AN INTRANASAL CHALLENGE MODEL FOR TESTING JAPANESE ENCEPHALITIS VACCINES IN RHESUS MONKEYS

Transcription

1 Am. J. Trop. Med. Hyg., 60(3), 1999, pp Copyright 1999 by The American Society of Tropical Medicine and Hygiene AN INTRANASAL CHALLENGE MODEL FOR TESTING JAPANESE ENCEPHALITIS VACCINES IN RHESUS MONKEYS BOONYOS RAENGSAKULRACH, ANAA NISALAK, MONTIP GETTAYACAMIN, VIPA THIRAWUTH, G. DAVID YOUNG, KHIN SAW AYE MYINT, LAURA M. FERGUSON, CHARLES H. HOKE JR., BRUCE L. INNIS, A DAVID W. VAUGHN Department of Virology, and Department of Veterinary Medicine, Armed Forces Research Institute of Medical Sciences, Bangkok, Thailand; Pasteur-Mérieux-Connaught, Swiftwater, Pennsylvania; Department of Virus Diseases, Walter Reed Army Institute of Research, Washington, District of Columbia Abstract. Placebo-controlled field efficacy trials of new Japanese encephalitis (JE) vaccines may be impractical. Therefore, an animal model to evaluate efficacy of candidate JE vaccines is sought. Previous work has shown that exposure of monkeys to JE virus (JEV) via the intranasal route results in encephalitis. Here we report the further development of this model and the availability of titered virus stocks to assess the protective efficacy of JE vaccines. To determine the effective dose of our JE challenge virus, dilutions of a stock JEV (KE-93 isolate) were inoculated into four groups of three rhesus monkeys. A dose-dependent response was observed and the 50% effective dose (ED 50 ) was determined to be plaque forming units (pfu). Among animals that developed encephalitis, clinical signs occurred 9 14 days postinoculation. Infection with JEV was confirmed by detection of JEV in nervous tissues and IgM to JEV in the cerebrospinal fluid. Viremia with JEV was also detected intermittently throughout infection. Validation of the model was performed using a known effective JE vaccine and saline control. One ED 90 of virus ( pfu) was used as a challenge dose. Four of four animals that received saline control developed encephalitis while one of four monkeys administered the JE vaccine did so. This study demonstrates that the virus strain, route of inoculation, dose, and the outcome measure (encephalitis) are suitable for assessment of protective efficacy of candidate JE vaccines. Japanese encephalitis (JE) is the leading cause of mosquito-borne viral encephalitis in the world, accounting for approximately 35,000 cases and 10,000 deaths each year in Asia. The JE virus (JEV) is transmitted by culicine mosquitoes through an epizootic cycle involving primarily pigs and birds. Most infections in humans are asymptomatic. It is estimated that only one in 20 to one in 1,000 infections result in encephalitis. However, once encephalitis occurs, it is fatal in 25% of the cases with 50% of the survivors experiencing neurologic sequelae. 1 3 Several JE vaccines are in use. Live attenuated vaccines have been developed and licensed for use only in China. 4 6 A more widely used vaccine is a highly purified formalininactivated virus preparation derived from infected mouse brain. In , an efficacy trial was conducted in 65,224 Thai children using monovalent (Nakayama strain) and bivalent (Nakayama and Beijing strain) mouse brain vaccines manufactured by BIKEN (Research Foundation of Microbial Diseases of Osaka University, Osaka, Japan). 7 The efficacy of both vaccines was 91% and no major adverse reactions were reported. This study contributed to licensure of the BIKEN vaccine in the United States in December Nevertheless, an increased incidence of hypersensitivity reactions has been reported in Denmark, 8 Australia, 9 Canada, 10 and the United Kingdom. Neurologic illness following JE vaccination has also recently been reported in Japan 12 and Denmark, 13 thus stimulating efforts to develop new JE vaccines. A new JE vaccine was recently created by the insertion of the premembrane (prm), envelope (E), nonstructural protein 1 (NS1), and NS2a genes of JEV into the genome of the NYVAC vector, which is a highly attenuated strain of vaccinia. 14 The vaccine candidate (NYVAC-JEV) was able to prevent viremia in swine after an intravenous challenge. A similar recombinant JE vaccine (ALVAC-JEV) has also been constructed using an avian poxvirus vector and has 329 been shown to protect mice from lethal JEV challenge. 15 In addition, several research groups are developing for international use formalin-inactivated JE virus vaccines prepared in cell cultures rather than mouse brains (Saluzzo JF, Eckels KH, unpublished data). Demonstrating protective efficacy of new JE vaccines in humans will be difficult. Challenge studies in humans are unethical due to the possibility of fatal encephalitis following challenge. Moreover, field trials comparing two vaccines (equivalency studies) would require large numbers of volunteers to detect differences between old and new vaccines due to a low attack rate. In addition, a scientific conundrum is presented by the conflicting need to have a placebo in the trials and the ethical unacceptability of doing so. Since protection against JE is highly correlated with neutralizing antibody, we propose that new JE vaccines can be evaluated in volunteers for immunogenicity equivalent to the licensed vaccine. Nevertheless, the evaluation will be incomplete without demonstration that vaccination can actually protect against the disease. Having two new JE vaccines (NYVAC-JEV and ALVAC- JEV) to evaluate, we therefore sought an animal model to demonstrate their protective efficacy. Previously, Harrington and others 16 demonstrated that JE can be produced in rhesus monkeys by intranasal administration of virus. Herein, we describe studies that confirmed their results and validate the suitability of this model for a test of vaccine protective efficacy. Results of the evaluation of the two new JE vaccines using this model are reported separately. 17 MATERIALS A METHODS Study protocol. This study consisted of two parts: determination of the challenge virus dose and validation of the model using a known effective JE vaccine. Animal procedures were performed under the United States National In-

2 330 RAENGSAKULRACH A OTHERS stitutes of Health guidelines for the humane use of laboratory animals. The study was approved by an Institutional Animal Care and Use Committee (United States Army Medical Component, Armed Force Research Institute of Medical Sciences) and by the Animal Use Review Office, United States Army Medical Research and Materiel Command. Monkeys. Rhesus monkeys (Macaca mulatta) 3 10 years of age were used. The animals were of either sex, weighed kg, and had no neutralizing antibody against JE and dengue viruses. The monkeys were housed individually in a double-screened housing facility. Challenge virus. The JEV isolate (KE-93) used in this study was obtained from the brain of a six-year old boy in Kamphaeng Phet Province, Thailand who died of JE in The virus was isolated in AP-61 cells and passaged once in C6/36 cells. The C6/36 cell culture supernatant was administered to several rhesus monkeys by the intranasal route. One monkey that became ill with encephalitis was selected. Virus was recovered from its brain and virus stocks were expanded by two additional passages in the brains of suckling mice. The infected mouse brains were harvested, made to 20% (weight by volume) in phosphate-buffered saline (PBS) with 1% bovine albumin, and homogenized. After centrifugation at 8,000 g for 30 min, the supernatant containing JEV was collected and kept in aliquots at 70C. The final stock had a titer of 2.0 x plaque-forming units (pfu)/ml in LLC-MK2 cells and was designated KE-93 (AP61-1, C6/36-1, Mm-1, SM-2) to indicate its passage history. The potency of this virus was determined in suckling and weanling mice (two and 21 days old, respectively). Ten-fold dilutions of the virus stock were inoculated intracerebrally into groups of eight suckling mice or intraperitoneally into groups of four weanling mice. The 50% lethal dose (LD 50 ) was calculated to be 13 pfu and pfu in suckling and weanling mice, respectively. Intranasal challenge. Each monkey was anesthetized intramuscularly with ketamine hydrochloride (5 10 mg/kg) and placed in dorsal recumbency. The head was tilted slightly back and 0.5 ml of JE virus inoculum was instilled slowly into each nostril. The head was held in that position for 5 min before the animal was returned to its cage. Clinical signs (i.e., anorexia, weakness, depression, tremor, paralysis, and levels of consciousness) were observed daily for a 28-day period. To provide specimens to determine if viremia occurred and to monitor the immune response, serum was collected by venipuncture on study days 0 just before virus inoculation, 1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 14, 21, 28, and at the time of killing. Cerebrospinal fluid (CSF) was collected by lumbar puncture 7 10 days before challenge and after challenge on days 14, 28, and at the time of necropsy. Determination of challenge virus dose. To determine the JEV dose that elicited overt encephalitis, four groups of three monkeys were inoculated intranasally with graded challenge doses of virus ( , , , and pfu). After the inoculation, the monkeys were followed for 28 days for the development of encephalitis, the primary endpoint measure. Monkeys that became comatose were killed by intravenous injection of pentobarbital sodium (100 mg/kg). Encephalitis due to JEV was defined as the occurrence of focal neurologic signs or depressed consciousness plus one of the following: the presence of JEV in the CSF, IgM to JEV in the CSF, or detection of JEV in tissues of the central nervous system (CNS) if killing was performed. Survival data were transformed to a Probit curve that allowed any effective dose (ED ) to be determined. An ED 90 was specified as the challenge dose of interest. Validation of the model by challenge of vaccinated animals. The BIKEN inactivated JE vaccine (Lot EJN 058; Research Foundation of Microbial Diseases of Osaka University) was used as a known effective vaccine. The vaccine (1.0 ml) was given subcutaneously to four monkeys. The second and third doses were given seven and 28 days later (as recommended for adult humans). A booster dose was administered months after the first vaccine dose. Along with the four vaccinated monkeys, another group of four monkeys was given physiological saline on the same vaccination schedule. The monkeys receiving either vaccine or saline were challenged with an ED 90 ( pfu) of JEV two months after the booster dose. All persons working with the animals and performing laboratory assays were masked to the test article administered. The rationale here was that in an acceptable challenge model, naive animals would develop overt encephalitis but animals vaccinated with JE vaccine known to be protective 7 would not. Necropsy. Just before killing, CSF was collected for an IgM to JEV assay, virus isolation, and detection of JEV RNA by the reverse transcription polymerase chain reaction (RT-PCR). At necropsy, nervous tissues (cerebral cortex, cerebellum, brain stem, and spinal cord) were collected for JEV isolation and JEV RT-PCR. Sterile technique was maintained during each necropsy and instruments were changed for each specimen collected to avoid cross-contamination. Immunologic assays. Antibody assays included an ELISA for IgG and IgM to JEV, 18 hemagglutination inhibition 19 (HAI), and a 50% plaque-reduction neutralization test. 20 A serum IgM value greater than 40 ELISA units in a single specimen or an increase from less than 15 to more than 30 ELISA units was considered positive. The mean SD of CSF IgM determined using 28 prechallenge monkey samples was ELISA units. A CSF IgM value greater than 10 ELISA units was considered positive. The ELISA and HAI assays were performed on sera collected on study days 0 to 8, 10, 12, 14, 21, and 28. In the experiment to determine the challenge virus dose, neutralizing antibody was determined only on sera from study days 0 and 28 postchallenge or the day of necropsy. In the experiment to validate the challenge model, neutralizing antibody was assayed in sera collected 10 days before the challenge and on days 0, 7, 12, 14, 21, and 28 or the day of necropsy. The ELISAs for IgG and IgM to JEV were performed on all CSF specimens. Virus isolation. To evaluate the viremia, 100 l of serum collected during each of the first 14 days following virus inoculation was mixed with 400 l of RPMI 1640 medium (containing 5% fetal bovine serum, 100 U/ml of penicillin, and 100 g/ml of streptomycin) and inoculated onto C6/36 cell monolayers. The cells were monitored for seven days for a cytopathic effect (CPE). If negative, the culture fluid was passed once on fresh C6/36 cells. If a CPE was detected, the presence of JEV in culture fluid was confirmed by an antigen-capture ELISA system 21 using JEV-specific mono-

3 INTRANASAL CHALLENGE MODEL FOR TESTING JE VACCINES 331 clonal antibody J Attempts were also made to isolate JEV from CSF and nervous tissues from monkeys with encephalitis. One hundred microliters of CSF was placed directly onto the C6/36 cells. Tissues were homogenized in four volumes of PBS before placing on the cell culture. The inoculated cell cultures were processed as described for serum samples. Reverse transcription polymerase chain reaction. The RNA was extracted from 100 l of sera, CSF, or tissue homogenate using an acid guanidinium isothiocyanate-phenolchloroform extraction method. 22 The JEV-specific primers used in the RT-PCR and nested PCR were selected from the conserved sequences in the NS3 region of the JEV genome. The RT-PCR was performed as follows. Five microliters of RNA extract (equivalent to 20 l of original specimen) was mixed with 10 pmol of the P5431 primer (5- CTGTTCGGTGACATCAGTCT-3) ina10-l volume. The mixture was heated at 68C for 3 min and cooled on ice. An RT-PCR was then set up in a 100-l volume containing following components: denatured RNA, 10 mm Tris-HCl (ph 8.3), 50 mm KCl, 3.5 mm MgCl 2, 200 M (each) of four deoxynucleoside triphosphates, 15 pmol of primers P5431, 25 pmol of primer P5050 (5-GCCTATACGGCA- ATGGAGTT-3), 40 units of RNAsin (Promega, Madison, WI), 1.8 unit of Rous associated virus-2 reverse transcriptase (Amersham, Arlington Heights, IL), and 2.5 units of AmpliTaq (Perkin Elmer, Norwalk, CT). The reaction mixture was incubated at 42C for 45 min for the RT reaction and then subjected to 35 cycles of PCR amplification: denaturation (94C for 1 min), annealing (55C for 1 min), and extension (72C for 1 min). Denaturation of the first cycle was set for 3 min and the extension step of the last cycle was 4 min. A nested PCR was performed on 5 l of diluted RT-PCR product (1:50 in distilled water). The 100-l reaction mixture contained the diluted RT-PCR product, 10 mm Tris-HCl (ph 8.3), 50 mm KCl, 3.5 mm MgCl 2, 200 um (each) of four deoxynucleotide triphosphates, 25 pmol each of primers P5073 (5-CTTGGCGATGGCTCATAC-3) and P5247 (5- TGTTCTTAGGCGCTGCTG-3), and 2.5 units of Ampli- Taq. The mixture was subjected to 35 PCR cycles as described for the RT-PCR except that the denaturation step of the first cycle was done for 1.5 min. Eighteen microliters of the nested product was then analyzed by agarose gel electrophoresis. The size of the JEV-specific nested product was 191 basepairs. Routinely, this PCR assay was able to detect 10 pfu of JEV in 100 ul of specimens used for RNA extraction. Negative controls were included in each extraction, RT-PCR, and nested steps to monitor the possibility of falsepositive results in each run. The assay yielded positive results only for JEV and negative results for other flaviviruses tested including dengue virus serotypes 1 to 4, Murray Valley encephalitis, West Nile, Kunjin, Langat, and Wesselsbron viruses. RESULTS Determination of challenge dose. Four groups of three monkeys were inoculated intranasally with graded doses of JEV ( , , , and pfu). Six of 12 animals developed encephalitis; the response was FIGURE 1. Determination of the Japanese encephalitis virus challenge dose in monkeys. Survival data was transformed by Probit analysis (A). Empty circles represent the probability of being affected by encephalitis in the monkeys at a given challenge dose (indicated on the top). A hypothetical surviving curve (solid line) and upper and lower 95 percent confidence intervals (dashed line) are calculated by a Probit computer program (SPSS, Inc., Chicago, IL). B, a Probit analysis that includes four additional monkeys that had received saline placebo and were challenged with a 90% effective dose ( plaque-forming units [pfu]). clearly dependent on dose (Figure 1A). The course of disease was similar in all six affected animals. Clinical signs appeared around nine days postinoculation and progressed from depression and anorexia, followed by tremors, paralysis, and then coma. Affected animals were killed by lethal injection days postinoculation. The mean duration of symptoms was 24 hr. The ED 50 was calculated to be pfu and the ED 90 of pfu was selected for the challenge vaccine experiments that followed. Infection of the CNS with JEV. Evidence of JEV infection in all fatal cases was obtained using CSF and CNS tissue specimens. Immunoglobulin M to JEV was detected in the CSF of only the three monkeys receiving the maxi-

4 332 RAENGSAKULRACH A OTHERS TABLE 1 Analysis of cerebrospinal fluid (CSF) and tissues of central nervous system (CNS) from monkeys at 14 and 28 days postchallenge or at necropsy Virus dose (plaque-forming units) Monkey Day of killing* DA470 DA525 DA DA314 DA322 DA DA161 DA304 DA DA228 DA324 DA CSF JEV isolated JE PCR JE IgM JEV isolated CNS tissues * survived. Japanese encephalitis virus (JEV) isolation and JE polymerase chain reaction (PCR) was performed on CSF collected on study day 14 or at the time of necropsy. The JE IgM level was determined in CSF collected on study days 14 and 28 or at the time of necropsy. negative; positive; not done. The CNS tissues include cerebral cortex, cerebellum, brain stem, and spinal cord (cervical, throacic, and lumbar). For all monkeys with encephalitis except DA470 and DA525, the plus sign denotes JEV isolated or JE PCR positive for all tissues listed above. Virus isolation and JE PCR were not performed on brain stems from monkeys DA470 and DA525. Only for monkey DA470, JEV was isolated from all CNS tissues listed in, except from cerebral cortex. JE PCR mum dose (Table 1). None of these monkeys had JEV isolated from or JEV RNA detected in their CSF. Among the remaining three fatal cases that did not have IgM to JEV in their CSF, JEV was isolated from the CSF of monkeys DA314 and DA304. Monkey DA322 did not have JEV or IgM to JEV in the CSF. However, all six monkeys with encephalitis had JEV isolated from all CNS tissues tested including cerebral cortex, cerebellum, brain stem, and spinal cord. Histopathology and immunocytochemistry of the infected CNS tissues were also performed and the results are reported elsewhere. 23 Infection of the CNS by JEV was also investigated in the six surviving monkeys. No JEV was isolated and no IgM to JEV was detected in the CSF of these monkeys. Immune response. Serum levels of IgM to JEV and HAI antibodies on study days 0 8, 10, 12, 14, 21, and 28 after JEV inoculation are shown for each monkey in Figure 2. For animals developing fatal encephalitis, two types of immune response were observed. In monkeys DA470, DA525, and DA526 that received the maximum JEV dose, an increase of HAI titer and IgM levels was seen around day 10 postinoculation. Neutralizing antibody was also detected at the day of necropsy in monkeys DA470 and DA526 but was not detected in monkey DA525. In contrast, no immune response was detected in monkeys DA314, DA322, and DA304 that received lower JEV doses. For surviving monkeys, two patterns of antibody response were also observed. Increasing levels of IgM to JEV and HAI antibodies were seen in five monkeys (DA534, DA161, DA537, DA324, and DA392) approximately two weeks postinoculation. Neutralizing antibody was also detected on study day 28 in all five monkeys. One monkey in the group receiving the lowest dose (DA228) had no immunologic evidence of JEV infection during the 28-day period. No significant increase in IgG to JEV was observed in any animals during this part of the study. Viremia. To determine the duration of viremia, virus isolation and the JEV RT-PCR were performed using sera collected on study days 0 8, 10, 12, 14, 21, and 28 postchallenge. Eight of 12 monkeys had evidence of JE viremia (Figure 2). Virus was isolated from two monkeys experiencing encephalitis (DA470 and DA526) and from four without encephalitis (DA534, DA161, DA537, DA324). Japanese encephalitis virus RNA was also detected in two monkeys with encephalitis (DA322 and DA304) and in two monkeys without encephalitis (DA161 and DA537) by RT-PCR. However, no single specimen was positive by both methods of virus detection. Validation of the model. Eight monkeys were randomly allocated to receive the BIKEN JE vaccine or the normal saline control. Two months after the final booster dose, all eight were challenged with one ED 90 of the challenge virus. All four monkeys who had received normal saline developed encephalitis, and all were killed days postchallenge. In contrast, all four vaccine recipients survived. However, one monkey (DA422) did show signs of encephalitis including depression, anorexia, and paralysis of limbs between days nine and 18 postchallenge. This monkey recovered fully without any specific treatments. Infection of the CNS in vaccine and saline recipients. Infection of the CNS by the challenge JEV was determined using CSF and CNS specimens (Table 2). In the challenged saline recipients, an increase in CSF IgM to JEV was detected in all four monkeys at the time of necropsy. None of the animals had JEV isolated from the CSF; however, JEV RNA was detected in the CSF of one monkey (DA378) at the time of necropsy. In the challenged vaccine recipients, no JEV was isolated from the CSF collected at 14 days postchallenge. No significant increase in CSF IgM to JEV was observed in three of the four monkeys. An increase in IgM to JEV in the CSF occurred in monkey DA422. Levels increased from 1 ELISA unit 10 days before the challenge to 10 and 29 ELISA units on study days 14 and 28, respectively. Monkey DA422 was the one that became ill and recovered. Immune response in vaccine and saline recipients after challenge. Following challenge, three of four saline control animals mounted a primary immune response as demonstrat-

5 INTRANASAL CHALLENGE MODEL FOR TESTING JE VACCINES 333 FIGURE 2. Immune response and Japanese encephalitis virus (JEV) viremia after inoculation with JEV. Four groups of three monkeys were inoculated intranasally with (A), (B), (C), (D) plaque-forming units of JEV. Neutralizing antibody titers (Neut), hemagglutination inhibition titers (HAI), and levels of IgM to JEV in ELISA units of each monkey are shown individually by study days postinoculation. Monkey identities (ID) are indicated on the top of each graph. Monkeys that developed encephalitis are shown in a shaded box. Detection of viremia by virus isolation and the polymerase chain reaction (PCR) was performed on study days 0 8, 10, 12, and 14. Only days with positive viremia are shown under the monkey ID. ed by an increased level of IgM rather than IgG to JEV (Figure 3A). No immune response was detected in monkey DA378. In contrast, all four vaccinated monkeys showed a marked anamnestic response as demonstrated by a rapid increase of JE-specific neutralizing antibodies in the serum on day 7 (Figure 3B). Immunoglobulin G to JEV and HAI antibodies also increased around 5 21 days postchallenge. Viremia in vaccine and saline recipients. Following challenge, no JEV was isolated from any monkey in either the vaccine or saline group on study days 0 8, 10, 12, and

6 334 RAENGSAKULRACH A OTHERS TABLE 2 Analysis of cerebrospinal fluid (CSF) and tissues of central nervous system (CNS) from vaccinated monkeys at 14 and 28 days postchallenge Monkey DA352 DA378 DA413 DA443 DA362 DA422 DA427 DA448 Vaccine Saline Saline Saline Saline Biken Biken Biken Biken Day of killing* * survived. and See Table 1 for description. JEV isolated CSF JE PCR JE IgM CNS tissues JEV isolated JE PCR 14. However, JEV RNA was detected by the RT-PCR in three of four saline recipients and two of four vaccine recipients on study days as indicated for each monkey in Figure 3. DISCUSSION Licensure for new JE vaccines may be sought on the basis of documented safety and a neutralizing antibody response comparable to the BIKEN vaccine. However, a neutralizing antibody response provides only presumptive evidence of protection. The challenge model described here would allow a candidate vaccine to demonstrate its protective efficacy in an animal model that closely mimics human disease. Efficacy of a JE vaccine can be evaluated in several animal systems. Mice have been widely used in preliminary tests for JE vaccine potency Testing in nonhuman primates is also possible but the outcomes are different depending on primate species, JEV strains, and routes of inoculation. Intradermal or subcutaneous inoculation of monkeys with JEV creates a situation that resembles natural infection via mosquitoes. However, this challenge model does not cause encephalitis in monkeys; therefore, it does not allow demonstration of protection against disease. 5, Direct intraspinal and intrathalamic inoculations of JEV in rhesus monkeys result in rapid fatal infection within seven days, 5 but a challenge model by these routes is very severe. The intranasal challenge used in this study represents a noninvasive peripheral route of inoculation that predictably produces clinical signs similar to those found in humans with JE. This challenge model has been used previously to demonstrate the protective efficacy of an inactivated vaccine 30 and of interferon inducers against JE. 16, 31 Compared with the previous studies, ours provides a more extensive characterization of the model. Because the challenge virus was carefully titered in monkeys, a high level of reproducibility of this model is likely. Hundreds of vials of the titered challenge virus have been stored for future studies. Following intranasal inoculation of monkeys with JEV, encephalitis occurred in a dose-related fashion. Nevertheless, the course of JE disease in all 10 affected monkeys was rather uniform and did not depend on the challenge dose. Clinical signs such as weakness usually occurred on day 9 and progressed rapidly to complete paralysis within 24 hr for all affected monkeys. This disease progression was similar to that seen in a previous study. 16 In that study, the onset occurred on days 5 9 and death ensued 12 days after intranasal inoculation. One striking difference between that study and ours is that the ED 50 of JEV (Peking strain) used in that study was pfu, which is 2,400 times lower than the ED 50 of pfu of our Thai strain. This may reflect the difference in virulence between the two JEV strains. No correlation was observed between the challenge doses and the extent of viremia. Most monkeys, both with and without encephalitis, had JEV or JE RNA detected intermittently during the first two weeks after virus inoculation. Monkey DA413 was an exception since JEV RNA was found every day during the first five days after challenge (Figure 3). It should be noted that there was no single serum specimen that yielded the positive results by both virus isolation and the RT-PCR. This may be due to the low levels of virus circulating in the blood; therefore, the possibility of detection was only by chance. The supporting evidence was that most monkeys did not have JEV culturable from their sera. In addition, the seven serum specimens from which JEV was isolated (Figure 2) were all negative by virus titration in LLC/MK2 cells, suggesting low levels of virus in the specimens. Loss of virus viability during storage of specimens (-70C) is possible but is unlikely to account for negative results for all seven specimens. This finding is in accordance with a human situation in which the viremic phase may not occur or is too short to be detected. JEV has usually been isolated from brain tissues and sometimes from CSF of fatal cases but rarely from blood or CSF of non-fatal cases. 32,33 Discordance in virus isolation and PCR results may also be explained by the effect of neutralizing antibody on the virus isolation but, to a lessor extent, on the PCR. Japanese encephalitis virus could be isolated from sera early after inoculation (days 1 6, mean 2.6, 95% confidence interval ), while virus genome could be detected longer by the PCR (days 1, mean 4.5, 95% confidence interval ). This suggests that increasing levels of JE neutralizing antibody may make virus isolation more difficult. Monkeys that received higher challenge doses were more likely to develop encephalitis and to have a more rapid increase in IgM to JEV in both serum and CSF. A total of monkeys experienced encephalitis (six monkeys in the titration experiment, four saline recipients, and one BIKEN recipient). Of these, an increase in serum IgM was seen only in the three monkeys who received the maximum challenge dose (DA470, DA525, and DA526 in Figure 2) and in three of four saline control monkeys who received one ED 90 of the virus (DA352, DA413, and DA443 in Figure 3). These six monkeys also had detectable IgM to JEV in their CSF. It is interesting to note that the amount of IgM to JEV in monkeys DA352, DA413, and DA443 was relatively equal both in CSF and sera (CSF/serum IgM to JEV: 154/108, 28/ 26, and 25/45 ELISA units, respectively). In contrast, the amount of IgM to JEV in the CSF of monkeys DA470, DA525, and DA526 was higher than those in the sera (CSF/ serum IgM to JEV: 301/121, 226/57, and 1/39 ELISA units, respectively). We speculate that inoculation with a higher virus dose may result in the extensive dissemination

7 INTRANASAL CHALLENGE MODEL FOR TESTING JE VACCINES 335 FIGURE 3. Efficacy testing of a commercially available Japanese encephalitis virus (JEV) vaccine using the intranasal challenge model. Groups of four monkeys receiving saline placebo (A) or BIKEN vaccine (B) were challenged with plaque-forming units of JEV (one 90% effective dose). Monkeys that developed encephalitis are shown in a shaded box. Viremia and antibody responses were determined as described in Figure 1. Levels of IgG to JEV instead of IgM are presented for the vaccine recipients. For definitions of other abbreviations, see Figure 2. of virus, causing a rapid antibody response both in the blood circulation and the CNS. Additionally, higher CSF IgM to JEV (compared with serum IgM to JEV) seen in monkeys that received the maximum dose may be due to IgM antibody produced locally in the CNS. Different from the six monkeys discussed above, monkey DA378 in the saline control group had no detectable IgM to JEV in the serum but had a low level of IgM ( units) in the CSF. This monkey also had JEV RNA detected in both CSF and blood at the time of necropsy (Figure 3A and Table 2). The antibody response in this monkey may have been low, thus allowing spread of virus in both blood and CSF. Dose-dependence characteristics of the model allowed us to create a survival curve (Figure 1A) that was used to select

8 336 RAENGSAKULRACH A OTHERS a challenge dose for the vaccine challenge experiments. The challenge dose of ED 90 was selected because it was thought that at least half of the control animals would experience encephalitis since the lower 95% confidence interval of the ED 90 (approximately pfu) was still higher than the ED 50 ( pfu). Indeed, the challenge was proved to be rigorous, affecting all control animals and causing illness in one vaccinated animal. Figure 1B adds the four saline control monkeys that developed encephalitis in the vaccine experiment to the Probit analysis. A curve with narrower 95% confidence intervals was obtained, thus providing a more accurate assessment of the challenge virus strength. In summary, inoculation of JEV into monkeys via an intranasal route produced a predictable outcome of encephalitis suitable for future vaccine efficacy testings. This model has been successfully validated using the licensed BIKEN JE vaccine. In addition, it had been used in recent efficacy evaluations of two new JE vaccines, NYVAC-JEV and AL- VAC-JEV. 17 With the practical limitations of conducting field efficacy trials of new JE vaccines, this model may prove useful to substantiate the protective efficacy of new JE vaccines. Acknowledgments: We thank members of the Department of Virology and the Department of Veterinary Medicine, Armed Forces Research Institute of Medical Sciences for expert technical assistance and Mala Chutinaswangporn for initial development of the JEV PCR. Financial support: This project was supported by Pasteur-Mérieux- Connaught Laboratories, Inc. and the U.S. Army Medical Research and Materiel Command. Disclaimer: The views of the authors do not necessarily reflect the position of the United States Department of the Army or the Department of Defense. Authors addresses: Boonyos Raengsakulrach, Ananda Nisalak, Vipa Thirawuth, and Khin Saw Aye Myint, Department of Virology, U.S. Army Medical Component-Armed Forces Research Institute of Medical Sciences (USAMC-AFRIMS), 315/6 Rajvithi Road, Bangkok 10400, Thailand. Montip Gettayacamin, Department of Veterinary Medicine, USAMC-AFRIMS, 315/6 Rajvithi Road, Bangkok 10400, Thailand. G. David Young, U.S. Army Center For Health Promotion and Preventive Medicine, 5158 Blackhawk Road, Aberdeen Proving Ground, MD Laura M. Ferguson, Pasteur-Mérieux-Connaught, Swiftwater, PA Charles H. Hoke, Jr., Military Infectious Disease Research Program, Medical Research and Materiel Command, 504 Scott Street, Fort Detrick, Frederick, MD Bruce L. Innis and David W. Vaughn, Department of Virus Diseases, Walter Reed Army Institute of Research, Washington, DC Reprint requests: Chief, Department of Virology, USAMC-AF- RIMS, APO AP REFERENCES 1. Vaughn DW, Hoke CH Jr, The epidemiology of Japanese encephalitis: prospects for prevention. Epidemiol Rev 14: Tsai TF, Yu YX, Japanese encephalitis vaccines. Plotkin SA, Mortimer EA, eds. Vaccines. Second edition. Philadelphia: W. B. Saunders, Innis BL, Japanese encephalitis. Porterfield JS, ed. Kass Handbook of Infectious Diseases: Exotic Viral Infections. London: Chapman & Hall Medical, Ho-Min L, Studies on attenuated live JE virus vaccine. Fukai K, ed. Virus Vaccines in Asian Countries. Tokyo: University of Tokyo Press, Eckels KH, Yong-Xin Y, Dubois DR, Marchette NJ, Trent DW, Johnson AJ, Japanese encephalitis virus live-attenuated vaccine, Chinese strain SA ; adaptation to primary canine kidney cell cultures and preparation of a vaccine for human use. Vaccine 6: Xin YY, Ming ZG, Peng GY, Jian A, Min LH, Safety of a live-attenuated Japanese encephalitis virus vaccine (SA ) for children. Am J Trop Med Hyg 39: Hoke CH, Nisalak A, Sangawhipa N, Jatanasen S, Laorakapongse T, Innis BL, Kotchasenee S, Gingrich JB, Latendresse J, Fukai K, Burke DS, Protection against Japanese encephalitis by inactivated vaccines. N Engl J Med 319: Andersen MM, Rrnne T, Side-effects with Japanese encephalitis vaccine. Lancet 337: Ruff TA, Eisen D, Fuller A, Kass R, Adverse reactions to Japanese encephalitis vaccine. Lancet 338: Robinson HC, Russell ML, Csokonay WM, Japanese encephalitis vaccine and adverse effects among travellers. Can Dis Wkly Rep 17: , Nazareth B, Levin J, Johnson H, Begg N, Systemic allergic reactions to Japanese encephalitis vaccines. Vaccine 12: Ohtaki E, Matsuishi T, Hirano Y, Maekawa K, Acute disseminated encephalomyelitis after treatment with Japanese B encephalitis vaccine (Nakayama-Yoken and Beijing strains). J Neurol Neurosurg Psychiatry 59: Plesner A, Arlien-Srborg P, Herning M, Neurological complications and Japanese encephalitis vaccination. Lancet 348: Konishi E, Pincus S, Paoletti E, Laegreid WW, Shope RE, Mason PW, A highly attenuated host range-restricted vaccinia virus strain, NYVAC, encoding the prm, E, and NS1 genes of Japanese encephalitis virus prevents JEV viremia in swine. Virology 190: Konishi E, Pincus S, Paoletti E, Shope RE, Mason PW, Avipox virus-vectored Japanese encephalitis virus vaccines: use as vaccine candidates in combination with purified subunit immunogens. Vaccine 12: Harrington DG, Hilmas DE, Elwell MR, Whitmire RE, Stephen EL, Intranasal infection of monkeys with Japanese encephalitis virus: clinical response and treatment with a nuclease-resistant derivative of poly(i)-poly(c). Am J Trop Med Hyg 26: Raengsakulrach B, Nisalak A, Gettayacamin M, Thirawuth V, Young GD, Myint KSA, Ferguson LM, Hoke CH Jr, Innis BL, Vaughn DW, Safety, immunogenicity, and protective efficacy of NYVAC-JEV and ALVAC-JEV recombinant Japanese encephalitis vaccines in rhesus monkeys. Am J Trop Med Hyg 60: Innis BL, Nisalak A, Nimmannitya S, Kusalerdchariya S, Chongswasdi V, Suntayakorn S, Puttisri P, Hoke CH, An enzyme-linked immunosorbent assay to characterize dengue infections where dengue and Japanese encephalitis cocirculate. Am J Trop Med Hyg 40: Clarke DH, Casals J, Techniques for hemagglutination and hemagglutination inhibition with arthropod-borne viruses. Am J Trop Med Hyg 7: Russell PK, Nisalak A, Sukhavachana P, Vivona S, A plaque reduction test for dengue virus neutralization antibodies. J Immunol 99: Kuno G, Gubler DJ, Santiago de Weil N, Antigen capture ELISA for the identification of dengue viruses. J Virol Methods 12: Chomczynski P, Sacchi N, Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162: Myint KSA, Raengsakulrach B, Young GD, Gettayacamin M, Ferguson LM, Innis BL, Hoke CH Jr, Vaughn DW, Production of lethal infection that resembles fatal human disease by intranasal inoculation of macaques with Japanese encephalitis virus. Am J Trop Med Hyg 60:

9 INTRANASAL CHALLENGE MODEL FOR TESTING JE VACCINES Miura T, A new method for a potency test of Japanese encephalitis vaccine. Direct challenge method on suckling mice by subcutaneous inoculation. Bull World Health Org 43: Yasuda A, Kimura-Kuroda J, Ogimoto M, Miyamoto M, Sata T, Sato T, Takamura C, Kurata T, Kojima A, Yasui K, Induction of protective immunity in animals vaccinated with recombinant vaccinia viruses that express PreM and E glycoproteins of Japanese encephalitis virus. J Virol 64: Mason PW, Pincus S, Fournier MJ, Mason TL, Shope RE, Paoletti E, Japanese encephalitis virus-vaccinia recombinants produce particulate forms of the structural membrane proteins and induce high levels of protection against lethal JEV infection. Virology 180: Lee HW, Scherer WF, The anamnestic antibody response to Japanese encephalitis virus in monkeys and its implications concerning naturally acquired immunity in man. J Immunol 86: Imam IZE, Hammon WM, Challenge of monkeys with Japanese B virus after immunization by West Nile infection plus Japanese B vaccine. J Immunol 79: Zhang M, Wang M, Jiang S, Ma W, Passive protection of mice, goats, and monkeys against Japanese encephalitis with monoclonal antibodies. J Med Virol 29: Lee GC, Grayston JT, Wang S, Protective studies in mice and monkeys with an inactivated Japanese encephalitis virus vaccine grown in hamster diploid cell culture. Proc Soc Exp Biol Med 125: Ghosh SN, Goverdhan MK, Sathe PS, Chelliah SC, Naik SV, Godbole PV, Banerjee K, Protective effect of 6-MFA, a fungal interferon inducer against Japanese encephalitis virus in bonnet macaques. Indian J Med Res 91: Leake CJ, Burke DS, Nisalak A, Hoke CH, Isolation of Japanese encephalitis virus from clinical specimens using a continuous mosquito cell line. Am J Trop Med Hyg 35: Kumar R, Mathur A, Kumar A, Sharma S, Saksena PN, Chaturvedi UC, Japanese encephalitis an important cause of acute childhood encephalopathy in Lucknow, India. Postgrad Med J 64: