Classical swine fever, a highly contagious disease of swine caused by a virus of the family Flaviviridae, is associated with a high mortality rate and seriously threatens the health of swine on farms.1,2 In China, vaccination against CSFV infection is generally used prophylactically to reduce morbidity and mortality rates among swine.2 Development of a CSFV vaccine has included assessment of early inactivated-virus vaccines, attenuated-virus vaccines, and genetically engineered vaccines.3,4 The attenuated-virus vaccine approved for use in China is a live attenuated–virus vaccine based on a Chinese (C) strain and developed by the Chinese Veterinary Drug Control Institute.5 The widespread use of this vaccine has caused variation in the antigens of the CSFV epidemic strain over time, resulting in atypical epidemic forms of disease, such as chronic or recessive infection.6,7 As a result, the CSF epidemic in China has been difficult to control, and immune failure after vaccination not only prevents the eradication of the disease in swine but also impedes their export.8,9 Therefore, a new genetically engineered marker vaccine with enhanced safety and efficiency that allows distinction of vaccinated swine from infected swine is urgently needed for the swine industry worldwide.
According to Veterinary Biologics Regulations of the People's Republic of China10 and People's Republic of China Veterinary Pharmacopoeia,11 the current test of efficacy of vaccines against the CSFV C strain relies mainly on the use of rabbits or swine. Swine are sensitive to subunit vaccines; consequently, results of efficacy testing of such vaccines can directly reflect those achievable with clinical application. No commercial SPF standard exists for testing swine in China, so it is difficult to purchase pigs with consistent health characteristics between batches. Additional disadvantages of vaccine efficacy testing in swine include the high cost and challenges of using swine, its time-consuming nature, and the possibility of a severe biosafety breach with the Shimen strain of CSFV. Rabbits are sensitive to the C strain, and efficacy testing during the production process for live attenuated–CSFV vaccines is mostly based on detection of pyrexia in rabbits.2 Domestic standards already exist for vaccine testing in rabbits, and the availability of rabbits suitable for such testing is high. Furthermore, testing in rabbits is easier, more reliable, and less expensive to conduct than testing in swine.
The purpose of the study reported here was to establish a method for evaluation of the efficacy of a CSFV subunit E2 vaccine in rabbits as determined via humoral immune responses after experimental challenge with the CSFV C strain 21 days after vaccination. Specifically, we sought to determine the immunoprotective efficacy and minimum immunoprotective dose of the vaccine in rabbits as assessed by rectal temperature, antibody response against CSFV, and detection of the C strain in blood samples by means of a qRT-PCR assay. A second objective was to determine whether the administered dose of the vaccine was correlated with the antibody response and with the rate of protection against viral challenge.
Materials and Methods
Animals
Two experiments were performed involving SPF female rabbits (body weight, 1.5 to 3.0 kg) purchased from a commercial breeder.a During the study, rabbits were housed in filter-top cages in an air-conditioned room of the Yebio experimental animal center. The study protocol and biosecurity methods were approved by the Animal Care and Use Committee of Yebio (permit No. 20110017) and performed in accordance with the Guidelines for Experimental Animals of the Department of Science and Technology of Shandong Province, China. At the end of the study, all rabbits were euthanized by IV administration of an overdose of pentobarbital.
Vaccine
The vaccine evaluated in the study was a genetically engineered 293T-E2 that was undergoing certification (approval document No. 2017007) by National Institutes for Food and Drug Control. To produce the vaccine, HEK-293T-E2 engineered cells were constructed as follows. The lentivector pCDH-CMV-MCS-EF1-GFP/Puro expressing the CSFV E2 protein was constructed, and a lentivirus carrying the CSFV E2 protein was packaged in HEK-293T cells.b Then HEK-293T cells were infected with a lentivirus carrying the CSFV E2 protein. The cells were monoclonalized by use of the limiting dilution method, expression levels of the cloned cells were detected by dot blot assay, and a cell clone capable of efficiently expressing the CSFV E2 protein was selected. Afterward, the selected cells were grown in flasks (surface area, 75 cm2) with Dulbecco modified Eagle mediumc containing 10% fetal bovine serumc for 24 hours at 37°C with 5% CO2.
When the monolayer appeared well grown, the original culture medium was discarded, expression mediumd was added, and cells were incubated at 37°C with 5% CO2 for 7 days. The cell culture medium was collected and centrifuged at 5,000 × g for 5 minutes. The supernatant, namely E2 protein solution, was then separated and processed via SDS-PAGE. The purity of E2 protein was analyzed by use of imaging equipment,e the total protein content in cell culture was detected with a Bradford protein detection kit,f and the concentration of E2 protein was determined to be 293.4 μg/mL. Thereafter, the E2 protein concentration was diluted to 50 μg/mL with sterile saline (0.9% NaCl) solution and mixed with adjuvantg (mass ratio of 1:1) to yield 3 batches of vaccine (E2 protein concentration, approx 23 μg/mL), numbered 01, 02, and 03. In accordance with the People's Republic of China Veterinary Pharmacopoeia,12 samples of the vaccine batches were submitted for sterility and Mycoplasma testing, with results indicating that the batches were free of bacteria and Mycoplasma growth.
Identification of the minimum dose of E2 protein required for immunization
In the first experiment (experiment 1), 40 rabbits were randomly assigned to 4 groups of 10 rabbits each by means of random number table. Rabbits assigned to receive the 01 batch of vaccine were inoculated SC in the neck region with vaccine doses representing 1.15, 2.3, or 4.6 μg of E2 protein/rabbit, which amounted to 0.05, 0.1, or 0.2 mL of vaccine, respectively. The fourth group of rabbits (negative control group) was inoculated SC in the neck region with saline solution.
At 21 days after vaccination, blood samples were collected for detection of serum antibody against CSFV. The tested rabbits in each group then underwent a viral challenge by injection with l mL of the CSFV C strainh (150 times the minimum infective dose for rabbits) in an ear vein.11 Rectal temperature was measured twice a day beginning 2 days before the challenge, then every 6 hours for 4 days after the challenge. The vaccine protection rate was quantified after challenge by monitoring of rabbits for pyrexia (ie, rectal temperature ≥ 40.5°C) and determining the proportion in each group free of pyrexia as well as by postmortem detection of viral loads of the C strain by means of qRT-PCR assay. From these results, the minimum dose of the vaccine required to achieve immunization against CSFV (ie, the minimum immunization dose) was determined.
Reproducibility of results achieved with the minimum immunization dose of vaccine
In the second experiment (experiment 2), the identified minimum immunization dose of the 01 batch of the vaccine was evaluated for reproducibility of the immune response. Another 30 SPF female rabbits (body weight, 1.5 to 3.0 kg) were randomly allocated to 3 groups (10 rabbits/group) for inoculation SC with the 02 or 03 batches of the vaccine at the identified minimum immunization dose or saline solution. The same procedures were followed as for the first experiment.
Serologic analyses
Serum was harvested from blood samples collected 21 days after vaccination for measurement of antibody against CSFV by use of a commercially available CSFV ELISAi and the manufacturer's instructions. Results are reported as the antibody blocking rate; a blocking rate ≥ 40% was considered positive for an antibody response, a rate between 30% and 40% was considered suggestive, and a rate ≤ 30% was considered negative. All samples were tested in triplicate and read at 450 nm with an ELISA plate reader.f
Serum samples also underwent FAVN testing for measurement of the protective humoral antibody titer.13 Briefly, serum samples were inactivated for 30 minutes at 56°C. They were then serially diluted 1:2 in 50 μL of cell culture medium (Earle minimal essential medium, 5% fetal bovine serum, and antimicrobials) in wells of a 96-well plate and mixed with 100 TCID50 of CSFV Shimen strain suspension. Plates were incubated for 1 hour at 37°C in 5% CO2, and 50 μL of PK-15 cell suspension (containing approx 2 × 105 cells/mL) was added to all wells. Plates were incubat for 3 days at 37°C in 5% CO2 until the cells reached 80% to 90% confluency. Afterward, plates were fixed in 20% acetone in PBSS for 10 minutes and thoroughly dried at 25°C to 30°C for 4 hours. For detection of virus neutralization, 50 μL of hyperimmune porcine CSFV antiserumj (1:100) was added to each well and incubated for 45 minutes at 37°C in 5% CO2. Thereafter, 50 μL of fluorescein antibodyk (1:2,000) was added to each well, and plates were incubated for 30 minutes at 37°C in 5% CO2. Samples of CSFV-positive and CSFV-negative seruml were used as control samples. Assay results were interpreted by use of a fluorescence microscope.m Antibody titers were calculated in accordance with a published equation by Karber.14 Fluorescence readings for each well were ≥ 1, indicating that the virus was not completely neutralized. A titer > 1:10 was considered positive for anti-CSFV antibody.
qRT-PCR assay
A qRT-PCR assay was used to detect CSFV C strain copies in spleen lysates obtained from rabbits after they had been euthanized at 4 days after viral challenge. For this assay, viral RNA was extracted from the spleen lysates by use of a commercial kit,n and cDNA was reverse transcribed from 1 μg of total RNA by use of a reverse transcription reagent kit.o For C strain-specific detection, a previously reported primer pair was used (forward sequence: 5′-GATCCT-CATACTGCCCACTTAC-3′; reverse sequence: 5′-GTATACCCCTTCACCAGCTTG-3′)15 that targeted a region corresponding to the NS2 gene.
The qRT-PCR assay was performedp,q 3 times/sample with the following parameters: 95°C for 5 seconds, followed by 40 cycles of 95°C for 5 seconds and 60°C for 30 seconds. A pCMV-myc plasmidr encoding the CSFV C strain NS2 protein was used to construct a fluorescence quantitative standard curve for calculating copy numbers of the C strain in the samples, and the mean of the 3 values/sample was used for statistical analysis.
Statistical analysis
Statistical softwares,t was used for data analysis. Results are reported as mean ± SD. The CVs of antibody responses within groups were calculated as SD/mean × 100. The mean number of virus copies in spleen lysates at 4 days after infection was compared among groups with the Student t test. Correlations among E2 protein concentration in the vaccine (ie, vaccine dose), antibody response (ie, ELISA antibody blocking rate), and protection rate and between antibody response and protection rate were determined. Values of P < 0.05 were considered significant.
Results
Experiment 1
At 21 days after inoculation of rabbits with various doses of the 01 batch of vaccine or saline solution, the proportion of rabbits with antibody against CSFV C strain for those that received 1.15 μg of E2 protein was 5 of 10, the ELISA antibody blocking rate was 26.4% to 55.1% (considered a positive antibody response), and the CV was 26.1%, which suggested high variability in the antibody response within that group. Although neutralizing antibody (1:16) was detected in 5 rabbits, the mean serum antibody titer as determined via FAVN testing was 1:5 (ie, ≤ 1:10, which was the cutoff used to indicate a positive antibody titer). In contrast, among rabbits that received 2.3 or 4.6 μg of E2 protein, the proportion with antibody against CSFV C strain was 10 of 10; the blocking rate was 64.3% to 69.3% and 68.1% to 73.8%, respectively; and the CV was 2.3% and 2.4%, respectively, indicating that the antibody responses were less variable among rabbits in those 2 groups. In addition, protective neutralizing antibody was identified in all serum samples from the 2 groups, with mean serum antibody titers of 1:30 and 1:37, respectively. For the 10 rabbits in the control group, results of FAVN testing were negative (mean titer < 1:2), the ELISA antibody blocking rate was 9.3% to 11.8% (considered a negative antibody response), and the CV was 8.6%.
Rectal temperatures in each group fluctuated little (ie, ≤ 1°C) before viral challenge. No (0/10) rabbits in the 2.3- and 4.6-μg vaccine dose groups developed pyrexia (ie, rectal temperature ≥ 40.5°C) after challenge, whereas half (5/10) of the rabbits in the 1.15-μg dose group and all rabbits (10/10) in the control group developed pyrexia (Figure 1). After challenge, CSFV RNA was detected in each group of rabbits, and the results appeared to correspond with the pyrexia data. The mean number of virus copies in spleen lysates was significantly (P < 0.001) less than that of the control group (mean ± SD, 18.75 ± 2.02 × 105 copies/mL) for the 2.3-μg/mL (0.18 ± 0.06 × 105 copies/mL) and 4.6-μg/mL (0.13 ± 0.03 × 105 copies/mL) dose groups but not for the 1.15-μg/mL dose group (6.82 ± 7.67 × 105 copies/mL). Generally, the highest amount of viral RNA was observed in 5 of the 10 rabbits in the 1.15-μg dose group as well as in all rabbits in the control group.
Mean rectal temperature at various points for rabbits inoculated SC with the 01 batch of CSFV subunit E2 vaccine at doses representing 1.15 μg (circles; n = 10), 2.3 μg (squares; 10), or 4.6 μg (triangles; 10) of E2 protein/rabbit or saline (0.9% NaCl) solution (diamonds; 10) and then challenged with the CSFV C strain 21 days later (ie, at 48 hours in this graph). Pyrexia, defined as a rectal temperature > 40.5°C, was used to assess whether the vaccine had a protective effect. Error bars represent SD.
Citation: American Journal of Veterinary Research 81, 6; 10.2460/ajvr.81.6.521
Vaccination with the 01 batch resulted in a specific antibody response against CSFV, and the vaccine dose of E2 protein was positively correlated with this response (R2 quadratic = 0.949; Figure 2). When the vaccine dose was ≥ 2.3 μg, the protection rate after viral challenge was 100% (10/10), whereas this rate was only 50% (5/10) with the 1.15-μg dose. Therefore, the minimum immunization dose was defined as 2.3 μg of E2 protein (or 1 mL of vaccine). The vaccine dose was positively correlated with protection rate (R2 linear = 0.751). When the ELISA antibody blocking rate was ≥ 47.6%, the protection rate was 100% (10/10), and the antibody response was strongly positively correlated with protection rate (R2 linear = 0.942).
Regression lines for correlations among the vaccine dose of CSFV E2 protein, antibody response (ie, ELISA antibody blocking rate [%]), and protection rate for the rabbits of Figure 1. Each data point in panels A and C represents the mean of 3 observations. The R2 values from these analyses are shown. In panel A, the horizontal dotted line represents the cutoff used to define a positive anti-CSFV antibody result (ie, blocking rate ≥ 40%).
Citation: American Journal of Veterinary Research 81, 6; 10.2460/ajvr.81.6.521
Experiment 2
At 21 days after vaccination, results for rabbits vaccinated with the 02 and 03 batches of vaccine administered at the identified minimum immunization dose (ie, 2.3 μg of E2 protein) indicated a positive antibody response for all rabbits. The ELISA antibody blocking rates were 63.1% to 69.4% for the 02 batch and 63.2% to 69.2% for the 03 batch. Respective CVs were 2.96% and 2.95%. Intersubject variability in the antibody response for vaccinated rabbits was small (Figure 3). Mean serum antibody titers as determined via FAVN testing were 1:27 and 1:28, with all vaccinated rabbits in both vaccinated groups having a protective titer.
Mean rectal temperature at various points and antibody response (ELISA antibody blocking rate [%]) 21 days after rabbits were inoculated SC with the 02 batch (circles; n = 10) or 03 batch (squares; 10) of CSFV subunit E2 vaccine at the identified minimum immunization dose (ie, 2.3 μg of E2 protein) or saline solution (diamonds; 10) and then challenged with the CSFV C strain 21 days later. *The difference between the indicated groups is significant (P < 0.001). See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 81, 6; 10.2460/ajvr.81.6.521
The protection rate was 100% (10/10) for both vaccinated groups. None of the vaccinated rabbits developed pyrexia. Again, the mean number of virus copies in spleen lysates was significantly (P < 0.001) less than that of the control group (mean, 19.83 ± 4.07 × 105 copies/mL) for the 02 vaccine batch (0.16 ± 0.07 × 105 copies/mL) and 03 vaccine batch (0.15 ± 0.07 × 105 copies/mL).
For the control group, no rabbit (0/10) had a positive antibody response. The antibody blocking rate was 8.4% to 11.6%, the intersubject CV was 10%, and results of FAVN testing were negative (mean titer < 1:2).
Discussion
Several vaccines against CSFV infection in swine have been developed on the basis of recombinant plasmids expressing proteins or synthetic peptides of the CSFV protective antigen E2.16–18 Therefore, quality control of these vaccines and establishment of their efficacy are important to enable eradication of CSF in China. According to the 3R principles (reduction, replacement, and refinement) for the use of animals in research stipulated by the animal welfare laws of various countries worldwide19–21 and the relevant provisions of the notice issued by the Ministry of Science and Technology of the People's Republic of China on the Guiding Opinions on Treating Experimental Animals (N.398) of 2006, animal species used in research that are easier to manage and handle should be used to replace larger species that are more difficult, or new alternative methods should be developed to achieve the same scientific aim instead of using living vertebrates.
Rabbits are susceptible to the CSFV C strain, and its effects are potent in this species. Rabbits lacking antibody against CSFV typically develop pyrexia following infection. In contrast, in rabbits with antibody against the virus, no pyrexia reaction occurs, and the antibody rapidly neutralizes its virulence.22 In the present study, correlations were identified among the dose of E2 protein in the evaluated vaccine, ELISA antibody blocking rate, and protection rate, suggesting that rabbits can be effectively used to evaluate the efficacy of the subunit vaccine. Furthermore, our findings indicated that the method of vaccine evaluation used in the study was reliable and repeatable.
According to the manufacturer's instructions for the ELISA test kiti used to detect serum antibody against CSFV, the anti-CSFV antibody in serum samples blocks the horseradish peroxidase–labeled monoclonal antibody against the CSFV E2 protein. Binding of the monoclonal antibody to the CSFV antigen can be determined by the degree of coloration produced by horseradish peroxidase and the substrate. Any antibody present that binds to the CSFV antigen competes with the labeled monoclonal antibody in the kit for binding to the CSFV antigen, regardless of the animal species in which the anti-CSFV antibody is produced. Results of both experiments in our study indicted that the evaluated vaccine produced a good antibody response in rabbits and that this response was positively correlated with the dose of E2 protein administered. When the vaccine dose reached 0.1 mL/rabbit (ie, 0.23 μg of E2 protein), the blocking rate of the ELISA antibody against CSFV was higher than 60% 21 days after vaccination, and intersubject variability was small (CV < 3%).
Overall, a good positive correlation was identified between the dose of the evaluated genetically engineered live attenuated–CSFV C strain vaccine and the antibody response in rabbits. The minimum immunization dose in rabbits was verified to be 2.3 μg of E2 protein/rabbit. The method used was more stable, simple, effective, time-saving, and cost-effective than would have been evaluation of the vaccine in swine.
Acknowledgments
Funded by the State Key Laboratory of Animal Genetic Engineering Vaccines, Yebio Bioengineering Co Ltd, Qingdao, China; and the Animal Health Products Industry Think Tank Joint Fund Phase II of the Bureau of Qingdao Science and Technology (grant No. 18-6-3-1-jch).
Dr. Cao, Heng Zhang, Hui Zhang, and Dr. Fan were employees of Yebio Bioengineering, the manufacturer of the evaluated vaccine, at the time of the study. The authors declare that there were no other conflicts of interest.
Dr. Cao and Heng Zhang contributed to conception and design of the study and contributed to acquisition, analysis, and interpretation of data. Dr. Yang, Hui Zhang, and Dr. Fan contributed to the design of the study and analysis of data. Dr. Cao drafted the manuscript. All authors critically revised the manuscript, gave final approval, and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work would be appropriately investigated and resolved.
ABBREVIATIONS
| CSF | Classical swine fever |
| CSFV | Classical swine fever virus |
| CV | Coefficient of variation |
| FAVN | Fluorescent antibody virus neutralization |
| HEK | Human embryonic kidney |
| qRT | Quantitative real-time |
| SPF | Specific pathogen–free |
Footnotes
Kangda Rabbit Industry Development Co Ltd, Qingdao, China.
Provided by professor Hongwei Li, Southern Medical University, Guangdong, China.
Gibco, Thermo Fisher Scientific, Waltham, Mass.
CD 293 TGE, ACROBiosystems, Newark, Del.
Image Lab software and a Bole EZ imager, Bio-Rad Laboratories, Hercules, Calif.
Thermo Fisher Scientific, Waltham, Mass.
ISA 201VG, Seppic, Shanghai, China.
CSFV C strain (batch No. 201601; 100 mL/bottle; valid until December 2017), Yebio Bioengineering Co Ltd, Qingdao, China.
CSFV antibody test, Idexx Laboratories, Westbrook, Me.
CSF antiserum, Yebio Bioengineering Co Ltd, Qingdao, China.
Goat anti-swine IgG (H+L)-fluorescein antibody, Merck KGaA, Darmstadt, Germany.
Provided by Professor Yanming Zhang, Northwest A&F University, Shaanxi, China.
Nikon, Tokyo, Japan.
RNeasy Mini kit, Qiagen, Valencia, Calif.
PrimeScript reagent kit with gDNA eraser, TaKaRa, Dalian, China.
Bio-Rad iQ5, Bio-Rad Laboratories, Hercules, Calif.
SYBR premix ex taq II, TaKaRa, Dalian, China.
Clontech Laboratories Inc, Mountain View, Calif.
Prism 6, GraphPad Software Inc, La Jolla, Calif.
SPSS, version 17.0, IBM Corp, Armonk, NY.
References
1. Edwards S, Fukusho A, Lefevre PC, et al. Classical swine fever: the global situation. Vet Microbiol 2000;73:103–119.
2. Luo Y, Li S, Sun Y, et al. Classical swine fever in China: a minireview. Vet Microbiol 2014;172:1–6.
3. Moennig V. Introduction to classical swine fever: virus, disease and control policy. Vet Microbiol 2000;73:93–102.
4. Zhou B. Classical swine fever in China—an update minireview. Front Vet Sci 2019;6:187.
5. Paton DJ, Greiser-Wilke I. Classical swine fever—an update. Res Vet Sci 2003;75:169–178.
6. Bensaude E. Classical swine fever virus induces proinflammatory cytokines and tissue factor expression and inhibits apoptosis and interferon synthesis during the establishment of long-term infection of porcine vascular endothelial cells. J Gen Virol 2004;85:1029–1037.
7. Sun J, Shi Z, Guo H, et al. Changes in the porcine peripheral blood mononuclear cell proteome induced by infection with highly virulent classical swine fever virus. J Gen Virol 2010;91:2254–2262.
8. Lund J, Sato A, Akira S, et al. Toll-like receptor 9–mediated recognition of herpes simplex virus-2 by plasmacytoid dendritic cells. J Exp Med 2003;198:513–520.
9. Tsung A, Sahai R, Tanaka H, et al. The nuclear factor HMGB1 mediates hepatic injury after murine liver ischemia-reperfusion. J Exp Med 2005;201:1135–1143.
10. Ministry of Agriculture of the People's Republic of China. The veterinary biological products standards of the People's Republic of China [in Chinese]. Beijing: Chemistry Press, 2000.
11. Commission of Chinese Veterinary Pharmacopoeia. Veterinary pharmacopoeia of the People's Republic of China [in Chinese]. Vol 3. Beijing: China Agricultureal Press, 2015;96–97.
12. Commission of Chinese Veterinary Pharmacopoeia. Veterinary pharmacopoeia of the People's Republic of China [in Chinese]. Vol 3. Beijing: China Agricultureal Press, 2015;19–21.
13. World Organisation for Animal Health. Chapter 3.8.3: classical swine fever (infection with classical swine fever virus). In: Manual of diagnostic tests and vaccines for terrestrial animals 2019. Available at: www.oie.int/standard-setting/terrestrial-manual/access-online. Accessed Dec 31, 2019.
14. Karber G. Beitrag zur kollektiven Behandlung pharmakologischer Reihenversuche. Arch Exp Pathol Pharmacol 1931;162:480–483.
15. Cao Z, Guo K, Zheng M, et al. A comparison of the impact of Shimen and C strains of classical swine fever virus on toll-like receptor expression. J Gen Virol 2015;96:1732–1745.
16. Bouma A, De Smit AJ, De Jong MCM, et al. Determination of the onset of the herd-immunity induced by the E2 subunit vaccine against classical swine fever virus. Vaccine 2000;18:1374–1381.
17. van Gennip HGP, Bouma A, van Rijn PA, et al. Experimental non-transmissible marker vaccines for classical swine fever (CSF) by trans-complementation of Erns or E2 of CSFV. Vaccine 2002;20:1544–1556.
18. Beer M, Reimann I, Hoffmann B, et al. Novel marker vaccines against classical swine fever. Vaccine 2007;25:5665–5670.
19. Osborne NJ, Payne D, Newman ML. Journal editorial policies, animal welfare, and the 3Rs. Am J Bioeth 2009;9:55–59.
20. Rusche B. The 3Rs and animal welfare—conflict or the way forward? ALTEX 2003;20(suppl 1):63–76.
21. Liebsch M, Grune B, Seiler A, et al. Alternatives to animal testing: current status and future perspectives. Arch Toxicol 2011;85:841–858.
22. Ge Y, Zhang XJ, Zhu QH, et al. Parallel relationship between real-time RT-PCR and fever reaction method in rabbits in testing hog cholera lapinized virus vaccines [in Chinese]. Chin J Prev Vet Med 2011;33:699–703.
