- Email: [email protected]
Induction of protective immunity by topic application of a recombinant adenovirus expressing rabies virus glycoprotein
Veterinary Microbiology 85 (2002) 295±303
Induction of protective immunity by topic application of a recombinant adenovirus expressing rabies virus glycoprotein Cammy Y. Leesa, Deborah J. Briggsa, Xianfu Wua,1, Rolan D. Davisa, Susan M. Moorea, Chandra Gordona, Zhiquan Xiangb, Hildegund C.J. Ertlb, De-chu C. Tangc, Zhen F. Fua,1,* a
Department of Diagnostic Medicine/Pathobiology, College of Veterinary Medicine, Kansas State University, Manhattan, KS 66506, USA b Wistar Institute, 3600 Spruce Street, Philadelphia, PA 19104, USA c Vaxin Inc., Birmingham, AL 35209, USA
Received 19 April 2001; received in revised form 8 November 2001; accepted 30 November 2001
Abstract The objective of this study was to determine if a replication defective recombinant adenovirus expressing rabies virus glycoprotein (Adrab.gp) given through a non-invasive vaccination route (by topical application) onto the skin (NIVS) could elicit an immune response and/or protection against rabies. Groups of mice were immunized by NIVS with various doses of Adrab.gp. For comparison, groups of mice were immunized intramuscularly, subcutaneously, or intradermally with Adrab.gp. Mice received two booster immunizations at 1 and 2 months after the ®rst immunization. Virus neutralizing antibody (VNA) titers were measured at day 21 after the ®rst and second immunizations and at day 14 after the third immunization. Fifty percent of the mice immunized by NIVS with 2 107 and 2 108 pfu Adrab.gp vaccine developed VNA, whereas none of the control mice or the mice immunized by NIVS with the lowest dose (2 106 pfu) of Adrab.gp virus developed VNA. However, this low dose induced high titers of VNA in mice immunized by parenteral routes. Two weeks after the last immunization, all the mice were challenged with a lethal dose of rabies virus. More than 70% of the animals immunized by NIVS
* Corresponding author. Tel.: 1-706-542-7021; fax: 1-706-542-5828. E-mail address: [email protected] (Z.F. Fu). 1 Present address: Department of Pathology, College of Veterinary Medicine, The University of Georgia, D.W. Brooks Drive, Athens, GA 30602-7388, USA.
0378-1135/02/$ ± see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 1 1 3 5 ( 0 1 ) 0 0 5 2 3 - 5
296
C.Y. Lees et al. / Veterinary Microbiology 85 (2002) 295±303
with 2 107 pfu Adrab.gp virus survived the challenge, whereas all the mice in the negative control group and the group immunized by NIVS with the lowest dose of Adrab.gp succumbed to rabies. Taken together, the results suggest that NIVS with Adrab.gp can induce VNA production and protection against lethal challenge with rabies virus in mice. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Rabies virus; Adenovirus; Virus neutralizing antibodies (VNAs); Protection; Non-invasive vaccination onto the skin (NIVS)
1. Introduction Rabies still presents a health threat to humans, especially in less developed nations with uncontrolled canine rabies endemic (Meslin and Fishbein, 1994). Worldwide more than four million people require post-exposure vaccination every year after receiving bites and scratches by domestic as well as wild animals (Meslin and Fishbein, 1994). Current recommendations for post-exposure prophylaxis by the World Health Organization include a series of vaccination (®ve to eight doses depending on the route of immunization) with inactivated rabies virus vaccines prepared from cell culture (CDC, 1999; WHO, 1992). These vaccinations are administered by either intramuscular or intradermal injections. Unfortunately, this is often traumatic and people may not come back to ®nish the series of vaccinations. Oral vaccination, although economical and easy to administer, has only been used to immunize wildlife animals against rabies (Brochier et al., 1991; Hanlon et al., 1998). The transdermal route has been used by the pharmaceutical industry for the delivery of various low molecular weight drugs (Babiuk et al., 2000). Recently, it has been reported that non-invasive vaccination onto the skin (NIVS), through topical application of vaccines (DNA vaccine, peptide vaccines or recombinant adenovirus vaccines) directly onto the bare skin, results in the induction of immunity to the target antigens (Tang et al., 1997; Glenn et al., 1999; Shi et al., 1999). The delivery of vaccines through the NIVS may have many advantages over the traditional parenteral routes of immunization. It is a relatively easy and painless procedure requiring no speci®cally trained personnel and equipment. The procedure will reduce many of the side effects associated with traditional immunizations. Furthermore, the elimination of the use of needles for immunization would minimize problems associated with the disposal of needles and undesirable transmission of diseases such as HIV and hepatitis C by reused needles, particularly in the developing world. A replication-defective recombinant adenovirus expressing rabies virus glycoprotein (Adrab.gp) of the Evelyn Rokitnicki Abelseth strain was developed (Xiang et al., 1996). The Adrab.gp vaccine is capable of inducing virus neutralizing antibodies (VNAs) in mice and dogs when given by intramuscular, subcutaneous, intradermal, or intranasal routes of immunization (Xiang et al., 1996; Wang et al., 1997; Tims et al., 2000). In the present study, we investigated the feasibility of using Adrab.gp by NIVS to elicit neutralizing antibody response to and protection against rabies virus. Mice immunized by NIVS with Adrab.gp produced VNA to, and were protected against lethal challenge with rabies virus.
C.Y. Lees et al. / Veterinary Microbiology 85 (2002) 295±303
297
2. Materials and methods 2.1. Animals, vaccine, and virus Female ICR mice were bred and kept in the temperature-controlled Animal Research Facility at Kansas State University. They had access to food and water ad libitum. Recombinant Adrab.gp was grown and puri®ed as described previously (Wang et al., 1997). The puri®ed Adrab.gp used in this study has a titer of 2 109 plaque forming units (pfu) per milliliter. The challenge standard rabies virus (CVS-24) was prepared from infected brains of neonatal mice (Fu et al., 1991). 2.2. Vaccination Seventy female ICR mice (6±7 weeks old) were divided into 7 groups of 10 mice each. Three groups of mice were immunized by NIVS with 2 106 , 2 107 and 2 108 pfu of Adrab.gp, respectively. These mice were anesthetized with sodium pentobarbital (50 mg/kg body weight) intraperitoneally, and an area (approximately 2 cm 1 cm) of the skin on the back of the mouse was shaved. After the shaved area was brushed to remove corni®ed epithelium, diluted Adrab.gp in a volume of 50 ml was applied topically to the shaved area, which then was covered with a Targederm (3 M) patch. After 1 h incubation, the Targederm patch was removed, and the vaccinated area of the skin was washed with tap water. Another three groups of mice were immunized with 2 106 pfu of Adrab.gp virus intramuscularly, subcutaneously, or intradermally. Mice received booster immunizations at 1 and 2 months after the ®rst immunization. One group of mice was left unvaccinated. Blood samples were collected via the tail vein at 21 days after the ®rst and second immunizations and 14 days after the third immunization to test for VNA. 2.3. Rapid fluoroscent focus inhibition test (RFFIT) The VNA was measured using the modi®ed RFFIT as described (Briggs et al., 1996). Antibody titers were expressed as the reciprocal dilution of the serum that neutralized 100% of the infecting virus. Antibody titers also were expressed as international units (IU), and animals showing IU equal to or greater than 0.5 were considered seroconverted. 2.4. Challenge Two weeks after the last immunization, mice were challenged intramuscularly in the hind leg with 10 MIMLD50 (mouse intramuscular (IM) median lethal dose) of CVS-24 virus as described (Fu et al., 1991). Animals were observed for 20 days for signs of rabies such as ataxia and paralysis. Once moribund, mice were euthanized. 3. Results Table 1 summarizes the rate of seroconversion and geometric mean titers (GMTs) in each of the immunized groups. After the ®rst immunization, 20 and 50% of the mice
298
Mouse group
First immunization
Second immunization
Third immunization
Seroconversion rate (%)
GMT
GMT IU
Seroconversion rate (%)
GMT
GMT IU
Seroconversion rate (%)
GMT
GMT IU
NIVS EE 6 NIVS EE 7 NIVS EE 8 IM SC ID Control
0 20 50 87.5 80 100 0
0 1/134 1/369 1/1282 1/228 1/1482 0
0 1.1 3.7 11.0 1.9 17.2 0
0 50 50 100 100 100 0
0 1/100 1/875 1/3056 1/1242 1/2453 0
0 0.925 8.3 29.2 11.9 23.4 0
0 50 50 100 100 100 0
0 1/178 1/356 12507 1/1047 1/2449 0
0 1.5 3.0 21.8 9.0 21.3 0
a NIVS, non-invasive vaccination onto the skin; EE n, 2 10n ; IM, intramuscular. SC, subcutaneous, ID, intradermal. The GMT is not significantly different within the groups via parental immunizations, or between the groups immunized via NIVS with 107 and 108 pfu of Adrab.gp at the 5% level of probability. However, the GMT in the groups immunized via NIVS with 107 and 108 pfu of Adrab.gp is significantly higher than that in the control group and in the group immunized via NIVS with 106 pfu of Adrab.gp. The GMT in the groups via parental immunizations is significantly higher than that in the groups immunized via NIVS with 107 and 108 pfu of Adrab.gp.
C.Y. Lees et al. / Veterinary Microbiology 85 (2002) 295±303
Table 1 The rate of seroconversion and GMT (IU) values for each mouse group after each immunizationa
C.Y. Lees et al. / Veterinary Microbiology 85 (2002) 295±303
299
immunized by NIVS with 2 107 and 2 108 pfu of Adrab.gp, respectively, developed VNA titers. After the second immunization, the rate of seroconversion increased to 50% in the group of mice immunized by NIVS with 2 107 pfu of Adrab.gp. The rate of seroconversion remained 50% in the group of mice immunized by NIVS with 2 108 pfu of Adrab.gp, but the GMT increased. After the third immunization, the GMTs for these two groups did not increase. None of the mice in the group immunized by NIVS with 2 106 pfu of Adrab.gp developed detectable VNA throughout the study (Table 1). In the groups of mice immunized intradermal (ID), 100% of the animals were seroconverted after the ®rst immunization (Table 1). In the groups of mice immunized IM and subcutaneous (SC), 87.5 and 80% of the animals seroconverted after the ®rst immunization and the rate of seroconversion then increased to 100% after the second immunization (Table 1). The GMTs for each of the groups after each immunization are summarized in Table 1. Overall, the GMT increased after the second immunization but remained the same or declined slightly after the third immunization. None of mice in the unvaccinated group had detectable VNA throughout the study. The daily survival rates of mice in each immunization group after challenge with rabies virus are shown in Fig. 1. The numbers of animals receiving challenge infection, the numbers of animals showing VNA titers above the 0.5 IU at the time of challenge, the numbers of animals succumbing to rabies after challenge, and the overall survival rates are summarized in Table 2. In the groups immunized by NIVS with 2 107 and 2 108 pfu of Adrab.gp, 70 and 75% of the mice were protected against challenge. All the animals immunized by the IM and ID routes were protected against challenge, and 71.4% of the mice immunized by SC survived the challenge. All the mice in the group immunized by NIVS with 2 106 pfu of Adrab.gp and in the non-vaccinated group succumbed to rabies.
Fig. 1. The daily survival rate (%) in each of the immunization groups after challenge. Two weeks after the last immunization, mice were challenged with 10 MIMLD50 of CVS-24 rabies virus. Animals were observed daily for 20 days after challenge. No more animals developed rabies after the 12th day.
300
C.Y. Lees et al. / Veterinary Microbiology 85 (2002) 295±303
Table 2 The number of mice present at the time of challenge, the number of mice with titers greater than 0.5 IU, the number of mice that succumb to challenge, and the survival rate for each of the immunization groups Mouse groupa,b
Number of mice challenged
Number of mice with VNA >0.5 IU
Number of mice succumb to rabies
Survival rate (%)
NIVS EE 6 NIVS EE 7 NIVS EE 8 IM SC ID Control
10 10 8c 10 7c 9c 10
0 5 4 10 7 9 0
10 3 2 0 2 0 10
0 70 75 100 71.4 100 0
a NIVS, non-invasive skin immunization; EE n, 2 10n ; IM, intramuscular. SC, subcutaneous; ID, intradermal. b The rate of seroconversion in the groups immunized via NIVS with 107 and 108 pfu of Adrab.gp is significantly higher than that in the control group and in the group immunized via NIVS with 106 pfu of Adrab.gp. The rate of seroconversion in the groups via parental immunizations is significantly higher than that in the groups immunized via NIVS with 107 and 108 pfu of Adrab.gp. However, the survival rate in the groups via NIVS immunization with 107 and 108 pfu of Adrab.gp is not significantly different from that in the groups via parental immunizations. c The numbers of animals at the time of challenge are different from the numbers of animals receiving immunization due to unrelated deaths.
Statistical analysis of the data demonstrates that the rate of seroconversion and the geometric mean antibody titers in the groups immunized by NIVS with 107 pfu of Adrab.gp are signi®cantly higher than the control group, but are signi®cantly lower than the groups of animals immunized by the parenteral routes. However, the survival rate in the groups immunized by NIVS with 107 pfu of Adrab.gp is not signi®cantly different from that in the groups of animals immunized by the parenteral routes. 4. Discussion The most important ®nding in the present study is that immunization of mice by a NIVS with a recombinant Adrab.gp induced a neutralizing antibody response that provided protection against a lethal challenge with rabies virus. The data from this study con®rmed previous ®ndings that direct immunization onto the skin is capable of inducing immune responses in mice (Tang et al., 1997; Shi et al., 1999). Our study showed further that animals immunized in such a non-invasive mode could be protected against a lethal challenge with rabies virus. The induction of immune responses and protection by NIVS depended on the dose of the vaccine. The VNA were detected in 0, 20 and 50% of the mice immunized by NIVS with 2 106 , 2 107 and 2 108 pfu Adrab.gp, respectively, after the ®rst immunization. The rate of seroconversion increased to 50% for the group of mice immunized by NIVS with 2 107 pfu Adrab.gp after the second immunization. The GMT is slightly higher in mice immunized with 2 108 pfu than in mice immunized with 2 107 pfu of Adrab.gp after each of the immunizations. In contrast, no VNA was detected in any of the mice immunized
C.Y. Lees et al. / Veterinary Microbiology 85 (2002) 295±303
301
with 2 106 pfu Adrab.gp throughout the study, and all the mice succumbed to rabies upon challenge. These data suggest that 2 107 pfu of Adrab.gp is the minimal dose required to induce an measurable immune response by the topical route of immunization. Previously, Tang et al. (1997) reported that 96% of mice immunized with 108 pfu of a recombinant adenovirus expressing human carcinoembryonic antigen (AdCMV-hcea) developed antibodies. In the present study, only 50% of the mice immunized by NIVS with 2 107 and 2 108 pfu of Adrab.gp developed antibodies to rabies virus. Yet, more than 70% of the mice were protected in these two groups. These results indicate that some of the mice may have been primed with the vaccine, although the VNA was undetectable. Upon challenge, these animals may have developed an anamnestic response that protected them from developing rabies. At the present time, the rate of seroconversion and the VNA titers in the groups of mice immunized by NIVS are signi®cantly lower than those in groups of animals immunized by parenteral routes. In our study, although 100% of animals immunized by the parenteral routes with 2 106 pfu of Adrab.gp developed VNA, only 50% of the animals immunized by NIVS seroconverted when as much as 2 108 pfu of Adrab.gp was given. A previous report indicated that 104 pfu of Adrab.gp were able to induce VNA in all of the immunized mice and to protect 100% of them against challenge (Xiang et al., 1996). These data suggest that the ability of current adenoviral vector to penetrate the skin is still not ef®cient. Future effort should be directed toward improving this ability. Another concern with the adenoviral vector is the immune responses directed to the adenoviral proteins (Wang et al., 1997). Pre-existing anti-adenoviral immunity may prevent the uptake of the vaccine by cells needed for expression of the target gene and, thus, impair the active immune response to the target antigen. Furthermore, revaccination or vaccination with adenoviral vector expressing a different target antigen may no longer be effective. However, Vos et al. (2001) reported recently that pre-existing immunity to adenovirus does not impair the ability of a similar recombinant adenovirus to elicit immune responses. In our study, mice were vaccinated three times 1 month apart. Overall, VNA titers increased in almost all the groups after the second immunization, indicating that prior immunization does not prevent an anamnestic response to a subsequent booster immunization. However, the VNA titers did not increase after the third immunization. The lack of VNA response after the third immunization may be due to the fact that blood was collected 2 weeks afterwards rather than 3 weeks after the ®rst and second immunizations. Nevertheless, the possibility that immunity to the adenovirus eventually may interfere with the stimulation of subsequent immune responses could not be ruled out completely (Wang et al., 1997). Traditional immunization is carried out by injection of vaccines into humans or animals with only a few exceptions such as the oral live poliovirus vaccine (Sabin and Boulger, 1973). Although oral vaccination is economical and easy to administer and, therefore, can achieve wide distribution for mass immunization, all currently used oral vaccines are either live-attenuated or live-recombinant viruses (Sabin and Boulger, 1973; Brochier et al., 1991; Hanlon et al., 1998). Some live-recombinant viruses such as the adenovirus have been shown not to induce an immune response via the oral route in experimental animals (Wang et al., 1997). In addition to the parenteral and the mucosal (oral) routes of immunization, the transdermal route also may be suitable for delivery of vaccines (Babiuk et al., 2000). This route has been used by the pharmaceutical industry for the delivery of
302
C.Y. Lees et al. / Veterinary Microbiology 85 (2002) 295±303
various low molecular weight drugs. Delivery of DNA-based vaccine via a gene gun has demonstrated that immunization through the transdermal route can induce immune responses comparable to those by parenteral immunizations (Johnston and Barry, 1997). Recently, topical application of DNA vaccine, peptide vaccines, or recombinant adenovirus vaccine directly onto the bare skin has resulted in the induction of immune responses (Tang et al., 1997; Glenn et al., 1999; Shi et al., 1999). In the present study, we con®rmed the previous ®ndings that direct immunization onto the skin with a recombinant Adrab.gp resulted in the induction of immune responses in mice. We also extended the previous studies and showed that animals immunized in such a way were protected against a lethal challenge with rabies virus. Together, these studies demonstrate the feasibility and potential of the NIVS as a novel strategy for future vaccination. Acknowledgement CYL was supported by a Howard Hughes Medical Institute Undergraduate Research Scholarship. References Babiuk, S., Baca-Estrada, M., Babiuk, L.A., Ewen, C., Foldvari, M., 2000. Cutaneous vaccination: the skin as an immunologically active tissue and the challenge of antigen delivery. J. Contr. Release 66, 199±214. Briggs, D.J., Dreesen, D.W., Morgan, P., Chin, J.E., Seedle, C.D., Cryz, L., Gluck, R., Cryz, S.J., 1996. Safety and immunogenicity of Lyssavac Berna human diploid cell rabies vaccine in healthy adults. Vaccine 14, 1361±1365. Brochier, B., Kieny, M.P., Costy, F., Coppens, P., Bauduin, B., Lecocq, J.P., Languet, B., Chappuis, G., Desmettre, P., Afiademanyo, K. et al., 1991. Large-scale eradication of rabies using recombinant vacciniarabies vaccine. Nature 354, 520±522. CDC, 1999. Human rabies preventionÐUnited States, 1999: recommendation of the Advisory Committee on Immunization Practices (ACIP). MMWR 48, 1±21. Fu, Z.F., Dietzschold, B., Schumacher, C.L., Wunner, W.H., Ertl, H.C., Koprowski, H., 1991. Rabies virus nucleoprotein expressed in and purified from insect cells is efficacious as a vaccine. Proc. Natl. Acad. Sci. USA 88, 2001±2005. Glenn, G.M., Scharton-Kersten, T., Vassell, R., Matyas, G.R., Alving, C.R., 1999. Transcutaneous immunization with bacterial ADP-ribosylating exotoxins as antigens and adjuvants. Infect. Immun. 67, 1100± 1106. Hanlon, C.A., Niezgoda, M., Hamir, A.N., Schumacher, C., Koprowski, H., Rupprecht, C.E., 1998. First North American field release of a vaccinia-rabies glycoprotein recombinant virus. J. Wildl. Dis. 34, 228±239. Johnston, S.A., Barry, M.A., 1997. Genetic to genomic vaccination. Vaccine 15, 808±809. Meslin, F.X., Fishbein, D.B., 1994. Matter HC. Rationale and prospects for rabies elimination in developing countries. Curr. Top. Microbiol. Immunol. 187, 1±26. Sabin, A.B., Boulger, L.R., 1973. History of the Sabin attenuated poliovirus oral live vaccine strain. J. Biol. Stand. 1, 15. Shi, Z., Curiel, D.T., Tang, D., 1999. DNA-based non-invasive vaccination onto the skin. Vaccine 17, 2136± 2141. Tang, D.C., Shi, Z., Curiel, D.T., 1997. Vaccination onto bare skin. Nature 388, 729±730. Tims, T., Briggs, D.J., Davis, R.D., Moore, S.M., Xiang, X., Ertl, H.C.J., Fu, Z.F., 2000. Adult dogs receiving a rabies booster dose with a recombinant adenovirus expressing rabies virus glycoprotein develop high titers of neutralizing antibodies. Vaccine 18, 2804±2807.
C.Y. Lees et al. / Veterinary Microbiology 85 (2002) 295±303
303
Vos, A., Neubert, A., Pommerening, E., Muller, T., Dohner, L., Neubert, L., Hughes, K., 2001. Immunogenicity of an E1-delected recombinant human adenovirus against rabies by different routes of administration. J. Gen. Virol. 82, 2191±2197. Wang, Y., Xiang, Z., Pasquini, S., Ertl, H.C., 1997. The use of an E1-deleted, replication-defective adenovirus recombinant expressing the rabies virus glycoprotein for early vaccination of mice against rabies. J. Virol. 72, 1790±1796. WHO, 1992. World Health Organization (WHO) Expert Committee on Rabies. Eighth Report. Technical Report Series No. 824. WHO, Geneva, 1992, pp. 1±84. Xiang, Z.Q., Yang, Y., Wilson, J.M., Ertl, H.C., 1996. A replication-defective human adenovirus recombinant serves as a highly efficacious vaccine carrier. Virology 219, 220±227.
Induction of protective immunity by topic application of a recombinant adenovirus expressing rabies virus glycoprotein Cammy Y. Leesa, Deborah J. Briggsa, Xianfu Wua,1, Rolan D. Davisa, Susan M. Moorea, Chandra Gordona, Zhiquan Xiangb, Hildegund C.J. Ertlb, De-chu C. Tangc, Zhen F. Fua,1,* a
Department of Diagnostic Medicine/Pathobiology, College of Veterinary Medicine, Kansas State University, Manhattan, KS 66506, USA b Wistar Institute, 3600 Spruce Street, Philadelphia, PA 19104, USA c Vaxin Inc., Birmingham, AL 35209, USA
Received 19 April 2001; received in revised form 8 November 2001; accepted 30 November 2001
Abstract The objective of this study was to determine if a replication defective recombinant adenovirus expressing rabies virus glycoprotein (Adrab.gp) given through a non-invasive vaccination route (by topical application) onto the skin (NIVS) could elicit an immune response and/or protection against rabies. Groups of mice were immunized by NIVS with various doses of Adrab.gp. For comparison, groups of mice were immunized intramuscularly, subcutaneously, or intradermally with Adrab.gp. Mice received two booster immunizations at 1 and 2 months after the ®rst immunization. Virus neutralizing antibody (VNA) titers were measured at day 21 after the ®rst and second immunizations and at day 14 after the third immunization. Fifty percent of the mice immunized by NIVS with 2 107 and 2 108 pfu Adrab.gp vaccine developed VNA, whereas none of the control mice or the mice immunized by NIVS with the lowest dose (2 106 pfu) of Adrab.gp virus developed VNA. However, this low dose induced high titers of VNA in mice immunized by parenteral routes. Two weeks after the last immunization, all the mice were challenged with a lethal dose of rabies virus. More than 70% of the animals immunized by NIVS
* Corresponding author. Tel.: 1-706-542-7021; fax: 1-706-542-5828. E-mail address: [email protected] (Z.F. Fu). 1 Present address: Department of Pathology, College of Veterinary Medicine, The University of Georgia, D.W. Brooks Drive, Athens, GA 30602-7388, USA.
0378-1135/02/$ ± see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 1 1 3 5 ( 0 1 ) 0 0 5 2 3 - 5
296
C.Y. Lees et al. / Veterinary Microbiology 85 (2002) 295±303
with 2 107 pfu Adrab.gp virus survived the challenge, whereas all the mice in the negative control group and the group immunized by NIVS with the lowest dose of Adrab.gp succumbed to rabies. Taken together, the results suggest that NIVS with Adrab.gp can induce VNA production and protection against lethal challenge with rabies virus in mice. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Rabies virus; Adenovirus; Virus neutralizing antibodies (VNAs); Protection; Non-invasive vaccination onto the skin (NIVS)
1. Introduction Rabies still presents a health threat to humans, especially in less developed nations with uncontrolled canine rabies endemic (Meslin and Fishbein, 1994). Worldwide more than four million people require post-exposure vaccination every year after receiving bites and scratches by domestic as well as wild animals (Meslin and Fishbein, 1994). Current recommendations for post-exposure prophylaxis by the World Health Organization include a series of vaccination (®ve to eight doses depending on the route of immunization) with inactivated rabies virus vaccines prepared from cell culture (CDC, 1999; WHO, 1992). These vaccinations are administered by either intramuscular or intradermal injections. Unfortunately, this is often traumatic and people may not come back to ®nish the series of vaccinations. Oral vaccination, although economical and easy to administer, has only been used to immunize wildlife animals against rabies (Brochier et al., 1991; Hanlon et al., 1998). The transdermal route has been used by the pharmaceutical industry for the delivery of various low molecular weight drugs (Babiuk et al., 2000). Recently, it has been reported that non-invasive vaccination onto the skin (NIVS), through topical application of vaccines (DNA vaccine, peptide vaccines or recombinant adenovirus vaccines) directly onto the bare skin, results in the induction of immunity to the target antigens (Tang et al., 1997; Glenn et al., 1999; Shi et al., 1999). The delivery of vaccines through the NIVS may have many advantages over the traditional parenteral routes of immunization. It is a relatively easy and painless procedure requiring no speci®cally trained personnel and equipment. The procedure will reduce many of the side effects associated with traditional immunizations. Furthermore, the elimination of the use of needles for immunization would minimize problems associated with the disposal of needles and undesirable transmission of diseases such as HIV and hepatitis C by reused needles, particularly in the developing world. A replication-defective recombinant adenovirus expressing rabies virus glycoprotein (Adrab.gp) of the Evelyn Rokitnicki Abelseth strain was developed (Xiang et al., 1996). The Adrab.gp vaccine is capable of inducing virus neutralizing antibodies (VNAs) in mice and dogs when given by intramuscular, subcutaneous, intradermal, or intranasal routes of immunization (Xiang et al., 1996; Wang et al., 1997; Tims et al., 2000). In the present study, we investigated the feasibility of using Adrab.gp by NIVS to elicit neutralizing antibody response to and protection against rabies virus. Mice immunized by NIVS with Adrab.gp produced VNA to, and were protected against lethal challenge with rabies virus.
C.Y. Lees et al. / Veterinary Microbiology 85 (2002) 295±303
297
2. Materials and methods 2.1. Animals, vaccine, and virus Female ICR mice were bred and kept in the temperature-controlled Animal Research Facility at Kansas State University. They had access to food and water ad libitum. Recombinant Adrab.gp was grown and puri®ed as described previously (Wang et al., 1997). The puri®ed Adrab.gp used in this study has a titer of 2 109 plaque forming units (pfu) per milliliter. The challenge standard rabies virus (CVS-24) was prepared from infected brains of neonatal mice (Fu et al., 1991). 2.2. Vaccination Seventy female ICR mice (6±7 weeks old) were divided into 7 groups of 10 mice each. Three groups of mice were immunized by NIVS with 2 106 , 2 107 and 2 108 pfu of Adrab.gp, respectively. These mice were anesthetized with sodium pentobarbital (50 mg/kg body weight) intraperitoneally, and an area (approximately 2 cm 1 cm) of the skin on the back of the mouse was shaved. After the shaved area was brushed to remove corni®ed epithelium, diluted Adrab.gp in a volume of 50 ml was applied topically to the shaved area, which then was covered with a Targederm (3 M) patch. After 1 h incubation, the Targederm patch was removed, and the vaccinated area of the skin was washed with tap water. Another three groups of mice were immunized with 2 106 pfu of Adrab.gp virus intramuscularly, subcutaneously, or intradermally. Mice received booster immunizations at 1 and 2 months after the ®rst immunization. One group of mice was left unvaccinated. Blood samples were collected via the tail vein at 21 days after the ®rst and second immunizations and 14 days after the third immunization to test for VNA. 2.3. Rapid fluoroscent focus inhibition test (RFFIT) The VNA was measured using the modi®ed RFFIT as described (Briggs et al., 1996). Antibody titers were expressed as the reciprocal dilution of the serum that neutralized 100% of the infecting virus. Antibody titers also were expressed as international units (IU), and animals showing IU equal to or greater than 0.5 were considered seroconverted. 2.4. Challenge Two weeks after the last immunization, mice were challenged intramuscularly in the hind leg with 10 MIMLD50 (mouse intramuscular (IM) median lethal dose) of CVS-24 virus as described (Fu et al., 1991). Animals were observed for 20 days for signs of rabies such as ataxia and paralysis. Once moribund, mice were euthanized. 3. Results Table 1 summarizes the rate of seroconversion and geometric mean titers (GMTs) in each of the immunized groups. After the ®rst immunization, 20 and 50% of the mice
298
Mouse group
First immunization
Second immunization
Third immunization
Seroconversion rate (%)
GMT
GMT IU
Seroconversion rate (%)
GMT
GMT IU
Seroconversion rate (%)
GMT
GMT IU
NIVS EE 6 NIVS EE 7 NIVS EE 8 IM SC ID Control
0 20 50 87.5 80 100 0
0 1/134 1/369 1/1282 1/228 1/1482 0
0 1.1 3.7 11.0 1.9 17.2 0
0 50 50 100 100 100 0
0 1/100 1/875 1/3056 1/1242 1/2453 0
0 0.925 8.3 29.2 11.9 23.4 0
0 50 50 100 100 100 0
0 1/178 1/356 12507 1/1047 1/2449 0
0 1.5 3.0 21.8 9.0 21.3 0
a NIVS, non-invasive vaccination onto the skin; EE n, 2 10n ; IM, intramuscular. SC, subcutaneous, ID, intradermal. The GMT is not significantly different within the groups via parental immunizations, or between the groups immunized via NIVS with 107 and 108 pfu of Adrab.gp at the 5% level of probability. However, the GMT in the groups immunized via NIVS with 107 and 108 pfu of Adrab.gp is significantly higher than that in the control group and in the group immunized via NIVS with 106 pfu of Adrab.gp. The GMT in the groups via parental immunizations is significantly higher than that in the groups immunized via NIVS with 107 and 108 pfu of Adrab.gp.
C.Y. Lees et al. / Veterinary Microbiology 85 (2002) 295±303
Table 1 The rate of seroconversion and GMT (IU) values for each mouse group after each immunizationa
C.Y. Lees et al. / Veterinary Microbiology 85 (2002) 295±303
299
immunized by NIVS with 2 107 and 2 108 pfu of Adrab.gp, respectively, developed VNA titers. After the second immunization, the rate of seroconversion increased to 50% in the group of mice immunized by NIVS with 2 107 pfu of Adrab.gp. The rate of seroconversion remained 50% in the group of mice immunized by NIVS with 2 108 pfu of Adrab.gp, but the GMT increased. After the third immunization, the GMTs for these two groups did not increase. None of the mice in the group immunized by NIVS with 2 106 pfu of Adrab.gp developed detectable VNA throughout the study (Table 1). In the groups of mice immunized intradermal (ID), 100% of the animals were seroconverted after the ®rst immunization (Table 1). In the groups of mice immunized IM and subcutaneous (SC), 87.5 and 80% of the animals seroconverted after the ®rst immunization and the rate of seroconversion then increased to 100% after the second immunization (Table 1). The GMTs for each of the groups after each immunization are summarized in Table 1. Overall, the GMT increased after the second immunization but remained the same or declined slightly after the third immunization. None of mice in the unvaccinated group had detectable VNA throughout the study. The daily survival rates of mice in each immunization group after challenge with rabies virus are shown in Fig. 1. The numbers of animals receiving challenge infection, the numbers of animals showing VNA titers above the 0.5 IU at the time of challenge, the numbers of animals succumbing to rabies after challenge, and the overall survival rates are summarized in Table 2. In the groups immunized by NIVS with 2 107 and 2 108 pfu of Adrab.gp, 70 and 75% of the mice were protected against challenge. All the animals immunized by the IM and ID routes were protected against challenge, and 71.4% of the mice immunized by SC survived the challenge. All the mice in the group immunized by NIVS with 2 106 pfu of Adrab.gp and in the non-vaccinated group succumbed to rabies.
Fig. 1. The daily survival rate (%) in each of the immunization groups after challenge. Two weeks after the last immunization, mice were challenged with 10 MIMLD50 of CVS-24 rabies virus. Animals were observed daily for 20 days after challenge. No more animals developed rabies after the 12th day.
300
C.Y. Lees et al. / Veterinary Microbiology 85 (2002) 295±303
Table 2 The number of mice present at the time of challenge, the number of mice with titers greater than 0.5 IU, the number of mice that succumb to challenge, and the survival rate for each of the immunization groups Mouse groupa,b
Number of mice challenged
Number of mice with VNA >0.5 IU
Number of mice succumb to rabies
Survival rate (%)
NIVS EE 6 NIVS EE 7 NIVS EE 8 IM SC ID Control
10 10 8c 10 7c 9c 10
0 5 4 10 7 9 0
10 3 2 0 2 0 10
0 70 75 100 71.4 100 0
a NIVS, non-invasive skin immunization; EE n, 2 10n ; IM, intramuscular. SC, subcutaneous; ID, intradermal. b The rate of seroconversion in the groups immunized via NIVS with 107 and 108 pfu of Adrab.gp is significantly higher than that in the control group and in the group immunized via NIVS with 106 pfu of Adrab.gp. The rate of seroconversion in the groups via parental immunizations is significantly higher than that in the groups immunized via NIVS with 107 and 108 pfu of Adrab.gp. However, the survival rate in the groups via NIVS immunization with 107 and 108 pfu of Adrab.gp is not significantly different from that in the groups via parental immunizations. c The numbers of animals at the time of challenge are different from the numbers of animals receiving immunization due to unrelated deaths.
Statistical analysis of the data demonstrates that the rate of seroconversion and the geometric mean antibody titers in the groups immunized by NIVS with 107 pfu of Adrab.gp are signi®cantly higher than the control group, but are signi®cantly lower than the groups of animals immunized by the parenteral routes. However, the survival rate in the groups immunized by NIVS with 107 pfu of Adrab.gp is not signi®cantly different from that in the groups of animals immunized by the parenteral routes. 4. Discussion The most important ®nding in the present study is that immunization of mice by a NIVS with a recombinant Adrab.gp induced a neutralizing antibody response that provided protection against a lethal challenge with rabies virus. The data from this study con®rmed previous ®ndings that direct immunization onto the skin is capable of inducing immune responses in mice (Tang et al., 1997; Shi et al., 1999). Our study showed further that animals immunized in such a non-invasive mode could be protected against a lethal challenge with rabies virus. The induction of immune responses and protection by NIVS depended on the dose of the vaccine. The VNA were detected in 0, 20 and 50% of the mice immunized by NIVS with 2 106 , 2 107 and 2 108 pfu Adrab.gp, respectively, after the ®rst immunization. The rate of seroconversion increased to 50% for the group of mice immunized by NIVS with 2 107 pfu Adrab.gp after the second immunization. The GMT is slightly higher in mice immunized with 2 108 pfu than in mice immunized with 2 107 pfu of Adrab.gp after each of the immunizations. In contrast, no VNA was detected in any of the mice immunized
C.Y. Lees et al. / Veterinary Microbiology 85 (2002) 295±303
301
with 2 106 pfu Adrab.gp throughout the study, and all the mice succumbed to rabies upon challenge. These data suggest that 2 107 pfu of Adrab.gp is the minimal dose required to induce an measurable immune response by the topical route of immunization. Previously, Tang et al. (1997) reported that 96% of mice immunized with 108 pfu of a recombinant adenovirus expressing human carcinoembryonic antigen (AdCMV-hcea) developed antibodies. In the present study, only 50% of the mice immunized by NIVS with 2 107 and 2 108 pfu of Adrab.gp developed antibodies to rabies virus. Yet, more than 70% of the mice were protected in these two groups. These results indicate that some of the mice may have been primed with the vaccine, although the VNA was undetectable. Upon challenge, these animals may have developed an anamnestic response that protected them from developing rabies. At the present time, the rate of seroconversion and the VNA titers in the groups of mice immunized by NIVS are signi®cantly lower than those in groups of animals immunized by parenteral routes. In our study, although 100% of animals immunized by the parenteral routes with 2 106 pfu of Adrab.gp developed VNA, only 50% of the animals immunized by NIVS seroconverted when as much as 2 108 pfu of Adrab.gp was given. A previous report indicated that 104 pfu of Adrab.gp were able to induce VNA in all of the immunized mice and to protect 100% of them against challenge (Xiang et al., 1996). These data suggest that the ability of current adenoviral vector to penetrate the skin is still not ef®cient. Future effort should be directed toward improving this ability. Another concern with the adenoviral vector is the immune responses directed to the adenoviral proteins (Wang et al., 1997). Pre-existing anti-adenoviral immunity may prevent the uptake of the vaccine by cells needed for expression of the target gene and, thus, impair the active immune response to the target antigen. Furthermore, revaccination or vaccination with adenoviral vector expressing a different target antigen may no longer be effective. However, Vos et al. (2001) reported recently that pre-existing immunity to adenovirus does not impair the ability of a similar recombinant adenovirus to elicit immune responses. In our study, mice were vaccinated three times 1 month apart. Overall, VNA titers increased in almost all the groups after the second immunization, indicating that prior immunization does not prevent an anamnestic response to a subsequent booster immunization. However, the VNA titers did not increase after the third immunization. The lack of VNA response after the third immunization may be due to the fact that blood was collected 2 weeks afterwards rather than 3 weeks after the ®rst and second immunizations. Nevertheless, the possibility that immunity to the adenovirus eventually may interfere with the stimulation of subsequent immune responses could not be ruled out completely (Wang et al., 1997). Traditional immunization is carried out by injection of vaccines into humans or animals with only a few exceptions such as the oral live poliovirus vaccine (Sabin and Boulger, 1973). Although oral vaccination is economical and easy to administer and, therefore, can achieve wide distribution for mass immunization, all currently used oral vaccines are either live-attenuated or live-recombinant viruses (Sabin and Boulger, 1973; Brochier et al., 1991; Hanlon et al., 1998). Some live-recombinant viruses such as the adenovirus have been shown not to induce an immune response via the oral route in experimental animals (Wang et al., 1997). In addition to the parenteral and the mucosal (oral) routes of immunization, the transdermal route also may be suitable for delivery of vaccines (Babiuk et al., 2000). This route has been used by the pharmaceutical industry for the delivery of
302
C.Y. Lees et al. / Veterinary Microbiology 85 (2002) 295±303
various low molecular weight drugs. Delivery of DNA-based vaccine via a gene gun has demonstrated that immunization through the transdermal route can induce immune responses comparable to those by parenteral immunizations (Johnston and Barry, 1997). Recently, topical application of DNA vaccine, peptide vaccines, or recombinant adenovirus vaccine directly onto the bare skin has resulted in the induction of immune responses (Tang et al., 1997; Glenn et al., 1999; Shi et al., 1999). In the present study, we con®rmed the previous ®ndings that direct immunization onto the skin with a recombinant Adrab.gp resulted in the induction of immune responses in mice. We also extended the previous studies and showed that animals immunized in such a way were protected against a lethal challenge with rabies virus. Together, these studies demonstrate the feasibility and potential of the NIVS as a novel strategy for future vaccination. Acknowledgement CYL was supported by a Howard Hughes Medical Institute Undergraduate Research Scholarship. References Babiuk, S., Baca-Estrada, M., Babiuk, L.A., Ewen, C., Foldvari, M., 2000. Cutaneous vaccination: the skin as an immunologically active tissue and the challenge of antigen delivery. J. Contr. Release 66, 199±214. Briggs, D.J., Dreesen, D.W., Morgan, P., Chin, J.E., Seedle, C.D., Cryz, L., Gluck, R., Cryz, S.J., 1996. Safety and immunogenicity of Lyssavac Berna human diploid cell rabies vaccine in healthy adults. Vaccine 14, 1361±1365. Brochier, B., Kieny, M.P., Costy, F., Coppens, P., Bauduin, B., Lecocq, J.P., Languet, B., Chappuis, G., Desmettre, P., Afiademanyo, K. et al., 1991. Large-scale eradication of rabies using recombinant vacciniarabies vaccine. Nature 354, 520±522. CDC, 1999. Human rabies preventionÐUnited States, 1999: recommendation of the Advisory Committee on Immunization Practices (ACIP). MMWR 48, 1±21. Fu, Z.F., Dietzschold, B., Schumacher, C.L., Wunner, W.H., Ertl, H.C., Koprowski, H., 1991. Rabies virus nucleoprotein expressed in and purified from insect cells is efficacious as a vaccine. Proc. Natl. Acad. Sci. USA 88, 2001±2005. Glenn, G.M., Scharton-Kersten, T., Vassell, R., Matyas, G.R., Alving, C.R., 1999. Transcutaneous immunization with bacterial ADP-ribosylating exotoxins as antigens and adjuvants. Infect. Immun. 67, 1100± 1106. Hanlon, C.A., Niezgoda, M., Hamir, A.N., Schumacher, C., Koprowski, H., Rupprecht, C.E., 1998. First North American field release of a vaccinia-rabies glycoprotein recombinant virus. J. Wildl. Dis. 34, 228±239. Johnston, S.A., Barry, M.A., 1997. Genetic to genomic vaccination. Vaccine 15, 808±809. Meslin, F.X., Fishbein, D.B., 1994. Matter HC. Rationale and prospects for rabies elimination in developing countries. Curr. Top. Microbiol. Immunol. 187, 1±26. Sabin, A.B., Boulger, L.R., 1973. History of the Sabin attenuated poliovirus oral live vaccine strain. J. Biol. Stand. 1, 15. Shi, Z., Curiel, D.T., Tang, D., 1999. DNA-based non-invasive vaccination onto the skin. Vaccine 17, 2136± 2141. Tang, D.C., Shi, Z., Curiel, D.T., 1997. Vaccination onto bare skin. Nature 388, 729±730. Tims, T., Briggs, D.J., Davis, R.D., Moore, S.M., Xiang, X., Ertl, H.C.J., Fu, Z.F., 2000. Adult dogs receiving a rabies booster dose with a recombinant adenovirus expressing rabies virus glycoprotein develop high titers of neutralizing antibodies. Vaccine 18, 2804±2807.
C.Y. Lees et al. / Veterinary Microbiology 85 (2002) 295±303
303
Vos, A., Neubert, A., Pommerening, E., Muller, T., Dohner, L., Neubert, L., Hughes, K., 2001. Immunogenicity of an E1-delected recombinant human adenovirus against rabies by different routes of administration. J. Gen. Virol. 82, 2191±2197. Wang, Y., Xiang, Z., Pasquini, S., Ertl, H.C., 1997. The use of an E1-deleted, replication-defective adenovirus recombinant expressing the rabies virus glycoprotein for early vaccination of mice against rabies. J. Virol. 72, 1790±1796. WHO, 1992. World Health Organization (WHO) Expert Committee on Rabies. Eighth Report. Technical Report Series No. 824. WHO, Geneva, 1992, pp. 1±84. Xiang, Z.Q., Yang, Y., Wilson, J.M., Ertl, H.C., 1996. A replication-defective human adenovirus recombinant serves as a highly efficacious vaccine carrier. Virology 219, 220±227.