Protective effect of polyclonal and . monoclonal antibodies against abortion in mice infected by Chlamydia psittaci Dominique Buzoni-Gatei *t, Fran~oise Bernard*, Arthur Andersen t and Annie Rodolakis* The role of antibody in preventing placental and fetal infection by Chlamydia psittaci was studied in mice. Pregnant mice were passively immunized with polyclonal sera or monoclonal antibodies (mAbs) at day 11 of gestation. The mice were intravenously challenged the following day with the virulent AB7 ovine abortion strain of C. psittaci. Mice were either killed on day 16 of gestation to determine placental and fetal chlamydial infection levels or were permitted to have and raise their young until 8 days old for comparison o f survival rates. Immune sera produced a decrease in both placental and fetal infection andreduced the number o f young dying in utero or shortly after birth. Polyclonal sera to the highly invasive AB7 and A B4 strains or to the invasive 1B strain were more effective than serum to the invasive AB13 strain. The B577]F3 and B577]A11 monoclonal antibodies gave almost complete protection, with only low levels of placental infection and no detectable fetal infection or decrease in survival rate. The study demonstrates that immune sera and type-specific mAbs can passively transfer resistance to placental and fetal colonization and to abortion and fetal loss in mice intravenously challenged with C. psittaci. Keywords: Chlamydia psittaci; abortion; monoclonal antibodies; polyclonal antibodies; mice
Introduction The genus Chlamydia includes three species, Chlamydia trachomatis, C. psittaci and C. pneumoniae. C. psittaci infects a wide variety of birds and mammals including man. One of the most common chlamydial diseases of ruminants is ovine abortion, which results from colonization of the placenta by C. psittaci after a systemic infection. An assessment of host protective mechanisms may aid the identification of chlamydial protective epitopes and the development of a new vaccine which would be safer, as efficient as those presently existing 1 and would enable a serological distinction between vaccinated and infected animals. Mice have been shown to be very susceptible to chlamydial infection especially during gestation and a mouse model was therefore chosen for the study of protection. It has already been demonstrated, in a model of non-pregnant mice, that adoptive transfer of immune T cells confers immunity to mice intravenously challenged with C. psittaci 2. The role of humoral components in protection is more ambiguous. A serological increase in specific antibodies is often observed at the time of ovine abortion. Research investigating antibody protective activity has produced contradictory results 3'4. In passive transfer experiments using non-pregnant mice, it has been demonstrated that transfer of specific polyclonal sera *INRA, Station de Pathologie de la Reproduction, 37380 Nouzilly, France. tPathology Laboratory, National Animal Disease Center, USDA, ARS, MWA, Ames, Iowa 50010, USA. tTo whom correspondence should be addressed. (Received 24 October 1989; revised 12 December 1989; accepted 2 February 1990) 0264-410X/90/040342--05 © 1990 Butterworth-Heinemann Ltd
342 Vaccine, Vol. 8, August 1990
induces significant immunity 2 although this was lower than for the cellular protection. Moreover, in a preliminary study, it has been observed that pregnant mice passively immunized with polyclonal C. psittaci antiserum shortly before an intraper.itoneal challenge were fully protected against fetal loss. T h e aim of the present work was to expand on this preliminary result by comparing the protection afforded by transfer of different polyclonal or monoclonal sera in pregnant mice intravenously challenged. Therefore placental and fetal colonization ,were used as criteria for protection, as well as abortion and fetal loss.
Materials and methods C. psittaci strains All the strains of C. psittaci were propagated in the yolk sacs of developing chick embryos and stored at - 70°C. The origins and invasive properties of the strains, measured by the capacity to colonize the popliteal lymph node and the spleen of mice inoculated in the footpad 5 are shown in Table 1. The titration of chlamydial infectivity for McCoy cells was performed by plaque assay 6. Briefly, the titration was carried out on McCoy monolayers under a solid overlay medium containing 1.5% Noble agar. Counts of plaque forming units (p.f.u.) were made on day 14, after staining with 1:10 000 neutral red. Anti-C. psittaci sera Immune sera against six strains of chlamydial were produced using 6-7-week-old non-pregnant Swiss mice (IFFA Credo, L'Arbresle, France) by two subcutaneous injections separated by one month with 1 x 105 p.f.u, of
Humoral protection in C. psittaci infected mice: D. Buzoni-Gatel et al. Table 1
Origin and invasiveness of C. psittaci strains used for protection of polyclonal sera and monoclonal antibodies ELISA antibody titres
Strains
Origin
Invasiveness"
Polyclonal serum AB7 AB4 1El AB13 MO907 iB1 Control
Ovine abortion" Ovine abortion Mutant of AB7" Ovine abortion Faeces of healthy ewe (Storz, USA) Faeces of healthy ewe Normal yolk sac
2 2 1 1 1 0
1:800 1:800 1:800 1:400 1:800 < 1:30 < 1:30
Monoclonal antibodies b B577 (mAb B577/D8) B577 (mAb B577/F3) B577 (mAb B577/A11)
Ovine abortion (Storz, USA) Ovine abortion Ovine abortion
ND ND ND
< 1:30 < 1:30 < 1:30
"lnvasiveness expressed by: 0, non-invasive, 1, invasive, 2, highly invasive strains 5 bThe three monoclonal antibodies were derived from the same C. psittaci strain B577
live chlamydiae. Sera were collected 15 days after the second injection and antibody titres (Table I) were determined by an ELISA technique using purified chlamydiae from AB7 strain as antigen. The titre was colorimetrically determined as the last dilution of serum giving an absorbance value > 1, with the value 1 being the upper limit of the absorbance range (mean + 2 s.d.) obtained with control sera. Control sera were obtained from mice inoculated twice with non-infected yolk sac.
Monoclonal antibodies Three different mAbs in ascite fluids were used: B577/All, B577/F3 and B577/D8. They were derived from Balb/c mice immunized with the B577 strain 7 of C. psittaci which was isolated from an aborted ewe and was characterized 7 previously. Briefly, mAbs were tested by the indirect fluorescent antibody (IFA) technique for isolate specificity and by the inclusion reduction method for their ability to neutralize homologous strains. The three mAbs used were shown to be type specific since they failed to react at a dilution of 1:16 with isolates of heterologous serovars (C. psittaci strains isolated .from psittacosis, ornithosis, polyarthritis), whereas they did react with homologous serovar isolates (strain isolates from abortions) at a dilution of 1:1024 and at higher dilutions. All the mAbs showed neutralizing ability
in vitro. B577/D8 was of IgG3 B577/A11 were of IgG2b The three ascites fluids and the ELISA antibody
isotype whereas B577/F3 and isotype 7. were diluted (1:10) before use titres are given in Table I.
Experimental design Groups of 15 ten-week-old pregnant Swiss mice were injected intravenously (0.1 ml) with either immune serum (six groups), control serum (one group) or mAbs (three groups) at day 11 of gestation. One day later the mice were intravenously challenged with 2 x l0 s p.f.u, from the virulent AB7 strain 8. Four days after inoculation (at day 16 of gestation) five mice in each group were killed by cervical dislocation. Protection was assessed as the decrease in placental and fetal infectivity. The uterus was removed aseptically and the placental discs were dissected from the fetuses. All placental tissues from the same uterine horn were pooled, weighed and ground in 1 ml sterile Dulbecco phosphate-buffered saline; the resulting samples were titrated by the plaque assay method.
Fetuses from the same uterine horn were similarly pooled and processed. The ten remaining mice were allowed to give birth and the numbers of live baby mice were recorded daily for 8 days. Placental tissues and fetuses were considered as colonized when at least 1 p.f.u, per pool was isolated. Infection was expressed as the average log number of chlamydiae per uterine horn for the placental tissue or for the fetuses. The barrier effect of the placenta was estimated by the ratio (%) of the number of uterine horns with noncolonized fetuses to the number of uterine horns with colonized placentas (ratio R).
Statistics Placental and fetal infections were compared by analysis of variance (F test), taking uterine horns from the right side or from the left side as a block. Individual comparisons were performed using the t-test. Results were termed significant when p < 0.05.
Results
Protection induced by transfer of polyclonal sera In the first series of experiments, the protective abilities of four different polyclonal sera were compared. Protection was first assessed by the decrease in average placental or fetal colonization after serum transfer. A significant degree of protection (p /50%. In contrast, there was no difference between the control group or the anti-AB13 experimental group. Protection was also assessed by the number of live young at birth and the number of young surviving up to day 8 following birth. In the groups receiving anti-AB13 or control sera, eight out of ten aborted, leaving no live offspring. The groups receiving anti-AB7, anti-lB and
Vaccine, Vol. 8, August 1990 343
Humoral protection in C. psittaci infected mice: D. Buzoni-Gatel et al.
;8 F E
4
and iB1), isolated from the faeces of apparently healthy ewes, were inoculated into mice to prepare immune sera (Table 1). The anti-MO907 serum was shown to be protective as it reduced the placental colonization by approximately 1.0 log p.f.u, and lowered the fetal colonization by 2.0 log p.f.u. The anti-iB 1 serum was ineffective; there was no reduction in the colonization rates of either the placenta or the fetus.
c;
2
Protection induced by transfer of monoclonal antibodies
+
6
0J ¢o
:5 >. to u
z 03 o
a
b
c
d
e
Sera t r a n s f e r r e d
Figure 1
Placental (11) and fetal ([]) colonization after transfer of different polyclonal sera control, anti-AB13, anti-AB7, anti-AB4 and anti-lB. Protection was assessed by the decrea~se in the number of chlamydiae (log) in the placenta and in the fetuses of each uterine horn. a, Control; b, anti-AB13; c, anti-AB7; d, anti-AB4; e, anti-lB
The protective abilities of three different mAbs derived from mice immunized with the B577 strain were compared. Despite low ELISA titres (Table 1), two mAbs, B577/F3 and B577/A1 I, conferred a strong protection: placental infection was decreased by 5 log p.f.u, and fetal colonization was completely prevented (Figure 2). The number of live baby mice was less than in the non-inoculated group but almost all of them remained alive after birth, in contrast with the high mortality rate observed in the inoculated control group (Table 4). In contrast, the protective activity of the mAb B577/D8 was low (Figure 2 and Table 4).
Table 2
Frequency of placental and fetal infection after immune polyclonal and monoclonal serum transfer. A measure of the placental barrier effect
Discussion The aim of this study was to compare the protection conferred on pregnant mice following transfer of different
Uterine horns presenting =
Serum
Uterine horns observed
Infected placenta
Infected fetus
R" (%)
Normal Anti-AB13 Anti-AB7 Anti-lB Anti-AB4 B577/D3 B577/A11 B577/D8
10 9 10 10 10 10 10 10
10 9 10 10 10 10 10 10
9 8 5 3 4 0 0 6
10 11 50 70 60 100 100 40
8
+
Table 3
Number of live baby mice at birth and on day 8 after birth after transfer of polyclonal sera to mice Mean number of live baby mice/litter_+ s.d." Treatments
At birth
Non-i noculated 11.8 -t- 0.8 Transferred with: Anti-1B 10.7 __+1.4 Anti-AB7 7.2+_. 1.7 Anti-AB4 6 . 8 1.7 Anti-AB13 0.5__+0.5 Non-immune serum 0.5+0.4
6
0J to :>.
E
=All placentas or fetuses from the same uterine horn were pooled together " "R=number of horns with non-colonized fetuses/number of horns with colonized placentas
r
oJ
to u
c;
z
q
2
Oh O
0
a
On day 8 after birth 10.1 Jr 0.6 5.1 _+1.0 5.0_ 1.4 4.0__. 1.3 0.5-t-0.5 0
b
c
d
Monoclonal a n t i b o d i e s t r a n s f e r r e d
Figure 2
Placental and fetal infection after passive immunization with three different monoclonal antibodies. The protection was assessed by the decrease in placental ( 1 ) and fetal (~1) colonization of each uterine horn. No fetal colonization was observed after transfer of B577/F3 and B577/A11 mAbs. a, Control; b, B577/D8; c, B577/F3; d, B577/A11
Table 4 •Ten mothers were used in each experimental group
Number of live baby mice at birth and on day 8 after birth after transfer of monoclonal antibodies to mice Mean number of live baby mice/litter+ s.d."
anti-AB4 sera produced an average of seven or more live baby mice (Table 3). In the following days some of the baby mice died. However, the number of live baby mice remained significantly higher than in the groups receiving anti-AB13 or normal sera (at day 8 after birth; for anti-AB7, p
Introduction The genus Chlamydia includes three species, Chlamydia trachomatis, C. psittaci and C. pneumoniae. C. psittaci infects a wide variety of birds and mammals including man. One of the most common chlamydial diseases of ruminants is ovine abortion, which results from colonization of the placenta by C. psittaci after a systemic infection. An assessment of host protective mechanisms may aid the identification of chlamydial protective epitopes and the development of a new vaccine which would be safer, as efficient as those presently existing 1 and would enable a serological distinction between vaccinated and infected animals. Mice have been shown to be very susceptible to chlamydial infection especially during gestation and a mouse model was therefore chosen for the study of protection. It has already been demonstrated, in a model of non-pregnant mice, that adoptive transfer of immune T cells confers immunity to mice intravenously challenged with C. psittaci 2. The role of humoral components in protection is more ambiguous. A serological increase in specific antibodies is often observed at the time of ovine abortion. Research investigating antibody protective activity has produced contradictory results 3'4. In passive transfer experiments using non-pregnant mice, it has been demonstrated that transfer of specific polyclonal sera *INRA, Station de Pathologie de la Reproduction, 37380 Nouzilly, France. tPathology Laboratory, National Animal Disease Center, USDA, ARS, MWA, Ames, Iowa 50010, USA. tTo whom correspondence should be addressed. (Received 24 October 1989; revised 12 December 1989; accepted 2 February 1990) 0264-410X/90/040342--05 © 1990 Butterworth-Heinemann Ltd
342 Vaccine, Vol. 8, August 1990
induces significant immunity 2 although this was lower than for the cellular protection. Moreover, in a preliminary study, it has been observed that pregnant mice passively immunized with polyclonal C. psittaci antiserum shortly before an intraper.itoneal challenge were fully protected against fetal loss. T h e aim of the present work was to expand on this preliminary result by comparing the protection afforded by transfer of different polyclonal or monoclonal sera in pregnant mice intravenously challenged. Therefore placental and fetal colonization ,were used as criteria for protection, as well as abortion and fetal loss.
Materials and methods C. psittaci strains All the strains of C. psittaci were propagated in the yolk sacs of developing chick embryos and stored at - 70°C. The origins and invasive properties of the strains, measured by the capacity to colonize the popliteal lymph node and the spleen of mice inoculated in the footpad 5 are shown in Table 1. The titration of chlamydial infectivity for McCoy cells was performed by plaque assay 6. Briefly, the titration was carried out on McCoy monolayers under a solid overlay medium containing 1.5% Noble agar. Counts of plaque forming units (p.f.u.) were made on day 14, after staining with 1:10 000 neutral red. Anti-C. psittaci sera Immune sera against six strains of chlamydial were produced using 6-7-week-old non-pregnant Swiss mice (IFFA Credo, L'Arbresle, France) by two subcutaneous injections separated by one month with 1 x 105 p.f.u, of
Humoral protection in C. psittaci infected mice: D. Buzoni-Gatel et al. Table 1
Origin and invasiveness of C. psittaci strains used for protection of polyclonal sera and monoclonal antibodies ELISA antibody titres
Strains
Origin
Invasiveness"
Polyclonal serum AB7 AB4 1El AB13 MO907 iB1 Control
Ovine abortion" Ovine abortion Mutant of AB7" Ovine abortion Faeces of healthy ewe (Storz, USA) Faeces of healthy ewe Normal yolk sac
2 2 1 1 1 0
1:800 1:800 1:800 1:400 1:800 < 1:30 < 1:30
Monoclonal antibodies b B577 (mAb B577/D8) B577 (mAb B577/F3) B577 (mAb B577/A11)
Ovine abortion (Storz, USA) Ovine abortion Ovine abortion
ND ND ND
< 1:30 < 1:30 < 1:30
"lnvasiveness expressed by: 0, non-invasive, 1, invasive, 2, highly invasive strains 5 bThe three monoclonal antibodies were derived from the same C. psittaci strain B577
live chlamydiae. Sera were collected 15 days after the second injection and antibody titres (Table I) were determined by an ELISA technique using purified chlamydiae from AB7 strain as antigen. The titre was colorimetrically determined as the last dilution of serum giving an absorbance value > 1, with the value 1 being the upper limit of the absorbance range (mean + 2 s.d.) obtained with control sera. Control sera were obtained from mice inoculated twice with non-infected yolk sac.
Monoclonal antibodies Three different mAbs in ascite fluids were used: B577/All, B577/F3 and B577/D8. They were derived from Balb/c mice immunized with the B577 strain 7 of C. psittaci which was isolated from an aborted ewe and was characterized 7 previously. Briefly, mAbs were tested by the indirect fluorescent antibody (IFA) technique for isolate specificity and by the inclusion reduction method for their ability to neutralize homologous strains. The three mAbs used were shown to be type specific since they failed to react at a dilution of 1:16 with isolates of heterologous serovars (C. psittaci strains isolated .from psittacosis, ornithosis, polyarthritis), whereas they did react with homologous serovar isolates (strain isolates from abortions) at a dilution of 1:1024 and at higher dilutions. All the mAbs showed neutralizing ability
in vitro. B577/D8 was of IgG3 B577/A11 were of IgG2b The three ascites fluids and the ELISA antibody
isotype whereas B577/F3 and isotype 7. were diluted (1:10) before use titres are given in Table I.
Experimental design Groups of 15 ten-week-old pregnant Swiss mice were injected intravenously (0.1 ml) with either immune serum (six groups), control serum (one group) or mAbs (three groups) at day 11 of gestation. One day later the mice were intravenously challenged with 2 x l0 s p.f.u, from the virulent AB7 strain 8. Four days after inoculation (at day 16 of gestation) five mice in each group were killed by cervical dislocation. Protection was assessed as the decrease in placental and fetal infectivity. The uterus was removed aseptically and the placental discs were dissected from the fetuses. All placental tissues from the same uterine horn were pooled, weighed and ground in 1 ml sterile Dulbecco phosphate-buffered saline; the resulting samples were titrated by the plaque assay method.
Fetuses from the same uterine horn were similarly pooled and processed. The ten remaining mice were allowed to give birth and the numbers of live baby mice were recorded daily for 8 days. Placental tissues and fetuses were considered as colonized when at least 1 p.f.u, per pool was isolated. Infection was expressed as the average log number of chlamydiae per uterine horn for the placental tissue or for the fetuses. The barrier effect of the placenta was estimated by the ratio (%) of the number of uterine horns with noncolonized fetuses to the number of uterine horns with colonized placentas (ratio R).
Statistics Placental and fetal infections were compared by analysis of variance (F test), taking uterine horns from the right side or from the left side as a block. Individual comparisons were performed using the t-test. Results were termed significant when p < 0.05.
Results
Protection induced by transfer of polyclonal sera In the first series of experiments, the protective abilities of four different polyclonal sera were compared. Protection was first assessed by the decrease in average placental or fetal colonization after serum transfer. A significant degree of protection (p /50%. In contrast, there was no difference between the control group or the anti-AB13 experimental group. Protection was also assessed by the number of live young at birth and the number of young surviving up to day 8 following birth. In the groups receiving anti-AB13 or control sera, eight out of ten aborted, leaving no live offspring. The groups receiving anti-AB7, anti-lB and
Vaccine, Vol. 8, August 1990 343
Humoral protection in C. psittaci infected mice: D. Buzoni-Gatel et al.
;8 F E
4
and iB1), isolated from the faeces of apparently healthy ewes, were inoculated into mice to prepare immune sera (Table 1). The anti-MO907 serum was shown to be protective as it reduced the placental colonization by approximately 1.0 log p.f.u, and lowered the fetal colonization by 2.0 log p.f.u. The anti-iB 1 serum was ineffective; there was no reduction in the colonization rates of either the placenta or the fetus.
c;
2
Protection induced by transfer of monoclonal antibodies
+
6
0J ¢o
:5 >. to u
z 03 o
a
b
c
d
e
Sera t r a n s f e r r e d
Figure 1
Placental (11) and fetal ([]) colonization after transfer of different polyclonal sera control, anti-AB13, anti-AB7, anti-AB4 and anti-lB. Protection was assessed by the decrea~se in the number of chlamydiae (log) in the placenta and in the fetuses of each uterine horn. a, Control; b, anti-AB13; c, anti-AB7; d, anti-AB4; e, anti-lB
The protective abilities of three different mAbs derived from mice immunized with the B577 strain were compared. Despite low ELISA titres (Table 1), two mAbs, B577/F3 and B577/A1 I, conferred a strong protection: placental infection was decreased by 5 log p.f.u, and fetal colonization was completely prevented (Figure 2). The number of live baby mice was less than in the non-inoculated group but almost all of them remained alive after birth, in contrast with the high mortality rate observed in the inoculated control group (Table 4). In contrast, the protective activity of the mAb B577/D8 was low (Figure 2 and Table 4).
Table 2
Frequency of placental and fetal infection after immune polyclonal and monoclonal serum transfer. A measure of the placental barrier effect
Discussion The aim of this study was to compare the protection conferred on pregnant mice following transfer of different
Uterine horns presenting =
Serum
Uterine horns observed
Infected placenta
Infected fetus
R" (%)
Normal Anti-AB13 Anti-AB7 Anti-lB Anti-AB4 B577/D3 B577/A11 B577/D8
10 9 10 10 10 10 10 10
10 9 10 10 10 10 10 10
9 8 5 3 4 0 0 6
10 11 50 70 60 100 100 40
8
+
Table 3
Number of live baby mice at birth and on day 8 after birth after transfer of polyclonal sera to mice Mean number of live baby mice/litter_+ s.d." Treatments
At birth
Non-i noculated 11.8 -t- 0.8 Transferred with: Anti-1B 10.7 __+1.4 Anti-AB7 7.2+_. 1.7 Anti-AB4 6 . 8 1.7 Anti-AB13 0.5__+0.5 Non-immune serum 0.5+0.4
6
0J to :>.
E
=All placentas or fetuses from the same uterine horn were pooled together " "R=number of horns with non-colonized fetuses/number of horns with colonized placentas
r
oJ
to u
c;
z
q
2
Oh O
0
a
On day 8 after birth 10.1 Jr 0.6 5.1 _+1.0 5.0_ 1.4 4.0__. 1.3 0.5-t-0.5 0
b
c
d
Monoclonal a n t i b o d i e s t r a n s f e r r e d
Figure 2
Placental and fetal infection after passive immunization with three different monoclonal antibodies. The protection was assessed by the decrease in placental ( 1 ) and fetal (~1) colonization of each uterine horn. No fetal colonization was observed after transfer of B577/F3 and B577/A11 mAbs. a, Control; b, B577/D8; c, B577/F3; d, B577/A11
Table 4 •Ten mothers were used in each experimental group
Number of live baby mice at birth and on day 8 after birth after transfer of monoclonal antibodies to mice Mean number of live baby mice/litter+ s.d."
anti-AB4 sera produced an average of seven or more live baby mice (Table 3). In the following days some of the baby mice died. However, the number of live baby mice remained significantly higher than in the groups receiving anti-AB13 or normal sera (at day 8 after birth; for anti-AB7, p
