- Email: [email protected]
The Opsonic Antibody Response of Female Rats to Type III Group B Streptococcal Immunization: a Model for Maternal Immunity
Veterinary Immunology and Immunopathology, 24 (1990) 79-89 Elsevier Science Publishers B.V., Amsterdam - - Printed in The Netherlands
79
The Opsonic A n t i b o d y R e s p o n s e of Female Rats to T y p e III Group B S t r e p t o c o c c a l I m m u n i z a t i o n : a Model for M a t e r n a l I m m u n i t y HOWARD S. HEIMAN and LEONARD E. WEISMAN
Departments of Pediatrics, Uniformed Services University, Bethesda, MD 20814-4799 (U.S.A.) Walter Reed Army Medical Center, Washington, DC 20307-5000 (U.S.A.) (Accepted 28 April 1989 )
ABSTRACT Heiman, H.S. and Weisman, L.E., 1990. The opsonic antibody response of female rats to type III group B streptococcal immunization: a model for maternal immunity. Vet. Immunol. Immu nopathol., 24: 79-89. Group B streptococcus (GBS) remains a major cause of neonatal infection. Maternal immunization-induced GBS antibody may protect neonates from GBS disease. Since the opsonophagocytosis assay correlates well with survival in GBS infected suckling rats, we sought to determine an immunization schedule which would induce type III GBS opsonic antibody in rat dams above a predetermined level of 10 dilution- 1 (dil- 1). This schedule could then be used for future studies of maternal-fetal immunity. Wistar rat dams (n = 12 ) were given killed GBS type III using three immunization schedules (primary injection, initial booster at 7, 14 or 22 days and then weekly boosters). Opsonic GBS type III antibody and total immunoglobulin (IgG) were measured. Only the schedule with a 7-day initial booster resulted in GBS type-specific opsonic antibody consistently above 10 dil-1. The IgG (467 _+83 mg/100 ml) remained constant while the opsonic antibody increased significantly to 50 and 63 dil- 1 (p < 0.01 compared to day 0 ) after boosters on day 7 and 14 respectively. Eight pregnant dams, who received a primary immunization and boosters at 7 and 14 days, developed GBS type III opsonic antibody titers (72 dil-1) similar to non-pregnant dams and potentially adequate to protect suckling rats from GBS disease. This model may now be used to study other adjuvants, immunogens, and maternal-fetal immunity using established suckling rat models of GBS disease. ABBREVIATIONS CFU, colony forming units; di1-1, inverse dilution; GBS, group B streptococcus; IgG, immunoglobulin G; ISG, immune serum globulin; IVIG, intravenous immunoglobulin.
INTRODUCTION
Group B streptococcus (GBS) infection remains a major cause of neonatal morbidity and mortality despite antibiotic therapy (Baker and Edwards, 1983; 0165-2427/90/$03.50
© 1990 Elsevier Science Publishers B.V.
SO
H.S. HEIMAN AND L.E, WEISMAN
Siegel, 1985; Wilson, 1986). It appears to overwhelm the neonate's immune system which is largely dependent on transplacental transfer of maternal IgG (Yeung and Hobbs, 1968; Wilson, 1986). Children of women with naturally elevated GBS type III antibody, however, have a decreased incidence of neonatal GBS type III disease (Baker and Kasper, 1976; Baker et al., 1981). Enhancement of maternal GBS type-specific immunity by vaccination, appears to be a logical step in controlling the deleterious effects of this organism in the neonate. Vaccination of non-pregnant women with purified GBS typespecific capsular polysaccharides, however, has resulted in variable responses to these antigens (Baker et al., 1978; DeCueninck et al., 1982; DeCueninck et al., 1983; Eisenstein et al., 1983; Fischer et al., 1983; Kasper et al., 1983; Baker and Kasper, 1985; Swenson et al., 1985). Although transplacental transfer of vaccine-induced GBS antibody has not been studied in human neonates, conflicting results have been obtained in newborn animals. Immunization of rhesus monkeys prior to pregnancy did not appear to protect their offspring from GBS infection (Hemming et al., 1985). Immunization of mice prior to pregnancy, however, protected their offspring from GBS infection (Itoh et al., 1986 ). Different methods of immunogen preparation, route and schedule of administration may account for these findings. An animal model, with mothers who respond to GBS antigens and offspring who are susceptible to GBS infection, may be helpful in clarifying these differences prior to human clinical trials. Suckling rats are susceptible to GBS and are used in many models of this disease (Fischer et al., 1978; DeCueninck et al., 1982; Fischer et al., 1982). Rabbit (Fischer et al., 1978) ( 100-200/~1) or human (DeCueninck et al., 1982 ) (50/~l) serum, which contains > 10 di1-1 of GBS type-specific opsonic antibody, injected into suckling rats protects them from a lethal inoculation of that specific type GBS. The effect of transplacental immunization-induced GBS antibody on suckling rats hinges on the ability of their mothers to adequately respond to GBS immunization. The opsonic antibody response of rat dams to GBS immunization has not been investigated. The goal of this study was to produce immunization-induced GBS opsonic antibody in rat dams above a level which has been shown to protect suckling rats when passively injected. These immunized dams and their pups could then be used to study maternal-fetal GBS immunity. Desirable characteristics of a model include effective maternal antibody levels with a minimum of injections and effective antibody levels during the last trimester. Three questions were addressed: (1) Does immunization of rat dams with killed GBS type III result in levels of opsonic antibody that could be protective ( > 10 dil- 1) ? (2) Will the initial immunization booster interval affect the final antibody levels independent of the total number of immunizations? (3) Will the number of boosters after the primary immunization response affect the final level of antibody achieved? (4) Will pregnancy adversely affect the response to an optimal immunization schedule?
81
OPSONIC ANTIBODY RESPONSE TO MATERNALGBS MATERIALS AND METHODS
Animals Twenty outbred Wistar rat dams from a commercial breeder (Charles River Breeding Laboratories, Wilmington, MA) were housed in stainless steel cages with hardwood litter and fed Purina rat chow (Ralston Purina, St. Louis, MO) and water ad libitum. All these dams were screened for GBS type III rat IgG by a whole cell ELISA assay (Rote et al., 1980) as modified by P. Smith, Department of Pediatrics, Uniformed Services University, Bethesda, MD, (personal communication, 1987 ) and found to be negative. Twelve dams had given birth to their first litter of pups 1 month prior to the experiment. Four dams were assigned to each of three schedules described in Table 1. The other eight dams were plug positive pregnant primigravida 6 to 8-week-old Wistar rats from the same source and were immunized according to the optimal schedule determined by the first 12 dams. The Walter Reed Army Medical Center Animal Use Committee approved the experimental protocol. Immunization preparation GBS type III (strain SS878), kindly supplied by Dr. Gerald Fischer of the Uniformed Services University of the Health Sciences, Bethesda, MD, was grown to mid-log phase in Todd Hewitt broth then frozen at - 70 ° C until ready for use. The organism was cultured overnight on Columbia sheep blood agar plates (BBL Microbiologic Systems, San Juan Capistrano, CA) and then incubated in 25 ml of Todd Hewitt broth at 37°C for 12 to 18 h. The concentration of colony forming units (CFU) was determined on a fluorometer (Turner, 1985) (Sequoia Turner model 450, Sequoia-Turner, Mountain View, CA) according to growth curves previously established in our laboratory. The sediment was washed twice with normal saline and then suspended in 0.2% formalin in normal saline. The suspension was stored at 4 °C until two consecutive daily subcultures each demonstrated no growth for 48 h. The formalin-killed TABLE 1 Initial rat d a m i m m u n i z a t i o n a n d serum sample schedules Schedule
Number of dams
A
4
B
4
C
4
Day 0
7
14
22
29
36
S,P S,P S,P
S,B
S,B S,B
S,B S,B S,B
S,B S,B S,B
S
S = serum, P = primary immunization, B = booster immunization.
S
S
82
H.S. HEIMAN AND L.E. WEISMAN
bacteria were resuspended in normal saline to a concentration of 2 × 109 CFU/ ml and stored at - 70 ° C.
Immunization administration Freund's adjuvant (Gibco Laboratories, Chagrin Falls, OH) was emulsified with an equal volume of the formalin-killed GBS suspension. Complete adjuvant was used for the primary immunization and incomplete adjuvant was used for all boosters. The final immunization concentration was 1 X 109 CFU/ml. The 1.0 ml primary immunization was administered to the footpads in four divided doses using ether anesthesia (Binaghi and Benacerrat, 1963 ). Each 1.0 ml booster immunization was equally divided and administered intramuscularly to the four quadrants of the back. Immunization schedule The initial immunization schedules were suggested by previous rat vaccine response times. The rat's antibody responses to other antigens peaks at 7 to 11 days, wanes after 14 days and rises significantly with a booster at 12 days (Binaghi and Benacerrat, 1963). Schedules A and B were designed to administer initial boosters during the estimated peak of the primary immunization response (days 7 to 14). Schedule C delivered the initial booster at the estimated trough (day 22) of the primary vaccine response. Subsequent boosters for all schedules were administered at weekly intervals to maintain maximum antibody levels since the serum half-life of rat IgG is 5 to 7 days. The 36-day duration of the schedules was used to ensure significant opsonic antibody during the 22 + 2 day rat gestation. The optimal immunization schedule tested in eight pregnant dams, was based on the results of these initial schedules. Pregnant dams received their primary immunization at 3 days and their boosters at 11 and 18 days post-impregnation. Serologic assays Serum was obtained at the intervals noted in Table 1 and at the time of delivery for the pregnant dams and stored at - 70 ° C until analyzed. Total rat IgG was measured in duplicate by radial immunodiffusion (with sheep antirat IgG) (ICN Immunobiologicals, Lisle, IL) (Mancini et al., 1965). An opsonic assay for functional GBS type III antibody was performed according to previously reported methods (Fischer et al., 1981). This assay measures antibody-mediated killing of GBS in presence of human polymorphonuclear cells and rabbit complement in vitro. Opsonic antibody is expressed as the highest dilution which will kill > 90% of GBS (type III, strain SS878). Prior to testing for opsonic antibody, all rat sera and commercial IgG preparations were heatin.hctivated at 56 ° C for 30 min. Controls in every experiment included nentrophils alone, complement alone, and neutrophils plus complement. Two preparations of human intravenous immunoglobulin (IVIG) (Sandoglobulin, San-
OPSONIC ANTIBODY RESPONSE TO MATERNAL GBS
83
doz, East Hanover, NJ) with previously established levels of opsonic antibody and demonstrated protection of suckling rats from GBS type III disease were used as positive controls and for comparison with rat sera (Fischer et al., 1986). One was a 5% hyperimmune IVIG preparation isolated from students immunized with a pentavalent capsular polysaccharide vaccine. The other was a 5% standard IVIG preparation. A 5% human immune serum globulin (ISG) (Cutter Biological Division of Miles Laboratories, Berkeley, CA) which previously had no GBS type III opsonic antibody and did not protect suckling rats from GBS type III disease was used as a negative control and for comparison with rat sera (Binaghi and Benacerrat, 1963). Statistical methods Opsonic antibody is expressed as geometric means of the reciprocal serum dilutions. Analysis of variance was used to compare the preimmunization antibody of all three initial schedules. The two-tailed Wilcoxon signed rank rest was used to compare schedule A opsonic antibody responses with preimmunization. The two-tailed Wilcoxon rank sum test was used for all other analyses of opsonic antibody. The two-tailed Student's t-test for paired data was used to compare the schedule A IgG responses with preimmunization values. All statistical analysis were performed on an IBM AT Personal Computer (International Business Machines, Boca Raton, FL) using the Epistat program (copyright Gustafson TL, Round Rock, TX, 1984, Version 3.0) or the Abstat program (copyright AndersonBell, Parker, CO, 1987, Version 5.03). RESULTS The rats appeared to tolerate the immunizations well. There was no tissue breakdown although the primary immunization caused some local swelling for 3 to 5 days. The rat dams consistently demonstrated normal feeding and activity patterns. The booster immunizations caused a small amount of fur loss over the injection sites in some dams. The opsonic antibody results of day 36 for all three initial schedules are shown in Fig. 1. Opsonic antibody in the schedule A dams was sixteen to twentytwo-fold higher (113 di1-1) than dams in schedules B (7 di1-1, P < 0.03) or C (5 dil - 1, p < 0.03 ). The opsonic antibody produced by schedule A was between that of the hyperimmune IVIG ( 1280 dil- 1) and the standard IVIG (24 dil- 1) preparations. ISG and all rat sera prior to immunization demonstrated, as expected, no opsonic GBS antibody. The sera from schedule A were analyzed in more detail because these opsonic antibody concentrations were the only ones above our desired minimum (Fig. 2). The day 0 to 29 sera of one dam were lost in a lab accident. This dam's opsonic antibody of 160 dil-1 on day 36 was not significantly different from the other three dams in that schedule. No significant opsonic antibody was
84
H.S. HEIMAN AND L.E. WEISMAN
2560
!280
6~C
3Z0
c ~
160
~ m m
S ~
80
z <
20
g
lo
o
\'x\ \ \ \ \ \ \
\ \ \ \ \ \
---
\ \ \ \ \ \ . . . . . . .
\,,,~-,<
\ \ \ \ \ \ \ \ \'\\ \ \ \ \ \ \ \ \ \ \ \ \
5 25
NO KILLING
HYPERIMMUNE IVIS
. . . . . . . .
\ \ \ \
tlii
\ \ \ \ \ \
SOHEDULE A DAY 36
STANDARD IVIS
SCHEDULE B DA'Y 36
SCHEDULE C DAv 36
IMMUNE SERUr~I GLDBUL[N
Fig. 1. Type III group B streptococcal opsonic antibody in individual rat dams immunized according to various schedules, hyperimmune and standard intravenous immunoglobulin (IVIG), and immune serum globulin (ISG). Individual data points (dark squares) and geometric means (shaded bar for rat sera and open bar for IVIG and ISG) are expressed as the highest dilution 1 with > 90% bacterial killing. 10 dilution- 1 ( . . . ) is our minimum desired antibody value. (*P < 0.03 compared to schedule B or C ).
detected prior to immunization (day 0). GBS opsonic antibody rose slightly after the primary immunization (8 dil-1 ), but was not significantly different from preimmunization. The first and second boosters produced opsonic antibody (50 and 63 dil-1) significantly greater than preimmunization ( P < 0.01 ) and above our desired minimum. The opsonic antibody values after day 7 represent the highest dilutions tested and several in fact may be higher. The schedule A dams had a mean preimmunization total IgG of 467 + 83 mg/ 100 ml (Fig. 3 ). The IgG during the study period (days 7 to 36 ) did not change significantly compared to the preimmunization IgG ( P > 0.8). The opsonic antibody response was seven to ten-fold higher in schedule A after the first booster, than in schedule C or B after the second or third booster respectively (Table 2 ). Thus, dams given their first booster on day 7 produced the highest amount of opsonic antibody independent of the total number of immunizations. The opsonic antibody concentration of pregnant dams at delivery was 72 dil-1. This was not significantly different from nonpregnant dams (63 dil 1,
O P S O N I C A N T I B O D Y R E S P O N S E T O M A T E R N A L GBS
320
PNJECTION
PRIMARY
85
BOOSTER
BOOSTER
BOOSTER
BOOSTER
150
'
80
5
40
so
2O
'
•
I
I
win
m
l i
' i
÷
93
10 ¢u 5 25 NO KILLING
0
7
14
22 T I M E (DAYS)
29
3g
Fig. 2. Type III group B streptococcal opsonic antibody in rats immunized according to schedule A. The antibody values of individual rat dams (dark squares) and geometric means (open bars) are expressed as the highest dilution -1 with >90% bacterial killing. 10 dilution -1 (.") is our minimum desired antibody value. The arrows indicate the timing of individual immunizations. (*P < 0.01 compared to day 0; +P < 0.001 compared to day 0).
700 600
E
500
8
400
m
300 200 100 0
Primary
Booster Booster
I
I
I
0
7
14
Booster I
21
Booster 1
28
I
35
Time (Days)
Fig. 3. Total IgG for rat dams in vaccine schedule A. The IgG values are expressed as mean _+s.e.m. ( n = 3 ). The arrows indicate the timing of primary and booster immunizations.
P> 0.8) who received a similar immunization schedule (schedule A after the second booster). Thus pregnancy appeared to have no significant impact on opsonic antibody production.
86
H.S. HEIMAN AND L.E. WEISMAN
TABLE2 O p s o n i c a n t i b o d y ~ r e s p o n s e after i m m u n i z a t i o n of rat d a m s w i t h type I I I group B s t r e p t o c o c c u s Schedule
A B C
Pre
None None None
Primary immunization
8 -
Booster immunization 1
2
3
4
50 -
63 5
63 7
80, 1132
1Expressed as geometric m e a n s e r u m d i l u t i o n - 1 w i t h > 90% killing. 280 w h e n assayed w i t h schedule A samples, 113 w h e n a s s a y e d w i t h schedule B a n d C samples. P r e = p r i o r i m m u n i z a t i o n , N o n e = n o n e detected, - = n o t done. DISCUSSION
Female mice have responded to intravenous immunization with heated or formalinized whole cell types Ia, Ia/c, III/c group B streptococci prior to pregnancy (Itoh et al., 1986). Female rhesus monkeys have responded to subcutaneous immunization with live GBS (type III, strain 878) prior to pregnancy (Hemming et al., 1985). Though opsonic antibody was not measured, both studies documented increased type-specific antibody following immunization as measured by enzyme-linked immunosorbent assays. Immunization with each serotype, appeared to protect neonatal mice from bacterial infection with the homologous organism. Some cross protectivity, however, was observed between serotype immunizations. There was no significant protection from intrauterine infection for the small number of newborn monkeys studied. Research using monkeys may be more applicable to the human experience, but cost and availability are sometimes prohibitive. Further investigation of maternal-fetal immunity in the well known suckling rat model of GBS disease may be helpful in sorting out these differences. Type III GBS was chosen for immunization since it causes 67% of all neonatal diseases (Baker and Edwards, 1983 ). The majority of the GBS antibody studies have been done with type III organisms, especially with their purified capsular polysaccharides. The whole-cell immunogen used in this study contains type and group-specific antigens. Future vaccination of the rat dam should include type-specific GBS antigens with or without other associated protein antigens. Wistar rat dams, when immunized with non-GBS antigens and Freund's adjuvant, developed greater amounts of antigen-specific antibody in less time than those vaccinated with whole organisms alone (Binaghi and Benacerrat 1963). Freund's adjuvant with whole-cell killed GBS type III has been used ir the successful immunization of mice (Shigeoka et al., 1984; Itoh et al., 1986) Using Freund's adjuvant in our vaccine we consistently produced significan!
OPSONIC ANTIBODY RESPONSE TO MATERNAL GBS
87
opsonic antibody in a relatively short time-period. Other adjuvants which are less toxic and more immunogenic should be considered in future studies. The amount of opsonic antibody correlates well with the in vivo protection of suckling rats (Fischer et al., 1982, 1986). Heat inactivation of all samples eliminated the opsonic activity of complement and the contribution of fibronectin to opsonic activity should not increase with immunization. It appears unlikely that these other factors contributed significantly to changes in opsonic activity. Enzyme-linked immunosorbent assays for GBS antibody are more precise, but do not accurately predict in vivo protection of suckling rats (Fischer et al., 1986). Prior to conducting neonatal impact studies, we chose to base our selection of an immunization schedule on a method that could potentially predict suckling rat survival following GBS disease. The standard IVIG preparation, with a 10 dil-1 opsonic antibody concentration, has been shown to protect suckling rat pups from lethal GBS inoculations (Fischer et al., 1981 ). The dams' opsonic antibody after the first booster in schedule A was five-fold higher than the levels at which this standard IVIG protected suckling rats. Though, this suggests that the rat dam immunizationinduced antibody concentrations produced in this study may protect suckling rats from type III GBS disease, extrapolation from human antibody may not be applicable. Previous exposure to GBS might have resulted in a higher antibody response to immunization. The lack of detectable GBS type III opsonic antibody and GBS type III IgG by whole cell enzyme-linked immunosorbent assay prior to immunization in these dams suggests they had no previous significant exposure to GBS. Though GBS type-specific opsonic antibody increased significantly in schedule A dams, total IgG did not change. This agrees with previous vaccine work and suggests that after vaccination a larger portion of the immunoglobulin may be GBS type-specific. The decreased amount of opsonic antibody produced following the 14 or 22day initial booster schedules could be the results of poor response to the initial immunization, to long a delay before the initial boosters were given, or consumption of antibody by the immunization antigens in the later boosters. Multiple boosters have been found to decrease antibody levels (Binaghi and Benacerrat, 1963). Since the schedule A levels did not decrease with multiple boosters, it is unlikely that the booster antigens caused the low antibody levels of schedules B and C. Leslie and Carwile (1973) have suggested that there may be some genetic non-responders to group A streptococcus. It is unlikely, however, that all of the Wistar rat dams assigned to schedules B and C were genetic non-responders to group B streptococcus. The lower concentration of opsonic antibody produced by these schedules would most likely be the result of too long an initial booster interval. The dams assigned to the initial immunization schedules, although recently postpartum, were not pregnant. Pregnant women respond to meningococcal
88
H.S. HEIMANANDL.E.WEISMAN
vaccine as well as non-pregnant women (McCormick et al., 1980). The pregnant rats who were immunized with the optimal schedule had antibody responses that were not significantly different from the non-pregnant dams who received a similar schedule. Pregnant and non-pregnant rat dams develop type III GBS opsonic antibody in response to a whole-cell killed immunization in Freund's adjuvant. Primary immunization on the third day of pregnancy followed by 2-weekly boosters should expose suckling rats to enough antibody in utero during the last trimester to potentially protect suckling rats from GBS infection. Development of this immunization schedule for pregnant rats should allow assesment of other adjuvants and antigens, and selected aspects of maternal-fetal immunity using established suckling rat models of GBS disease. ACKNOWLEDGEMENTS
This work was supported in part by the Walter Reed Army Medical Center, Department of Clinical Investigation Grant # 6055. We wish to thank Sam Wilson, Gerald W. Fischer MD and Patrick Smith PhD, of the Department of Pediatrics, Uniformed Services University for sharing their knowledge, and resources with us. Reprint requests to (HSH) Department of Pediatrics, Brooke Army Medical Center, San Antonio, Texas, TX 78234-6200. The opinions and assertations contained herein are those of the author and do not necessarily reflect those of the Department of the Army or the Department of Defense.
REFERENCES Baker, C.J. and Edwards, M.S., 1983. Group B streptococcal infections. In: J.D. Remington and J.O. Klein (Editors), Infectious Diseases of the Fetus and Newborn Infant, 2nd Edn. W.B. Saunders Co., Philadelphia, PA, pp. 820-881. Baker, C.J. and Kasper, D.L., 1976. Correlation of maternal antibody deficiency with susceptibility to neonatal group B streptococcal infection. N. Engl. J. Med., 294: 753-756. Baker, C.J. and Kasper, D.L., 1985. Group B streptococcal vaccines. Rev. Infect. Dis., 7: 458-467. Baker, C.J., Edwards, M.S. and Kasper, D.L., 1978. Immunogenicityof polysaccharides from type III group B streptococcus. J. Clin. Invest., 61: 1107-1110. Baker, C.J., Edwards, W.S. and Kasper, D.L., 1981. Role of antibody to native type III polysaccharide of group B streptococcus in infant infection. Pediatrics, 68: 544-549. Binaghi, R.A. and Benacerrat, B., 1963. The production of anaphylactic antibody in the rat. J. Immunol., 92:920 926. DeCueninck, B.J., Eisenstein, T.K., McIntosh, T.S., Shockman, G.D. and Swenson, R.M., 1982. Type-specific protection of neonatal rats from lethal group B streptococcal infection by immune sera obtained from human volunteers vaccinated with type III-specific polysaccharide. Infect. Immun., 37: 961-965. DeCueninck, B.J., Eisenstein, T.K., McIntosh, T.S., Shockman, G.D. and Swenson, R., 1983.
OPSONICANTIBODYRESPONSETO MATERNALGBS
89
Quantitation of in vitro opsonic activity of human antibody induced by a vaccine consisting of the type III-specific polysaccharide of group B streptococcus. Infect. Immun., 39: 1155-1160. Eisenstein, T.K., DeCueninck, B.J., Resavy, D., Shockman, G.D., Carey, R.B. and Swenson, R.M., 1983. Quantitative determination in human sera of vaccine-induced antibody to type-specific polysaccharides of group B streptococci using an enzyme-linked immunosorbent assay. J. Infect. Dis., 147: 847-856. Fischer, G.W., Lowell, G.H., Crumrine, M.H. and Bass, J.W., 1978. Demonstration of opsonic activity and in vivo protection against group B streptococci type III by streptococcus pneumoniae type 14 antisera. J. Exp. Med., 148: 776-786. Fischer, G.W., Hunter, K.W. and Wilson, S.R., 1981. Type III group B streptococcal strain differences in susceptibility to opsonization with human serum. Pediatr. Res., 15: 1525-1529. Fischer, G.W., Hunter, K.W. and Wilson, S.R., 1982. Modified human immune serum globulin for intravenous administration: in vitro opsonic activity and in vivo protection against group B streptococcal disease in suckling rats. Acta Paediatr. Scand., 71: 639-644. Fischer, G.W., Horton, R.E. and Edelman, R., 1983. Summary of the National Institutes of Health workshop on group B streptococcal infection. J. Infect. Dis., 148: 163-166. Fischer, G.W., Hemming, V.G., Hunter, K.W., Gloser, H., Bachmayer, H., Von Pilar, C.E., Helting, T., Weisman, L.E., Wilson, S.R. and Baron, P.A., 1986. Intravenous immunoglobulin in the treatment of neonatal sepsis: therapeutic strategies and laboratory studies. Pediatr. Infect. Dis., 5: $171-$175. Hemming, V.G., London, W.T., Curfman, B.L., Patrick, D.F. and Fischer, G.W., 1985. Maternal immunity and neonatal GBS infections: studies in a primate model. Antibiot. Chemother., 35: 194-200. Itoh, T., Yan, X., Nakano, H. and Yoshioka, M., 1986. Protective efficacy against group B streptococcal infection in neonatal mice delivered from preimmunized pregnants. Microbiol. Immunol., 30: 297-305, Kasper, D.L., Baker, C.J., Galdes, B., Katzenellenbogen, E. and Jennings, H.J., 1983. Immunochemical analysis and immunogenicity of the type II group B streptococcal capsular polysaccharide. J. Clin. Invest., 72: 260-269. Leslie, G.A. and Carwile, H.F., 1973. Immune response of rats to group A streptococcal vaccine, Infect. Immun., 7: 781-785. Mancini, G., Carbonara, A.O. and Heremans, J.F., 1965. Immunochemical quantitation of antigens by single radial immunodiffusion. Int. J. Immunochem., 2: 235-254. McCormick, J.B., Hermalino, H.G., Shigeo, N., Freire, J.B., Veras, J., Gorman, G., Teeley, T.C. and Wingo, P., 1980. Antibody response to serogroup A and C meningococcal polysaccharide vaccines in infants born to mothers vaccinated during pregnancy. J. Clin. Invest., 65:11411144. Rote, N.S., Taylor, N.L., Shigeoka, A.O., Scott, J.R. and Hill, H.R., 1980. Enzyme-linked immunosorbent assay for group B streptococcal antibodies. Infect. Immun., 27: 116-123. Shigeoka, A.O., Pincus, S.H., Rote, N.S. and Hill, H.R., 1984. Protective efficacy of hybridoma type-specific antibody against experimental infection with group B streptococcus. J. Infect. Dis., 149: 363-372. Siegel, J., 1985. Prevention and treatment of group B streptococcal infections. Pediatr. Infect. Dis., 4: $33-$36. Swenson, R.M., Eisenstein, T.K. and Shockman, G.D., 1985. The soluble antigens of group B streptococcus as human vaccines. Antibiot. Chemother., 35: 291-295. Turner, G.K., 1985. Measurement of light from chemical or biochemical reactions. In: K. Van Dyke (Editor), Bioluminescence and Chemiluminescence Instruments and Applications, Volume 1. C.R.C. Press Inc., Boca Raton, FL, pp. 47-51. Wilson, C.B., 1986. Immunologic basis for increased susceptibility of the neonate to infection. J. Pediatr., 108: 1-11. Yeung, C.Y. and Hobbs, J.R., 1968. Serum-gamma g-globulin levels in normal premature, postmature, and "small-for-dates" newborn babies. Lancet, 1: 1167-1172.
79
The Opsonic A n t i b o d y R e s p o n s e of Female Rats to T y p e III Group B S t r e p t o c o c c a l I m m u n i z a t i o n : a Model for M a t e r n a l I m m u n i t y HOWARD S. HEIMAN and LEONARD E. WEISMAN
Departments of Pediatrics, Uniformed Services University, Bethesda, MD 20814-4799 (U.S.A.) Walter Reed Army Medical Center, Washington, DC 20307-5000 (U.S.A.) (Accepted 28 April 1989 )
ABSTRACT Heiman, H.S. and Weisman, L.E., 1990. The opsonic antibody response of female rats to type III group B streptococcal immunization: a model for maternal immunity. Vet. Immunol. Immu nopathol., 24: 79-89. Group B streptococcus (GBS) remains a major cause of neonatal infection. Maternal immunization-induced GBS antibody may protect neonates from GBS disease. Since the opsonophagocytosis assay correlates well with survival in GBS infected suckling rats, we sought to determine an immunization schedule which would induce type III GBS opsonic antibody in rat dams above a predetermined level of 10 dilution- 1 (dil- 1). This schedule could then be used for future studies of maternal-fetal immunity. Wistar rat dams (n = 12 ) were given killed GBS type III using three immunization schedules (primary injection, initial booster at 7, 14 or 22 days and then weekly boosters). Opsonic GBS type III antibody and total immunoglobulin (IgG) were measured. Only the schedule with a 7-day initial booster resulted in GBS type-specific opsonic antibody consistently above 10 dil-1. The IgG (467 _+83 mg/100 ml) remained constant while the opsonic antibody increased significantly to 50 and 63 dil- 1 (p < 0.01 compared to day 0 ) after boosters on day 7 and 14 respectively. Eight pregnant dams, who received a primary immunization and boosters at 7 and 14 days, developed GBS type III opsonic antibody titers (72 dil-1) similar to non-pregnant dams and potentially adequate to protect suckling rats from GBS disease. This model may now be used to study other adjuvants, immunogens, and maternal-fetal immunity using established suckling rat models of GBS disease. ABBREVIATIONS CFU, colony forming units; di1-1, inverse dilution; GBS, group B streptococcus; IgG, immunoglobulin G; ISG, immune serum globulin; IVIG, intravenous immunoglobulin.
INTRODUCTION
Group B streptococcus (GBS) infection remains a major cause of neonatal morbidity and mortality despite antibiotic therapy (Baker and Edwards, 1983; 0165-2427/90/$03.50
© 1990 Elsevier Science Publishers B.V.
SO
H.S. HEIMAN AND L.E, WEISMAN
Siegel, 1985; Wilson, 1986). It appears to overwhelm the neonate's immune system which is largely dependent on transplacental transfer of maternal IgG (Yeung and Hobbs, 1968; Wilson, 1986). Children of women with naturally elevated GBS type III antibody, however, have a decreased incidence of neonatal GBS type III disease (Baker and Kasper, 1976; Baker et al., 1981). Enhancement of maternal GBS type-specific immunity by vaccination, appears to be a logical step in controlling the deleterious effects of this organism in the neonate. Vaccination of non-pregnant women with purified GBS typespecific capsular polysaccharides, however, has resulted in variable responses to these antigens (Baker et al., 1978; DeCueninck et al., 1982; DeCueninck et al., 1983; Eisenstein et al., 1983; Fischer et al., 1983; Kasper et al., 1983; Baker and Kasper, 1985; Swenson et al., 1985). Although transplacental transfer of vaccine-induced GBS antibody has not been studied in human neonates, conflicting results have been obtained in newborn animals. Immunization of rhesus monkeys prior to pregnancy did not appear to protect their offspring from GBS infection (Hemming et al., 1985). Immunization of mice prior to pregnancy, however, protected their offspring from GBS infection (Itoh et al., 1986 ). Different methods of immunogen preparation, route and schedule of administration may account for these findings. An animal model, with mothers who respond to GBS antigens and offspring who are susceptible to GBS infection, may be helpful in clarifying these differences prior to human clinical trials. Suckling rats are susceptible to GBS and are used in many models of this disease (Fischer et al., 1978; DeCueninck et al., 1982; Fischer et al., 1982). Rabbit (Fischer et al., 1978) ( 100-200/~1) or human (DeCueninck et al., 1982 ) (50/~l) serum, which contains > 10 di1-1 of GBS type-specific opsonic antibody, injected into suckling rats protects them from a lethal inoculation of that specific type GBS. The effect of transplacental immunization-induced GBS antibody on suckling rats hinges on the ability of their mothers to adequately respond to GBS immunization. The opsonic antibody response of rat dams to GBS immunization has not been investigated. The goal of this study was to produce immunization-induced GBS opsonic antibody in rat dams above a level which has been shown to protect suckling rats when passively injected. These immunized dams and their pups could then be used to study maternal-fetal GBS immunity. Desirable characteristics of a model include effective maternal antibody levels with a minimum of injections and effective antibody levels during the last trimester. Three questions were addressed: (1) Does immunization of rat dams with killed GBS type III result in levels of opsonic antibody that could be protective ( > 10 dil- 1) ? (2) Will the initial immunization booster interval affect the final antibody levels independent of the total number of immunizations? (3) Will the number of boosters after the primary immunization response affect the final level of antibody achieved? (4) Will pregnancy adversely affect the response to an optimal immunization schedule?
81
OPSONIC ANTIBODY RESPONSE TO MATERNALGBS MATERIALS AND METHODS
Animals Twenty outbred Wistar rat dams from a commercial breeder (Charles River Breeding Laboratories, Wilmington, MA) were housed in stainless steel cages with hardwood litter and fed Purina rat chow (Ralston Purina, St. Louis, MO) and water ad libitum. All these dams were screened for GBS type III rat IgG by a whole cell ELISA assay (Rote et al., 1980) as modified by P. Smith, Department of Pediatrics, Uniformed Services University, Bethesda, MD, (personal communication, 1987 ) and found to be negative. Twelve dams had given birth to their first litter of pups 1 month prior to the experiment. Four dams were assigned to each of three schedules described in Table 1. The other eight dams were plug positive pregnant primigravida 6 to 8-week-old Wistar rats from the same source and were immunized according to the optimal schedule determined by the first 12 dams. The Walter Reed Army Medical Center Animal Use Committee approved the experimental protocol. Immunization preparation GBS type III (strain SS878), kindly supplied by Dr. Gerald Fischer of the Uniformed Services University of the Health Sciences, Bethesda, MD, was grown to mid-log phase in Todd Hewitt broth then frozen at - 70 ° C until ready for use. The organism was cultured overnight on Columbia sheep blood agar plates (BBL Microbiologic Systems, San Juan Capistrano, CA) and then incubated in 25 ml of Todd Hewitt broth at 37°C for 12 to 18 h. The concentration of colony forming units (CFU) was determined on a fluorometer (Turner, 1985) (Sequoia Turner model 450, Sequoia-Turner, Mountain View, CA) according to growth curves previously established in our laboratory. The sediment was washed twice with normal saline and then suspended in 0.2% formalin in normal saline. The suspension was stored at 4 °C until two consecutive daily subcultures each demonstrated no growth for 48 h. The formalin-killed TABLE 1 Initial rat d a m i m m u n i z a t i o n a n d serum sample schedules Schedule
Number of dams
A
4
B
4
C
4
Day 0
7
14
22
29
36
S,P S,P S,P
S,B
S,B S,B
S,B S,B S,B
S,B S,B S,B
S
S = serum, P = primary immunization, B = booster immunization.
S
S
82
H.S. HEIMAN AND L.E. WEISMAN
bacteria were resuspended in normal saline to a concentration of 2 × 109 CFU/ ml and stored at - 70 ° C.
Immunization administration Freund's adjuvant (Gibco Laboratories, Chagrin Falls, OH) was emulsified with an equal volume of the formalin-killed GBS suspension. Complete adjuvant was used for the primary immunization and incomplete adjuvant was used for all boosters. The final immunization concentration was 1 X 109 CFU/ml. The 1.0 ml primary immunization was administered to the footpads in four divided doses using ether anesthesia (Binaghi and Benacerrat, 1963 ). Each 1.0 ml booster immunization was equally divided and administered intramuscularly to the four quadrants of the back. Immunization schedule The initial immunization schedules were suggested by previous rat vaccine response times. The rat's antibody responses to other antigens peaks at 7 to 11 days, wanes after 14 days and rises significantly with a booster at 12 days (Binaghi and Benacerrat, 1963). Schedules A and B were designed to administer initial boosters during the estimated peak of the primary immunization response (days 7 to 14). Schedule C delivered the initial booster at the estimated trough (day 22) of the primary vaccine response. Subsequent boosters for all schedules were administered at weekly intervals to maintain maximum antibody levels since the serum half-life of rat IgG is 5 to 7 days. The 36-day duration of the schedules was used to ensure significant opsonic antibody during the 22 + 2 day rat gestation. The optimal immunization schedule tested in eight pregnant dams, was based on the results of these initial schedules. Pregnant dams received their primary immunization at 3 days and their boosters at 11 and 18 days post-impregnation. Serologic assays Serum was obtained at the intervals noted in Table 1 and at the time of delivery for the pregnant dams and stored at - 70 ° C until analyzed. Total rat IgG was measured in duplicate by radial immunodiffusion (with sheep antirat IgG) (ICN Immunobiologicals, Lisle, IL) (Mancini et al., 1965). An opsonic assay for functional GBS type III antibody was performed according to previously reported methods (Fischer et al., 1981). This assay measures antibody-mediated killing of GBS in presence of human polymorphonuclear cells and rabbit complement in vitro. Opsonic antibody is expressed as the highest dilution which will kill > 90% of GBS (type III, strain SS878). Prior to testing for opsonic antibody, all rat sera and commercial IgG preparations were heatin.hctivated at 56 ° C for 30 min. Controls in every experiment included nentrophils alone, complement alone, and neutrophils plus complement. Two preparations of human intravenous immunoglobulin (IVIG) (Sandoglobulin, San-
OPSONIC ANTIBODY RESPONSE TO MATERNAL GBS
83
doz, East Hanover, NJ) with previously established levels of opsonic antibody and demonstrated protection of suckling rats from GBS type III disease were used as positive controls and for comparison with rat sera (Fischer et al., 1986). One was a 5% hyperimmune IVIG preparation isolated from students immunized with a pentavalent capsular polysaccharide vaccine. The other was a 5% standard IVIG preparation. A 5% human immune serum globulin (ISG) (Cutter Biological Division of Miles Laboratories, Berkeley, CA) which previously had no GBS type III opsonic antibody and did not protect suckling rats from GBS type III disease was used as a negative control and for comparison with rat sera (Binaghi and Benacerrat, 1963). Statistical methods Opsonic antibody is expressed as geometric means of the reciprocal serum dilutions. Analysis of variance was used to compare the preimmunization antibody of all three initial schedules. The two-tailed Wilcoxon signed rank rest was used to compare schedule A opsonic antibody responses with preimmunization. The two-tailed Wilcoxon rank sum test was used for all other analyses of opsonic antibody. The two-tailed Student's t-test for paired data was used to compare the schedule A IgG responses with preimmunization values. All statistical analysis were performed on an IBM AT Personal Computer (International Business Machines, Boca Raton, FL) using the Epistat program (copyright Gustafson TL, Round Rock, TX, 1984, Version 3.0) or the Abstat program (copyright AndersonBell, Parker, CO, 1987, Version 5.03). RESULTS The rats appeared to tolerate the immunizations well. There was no tissue breakdown although the primary immunization caused some local swelling for 3 to 5 days. The rat dams consistently demonstrated normal feeding and activity patterns. The booster immunizations caused a small amount of fur loss over the injection sites in some dams. The opsonic antibody results of day 36 for all three initial schedules are shown in Fig. 1. Opsonic antibody in the schedule A dams was sixteen to twentytwo-fold higher (113 di1-1) than dams in schedules B (7 di1-1, P < 0.03) or C (5 dil - 1, p < 0.03 ). The opsonic antibody produced by schedule A was between that of the hyperimmune IVIG ( 1280 dil- 1) and the standard IVIG (24 dil- 1) preparations. ISG and all rat sera prior to immunization demonstrated, as expected, no opsonic GBS antibody. The sera from schedule A were analyzed in more detail because these opsonic antibody concentrations were the only ones above our desired minimum (Fig. 2). The day 0 to 29 sera of one dam were lost in a lab accident. This dam's opsonic antibody of 160 dil-1 on day 36 was not significantly different from the other three dams in that schedule. No significant opsonic antibody was
84
H.S. HEIMAN AND L.E. WEISMAN
2560
!280
6~C
3Z0
c ~
160
~ m m
S ~
80
z <
20
g
lo
o
\'x\ \ \ \ \ \ \
\ \ \ \ \ \
---
\ \ \ \ \ \ . . . . . . .
\,,,~-,<
\ \ \ \ \ \ \ \ \'\\ \ \ \ \ \ \ \ \ \ \ \ \
5 25
NO KILLING
HYPERIMMUNE IVIS
. . . . . . . .
\ \ \ \
tlii
\ \ \ \ \ \
SOHEDULE A DAY 36
STANDARD IVIS
SCHEDULE B DA'Y 36
SCHEDULE C DAv 36
IMMUNE SERUr~I GLDBUL[N
Fig. 1. Type III group B streptococcal opsonic antibody in individual rat dams immunized according to various schedules, hyperimmune and standard intravenous immunoglobulin (IVIG), and immune serum globulin (ISG). Individual data points (dark squares) and geometric means (shaded bar for rat sera and open bar for IVIG and ISG) are expressed as the highest dilution 1 with > 90% bacterial killing. 10 dilution- 1 ( . . . ) is our minimum desired antibody value. (*P < 0.03 compared to schedule B or C ).
detected prior to immunization (day 0). GBS opsonic antibody rose slightly after the primary immunization (8 dil-1 ), but was not significantly different from preimmunization. The first and second boosters produced opsonic antibody (50 and 63 dil-1) significantly greater than preimmunization ( P < 0.01 ) and above our desired minimum. The opsonic antibody values after day 7 represent the highest dilutions tested and several in fact may be higher. The schedule A dams had a mean preimmunization total IgG of 467 + 83 mg/ 100 ml (Fig. 3 ). The IgG during the study period (days 7 to 36 ) did not change significantly compared to the preimmunization IgG ( P > 0.8). The opsonic antibody response was seven to ten-fold higher in schedule A after the first booster, than in schedule C or B after the second or third booster respectively (Table 2 ). Thus, dams given their first booster on day 7 produced the highest amount of opsonic antibody independent of the total number of immunizations. The opsonic antibody concentration of pregnant dams at delivery was 72 dil-1. This was not significantly different from nonpregnant dams (63 dil 1,
O P S O N I C A N T I B O D Y R E S P O N S E T O M A T E R N A L GBS
320
PNJECTION
PRIMARY
85
BOOSTER
BOOSTER
BOOSTER
BOOSTER
150
'
80
5
40
so
2O
'
•
I
I
win
m
l i
' i
÷
93
10 ¢u 5 25 NO KILLING
0
7
14
22 T I M E (DAYS)
29
3g
Fig. 2. Type III group B streptococcal opsonic antibody in rats immunized according to schedule A. The antibody values of individual rat dams (dark squares) and geometric means (open bars) are expressed as the highest dilution -1 with >90% bacterial killing. 10 dilution -1 (.") is our minimum desired antibody value. The arrows indicate the timing of individual immunizations. (*P < 0.01 compared to day 0; +P < 0.001 compared to day 0).
700 600
E
500
8
400
m
300 200 100 0
Primary
Booster Booster
I
I
I
0
7
14
Booster I
21
Booster 1
28
I
35
Time (Days)
Fig. 3. Total IgG for rat dams in vaccine schedule A. The IgG values are expressed as mean _+s.e.m. ( n = 3 ). The arrows indicate the timing of primary and booster immunizations.
P> 0.8) who received a similar immunization schedule (schedule A after the second booster). Thus pregnancy appeared to have no significant impact on opsonic antibody production.
86
H.S. HEIMAN AND L.E. WEISMAN
TABLE2 O p s o n i c a n t i b o d y ~ r e s p o n s e after i m m u n i z a t i o n of rat d a m s w i t h type I I I group B s t r e p t o c o c c u s Schedule
A B C
Pre
None None None
Primary immunization
8 -
Booster immunization 1
2
3
4
50 -
63 5
63 7
80, 1132
1Expressed as geometric m e a n s e r u m d i l u t i o n - 1 w i t h > 90% killing. 280 w h e n assayed w i t h schedule A samples, 113 w h e n a s s a y e d w i t h schedule B a n d C samples. P r e = p r i o r i m m u n i z a t i o n , N o n e = n o n e detected, - = n o t done. DISCUSSION
Female mice have responded to intravenous immunization with heated or formalinized whole cell types Ia, Ia/c, III/c group B streptococci prior to pregnancy (Itoh et al., 1986). Female rhesus monkeys have responded to subcutaneous immunization with live GBS (type III, strain 878) prior to pregnancy (Hemming et al., 1985). Though opsonic antibody was not measured, both studies documented increased type-specific antibody following immunization as measured by enzyme-linked immunosorbent assays. Immunization with each serotype, appeared to protect neonatal mice from bacterial infection with the homologous organism. Some cross protectivity, however, was observed between serotype immunizations. There was no significant protection from intrauterine infection for the small number of newborn monkeys studied. Research using monkeys may be more applicable to the human experience, but cost and availability are sometimes prohibitive. Further investigation of maternal-fetal immunity in the well known suckling rat model of GBS disease may be helpful in sorting out these differences. Type III GBS was chosen for immunization since it causes 67% of all neonatal diseases (Baker and Edwards, 1983 ). The majority of the GBS antibody studies have been done with type III organisms, especially with their purified capsular polysaccharides. The whole-cell immunogen used in this study contains type and group-specific antigens. Future vaccination of the rat dam should include type-specific GBS antigens with or without other associated protein antigens. Wistar rat dams, when immunized with non-GBS antigens and Freund's adjuvant, developed greater amounts of antigen-specific antibody in less time than those vaccinated with whole organisms alone (Binaghi and Benacerrat 1963). Freund's adjuvant with whole-cell killed GBS type III has been used ir the successful immunization of mice (Shigeoka et al., 1984; Itoh et al., 1986) Using Freund's adjuvant in our vaccine we consistently produced significan!
OPSONIC ANTIBODY RESPONSE TO MATERNAL GBS
87
opsonic antibody in a relatively short time-period. Other adjuvants which are less toxic and more immunogenic should be considered in future studies. The amount of opsonic antibody correlates well with the in vivo protection of suckling rats (Fischer et al., 1982, 1986). Heat inactivation of all samples eliminated the opsonic activity of complement and the contribution of fibronectin to opsonic activity should not increase with immunization. It appears unlikely that these other factors contributed significantly to changes in opsonic activity. Enzyme-linked immunosorbent assays for GBS antibody are more precise, but do not accurately predict in vivo protection of suckling rats (Fischer et al., 1986). Prior to conducting neonatal impact studies, we chose to base our selection of an immunization schedule on a method that could potentially predict suckling rat survival following GBS disease. The standard IVIG preparation, with a 10 dil-1 opsonic antibody concentration, has been shown to protect suckling rat pups from lethal GBS inoculations (Fischer et al., 1981 ). The dams' opsonic antibody after the first booster in schedule A was five-fold higher than the levels at which this standard IVIG protected suckling rats. Though, this suggests that the rat dam immunizationinduced antibody concentrations produced in this study may protect suckling rats from type III GBS disease, extrapolation from human antibody may not be applicable. Previous exposure to GBS might have resulted in a higher antibody response to immunization. The lack of detectable GBS type III opsonic antibody and GBS type III IgG by whole cell enzyme-linked immunosorbent assay prior to immunization in these dams suggests they had no previous significant exposure to GBS. Though GBS type-specific opsonic antibody increased significantly in schedule A dams, total IgG did not change. This agrees with previous vaccine work and suggests that after vaccination a larger portion of the immunoglobulin may be GBS type-specific. The decreased amount of opsonic antibody produced following the 14 or 22day initial booster schedules could be the results of poor response to the initial immunization, to long a delay before the initial boosters were given, or consumption of antibody by the immunization antigens in the later boosters. Multiple boosters have been found to decrease antibody levels (Binaghi and Benacerrat, 1963). Since the schedule A levels did not decrease with multiple boosters, it is unlikely that the booster antigens caused the low antibody levels of schedules B and C. Leslie and Carwile (1973) have suggested that there may be some genetic non-responders to group A streptococcus. It is unlikely, however, that all of the Wistar rat dams assigned to schedules B and C were genetic non-responders to group B streptococcus. The lower concentration of opsonic antibody produced by these schedules would most likely be the result of too long an initial booster interval. The dams assigned to the initial immunization schedules, although recently postpartum, were not pregnant. Pregnant women respond to meningococcal
88
H.S. HEIMANANDL.E.WEISMAN
vaccine as well as non-pregnant women (McCormick et al., 1980). The pregnant rats who were immunized with the optimal schedule had antibody responses that were not significantly different from the non-pregnant dams who received a similar schedule. Pregnant and non-pregnant rat dams develop type III GBS opsonic antibody in response to a whole-cell killed immunization in Freund's adjuvant. Primary immunization on the third day of pregnancy followed by 2-weekly boosters should expose suckling rats to enough antibody in utero during the last trimester to potentially protect suckling rats from GBS infection. Development of this immunization schedule for pregnant rats should allow assesment of other adjuvants and antigens, and selected aspects of maternal-fetal immunity using established suckling rat models of GBS disease. ACKNOWLEDGEMENTS
This work was supported in part by the Walter Reed Army Medical Center, Department of Clinical Investigation Grant # 6055. We wish to thank Sam Wilson, Gerald W. Fischer MD and Patrick Smith PhD, of the Department of Pediatrics, Uniformed Services University for sharing their knowledge, and resources with us. Reprint requests to (HSH) Department of Pediatrics, Brooke Army Medical Center, San Antonio, Texas, TX 78234-6200. The opinions and assertations contained herein are those of the author and do not necessarily reflect those of the Department of the Army or the Department of Defense.
REFERENCES Baker, C.J. and Edwards, M.S., 1983. Group B streptococcal infections. In: J.D. Remington and J.O. Klein (Editors), Infectious Diseases of the Fetus and Newborn Infant, 2nd Edn. W.B. Saunders Co., Philadelphia, PA, pp. 820-881. Baker, C.J. and Kasper, D.L., 1976. Correlation of maternal antibody deficiency with susceptibility to neonatal group B streptococcal infection. N. Engl. J. Med., 294: 753-756. Baker, C.J. and Kasper, D.L., 1985. Group B streptococcal vaccines. Rev. Infect. Dis., 7: 458-467. Baker, C.J., Edwards, M.S. and Kasper, D.L., 1978. Immunogenicityof polysaccharides from type III group B streptococcus. J. Clin. Invest., 61: 1107-1110. Baker, C.J., Edwards, W.S. and Kasper, D.L., 1981. Role of antibody to native type III polysaccharide of group B streptococcus in infant infection. Pediatrics, 68: 544-549. Binaghi, R.A. and Benacerrat, B., 1963. The production of anaphylactic antibody in the rat. J. Immunol., 92:920 926. DeCueninck, B.J., Eisenstein, T.K., McIntosh, T.S., Shockman, G.D. and Swenson, R.M., 1982. Type-specific protection of neonatal rats from lethal group B streptococcal infection by immune sera obtained from human volunteers vaccinated with type III-specific polysaccharide. Infect. Immun., 37: 961-965. DeCueninck, B.J., Eisenstein, T.K., McIntosh, T.S., Shockman, G.D. and Swenson, R., 1983.
OPSONICANTIBODYRESPONSETO MATERNALGBS
89
Quantitation of in vitro opsonic activity of human antibody induced by a vaccine consisting of the type III-specific polysaccharide of group B streptococcus. Infect. Immun., 39: 1155-1160. Eisenstein, T.K., DeCueninck, B.J., Resavy, D., Shockman, G.D., Carey, R.B. and Swenson, R.M., 1983. Quantitative determination in human sera of vaccine-induced antibody to type-specific polysaccharides of group B streptococci using an enzyme-linked immunosorbent assay. J. Infect. Dis., 147: 847-856. Fischer, G.W., Lowell, G.H., Crumrine, M.H. and Bass, J.W., 1978. Demonstration of opsonic activity and in vivo protection against group B streptococci type III by streptococcus pneumoniae type 14 antisera. J. Exp. Med., 148: 776-786. Fischer, G.W., Hunter, K.W. and Wilson, S.R., 1981. Type III group B streptococcal strain differences in susceptibility to opsonization with human serum. Pediatr. Res., 15: 1525-1529. Fischer, G.W., Hunter, K.W. and Wilson, S.R., 1982. Modified human immune serum globulin for intravenous administration: in vitro opsonic activity and in vivo protection against group B streptococcal disease in suckling rats. Acta Paediatr. Scand., 71: 639-644. Fischer, G.W., Horton, R.E. and Edelman, R., 1983. Summary of the National Institutes of Health workshop on group B streptococcal infection. J. Infect. Dis., 148: 163-166. Fischer, G.W., Hemming, V.G., Hunter, K.W., Gloser, H., Bachmayer, H., Von Pilar, C.E., Helting, T., Weisman, L.E., Wilson, S.R. and Baron, P.A., 1986. Intravenous immunoglobulin in the treatment of neonatal sepsis: therapeutic strategies and laboratory studies. Pediatr. Infect. Dis., 5: $171-$175. Hemming, V.G., London, W.T., Curfman, B.L., Patrick, D.F. and Fischer, G.W., 1985. Maternal immunity and neonatal GBS infections: studies in a primate model. Antibiot. Chemother., 35: 194-200. Itoh, T., Yan, X., Nakano, H. and Yoshioka, M., 1986. Protective efficacy against group B streptococcal infection in neonatal mice delivered from preimmunized pregnants. Microbiol. Immunol., 30: 297-305, Kasper, D.L., Baker, C.J., Galdes, B., Katzenellenbogen, E. and Jennings, H.J., 1983. Immunochemical analysis and immunogenicity of the type II group B streptococcal capsular polysaccharide. J. Clin. Invest., 72: 260-269. Leslie, G.A. and Carwile, H.F., 1973. Immune response of rats to group A streptococcal vaccine, Infect. Immun., 7: 781-785. Mancini, G., Carbonara, A.O. and Heremans, J.F., 1965. Immunochemical quantitation of antigens by single radial immunodiffusion. Int. J. Immunochem., 2: 235-254. McCormick, J.B., Hermalino, H.G., Shigeo, N., Freire, J.B., Veras, J., Gorman, G., Teeley, T.C. and Wingo, P., 1980. Antibody response to serogroup A and C meningococcal polysaccharide vaccines in infants born to mothers vaccinated during pregnancy. J. Clin. Invest., 65:11411144. Rote, N.S., Taylor, N.L., Shigeoka, A.O., Scott, J.R. and Hill, H.R., 1980. Enzyme-linked immunosorbent assay for group B streptococcal antibodies. Infect. Immun., 27: 116-123. Shigeoka, A.O., Pincus, S.H., Rote, N.S. and Hill, H.R., 1984. Protective efficacy of hybridoma type-specific antibody against experimental infection with group B streptococcus. J. Infect. Dis., 149: 363-372. Siegel, J., 1985. Prevention and treatment of group B streptococcal infections. Pediatr. Infect. Dis., 4: $33-$36. Swenson, R.M., Eisenstein, T.K. and Shockman, G.D., 1985. The soluble antigens of group B streptococcus as human vaccines. Antibiot. Chemother., 35: 291-295. Turner, G.K., 1985. Measurement of light from chemical or biochemical reactions. In: K. Van Dyke (Editor), Bioluminescence and Chemiluminescence Instruments and Applications, Volume 1. C.R.C. Press Inc., Boca Raton, FL, pp. 47-51. Wilson, C.B., 1986. Immunologic basis for increased susceptibility of the neonate to infection. J. Pediatr., 108: 1-11. Yeung, C.Y. and Hobbs, J.R., 1968. Serum-gamma g-globulin levels in normal premature, postmature, and "small-for-dates" newborn babies. Lancet, 1: 1167-1172.