Inhibition of immune responses against rabies virus by monoclonal antibodies directed against rabies virus antigens Carolin L. Schumacher*, Hildegund C.J. Ertl*, Hilary Koprowski* and Bernhard Dietzschold t* Treatment of mice with a cocktail o f murine anti-rabies monoclonal antibodies (mAb-C) interfered with the ability o f these animals to mount a virus-neutralizing antibody response to rabies vaccine. Administered mAb-C did not affect the induction of rabies virus-specific T-helper cells. The magnitude of the inhibition o f rabies virus-specific B-cell response was dependent on the concentration of the mAb-C and the duration of the mAb-mediated interference was inversely proportional to the biological half-life of the mAb. As long as the serum titres were above a critical threshold, the suppression could not be overcome even by multiple vaccinations. Since injection of mice with immunocomplexes consisting of inactivated rabies virus and mAb rendered the animals non-responsive to a subsequent vaccination with inactivated rabies virus, it is concluded that the mAb-induced suppression might be caused by the formation o f antigen-antibody complexes which exert a negative signalling effect to premature B cells. Keywords: Postexposure treatment; mouse anti-rabies monoclonal antibodies; inhibition of immune response; antigen-antibody complexes
INTRODUCTION Protection against rabies virus infection is the result of a number of specific and non-specific immune mechanisms; however, virus neutralizing antibodies (VNA) are believed to play a major role in immunoprotection against rabies 1. VNA may offer protection, especially in the early critical phase after exposure to rabies virus before the virus reaches the central nervous system 2, It has been shown that in postexposure situations of human rabies, the combined treatment with vaccine and human rabies immunoglobulin was more effective in preventing lethal rabies than was treatment with vaccine alone 3'4. Therefore, a combination of serum and vaccine became the standard treatment of humans for severe exposure to rabies virus 5. Recently, it has been demonstrated that passive immunization of mice and hamsters, with a cocktail of mouse monoclonal antibodies (mAb-C) specific for rabies ribonucleocapsid protein and glycoprotein, protects animals not only when challenged peripherally with a lethal dose of rabies virus after immunization, but also in postexposure situations 6. In addition, a human rabies virus-neutralizing mAb
*The Wistar Institute of Anatomy and Biology, 3601 Spruce Street, Philadelphia, PA 19104, USA. tDepartment of Microbiology, 459 Alumni Hall, Thomas Jefferson University, 1020 Locust Street, Philadelphia, PA 19107, USA. ;To whom correspondence should be addressed. (Received 9 October 1991; revised 12 March 1992; accepted 12 March 1992) 0264-410X/92/110754-07 © 1992Butterworth-HeinemannLtd 754
Vaccine, Vol. 10, Issue 11, 1992
protected laboratory rodents against lethal rabies virus infection v. These protection experiments with mAb-C indicate that rabies virus-specific mAb-C may be utilized for the postexposure treatment of human rabies. One of the major problems connected with the administration of anti-rabies antibodies is the adverse effect of immunoglobulins on the humoral and cellular immune response to rabies vaccine 8'9. It has been shown that injection of rabies antiserum resulted in suppression of the immune response to rabies vaccine 1°. Decrease of the antibody responses by passive immunization has been extensively studied with sheep erythrocytes, where it has been shown that small amounts of antibody against the red blood cells specifically suppress the IgM and IgG antibody response 11. The duration of this suppression was a function of the biological half-life of the antibody and depended on the amount and dose of antibody given 11. Moreover, passive immunization had little or no effect on animals with previous active immunizations ~t. Although the mechanism of antibody-induced unresponsiveness is not fully understood, it is believed that antibody suppresses the development of B cells by forming immunocomplexes with the antigen. The suppressive effect of immunocomplexes formed in vitro on humoral immune responses has been described 11. The present studies were undertaken to determine the suppressive effect of administered anti-rabies virus mAb on the induction of antibodies and T-helper (TH) cells, and thus the protective immune response following immunization with inactivated rabies virus.
Rabies virus-specific monoclonal antibodies cause interference: C.L. Schumacher et al.
MATERIALS AND METHODS
Viruses. The propagation of the fixed rabies virus strains Evelyn-Rokitniki-Abelseth (ERA) and the challenge virus standard (CVS) in BHK-21 cell monolayer cultures has been described previously ~2. Vaccine preparation. Rabies vaccine (ERA-fl-propriolactone [ B P L ] ) was prepared from gradient-purified ERA virus. Inactivation of concentrated ERA virus was carried out with BPL treatment according to the method described by Wiktor et al. 12. Preparation of antibodies. MAb-secreting hybridomas were produced through fusion of splenocytes from rabies-immunized mice with P3 × 63 Ag8 or 654 mouse myeloma cells according to the protocol described by Wiktor and Koprowski 13. MAb were purified from ascites fluid as previously described 6. The production and characterization of the human mAb 57 has been described by Dietzschold et al. 7. Three mAb specific for the glycoprotein (G) and two mAb specific for epitopes on the nucleoprotein complex of rabies virus were used to prepare an mAb-C, as described previously 6. MAb-C represents a mixture of 112.5 IU ml-1 ofmAb 509-6, 10 IU ml-1 ofmAb 1112-1, 800IU m1-1 ofmAb 523-11,0.04 mg m1-1 ofmAb 802-2, and 0.6 mg ml- 1 of mAb 502-2. The neutralizing activity of the cocktail was ~ 1376 IU mg -~ protein. Normal mouse IgG was purified from serum of untreated Balb/c mice by Protein-A-Sepharose chromatography, as previously described 6. Preparation of immunocomplexes. Immunocomplexes were produced by in vitro incubation of 200 IU murine anti-rabies mAb-C with 100 #g E R A - B P L virus in 5 ml phosphate-buffered saline. Incubation was carried out for 2 h at 37°C and overnight at 4°C. Immunocomplexes were separated from free mAb-C by repeated ultracentrifugation for 1 h at 100 000g. The pellet containing the immunocomplex was reconstituted in phosphatebuffered saline. Protection experiments with mAb and E R A - B P L vaccine. Six-week-old female white Swiss mice (ICR) or 6-week-old female Balb/c mice (Harlan SpragueDawley, Inc., Indianapolis, IN) were injected intramuscularly (i.m.) in the gastrocnemius muscle with 0.1 ml of serial dilutions of antibodies. Mice were vaccinated intraperitoneally (i.p.) at different times after mAb treatment with E R A - B P L vaccine. Blood was withdrawn from these animals at different time intervals for determination of neutralizing antibodies as described ~. Seven days after vaccination, the mice were challenged intracerebrally (i.c.) with 25 MICLDso (MICLDso is the i.c. dose which is lethal for 50% of naive age-matched mice) of the CVS-11 rabies virus strain. Animals were observed for 3 weeks after challenge for development of symptoms and death. Tn cellassay. Groups of three 6-8-week-old Balb/c mice were treated i.m. with either mAb-C or the equivalent amount of normal Balb/c IgG and subsequently vaccinated i.m. with 1/~g E R A - B P L virus at different time points. Untreated groups of three mice were vaccinated with 1/zg E R A - B P L virus on the same days
as mAb-treated animals. Seven days after vaccination, lymphocytes (4 x 106) from the popliteal and medial iliac lymph nodes of immune animals were cultured at 37°C in 1.6 #1 of Dulbecco's modified Eagle's medium, supplemented with 2% fetal bovine serum and 10 -5 M 2-mercaptoethanol, in flat-bottomed 24-well plates with different dilutions of E R A - B P L virus or medium. Supernatants were harvested 20-24 h later and tested on lymphokine (interleukin 2 and 4)-sensitive HT-2 cells. Quadruplets of 2 x 103 HT-2 cells in 100 #1 of medium supplemented with 10% fetal bovine serum were incubated at 37°C in 96-well round-bottomed plates with 50/~1 of cell culture supernatants, medium (negative control), or 10% rat concanavalin A supernatant (positive control). Proliferation was measured 36-48 h later after a 6 8 h [3H]methyl-thymidine pulse (0.7 #Ci/ well) by liquid scintillation counting. A threefold counts min- 1 increase over the medium control was considered a significant proliferation. Values are presented as the mean of four single counts ___the standard deviation.
Virus neutralization assay. VNA titres of serum samples were tested with a modified version of the rapid fluorescent focus inhibiting test as described by Zalan et al. 14. The reciprocal of the serum dilution resulting in a 50% reduction of infected foci was considered the neutralization titre of the serum sample. RESULTS
Effect of administered mAb-C on vaccine-mediated protective immunity To investigate whether administration of anti-rabies mAb-C interferes with the production of protective immunity induced by a rabies vaccine, groups of ten ICR mice were treated i.m. with 10 IU per mouse of mAb-C. After 24 h, mAb-C treated and untreated control animals were vaccinated i.p. with 0.016 to 2.0/~g E R A - B P L vaccine. Seven days after the primary vaccination, the groups of mice were revaccinated with the same amounts of vaccine, and 7 days after the second vaccination, all animals were infected i.e. with 25 MICLDso of CVS-11 virus. The i.e. route of infection was chosen because, as previously shown 6, only active immunized animals and not passive immunized animals are protected against an i.e. challenge infection. Therefore, protection from an i.e. infection is an indicator of vaccine-mediated immunity. Table I shows that, depending on the vaccine dose, Table 1 Effect of mAb-C treatment on vaccine-induced protective immunity
Mortality ( % )a after Vaccine concentration b (/~g)
Vaccination ~ only
mAb-C treatmen~ and vaccination
2 0.4 0.08 0.016 None
10 30 80 90 100
100 90 90 100 100
aMice were challenged i.c. 7 days after second vaccination with 25 MICLDso of CVS-11 virus ~Mice were vaccinated with ERA-BPL virus on days 1 and 10 CGroups of ten ICR mice were pretreated with 10 IU of mAb-C 24 h before vacci nation
Vaccine, Vol. 10, Issue 11, 1992
755
Rabies virus-specific monoclonal antibodies cause interference: C.L. Schumacher et al.
9 0 - 1 0 0 % of vaccinated animals survived the lethal challenge infection. However, the protective effect of even the highest doses of vaccine was almost totally abolished when the mice were pretreated with 10 IU mAb-C prior to vaccination.
Dependence of interference induction on the serum concentration of administered antibodies To demonstrate that the induction and duration of interference depends on the serum concentration of the administered antibodies, groups of ten Balb/c mice were treated with different amounts of mAb-C (10-0.04 IU). The mice received four vaccinations with 5 m g E R A - B P L vaccine per mouse 2, 12, 62 and 92 days after mAb-C treatment, and were bled 2 days after mAb-C treatment and 10 days after each vaccination. The results of this experiment are summarized in Table 2. In contrast to the groups of untreated control animals, which had already exhibited a vaccine-induced B-cell response after the first vaccination (geometric mean titre [ G M T ] 1:103), animals pretreated with I0 or 5 IU of mAb-C showed an increase in VNA titres only after the third vaccination. Mice pretreated with 1 or 0.2 IU of mAb-C showed an increase in VNA titres after the second vaccination, while animals pretreated with 0.04IU exhibited, like untreated animals, a vaccine-induced immune response after the first vaccination. However, the VNA G M T in animals pretreated with as little as 0.04IU of mAb-C were, after each vaccination, significantly lower than the corresponding titres in the control group. These results indicate that the inhibition of a vaccine-induced VNA response is dependent on the serum concentration of the administered mAb. Production of VNA is apparently only possible when the serum concentration of the administered mAb is below a certain threshold level ( G M T of 1:10 to 1:30). The time interval between mAb-C treatment and vaccination seems to play a role in the induction of interference only in so far as the titre of a particular administered mAb decreases with time. In order to demonstrate a correlation between the mAb-induced non-responsiveness of rabies-specific B cells and the lack of protective immunity in response to
Table 2 Dependence of interference on serum concentrations of administered mAb-C VNA GMT" Days after mAb-C treatment Treatmentb
mAb dose
2
12
22
72
101
mAb-C + vaccine mAb-C + vaccine mAb-C + vaccine mAb-C + vaccine mAb-C + vaccine Vaccine only
10 5 1 0.2 0.04 0
110 52 24 8 7 3
82 23 10 12 50 103
35 16 30 141 269 479
304 412 772 1078 1102 1812
710 509 1373 2193 3175 6139
aMice were bled 2 days after mAb-C treatment and 10 days after each vaccination. VNA were determined as described in Materials and methods. Underlined numbers represent an increase of VNA after vaccination bGroups of ten Balb/c mice were treated i.m. with different amounts of mAImC and then vaccinated four times with 5/~g ERA-BPL virus i.p. 2, 12, 62 and 92 days after mAb-C treatment
756
Vaccine, Vol. 10, Issue 11, 1992
Table 3
Duration of mAb-C-induced non-responsiveness to vaccination
Treatment"
Time interval between mAb-C treatment and vaccination (days)
Mortality ( % ).
mAb-C + vaccine
0
86
mAb-C + vaccine
7
100
mAb-C + vaccine
14
70
mAb-C + vaccine
21
90
Vaccine only
-
20
None
-
100
VNA GMT° (range) 133 (90 135) 130 (90-135) 102 (90-135) 84 (90-135) 1356 (810-3645) 0 (-)
"Groups of ten Balb/c mice were treated i.m. with 10 IU of mAb-C and vaccinated after the indicated time intervals with 2 #g ERA-BPL virus ~Mice were challenged i.c. 7 days after the second vaccination with 25 MICLDso of CVS-11 virus ~Mice were bled 1 day before challenge infection, and VNA titres were determined as described in Materials and methods
vaccine, part of the experiment was repeated as follows: groups of ten Balb/c mice were treated with 10 IU per mouse of mAb-C and then received after different time intervals (0-21 days) 2/tg of E R A - B P L vaccine. The mice were then revaccinated 7 days after the first vaccination with the same amount of vaccine. Seven days after the second vaccination, the mice were bled and then infected i.c. with 25 MICLDso ofCVS-11 virus. As shown in Table 3, the group of mice which received vaccine alone developed a VNA G M T of 1:1356 and 80% of the animals in this group survived the challenge infection. In contrast, the serum VNA G M T in all groups of mice which were pretreated with mAb-C were at least ten times lower ( 1 : 84 to 1 : 133 ) regardless of the time lapse between mAb-C treatment and vaccination. The lower VNA titres in the mAb-C pretreated mice are reflected by a strong decrease in protective immunity, as indicated by the high mortality rates in these groups of mice. Even when the mice were vaccinated 21 days after mAb-C administration, 90% of these vaccinated animals succumbed to a rabies virus infection. Dependence of interference induction on biological half-life of the antibody To investigate the relationship between duration of mAb-induced interference and the half-life of a particular antibody, the half-lives of administered murine and human mAb in Balb/c mice were determined. Groups often Balb/c mice were treated i.m. with either 50 IU (6 #g) per mouse of homologous murine mAb 523-11, or with 50 IU (4 ~g) of heterologous human mAb 57. The mice were bled 1, 3, 7, 14, 21 and 28 days after mAb administration, to determine the circulating VNA titres. Figure 1 shows the time-dependent decrease of VNA titres of the two mAbs. To compare the titres at different time points, the serum VNA titres of each antibody preparation 24 h after administration were taken as 100%. According to this calculation, the half-life of the murine mAb is approximately 6 days compared with only 4 days for the human mAb. Of the murine mAb, 25% were still circulating in the serum 14 days
Rabies virus-specific monoclonal antibodies cause interference: C.L. Schumacher et al.
'°°I 8o
g
!
0
5
I0
15
20
Days a f t e r mAb-treatment Figure I Clearance of murine mAb 523-11 ( 0 ) and human mAb 57 (FI) in Balb/c mice. Groups of ten Balb/c mice were treated i.m. with either 50 IU (6 #g) per mouse of murine mAb 523-11, or with 50 IU (4 #g) of human mAb-57. Mice were bled 3, 7, 14, 21 and 28 days after mAb administration and VNA GMT were determined. VNA GMT from 3 to 28 days are expressed as percentage of the VNA GMT determined 24 h after mAb administration. Data are plotted using a Stalview SE + Graphics T H program
after injection, while the human mAb was no longer detectable. The kinetics of the single murine mAb 523-11 and mAb-C were identical (data not shown). To demonstrate that the duration of antibodymediated interference is a function of the half-life of a particular administered mAb, groups of Balb/c mice were treated i.m. with 5 IU of murine mAb 523-11, or 5 IU of human mAb 57. The animals were then vaccinated three times with 5/~g of E R A - B P L vaccine 1, 10 and 20 days after mAb administration. The mice were bled 24 h after mAb administration and 10 days after each vaccination, and VNA G M T were determined as described in Materials and methods. The VNA G M T increased after the second immunization in mice treated with the human mAb, while the VNA titres in mice treated with similar amounts of murine mAb increased only after the third vaccination (Table 4). These results indicated that the duration of mAb-mediated interference is reciprocally proportional to the biological half-life of the mAb.
antigen-specific Ta cell responses between mice pretreated with mAb-C or with normal Balb/c IgG. Ta cells from ERA-BPL-vaccinated mice not pretreated or pretreated with normal Balb/c IgG secreted similar amounts of interleukin as T H cells from mice that were pretreated with mAb-C. Therefore, the suppression of a B-cell response in mAb-C pretreated mice is not due to suppression of rabies virus-specific T n cell activation. (ii) Effect of immunocomplexes on production of rabies virus-specific antibodies: To investigate the effect of immunocomplexes on activation of rabies virusspecific B cells, groups of Balb/c mice were immunized with immunocomplexes ( E R A - B P L virus-mAb complexes) in combination with free E R A - B P L virus as follows: group 1 received a primary vaccination with 5/~g immunocomplexes i.p. per mouse, and 10 and 20 days later two booster immunizations i.p., each with 5 #g E R A - B P L vaccine per mouse; the second group received a primary immunization with 5/~g E R A - B P L vaccine, followed by a booster immunization with 5/~g immunocomplexes i.p. on day 10, and a booster immunization with 5/~g E R A - B P L vaccine on day 20; the third group of mice received a primary immunization with 5 #g E R A - B P L vaccine per mouse, followed by two booster immunizations with 5/~g E R A - B P L vaccine 10 and 20 days after primary immunization. Ten days after each immunization, the mice were bled to determine the circulating VNA titres. Table 5 shows that mice of group 1 did not develop significant VNA titres after primary immunization with
Table 4
Effect of mAb half-life on the duration of interference VNA GMT" Days after mAb treatment
Treatment ~
1
11
21
31
mAb 523-11 + vaccine mAb 57 + vaccine Vaccine only
84 171 0
15 10 73
10 89 354
68 396 1918
aMice were bled 1 day after mAb treatment and 10 days after each vaccination. VNA titres were determined as described in Materials and methods bGroups of ten Balb/c mice were treated i.m. with either 5 IU of murine mAb 523-11 or 5 IU of human mAb 57, and then vaccinated three times with 5 #g ERA-BPL virus 1, 10 and 20 days after mAb treatment
Study of mechanisms of mAb-mediated interference (i) Effect of administered mAb-C on the induction of virus-specific T H cells: To determine whether the suppression of a B-cell response by administered mAb is the result of an.inhibition of T H cell activation, Balb/c mice received i.m. either 10 IU ofmAb-C or an equivalent amount (i.e. 7/~g) of normal Balb/c IgG. The mice were immunized i.m. with 1 #g of E R A - B P L vaccine 1, 2, 7, 14 or 28 days after mAb-C administration. In addition, mice that were not pretreated with mAb-C were immunized with 1 #g E R A - B P L vaccine on the same days as mAb-C-treated mice. One week after immunization, the mice were killed humanely, and lymphocytes were isolated from draining lymph nodes and then cultured at 37°C for 24 h in the presence of 5/tg or 1 #g of E R A - B P L vaccine or without antigen. The amounts of lymphokines produced in vitro by these lymphocytes in response to rabies antigen are illustrated in Figure 2. No significant differences were observed in the rabies
Table 5 Effect of immunization vaccine-induced VNA responsea
with immunocomplexes
(IC) on
immunization schedule Experimental group
Day 0
Day 10
Day 20
1 VNA GMT (range)
IC 7 (5-10)
ERA-BPL 18 (10-30)
ERA-BPL 78 (45-405)
2 VNA GMT (range)
ERA-BPL 77 (45-270)
IC 163 (45-405)
ERA-BPL 1997 (405-7290)
3 VNA GMT (range)
ERA-BPL 74 (45-270)
ERA-BPL 176 (90-405)
ERA-BPL 1021 (270-2430)
°Groups of ten Balb/c mice were immunized at indicated times with either 5/~g IC or 5/~g ERA-BPL virus. Ten days after each immunization, mice were bled and VNA titres were determined, as described in Materials and methods
Vacci ne, Vol. 10, I ssue 11, 1992
757
Rabies virus-specific monoclonal antibodies cause interference: C.L. Schumacher et al. 8000
8000
6000
6000
6000
4000
4000
4000
2000
2000
2000
8ooo I
0
Day 1
0
0
I
0
ERA--BPL (!.Jg ml - I )
0
1
1 ERA--BPL (IJg m1-1)
ERA--BPL (IJg m1-1)
8000
0
8000
Day 28
Day 14
6000
6000
4000
4000
2000
2000
0
I
5
ERA--BPL (~g ml - I )
0
I
5
ERA--BPL (#g rnl - I )
Figure 2 Effect of passively administered mAb-C on induction of rabies virus-specific TH cells. Groups of Balb/c mice were injected with 10 IU of the mAb-C (m), or an equivalent amount of normal Balb/c IgG ([]). Mice were inoculated 1, 2, 7, 14 or 28 days later with 1/~g of ERA-BPL virus. Lymphocytes were tested 7 days after injection of ERA BPL virus for lymphokine release to medium (0), 1/lg of ERA-BPL virus (1) or 5#g of ERA-BPL virus (5). Controls: in each group the HT-2 indicator cells were tested for proliferation to medium and a 10% rat concanavalin A supernatant (RCAS); the following data (counts min -1 _+ s.d.) were obtained: day 1: medium 27 -I- 5, RCAS 6529 _+ 352; day 2: medium 40 _+ 20, RCAS 5951 4- 265; day 7: medium 66 ± 31, RCAS 12046 _+ 2939; day 14: medium 119 ± 40, RCAS 11110 -I- 1145; day 28: medium 56 _+ 11, RCAS 4449 _+ 995
immunocomplexes. Also, after the first booster immunization with ERA BPL vaccine, these animals developed only very low VNA titres (1:18). In contrast, mice of group 2 responded to a secondary immunization with immunocomplexes with a significant increase in VNA titres (1:77 to 1:163), which was comparable to the titre increase in animals of group 3 after the first booster immunization. This tendency in the development of VNA titres in groups 1 to 3 continued after the third booster immunization. After the second booster immunization, the VNA titres in group 1 were 12 to 25 times lower than the corresponding titres in groups 2 and 3. These results clearly indicate that after primary immunization with immunocomplexes, the production of VNA in response to E R A - B P L vaccine is strongly inhibited. However, when mice were primed with ERA BPL vaccine and then boosted with immunocomplexes, no inhibition of the VNA response was observed. In contrast, under these conditions, the immunocomplexes stimulated a secondary VNA response (1:163) which was similar to the secondary response in mice that received ERA BPL vaccine for both immunizations (1:176). DISCUSSION The impairment of an immune response to vaccine by administered virus-specific antibodies is a well known phenomenon which has been described for respiratory syncytial virus is and rabies virus 8'9. In this paper investigations have been reported on the suppressive effect of an administered cocktail of mouse anti-rabies mAb. This mAb-C, which consists of murine
758
Vaccine, Vol. 10, Issue 11, 1992
mAb to the rabies G and N protein, has been shown to protect mice and hamsters in pre- and postexposure situations against a lethal peripheral challenge infection with rabies virus 6. Mice were pretreated with different concentrations of the mAb-C, subsequently immunized with a protective dose of rabies vaccine, and then infected i.c. with a lethal dose of rabies virus. Since only actively immunized mice and not mice with passive immunity from administered mAb-C are protected against an i.c. challenge infection, the survival from an i.c. challenge infection indicates that the protective immunity had been induced by the vaccine. Our data clearly show that administration of mAb-C prior to vaccination interfered with the induction of a protective immune response upon vaccination. The lack of protection in animals pretreated with mAb-C correlated with the inability of these animals to mount a VNA response. The magnitude of the inhibition of the B-cell response was dependent on the serum concentration of the passively administered antibody at the time of the vaccination. The dose of the mAb-C also determined the time interval required to achieve a vaccine-specific B-cell response. While animals that received less than 1 IU of mAb-C were able to generate an antibody response to vaccine given 7 or 14 days later, mice that received 10 I U of mAb-C failed to respond until the fourth vaccination given 2 months after mAb-C treatment. At identical serum concentrations, homologous species-specific mouse mAb or heterologous human mAb cause the same magnitude of suppression. However, because of the shorter half-life of the heterologous mAb, the serum
Rabies virus-specific m o n o c l o n a l antibodies cause interference: C.L. S c h u m a c h e r et al.
concentration decreased faster resulting in a shorter duration of suppression of the VNA response to vaccine. There are reports stating that the non-responsiveness in humans that received anti-rabies hyperimmunoglobulin in addition to vaccine can be overcome by repeated vaccinations x6. Our data demonstrated that, in mice, the antibody-mediated interference phenomenon disappeared only if the serum titre of the administered mAb had decreased below a certain threshold. As long as the serum titres were above this level, the suppression could not be overcome even by multiple vaccinations. The mechanisms by which passively administered anti-rabies mAb induce a suppression of a protective immune response to rabies vaccine are not fully understood. Since the interference could be interrupted neither by multiple immunizations with large concentrations of vaccine nor by immunization with a live vaccinia rabies glycoprotein recombinant virus (data not shown), it seems unlikely that the non-responsiveness to vaccine is due to removal of antigen by circulating mAb. Another possible mechanism of interference of a B-cell response to vaccine might have been inhibition at the level of the T n cells needed to generate a rabies virus-specific IgG response. The mechanism seems unlikely because lymphocytes of Balb/c mice that received a high dose (i.e. 10 IU) of the mAb-C prior to vaccination release in vitro the same amount of lymphokines (in response to inactivated rabies virus) as do lymphocytes from animals that receive vaccine only or a control antibody. A more plausible mechanism is the formation of antigen-antibody complexes between the rabies virus and the antibody vaccine. It has been described in other experimental systems that antigenantibody complexes inhibit a primary B-cell response without affecting a secondary memory response ix. We obtained comparable results; injection of mice with antigen-antibody complexes consisting of inactivated rabies virus and mAb rendered the animals nonresponsive to subsequent vaccinations with virus. In contrast, injection of antigen-antibody complexes into mice that had been vaccinated previously had a booster effect that was even slightly superior to that of vaccine only. These results suggest that antigen-antibody complexes possibly exert a negative signalling effect on immature IgM receptors bearing B cells preventing them from differentiation into antibody-secreting plasma cells, although the possibility cannot be excluded that small amounts of free antibody that dissociated from the virus rather than immunocomplexes caused the impairment of the B-cell response. It has been postulated that multivalent antigens, such as antigen-antibody complexes, crosslink the IgM receptors and initiate a patching capping-endocytosis cycle which prevents further B-cell development by reducing the IgM receptor molecules on the cell surface xv. More mature B cells which possess IgD receptors in addition to IgM receptors are less susceptible to this antigen-antibody complexmediated immunosuppression 18. Based on the assumption that the combined treatment of animals with vaccine and antibodies will result in the formation of antigenantibody complexes, the observed immunosuppression might be a specific B-cell tolerance which persists until the concentration of mAb has decreased below a critical threshold at which level immunocomplex formation is either no longer possible or happens at an ineffective frequency.
Because the incubation times of rabies in humans can vary from only a few days to several years 19, the administration of antibody together with vaccine is the treatment of choice against rabies in humans. The main purpose of the antibody treatment is to neutralize the majority of virus particles at the site of infection, which would probably not result in a complete protection, but would at least increase the incubation time 2°. The remaining infectious virus particles or infected cells would then be eliminated by immune effectors induced by the vaccine. Because of their high specific activity and short half-life in primates (unpublished data), mouse mAb could have an advantage over the currently used human hyperimmunoglobulin preparations.
ACKNOWLEDGEMENTS The authors are grateful to Stacy Cohen for her expert technical assistance. This work was partly supported by Public Health Service grants AI-09706 and AI-23504 to H.C.J.E. and B.D.
REFERENCES 1
2
3
4 5
6
7
8
9
10
11
12
13
14 15
Dietzschold, B., Tollis, M., Lafon, M., Wunner, W.H. and Koprowski, H. Mechanisms of rabies virus neutralization by glycoprotein-specific monoclonal antibodies. Virology 1987,161,29 Rubin, R.H., Gough, P., Gerlach, E.H., Dierks, R.E., Gregg, M.B. and Sikes, R.K. Immunoglobulin response to rabies vaccine in man. Lancet 1971, ii, 625 Bahmanyar, F., Nour-Saleh, M. and Koprowski, H. Successful protection of humans exposed to rabies infection. J. Am. Med. Assoc. 1976, 236, 2751 Baer, G.H. and Fishbein, D.B. Rabies postexposure prophylaxis. N. Engl. J. Med. 1987, 316, 1270 WHO Technical Report Series. Prevention of rabies in man. In: 7th WHO Expert Report on Rabies, World Health Organization, Geneva, 1984, p. 27 Schumacher, C.L., Dietzschold, B., Ertl, H.C.J., Niu, H.-S., Rupprecht, C.E. and Koprowski, H. Use of mouse anti-rabies monoclonal antibodies in postexposure treatment of rabies. J. Clin. Invest. 1989, 84, 971 Dietzschold, B., Gore,M., Casali, P., Ueki, Y., Rupprecht, C.E., Notkins, A.L. and Koprowski, H. Biological characterization of human monoclonal antibodies to rabies virus. J. Virol. 1990, 64, 3087 Wiktor, T.J., Lerner, R.A. and Koprowski, H. Inhibitory effect of passive antibody on active immunity induced against rabies by vaccination. Buff. WHO 1971, 45, 747 Wiktor, T.J., Doherty, P.C. and Koprowski, H. In vitro evidence of cell-mediated immunity after exposure of mice to both live and inactivated rabies virus. Proc. Natl Acad. Sci. USA 1977, 74, 334 Vodopija, I., Sureau, P., Smerdel, S., Lafon, M., Baklaic, Z., Ljubicic, H. and Svjetlicic, M. Interaction of rabies vaccine with human rabies immunoglobulin and reliability of a 2-1-1 schedule application for postexposure treatment. Vaccine 1988, 6, 283 Rowley, D.A., Fitch, F.W., Stuart, F.P., Kohler, H. and Cosenza, H. Specific suppression of immune responses by antibody directed against either antigen or receptors for antigen can suppress immunity specifically. Science 1973, 181, 1133 Wiktor, T.J. Tissue culture methods. In: Laboratory Techniques in Rabies (Eds Kaplan, M. and Koprowski, H.), World Health Organization Monograph, 2nd edn, no. 23, World Health Organization, Geneva, 1973, p. 101 Wiktor, T.J. and Koprowski, H. Monoclonal antibodies against rabies virus produced by somatic cell hybridization detection of antigenic variants. Proc. NaU Acad. Sci. USA 1978, 75, 3938 Zalan, E., Wilson, C. and Pukitis, D. A microtest for the quantitation of rabies virus neutralizing antibodies. J. Biol. Stand. 1979, 7, 213 Murphy, B.R., Collins, P.L., Lawrence, L., Zubak, J., Chanock, R.M. and Prince, G.A. Immunosuppression of the antibody response to respiratory syncytial virus (RSV) by pre-existing serum antibodies: partial prevention by topical infection of the respiratory tract with vaccinia virus-RSV recombinants. J. Gen. Virol. 1989, 70, 2185
Vaccine, Vol. 10, Issue 11, 1992
759
R a b i e s virus-specific m o n o c l o n a l a n t i b o d i e s cause interference: C.L. S c h u m a c h e r et al. 16
Selimov, M.A., Klyueva, E.V., Aksenova, T.A., Labedeva, I.R. and Gribencha, L.F. Treatment of patients bitten by rabid or suspected rabid wolves with inactivated tissue culture rabies vaccine and rabies gammaglobulin. Dev. Biol. Stand. 1978, 40, 141 17 Nossal, G.J.V. Cellular mechanisms of immunologic tolerance. Annu. Rev. Immunol. 1983, 1, 33 18 Metcalf, E.S., Schrater, A.F. and Klinman, N.R. Murine models of
760
Vaccine, Vol. 10, Issue 11, 1992
19 20
tolerance induction in developing and mature B cells. Immunol. Rev. 1979, 4,1, 143 Rupprecht, C.D. and Dietzschold, B. Editorial: Perspectives on rabies virus pathogenesis. Lab. Invest. 1987, 5"7, 603 Baer, G.M. and Yager, P.A. A mouse model for post-exposure rabies prophylaxis: the comparative efficacy of two vaccines and of antiserum administration. J. Gen. Virol. 1977, 36, 51
INTRODUCTION Protection against rabies virus infection is the result of a number of specific and non-specific immune mechanisms; however, virus neutralizing antibodies (VNA) are believed to play a major role in immunoprotection against rabies 1. VNA may offer protection, especially in the early critical phase after exposure to rabies virus before the virus reaches the central nervous system 2, It has been shown that in postexposure situations of human rabies, the combined treatment with vaccine and human rabies immunoglobulin was more effective in preventing lethal rabies than was treatment with vaccine alone 3'4. Therefore, a combination of serum and vaccine became the standard treatment of humans for severe exposure to rabies virus 5. Recently, it has been demonstrated that passive immunization of mice and hamsters, with a cocktail of mouse monoclonal antibodies (mAb-C) specific for rabies ribonucleocapsid protein and glycoprotein, protects animals not only when challenged peripherally with a lethal dose of rabies virus after immunization, but also in postexposure situations 6. In addition, a human rabies virus-neutralizing mAb
*The Wistar Institute of Anatomy and Biology, 3601 Spruce Street, Philadelphia, PA 19104, USA. tDepartment of Microbiology, 459 Alumni Hall, Thomas Jefferson University, 1020 Locust Street, Philadelphia, PA 19107, USA. ;To whom correspondence should be addressed. (Received 9 October 1991; revised 12 March 1992; accepted 12 March 1992) 0264-410X/92/110754-07 © 1992Butterworth-HeinemannLtd 754
Vaccine, Vol. 10, Issue 11, 1992
protected laboratory rodents against lethal rabies virus infection v. These protection experiments with mAb-C indicate that rabies virus-specific mAb-C may be utilized for the postexposure treatment of human rabies. One of the major problems connected with the administration of anti-rabies antibodies is the adverse effect of immunoglobulins on the humoral and cellular immune response to rabies vaccine 8'9. It has been shown that injection of rabies antiserum resulted in suppression of the immune response to rabies vaccine 1°. Decrease of the antibody responses by passive immunization has been extensively studied with sheep erythrocytes, where it has been shown that small amounts of antibody against the red blood cells specifically suppress the IgM and IgG antibody response 11. The duration of this suppression was a function of the biological half-life of the antibody and depended on the amount and dose of antibody given 11. Moreover, passive immunization had little or no effect on animals with previous active immunizations ~t. Although the mechanism of antibody-induced unresponsiveness is not fully understood, it is believed that antibody suppresses the development of B cells by forming immunocomplexes with the antigen. The suppressive effect of immunocomplexes formed in vitro on humoral immune responses has been described 11. The present studies were undertaken to determine the suppressive effect of administered anti-rabies virus mAb on the induction of antibodies and T-helper (TH) cells, and thus the protective immune response following immunization with inactivated rabies virus.
Rabies virus-specific monoclonal antibodies cause interference: C.L. Schumacher et al.
MATERIALS AND METHODS
Viruses. The propagation of the fixed rabies virus strains Evelyn-Rokitniki-Abelseth (ERA) and the challenge virus standard (CVS) in BHK-21 cell monolayer cultures has been described previously ~2. Vaccine preparation. Rabies vaccine (ERA-fl-propriolactone [ B P L ] ) was prepared from gradient-purified ERA virus. Inactivation of concentrated ERA virus was carried out with BPL treatment according to the method described by Wiktor et al. 12. Preparation of antibodies. MAb-secreting hybridomas were produced through fusion of splenocytes from rabies-immunized mice with P3 × 63 Ag8 or 654 mouse myeloma cells according to the protocol described by Wiktor and Koprowski 13. MAb were purified from ascites fluid as previously described 6. The production and characterization of the human mAb 57 has been described by Dietzschold et al. 7. Three mAb specific for the glycoprotein (G) and two mAb specific for epitopes on the nucleoprotein complex of rabies virus were used to prepare an mAb-C, as described previously 6. MAb-C represents a mixture of 112.5 IU ml-1 ofmAb 509-6, 10 IU ml-1 ofmAb 1112-1, 800IU m1-1 ofmAb 523-11,0.04 mg m1-1 ofmAb 802-2, and 0.6 mg ml- 1 of mAb 502-2. The neutralizing activity of the cocktail was ~ 1376 IU mg -~ protein. Normal mouse IgG was purified from serum of untreated Balb/c mice by Protein-A-Sepharose chromatography, as previously described 6. Preparation of immunocomplexes. Immunocomplexes were produced by in vitro incubation of 200 IU murine anti-rabies mAb-C with 100 #g E R A - B P L virus in 5 ml phosphate-buffered saline. Incubation was carried out for 2 h at 37°C and overnight at 4°C. Immunocomplexes were separated from free mAb-C by repeated ultracentrifugation for 1 h at 100 000g. The pellet containing the immunocomplex was reconstituted in phosphatebuffered saline. Protection experiments with mAb and E R A - B P L vaccine. Six-week-old female white Swiss mice (ICR) or 6-week-old female Balb/c mice (Harlan SpragueDawley, Inc., Indianapolis, IN) were injected intramuscularly (i.m.) in the gastrocnemius muscle with 0.1 ml of serial dilutions of antibodies. Mice were vaccinated intraperitoneally (i.p.) at different times after mAb treatment with E R A - B P L vaccine. Blood was withdrawn from these animals at different time intervals for determination of neutralizing antibodies as described ~. Seven days after vaccination, the mice were challenged intracerebrally (i.c.) with 25 MICLDso (MICLDso is the i.c. dose which is lethal for 50% of naive age-matched mice) of the CVS-11 rabies virus strain. Animals were observed for 3 weeks after challenge for development of symptoms and death. Tn cellassay. Groups of three 6-8-week-old Balb/c mice were treated i.m. with either mAb-C or the equivalent amount of normal Balb/c IgG and subsequently vaccinated i.m. with 1/~g E R A - B P L virus at different time points. Untreated groups of three mice were vaccinated with 1/zg E R A - B P L virus on the same days
as mAb-treated animals. Seven days after vaccination, lymphocytes (4 x 106) from the popliteal and medial iliac lymph nodes of immune animals were cultured at 37°C in 1.6 #1 of Dulbecco's modified Eagle's medium, supplemented with 2% fetal bovine serum and 10 -5 M 2-mercaptoethanol, in flat-bottomed 24-well plates with different dilutions of E R A - B P L virus or medium. Supernatants were harvested 20-24 h later and tested on lymphokine (interleukin 2 and 4)-sensitive HT-2 cells. Quadruplets of 2 x 103 HT-2 cells in 100 #1 of medium supplemented with 10% fetal bovine serum were incubated at 37°C in 96-well round-bottomed plates with 50/~1 of cell culture supernatants, medium (negative control), or 10% rat concanavalin A supernatant (positive control). Proliferation was measured 36-48 h later after a 6 8 h [3H]methyl-thymidine pulse (0.7 #Ci/ well) by liquid scintillation counting. A threefold counts min- 1 increase over the medium control was considered a significant proliferation. Values are presented as the mean of four single counts ___the standard deviation.
Virus neutralization assay. VNA titres of serum samples were tested with a modified version of the rapid fluorescent focus inhibiting test as described by Zalan et al. 14. The reciprocal of the serum dilution resulting in a 50% reduction of infected foci was considered the neutralization titre of the serum sample. RESULTS
Effect of administered mAb-C on vaccine-mediated protective immunity To investigate whether administration of anti-rabies mAb-C interferes with the production of protective immunity induced by a rabies vaccine, groups of ten ICR mice were treated i.m. with 10 IU per mouse of mAb-C. After 24 h, mAb-C treated and untreated control animals were vaccinated i.p. with 0.016 to 2.0/~g E R A - B P L vaccine. Seven days after the primary vaccination, the groups of mice were revaccinated with the same amounts of vaccine, and 7 days after the second vaccination, all animals were infected i.e. with 25 MICLDso of CVS-11 virus. The i.e. route of infection was chosen because, as previously shown 6, only active immunized animals and not passive immunized animals are protected against an i.e. challenge infection. Therefore, protection from an i.e. infection is an indicator of vaccine-mediated immunity. Table I shows that, depending on the vaccine dose, Table 1 Effect of mAb-C treatment on vaccine-induced protective immunity
Mortality ( % )a after Vaccine concentration b (/~g)
Vaccination ~ only
mAb-C treatmen~ and vaccination
2 0.4 0.08 0.016 None
10 30 80 90 100
100 90 90 100 100
aMice were challenged i.c. 7 days after second vaccination with 25 MICLDso of CVS-11 virus ~Mice were vaccinated with ERA-BPL virus on days 1 and 10 CGroups of ten ICR mice were pretreated with 10 IU of mAb-C 24 h before vacci nation
Vaccine, Vol. 10, Issue 11, 1992
755
Rabies virus-specific monoclonal antibodies cause interference: C.L. Schumacher et al.
9 0 - 1 0 0 % of vaccinated animals survived the lethal challenge infection. However, the protective effect of even the highest doses of vaccine was almost totally abolished when the mice were pretreated with 10 IU mAb-C prior to vaccination.
Dependence of interference induction on the serum concentration of administered antibodies To demonstrate that the induction and duration of interference depends on the serum concentration of the administered antibodies, groups of ten Balb/c mice were treated with different amounts of mAb-C (10-0.04 IU). The mice received four vaccinations with 5 m g E R A - B P L vaccine per mouse 2, 12, 62 and 92 days after mAb-C treatment, and were bled 2 days after mAb-C treatment and 10 days after each vaccination. The results of this experiment are summarized in Table 2. In contrast to the groups of untreated control animals, which had already exhibited a vaccine-induced B-cell response after the first vaccination (geometric mean titre [ G M T ] 1:103), animals pretreated with I0 or 5 IU of mAb-C showed an increase in VNA titres only after the third vaccination. Mice pretreated with 1 or 0.2 IU of mAb-C showed an increase in VNA titres after the second vaccination, while animals pretreated with 0.04IU exhibited, like untreated animals, a vaccine-induced immune response after the first vaccination. However, the VNA G M T in animals pretreated with as little as 0.04IU of mAb-C were, after each vaccination, significantly lower than the corresponding titres in the control group. These results indicate that the inhibition of a vaccine-induced VNA response is dependent on the serum concentration of the administered mAb. Production of VNA is apparently only possible when the serum concentration of the administered mAb is below a certain threshold level ( G M T of 1:10 to 1:30). The time interval between mAb-C treatment and vaccination seems to play a role in the induction of interference only in so far as the titre of a particular administered mAb decreases with time. In order to demonstrate a correlation between the mAb-induced non-responsiveness of rabies-specific B cells and the lack of protective immunity in response to
Table 2 Dependence of interference on serum concentrations of administered mAb-C VNA GMT" Days after mAb-C treatment Treatmentb
mAb dose
2
12
22
72
101
mAb-C + vaccine mAb-C + vaccine mAb-C + vaccine mAb-C + vaccine mAb-C + vaccine Vaccine only
10 5 1 0.2 0.04 0
110 52 24 8 7 3
82 23 10 12 50 103
35 16 30 141 269 479
304 412 772 1078 1102 1812
710 509 1373 2193 3175 6139
aMice were bled 2 days after mAb-C treatment and 10 days after each vaccination. VNA were determined as described in Materials and methods. Underlined numbers represent an increase of VNA after vaccination bGroups of ten Balb/c mice were treated i.m. with different amounts of mAImC and then vaccinated four times with 5/~g ERA-BPL virus i.p. 2, 12, 62 and 92 days after mAb-C treatment
756
Vaccine, Vol. 10, Issue 11, 1992
Table 3
Duration of mAb-C-induced non-responsiveness to vaccination
Treatment"
Time interval between mAb-C treatment and vaccination (days)
Mortality ( % ).
mAb-C + vaccine
0
86
mAb-C + vaccine
7
100
mAb-C + vaccine
14
70
mAb-C + vaccine
21
90
Vaccine only
-
20
None
-
100
VNA GMT° (range) 133 (90 135) 130 (90-135) 102 (90-135) 84 (90-135) 1356 (810-3645) 0 (-)
"Groups of ten Balb/c mice were treated i.m. with 10 IU of mAb-C and vaccinated after the indicated time intervals with 2 #g ERA-BPL virus ~Mice were challenged i.c. 7 days after the second vaccination with 25 MICLDso of CVS-11 virus ~Mice were bled 1 day before challenge infection, and VNA titres were determined as described in Materials and methods
vaccine, part of the experiment was repeated as follows: groups of ten Balb/c mice were treated with 10 IU per mouse of mAb-C and then received after different time intervals (0-21 days) 2/tg of E R A - B P L vaccine. The mice were then revaccinated 7 days after the first vaccination with the same amount of vaccine. Seven days after the second vaccination, the mice were bled and then infected i.c. with 25 MICLDso ofCVS-11 virus. As shown in Table 3, the group of mice which received vaccine alone developed a VNA G M T of 1:1356 and 80% of the animals in this group survived the challenge infection. In contrast, the serum VNA G M T in all groups of mice which were pretreated with mAb-C were at least ten times lower ( 1 : 84 to 1 : 133 ) regardless of the time lapse between mAb-C treatment and vaccination. The lower VNA titres in the mAb-C pretreated mice are reflected by a strong decrease in protective immunity, as indicated by the high mortality rates in these groups of mice. Even when the mice were vaccinated 21 days after mAb-C administration, 90% of these vaccinated animals succumbed to a rabies virus infection. Dependence of interference induction on biological half-life of the antibody To investigate the relationship between duration of mAb-induced interference and the half-life of a particular antibody, the half-lives of administered murine and human mAb in Balb/c mice were determined. Groups often Balb/c mice were treated i.m. with either 50 IU (6 #g) per mouse of homologous murine mAb 523-11, or with 50 IU (4 ~g) of heterologous human mAb 57. The mice were bled 1, 3, 7, 14, 21 and 28 days after mAb administration, to determine the circulating VNA titres. Figure 1 shows the time-dependent decrease of VNA titres of the two mAbs. To compare the titres at different time points, the serum VNA titres of each antibody preparation 24 h after administration were taken as 100%. According to this calculation, the half-life of the murine mAb is approximately 6 days compared with only 4 days for the human mAb. Of the murine mAb, 25% were still circulating in the serum 14 days
Rabies virus-specific monoclonal antibodies cause interference: C.L. Schumacher et al.
'°°I 8o
g
!
0
5
I0
15
20
Days a f t e r mAb-treatment Figure I Clearance of murine mAb 523-11 ( 0 ) and human mAb 57 (FI) in Balb/c mice. Groups of ten Balb/c mice were treated i.m. with either 50 IU (6 #g) per mouse of murine mAb 523-11, or with 50 IU (4 #g) of human mAb-57. Mice were bled 3, 7, 14, 21 and 28 days after mAb administration and VNA GMT were determined. VNA GMT from 3 to 28 days are expressed as percentage of the VNA GMT determined 24 h after mAb administration. Data are plotted using a Stalview SE + Graphics T H program
after injection, while the human mAb was no longer detectable. The kinetics of the single murine mAb 523-11 and mAb-C were identical (data not shown). To demonstrate that the duration of antibodymediated interference is a function of the half-life of a particular administered mAb, groups of Balb/c mice were treated i.m. with 5 IU of murine mAb 523-11, or 5 IU of human mAb 57. The animals were then vaccinated three times with 5/~g of E R A - B P L vaccine 1, 10 and 20 days after mAb administration. The mice were bled 24 h after mAb administration and 10 days after each vaccination, and VNA G M T were determined as described in Materials and methods. The VNA G M T increased after the second immunization in mice treated with the human mAb, while the VNA titres in mice treated with similar amounts of murine mAb increased only after the third vaccination (Table 4). These results indicated that the duration of mAb-mediated interference is reciprocally proportional to the biological half-life of the mAb.
antigen-specific Ta cell responses between mice pretreated with mAb-C or with normal Balb/c IgG. Ta cells from ERA-BPL-vaccinated mice not pretreated or pretreated with normal Balb/c IgG secreted similar amounts of interleukin as T H cells from mice that were pretreated with mAb-C. Therefore, the suppression of a B-cell response in mAb-C pretreated mice is not due to suppression of rabies virus-specific T n cell activation. (ii) Effect of immunocomplexes on production of rabies virus-specific antibodies: To investigate the effect of immunocomplexes on activation of rabies virusspecific B cells, groups of Balb/c mice were immunized with immunocomplexes ( E R A - B P L virus-mAb complexes) in combination with free E R A - B P L virus as follows: group 1 received a primary vaccination with 5/~g immunocomplexes i.p. per mouse, and 10 and 20 days later two booster immunizations i.p., each with 5 #g E R A - B P L vaccine per mouse; the second group received a primary immunization with 5/~g E R A - B P L vaccine, followed by a booster immunization with 5/~g immunocomplexes i.p. on day 10, and a booster immunization with 5/~g E R A - B P L vaccine on day 20; the third group of mice received a primary immunization with 5 #g E R A - B P L vaccine per mouse, followed by two booster immunizations with 5/~g E R A - B P L vaccine 10 and 20 days after primary immunization. Ten days after each immunization, the mice were bled to determine the circulating VNA titres. Table 5 shows that mice of group 1 did not develop significant VNA titres after primary immunization with
Table 4
Effect of mAb half-life on the duration of interference VNA GMT" Days after mAb treatment
Treatment ~
1
11
21
31
mAb 523-11 + vaccine mAb 57 + vaccine Vaccine only
84 171 0
15 10 73
10 89 354
68 396 1918
aMice were bled 1 day after mAb treatment and 10 days after each vaccination. VNA titres were determined as described in Materials and methods bGroups of ten Balb/c mice were treated i.m. with either 5 IU of murine mAb 523-11 or 5 IU of human mAb 57, and then vaccinated three times with 5 #g ERA-BPL virus 1, 10 and 20 days after mAb treatment
Study of mechanisms of mAb-mediated interference (i) Effect of administered mAb-C on the induction of virus-specific T H cells: To determine whether the suppression of a B-cell response by administered mAb is the result of an.inhibition of T H cell activation, Balb/c mice received i.m. either 10 IU ofmAb-C or an equivalent amount (i.e. 7/~g) of normal Balb/c IgG. The mice were immunized i.m. with 1 #g of E R A - B P L vaccine 1, 2, 7, 14 or 28 days after mAb-C administration. In addition, mice that were not pretreated with mAb-C were immunized with 1 #g E R A - B P L vaccine on the same days as mAb-C-treated mice. One week after immunization, the mice were killed humanely, and lymphocytes were isolated from draining lymph nodes and then cultured at 37°C for 24 h in the presence of 5/tg or 1 #g of E R A - B P L vaccine or without antigen. The amounts of lymphokines produced in vitro by these lymphocytes in response to rabies antigen are illustrated in Figure 2. No significant differences were observed in the rabies
Table 5 Effect of immunization vaccine-induced VNA responsea
with immunocomplexes
(IC) on
immunization schedule Experimental group
Day 0
Day 10
Day 20
1 VNA GMT (range)
IC 7 (5-10)
ERA-BPL 18 (10-30)
ERA-BPL 78 (45-405)
2 VNA GMT (range)
ERA-BPL 77 (45-270)
IC 163 (45-405)
ERA-BPL 1997 (405-7290)
3 VNA GMT (range)
ERA-BPL 74 (45-270)
ERA-BPL 176 (90-405)
ERA-BPL 1021 (270-2430)
°Groups of ten Balb/c mice were immunized at indicated times with either 5/~g IC or 5/~g ERA-BPL virus. Ten days after each immunization, mice were bled and VNA titres were determined, as described in Materials and methods
Vacci ne, Vol. 10, I ssue 11, 1992
757
Rabies virus-specific monoclonal antibodies cause interference: C.L. Schumacher et al. 8000
8000
6000
6000
6000
4000
4000
4000
2000
2000
2000
8ooo I
0
Day 1
0
0
I
0
ERA--BPL (!.Jg ml - I )
0
1
1 ERA--BPL (IJg m1-1)
ERA--BPL (IJg m1-1)
8000
0
8000
Day 28
Day 14
6000
6000
4000
4000
2000
2000
0
I
5
ERA--BPL (~g ml - I )
0
I
5
ERA--BPL (#g rnl - I )
Figure 2 Effect of passively administered mAb-C on induction of rabies virus-specific TH cells. Groups of Balb/c mice were injected with 10 IU of the mAb-C (m), or an equivalent amount of normal Balb/c IgG ([]). Mice were inoculated 1, 2, 7, 14 or 28 days later with 1/~g of ERA-BPL virus. Lymphocytes were tested 7 days after injection of ERA BPL virus for lymphokine release to medium (0), 1/lg of ERA-BPL virus (1) or 5#g of ERA-BPL virus (5). Controls: in each group the HT-2 indicator cells were tested for proliferation to medium and a 10% rat concanavalin A supernatant (RCAS); the following data (counts min -1 _+ s.d.) were obtained: day 1: medium 27 -I- 5, RCAS 6529 _+ 352; day 2: medium 40 _+ 20, RCAS 5951 4- 265; day 7: medium 66 ± 31, RCAS 12046 _+ 2939; day 14: medium 119 ± 40, RCAS 11110 -I- 1145; day 28: medium 56 _+ 11, RCAS 4449 _+ 995
immunocomplexes. Also, after the first booster immunization with ERA BPL vaccine, these animals developed only very low VNA titres (1:18). In contrast, mice of group 2 responded to a secondary immunization with immunocomplexes with a significant increase in VNA titres (1:77 to 1:163), which was comparable to the titre increase in animals of group 3 after the first booster immunization. This tendency in the development of VNA titres in groups 1 to 3 continued after the third booster immunization. After the second booster immunization, the VNA titres in group 1 were 12 to 25 times lower than the corresponding titres in groups 2 and 3. These results clearly indicate that after primary immunization with immunocomplexes, the production of VNA in response to E R A - B P L vaccine is strongly inhibited. However, when mice were primed with ERA BPL vaccine and then boosted with immunocomplexes, no inhibition of the VNA response was observed. In contrast, under these conditions, the immunocomplexes stimulated a secondary VNA response (1:163) which was similar to the secondary response in mice that received ERA BPL vaccine for both immunizations (1:176). DISCUSSION The impairment of an immune response to vaccine by administered virus-specific antibodies is a well known phenomenon which has been described for respiratory syncytial virus is and rabies virus 8'9. In this paper investigations have been reported on the suppressive effect of an administered cocktail of mouse anti-rabies mAb. This mAb-C, which consists of murine
758
Vaccine, Vol. 10, Issue 11, 1992
mAb to the rabies G and N protein, has been shown to protect mice and hamsters in pre- and postexposure situations against a lethal peripheral challenge infection with rabies virus 6. Mice were pretreated with different concentrations of the mAb-C, subsequently immunized with a protective dose of rabies vaccine, and then infected i.c. with a lethal dose of rabies virus. Since only actively immunized mice and not mice with passive immunity from administered mAb-C are protected against an i.c. challenge infection, the survival from an i.c. challenge infection indicates that the protective immunity had been induced by the vaccine. Our data clearly show that administration of mAb-C prior to vaccination interfered with the induction of a protective immune response upon vaccination. The lack of protection in animals pretreated with mAb-C correlated with the inability of these animals to mount a VNA response. The magnitude of the inhibition of the B-cell response was dependent on the serum concentration of the passively administered antibody at the time of the vaccination. The dose of the mAb-C also determined the time interval required to achieve a vaccine-specific B-cell response. While animals that received less than 1 IU of mAb-C were able to generate an antibody response to vaccine given 7 or 14 days later, mice that received 10 I U of mAb-C failed to respond until the fourth vaccination given 2 months after mAb-C treatment. At identical serum concentrations, homologous species-specific mouse mAb or heterologous human mAb cause the same magnitude of suppression. However, because of the shorter half-life of the heterologous mAb, the serum
Rabies virus-specific m o n o c l o n a l antibodies cause interference: C.L. S c h u m a c h e r et al.
concentration decreased faster resulting in a shorter duration of suppression of the VNA response to vaccine. There are reports stating that the non-responsiveness in humans that received anti-rabies hyperimmunoglobulin in addition to vaccine can be overcome by repeated vaccinations x6. Our data demonstrated that, in mice, the antibody-mediated interference phenomenon disappeared only if the serum titre of the administered mAb had decreased below a certain threshold. As long as the serum titres were above this level, the suppression could not be overcome even by multiple vaccinations. The mechanisms by which passively administered anti-rabies mAb induce a suppression of a protective immune response to rabies vaccine are not fully understood. Since the interference could be interrupted neither by multiple immunizations with large concentrations of vaccine nor by immunization with a live vaccinia rabies glycoprotein recombinant virus (data not shown), it seems unlikely that the non-responsiveness to vaccine is due to removal of antigen by circulating mAb. Another possible mechanism of interference of a B-cell response to vaccine might have been inhibition at the level of the T n cells needed to generate a rabies virus-specific IgG response. The mechanism seems unlikely because lymphocytes of Balb/c mice that received a high dose (i.e. 10 IU) of the mAb-C prior to vaccination release in vitro the same amount of lymphokines (in response to inactivated rabies virus) as do lymphocytes from animals that receive vaccine only or a control antibody. A more plausible mechanism is the formation of antigen-antibody complexes between the rabies virus and the antibody vaccine. It has been described in other experimental systems that antigenantibody complexes inhibit a primary B-cell response without affecting a secondary memory response ix. We obtained comparable results; injection of mice with antigen-antibody complexes consisting of inactivated rabies virus and mAb rendered the animals nonresponsive to subsequent vaccinations with virus. In contrast, injection of antigen-antibody complexes into mice that had been vaccinated previously had a booster effect that was even slightly superior to that of vaccine only. These results suggest that antigen-antibody complexes possibly exert a negative signalling effect on immature IgM receptors bearing B cells preventing them from differentiation into antibody-secreting plasma cells, although the possibility cannot be excluded that small amounts of free antibody that dissociated from the virus rather than immunocomplexes caused the impairment of the B-cell response. It has been postulated that multivalent antigens, such as antigen-antibody complexes, crosslink the IgM receptors and initiate a patching capping-endocytosis cycle which prevents further B-cell development by reducing the IgM receptor molecules on the cell surface xv. More mature B cells which possess IgD receptors in addition to IgM receptors are less susceptible to this antigen-antibody complexmediated immunosuppression 18. Based on the assumption that the combined treatment of animals with vaccine and antibodies will result in the formation of antigenantibody complexes, the observed immunosuppression might be a specific B-cell tolerance which persists until the concentration of mAb has decreased below a critical threshold at which level immunocomplex formation is either no longer possible or happens at an ineffective frequency.
Because the incubation times of rabies in humans can vary from only a few days to several years 19, the administration of antibody together with vaccine is the treatment of choice against rabies in humans. The main purpose of the antibody treatment is to neutralize the majority of virus particles at the site of infection, which would probably not result in a complete protection, but would at least increase the incubation time 2°. The remaining infectious virus particles or infected cells would then be eliminated by immune effectors induced by the vaccine. Because of their high specific activity and short half-life in primates (unpublished data), mouse mAb could have an advantage over the currently used human hyperimmunoglobulin preparations.
ACKNOWLEDGEMENTS The authors are grateful to Stacy Cohen for her expert technical assistance. This work was partly supported by Public Health Service grants AI-09706 and AI-23504 to H.C.J.E. and B.D.
REFERENCES 1
2
3
4 5
6
7
8
9
10
11
12
13
14 15
Dietzschold, B., Tollis, M., Lafon, M., Wunner, W.H. and Koprowski, H. Mechanisms of rabies virus neutralization by glycoprotein-specific monoclonal antibodies. Virology 1987,161,29 Rubin, R.H., Gough, P., Gerlach, E.H., Dierks, R.E., Gregg, M.B. and Sikes, R.K. Immunoglobulin response to rabies vaccine in man. Lancet 1971, ii, 625 Bahmanyar, F., Nour-Saleh, M. and Koprowski, H. Successful protection of humans exposed to rabies infection. J. Am. Med. Assoc. 1976, 236, 2751 Baer, G.H. and Fishbein, D.B. Rabies postexposure prophylaxis. N. Engl. J. Med. 1987, 316, 1270 WHO Technical Report Series. Prevention of rabies in man. In: 7th WHO Expert Report on Rabies, World Health Organization, Geneva, 1984, p. 27 Schumacher, C.L., Dietzschold, B., Ertl, H.C.J., Niu, H.-S., Rupprecht, C.E. and Koprowski, H. Use of mouse anti-rabies monoclonal antibodies in postexposure treatment of rabies. J. Clin. Invest. 1989, 84, 971 Dietzschold, B., Gore,M., Casali, P., Ueki, Y., Rupprecht, C.E., Notkins, A.L. and Koprowski, H. Biological characterization of human monoclonal antibodies to rabies virus. J. Virol. 1990, 64, 3087 Wiktor, T.J., Lerner, R.A. and Koprowski, H. Inhibitory effect of passive antibody on active immunity induced against rabies by vaccination. Buff. WHO 1971, 45, 747 Wiktor, T.J., Doherty, P.C. and Koprowski, H. In vitro evidence of cell-mediated immunity after exposure of mice to both live and inactivated rabies virus. Proc. Natl Acad. Sci. USA 1977, 74, 334 Vodopija, I., Sureau, P., Smerdel, S., Lafon, M., Baklaic, Z., Ljubicic, H. and Svjetlicic, M. Interaction of rabies vaccine with human rabies immunoglobulin and reliability of a 2-1-1 schedule application for postexposure treatment. Vaccine 1988, 6, 283 Rowley, D.A., Fitch, F.W., Stuart, F.P., Kohler, H. and Cosenza, H. Specific suppression of immune responses by antibody directed against either antigen or receptors for antigen can suppress immunity specifically. Science 1973, 181, 1133 Wiktor, T.J. Tissue culture methods. In: Laboratory Techniques in Rabies (Eds Kaplan, M. and Koprowski, H.), World Health Organization Monograph, 2nd edn, no. 23, World Health Organization, Geneva, 1973, p. 101 Wiktor, T.J. and Koprowski, H. Monoclonal antibodies against rabies virus produced by somatic cell hybridization detection of antigenic variants. Proc. NaU Acad. Sci. USA 1978, 75, 3938 Zalan, E., Wilson, C. and Pukitis, D. A microtest for the quantitation of rabies virus neutralizing antibodies. J. Biol. Stand. 1979, 7, 213 Murphy, B.R., Collins, P.L., Lawrence, L., Zubak, J., Chanock, R.M. and Prince, G.A. Immunosuppression of the antibody response to respiratory syncytial virus (RSV) by pre-existing serum antibodies: partial prevention by topical infection of the respiratory tract with vaccinia virus-RSV recombinants. J. Gen. Virol. 1989, 70, 2185
Vaccine, Vol. 10, Issue 11, 1992
759
R a b i e s virus-specific m o n o c l o n a l a n t i b o d i e s cause interference: C.L. S c h u m a c h e r et al. 16
Selimov, M.A., Klyueva, E.V., Aksenova, T.A., Labedeva, I.R. and Gribencha, L.F. Treatment of patients bitten by rabid or suspected rabid wolves with inactivated tissue culture rabies vaccine and rabies gammaglobulin. Dev. Biol. Stand. 1978, 40, 141 17 Nossal, G.J.V. Cellular mechanisms of immunologic tolerance. Annu. Rev. Immunol. 1983, 1, 33 18 Metcalf, E.S., Schrater, A.F. and Klinman, N.R. Murine models of
760
Vaccine, Vol. 10, Issue 11, 1992
19 20
tolerance induction in developing and mature B cells. Immunol. Rev. 1979, 4,1, 143 Rupprecht, C.D. and Dietzschold, B. Editorial: Perspectives on rabies virus pathogenesis. Lab. Invest. 1987, 5"7, 603 Baer, G.M. and Yager, P.A. A mouse model for post-exposure rabies prophylaxis: the comparative efficacy of two vaccines and of antiserum administration. J. Gen. Virol. 1977, 36, 51
