, Ferguson

, and Fitzgerald

. Rabies vaccine standardization: International Collaborative Study for the Characterization of the fifth International Standard for Rabies Vaccine. Biologicals, 1992; 20:301–313.

18.

Meslin

F-X

, Kaplan

, and Koprowski

. Laboratory Techniques in Rabies. 4th ed. Geneva: World Health Organization, 1996, 476 pp.

19.

Moore

, Davis

, Kang

, et al. Comparison of anamnestic responses to rabies vaccination in dogs and cats with current and out-of-date vaccination status. J Am Vet Med Assoc, 2015; 246:205–211.

20.

Muhamuda

, Madhusudana

, and Ravi

. Development and evaluation of a competitive ELISA for estimation of rabies neutralizing antibodies after post-exposure rabies vaccination in humans. Int J Infect Dis, 2007; 11:441–445.

21.

Perrin

, Morgeaux

, and Sureau

. In vitro rabies vaccine potency appraisal by ELISA: advantages of the immunocapture method with a neutralizing anti-glycoprotein monoclonal antibody. Biologicals, 1990; 18:321–330.

22.

Qian

, Aguilar

, Wang

, and Qiu

. Secretion of truncated recombinant rabies virus glycoprotein with preserved antigenic properties using a co-expression system in Hansenula polymorpha. J Microbiol, 2013; 51:234–240.

23.

Seligmann

Jr.

Laboratory techniques in rabies: the NIH test for potency. Monogr Ser World Health Organ, 1973:279–286.

24.

Servat

, Kempff

, Brogat

, et al. A step forward in the quality control testing of inactivated rabies vaccines—extensive evaluation of European vaccines by using alternative methods to the in vivo potency tests. Altern Lab Anim, 2015; 43:19–27.

25.

Sigoillot-Claude

, Battaglio

, Fiorucci

, et al. A versatile in vitro ELISA test for quantification and quality testing of infectious, inactivated and formulated rabies virus used in veterinary monovalent or combination vaccine. Vaccine, 2015; 33:3843–3849.

26.

Stokes

, McFarland

, Kulpa-Eddy

, et al. Report on the international workshop on alternative methods for human and veterinary rabies vaccine testing: state of the science and planning the way forward. Biologicals, 2012; 40:369–381.

27.

Ukelis

, Kramer

, Olejniczak

, and Mueller

. Replacement of in vivo acute oral toxicity studies by in vitro cytotoxicity methods: opportunities, limits and regulatory status. Regul Toxicol Pharmacol, 2008; 51:108–118.

28.

Welch

, Anderson

, and Litwin

. An evaluation of two commercially available ELISAs and one in-house reference laboratory ELISA for the determination of human anti-rabies virus antibodies. J Med Microbiol, 2009; 58:806–810.

29.

WHO.. Fact Sheet; Rabies [Online]. March. 2016. WHO Media centre. Available at www.who.int/mediacentre/factsheets/fs099/en/ (accessed December 7, 2016 ).

30.

Wiktor

, György

, Schlumberger

, Sokol

, and Koprowski

. Antigenic properties of rabies virus components. J Immunol, 1973; 110:269–276.