Eur. J. Immunol. 1990. 20: 2149-2152
Maternal effect on autoimmunity
2149
Short paper Ronald F. Van Vollenhovenov, G. Jeanette ThorbeckeO and Gregory W. Siskind Division of Allergy and Immunology, Department of Medicine, Cornell University Medical College and Department of Pathologyo, New York University School of Medicine, New York
Exposure of female mice to type I1 collagen reduces susceptibility to collagen-induced arthritis in offspring* The effect of exposure of female DBA/1 mice to collagen11 (CII) prior to breeding on the susceptibility of their offspring to CII-induced arthritis (CIA) was investigated. It was found that female offspring, born within 3 months after exposure of the mothers to CII, had a significantly reduced incidence of CIA, following immunization with CII. Just prior to this immunization, no anti-CII could be detected in the offspring. Offspring born more than 3 months after exposure of the mothers t o CII showed no differences in susceptibility to induction of CIA, if optimal conditions for induction were used. However, when suboptimal conditions for induction of CIA were used, offspring of females that had been exposed to CII developed less severe arthritis and had a delayed onset of arthritis as compared with controls. It is concluded that exposure of female mice to CII prior to mating results in changes in the immune response to CII in the offspring, leading to a subtle decrease in susceptibility t o CIA.
1 Introduction Exposure of neonatal animals to antigen usually results in tolerance to the antigen [l-31. However, depending on various factors including genetic background, type of antigen, dosage and route of administration, such exposure can result in immunity [4-61. Similarly, administration of antibody or anti-idiotypic antibody during the neonatal period may have long-lasting effects on the immune response to the corresponding antigen. For example, increased production of antibody of the same idiotype as that administered neonatally has been reported [7, 81. Keamey et al. [8] identified a network of idiotypes and anti-idiotypes in neonatal BALB/c mice which was presumably encoded in the germ line, and which was readily modulated by exposure to idiotypic or anti-idiotypic antibody. Such factors were felt to play an important role in shaping the adult B cell repertoire. Thus, it has become clear that neonatal mice are exquisitely sensitive to administration of minute quantities of anti-idiotypic antibody, which, depending upon details of experimental design, can lead to augmentation [9, 101 or suppression [ll]of subsequent production, in response to antigen, of antibody bearing the corresponding idiotype.
[I 76641
*
This work was supported in part by grants from the National Institutes of Health U.S.P.H.S. AG 04980 and A1 11694. Recipient of a Research Fellowship the Charles H. Revson Foundation. Current address: Department of Medicine, Stanford University, Stanford, CA 94305, USA.
Correspondence: George W. Siskind, Cornell University Medical College, Department of Medicine, 1300 York Avenue, New York, NY 10021, USA Abbreviations: CIA: Collagen-induced arthritis CII: Type I1 collagen 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990
Findings such as those described above can be regarded as reflecting an experimental equivalent of the role which the transfer of maternal antigen, antibody or anti-idiotypic antibody through the placenta or via colostrum might play in the normal ontogeny of the immune response. One would predict that previous immunologic experience of female animals might influence the immunologic responses of their progeny. Weiler et al. [12] showed that immunization of female SJL mice with a major idiotype of antidextran antibodies resulted in decreased immune responsiveness to dextran in offspring born up to 5 months after immunization. Similar results were obtained in a different system by BernabC et al. [13]. Gorczynski et al. [14] demonstrated decreased immune mediated rejection of allogeneic skin grafts in the progeny of animals previously tolerized to cells with the haplotype of the skin graft. Treatment of the mother so as to modify an immune response which is linked to the induction of an experimental autoimmune disease in progeny has not been reported. In DBA/1 mice, an immune response to collagen type I1 (CII) is linked to the development of arthritis [15]. In the present study, we demonstrate decreased incidence and severity of collagen-induced arthritis (CIA) and delayed onset of arthritis in offspring of female mice that had been exposed to CII.
2 Materials and methods 2.1 Animals
Six- to ten-week-old DBA/1 mice were purchased from Jackson Laboratories (Bar Harbor, ME). For breeding purposes, two or three female DBA/1 mice were housed with one DBA/1 male.When a female was visibly pregnant she was moved to an individual cage. Five weeks after birth the young were separated according to gender and the mother was again allowed to mate.
+
0014-2980/90/0909-2149$3.50 .25/0
2150
Eur. J. Immunol. 1990.20: 2149-2152
R.Van Vollenhoven, G. J. Thorbecke and G. W. Siskind
2.2 Collagen Chick collagen type I1 (CII) was purchased from Genzyme Corp. (Boston, MA) and stored at -70°C. It was dissolved at 1mg/ml in 0.01 M acetic acid with gentle stirring at 4 "C.
in PBS; the plates were incubated for 90 min at 37°C and washed. p-Nitrophenyl phosphate (Sigma) was added as enzyme substrate and after color development the absorbance at 405 nm was determined. Dilutions of affinitypurified anti-CII antibodies were included on each plate as a standard, allowing expression of the results in pg anti-CII antibodiedml of serum.
2.3 Induction and assessment of arthritis 2.5 Statistical analysis CFA was prepared by mixing pulverized, lyophilized, heat-killed Mycobacreria (strains C, DTand PN, Ministry of Agriculture, Fisheries and Food, Weybridge, Surrey, GB) into IFA (Difco Labs., Detroit, MI) at 8 mg/ml. CII at 1 mg/ml was emulsified with an equal volume of CFA, and 0.2 ml of emulsion was injected i.d. divided over four sites on the back. All mice were boosted on day 28 after primary immunization, with the exception of those females that were used for breeding after the primary immunization. Boosting was by i.p. injection of 100 pg CII in aqueous solution. All animals were inspected for arthritis 2 or 3 times each week. Distal joint swelling was quantitated by measuring the thickness of the foot and the width of the ankle with a constant tension caliper (Dyer Co., Lancaster, PA). Each paw was scored on a scale from 0 to 2: 0 = no sign of inflammation, 1= mild swelling and erythema, 2 = pronounced swelling and/or ankylosis. Mean arthritis severity for a group of animals was calculated as the sum of the highest scores observed for each animal (that is a maximum of 2 for each paw x 4 paws or 8 per animal) divided by the number of arthritic animals. The mean arthritis index used in Fig. 1was calculated as the sum of the scores at a given time point divided by the total number of animals in the group. 2.4 Anti-CII antibody levels Serum anti-CII levels were assayed using an ELISA. Immulon II plates (Dynatech Labs., Alexandria,VA) were coated with CII (0.01 mg/ml, 100 pVwel1) dissolved in PBS (0.15 M NaCl, 0.01 M phosphate buffer, pH 7.2), at 4°C overnight. After washing with 0.1% Tween 20 (Sigma Chemical Co., St. Louis, MO) in PBS (PBS-Tween), blocking with 2% BSA (Miles Scientific, Naperville, IL) in PBS for 90 min at 37 "C, and washing, serial dilutions of the sera were added. The plates were incubated for 90 min at 37°C and were washed. Alkaline phosphatase (Sigma) conjugated goat anti-mouse IgG (Cappel, Cochranville, PA) was added at a previously determined optimal dilution
Comparison of arthritis incidence was done by two-tailed Fisher exact test; means of arthritis severity and arthritis index were analyzed by the nonparametric Wilcoxon rank-sum test; day of onset of arthritis and serum anti-CII levels were analyzed by Student's r-test.
3 Results 3.1 Effect of exposure of female mice to CII on the incidence of CIA in their offspring in response to an optimal CIA induction procedure Groups of 6-8-week-old female DBA/1 mice were treated in one of three ways: (a) immunization with 100 pg CII in CFA i d . , a procedure shown previously to optimally induce CIA; (b) tolerization with 100 pg CII in aqueous solution i.v., a procedure shown previously to render DBA/1 mice unresponsive to CII and resistant to induction of CIA, followed after 1 week by immunization with 100 pg CII in CFA i.d.; and (c) solvent (0.01 M acetic acid) i.v. followed after 1 week by 0.01 M acetic acid emulsified with CFA i.d. These mice were then mated. Since sex-related differences have been described in CIA, male and female progeny were used simultaneously but analyzed separately. As expected the female parents which had been immunized with CII i.d. (without i.v. pretreatment) all developed arthritis but this did not appear to interfere with reproduction. Control mice or mice injected with CII i.v. before immunization with CII i.d. did not develop arthritis. Female mice born within 3 months after the mothers had been injected with CII i.d. with or without a prior i.v. injection had a significantly lower incidence of CIA than did offspring of control mothers (Table 1). In contrast, male off spring showed no significant influence of maternal exposure to CII, and neither male nor female mice, born more than three months after CII treatment of mothers, demonstrated a significantly reduced CIA incidence under these challenge conditions, although the offspring from
Table 1. Incidence of CIA in progeny born < 3 months after CII injections of female mice@ Treatment of
mothers Control CII i.d. CII i.v. i.d.
+
Arthritis indicence in pro- Serum anti-CII antibodies in geny (n) female progeny (pg/ml? SD Female (n) Male (n)
(8) 77% (13) 33%b) (9) 92% (12)
100%
40%C)(10)
80%
(5)
2615 k 1752 1894 f 1231 1255f 626
a) Female DBAll mice were either immunized with 100 pg CII in CFA i.d., or tolerized with 100 pg CII in acqueous solution i.v. followed after 1 week by CII in CFA i.c. Control mothers received CFA alone. After immunization mice were mated with untreated syngeneic mice and the progeny were injected with 1OOpg CII in CFA i.d. at age 8 to 12 weeks.The mice were observed for at least 90 days to assess the incidence of CIA. Normal untreated male DBA/1 mice exhibit an indicence of 82% (n = 16).
b) Statistically significantly different from control group by Fisher exact test (p < 0.01). c) Statistically significantly different from control group by Fisher exact test (p < 0.02).
Eur. J. Immunol. 1990.20: 2149-2152
Maternal effect on autoimmunity
mice only immunized with CII i.d. did show a slight reduction (data not shown). Anti-CII serum levels were determined by ELISA and are presented in Table 1.The offspring of CII-treated mothers tended to have somewhat lower anti-CII antibody titers although the differences were not significant. No anti-CII antibody was detected (< 10 pg/ml) in sera obtained from the offspring prior to challenge with CII.
x
2151
3.2 Effect of exposure of female mice to CII on the development of CIA in their offspring in response to a suboptimal CIA induction procedure
Male progeny, born 6 to 8 months after treatment of the mothers, were subjected to a regimen previously shown to yield an approximately 50% incidence of CIA. This treatment consisted of 25 pg CII in aqueous solution i.v., followed one week later by 100 pg CII in CFA i.d. As shown in Table 2, the control mice did, in fact, have a 50% incidence of CIA. Although the incidence of CIA was unaltered by exposure of the mothers to CII, the offspring from CII-immunized or -tolerized/challenged mice exhibited a significantly decreased average severity of arthritis and a delayed onset of arthritis. The time course of CIA severity is illustrated in Fig. 1. It is seen that offspring of CII-treated mice show a significantly reduced mean arthritis index as compared to control mice, following this suboptimal challenge procedure. As indicated in Table 2 the difference in seventy and onset of arthritis was not associated with any change in anti-CII antibody titer in the arthritic animals. Anti-CII antibody levels were also assessed in sera of the progeny before challenge with CII; no anti-CII antibody was detected (< 10 pg/ml).
4 Discussion TIME
AFTER
IMMUNIZATION
( DAYS )
Figure 1. Course of CIA following suboptimal induction in male progeny of female mice previously exposed to CII. Female DBAll mice were either immunized with 100 pg CII with CFA i.d., or tolerized with 100 pg CII in aqueous solution i.v. followed after 1 week by CII in CFA i.d. After immunization they were mated with untreated syngeneic mice. Male progeny born 6 to 8 months after treatment of the mothers were injected at age 8 to 12 weeks with 25 pg CII in aqueous solution i.v. followed after 1week by 100 pg CII in CFA i.d. Arthritis severity was expressed for each entire group of mice using the mean arthritis index (max. = 8) which takes into account both severity and incidence as described in Sect. 2.3.Treatment of maternal mice: (a) no treatment; (b) CII i.v.. i.d.; (c) CII i.d. Data presented as mean k SD. * Statistically significantly different from control group, evaluated by Wilcoxon rank-sum test (p < 0.05).
+
The present results show that injection of female mice with CII prior to conception modifies the susceptibility of their offspring to induction of CIA. Such offspring, after immunization with CII, have a reduced incidence of CIA, and/or reduced severity and delayed onset of arthritis. These effects were observed regardless of whether or not the mothers had arthritis. Based upon previous studies [ 161 we would expect females immunized with CII in CFA to have 500-1500 pg/ml of anti-CII antibodies, while mice given 100 pg CII i.v. prior to i.d. immunization have about one-third that level. It is thus reasonable to assume that offspring of such mice were exposed to anti-CII antibodies of maternal origin in urero and/or via colostrum. It is less likely that they were exposed to significant amounts of CII, since most of the litters were conceived > 2 weeks after the last injection of CII, and effects could still be demonstrated
Table 2. Incidence, seventy and time of onset of CIA following a suboptimal CIA induction procedure in male progeny of female mice which had been previously exposed to CIIa)
Treatment of mothers
n
Control CII i.d. CII i.v. i.d.
12 8
+
11
Incidence of CIA
50%
50% 45%
Mean arthritis
Average day of
severity
onset
Anti-CII antibodies in arthritic mice (kg/ml* SD)
32.0f 5.3 45.5 7.W) 38.0 f 3.3c)
1443f 970 1370 & 1227 1469 f 1476
5.7k 1.2 3.0 & 1.2b) 3.2f l.lb)
*
Female DBA/1 mice were either immunized with 100 pg CII in CFA i.d., or tolerized with 100 pg CII in acqueous solution i.v. followed after 1 week by CII in CFA i.d. After immunization mice were mated with untreated syngeneic mice. Male progeny born 6 to 8 months after treatment of the mothers were injected at age 8 to 12 weeks with 25 pg CII in aqueous solution i.v. followed after 1week by 100 pg CII in CFA i.d.The mice were observed for at least 90 days to assess the incidence and severity of CIA. Arthritis severity was quantified for those animals that developed arthritis using the mean arthritis severity index (max. = 8 or 2 per paw X 4 paws) as described in Sect. 2.3. Data are presented as mean f SD. Statistically significantly different from control group by Wilcoxon rank-sum test (p < 0.05). Statistically significantly different from control group by Student’s t-test 0,< 0.04).
2152
Eur. J. Immunol. 1990. 20: 2149-2152
R. Van Vollenhoven, G. J. Thorbecke and G. W. Siskind
in offspring conceived > 3 months after antigen injection. Offspring born early after exposure of the mothers to CII showed a resistance to induction of CIA even when an optimal CII immunization procedure was used. This effect was seen only in female mice. Although in our hands female animals are equally susceptible to CIA as males, some authors who used CII of a bovine source found a gender difference ,female animals being less susceptible than males [15].Thus, it is possible that a subtle difference exists, and that a very small decrease in susceptibility to CIA induction might be more readily detected in female mice. However, when a CIA induction procedure which yielded disease in only 50% of control mice was used, a reduced severity and delayed onset of arthritis could be demonstrated even in male offspring of pretreated females. Assuming that antibody transfer from mother to offspring is the basis for the effects observed, it is unlikely that antibody levels in the young are sufficiently high to directly block an immune response 6 t o 8 weeks after birth when mice were immunized with CII. Rather an effect of these antibodies on the developing idiotypic network of the newborn can be reasonably postulated. Influences on the B cell idiotype network resulting from exposure of neonates to idiotypes or anti-idiotypes have been documented [7-11,171; however, the effects reported in this report were not associated with signficiant changes in anti-CII antibody titers. An alternative hypothesis could be that changes inT cell-mediated immune responsiveness to CII are involved, but such responses were not assessed in these experiments. Acquisition of the immunological repertoire has generally been seen as taking place at the level of the individual. Observations such as those reported here suggest that longitudinal transmission of immunological information
may take place, and that such information may therefore be a property of the species rather than of merely the individual. Received May 5, 1989; in revised form May 2, 1990.
5 References 1 Wells, H. G. and Osborne,T. B., J. Infect. Dis. 1911. 8: 66. 2 Owen, R. D., Nature 1945. 102: 400. 3 Billingham, R. E., Brent, L. and Medewar, €? B., Nature 1953. 112: 602. 4 Thorbecke, G. J., Siskind, G. W. and Goldberger, N. J., Immunology 1961. 87: 147. 5 Mitchison, N. A., Immunology 1962. 5: 341. 6 Siskind, G. W., Paterson, P. Y. and Thomas, L., J. Immunol. 1963. 90: 929. 7 Rubinstein, L. J.,Yeh, M. and Bona, C. A., J. Exp. Med. 1982. 156: 506. 8 Keamey, J. F. and Vakil, M., Immunol. Rev. 1986. 94: 39. 9 Rubinstein, L. J., Goldberg, B., Hiernaux, J., Stein, K. E. and Bona, C. A., J. Exp. Med. 1983. 158: 1129. J. Exp. Med. 1981. 10 Hiernaux, J., Bona, C. A. and Baker, €?, 153: 1004. 11 Kelsoe, G., Reth, M. and Rajewsky, K., Immunol. Rev. 1980. 52: 75. 12 Weiler, I. J.,Weiler, E., Sprenger, R. and Cosenza, H., Eur. J. Immunol. 1977. 7: 591. 13 BemabC, R. R., Coutinho, A., Cazenave, €?-A.and Forni, L., Proc. Natl. Acad. Sci. USA 1981. 78: 6416. 14 Gorczynski, R. M., Kennedy, M., McRae, S. and Ciampi, A., J. lmmunol. 1983. 131: 1115. 15 Courtenay, J. S., Dallman, M. J., Dayan, A. D., Martin, A. and Mosedale, B., Nature 1980. 283: 666. 16 Van Vollenhoven, R. F., Nagler-Anderson, C., Soriano, A., Siskind, G.W. and Thorbecke, G. J., Cell. Immunol. 1988.115: 146. 17 Stein, K. E. and Soderstrom, T., J. Exp. Med. 1984. 160: 1001.
Maternal effect on autoimmunity
2149
Short paper Ronald F. Van Vollenhovenov, G. Jeanette ThorbeckeO and Gregory W. Siskind Division of Allergy and Immunology, Department of Medicine, Cornell University Medical College and Department of Pathologyo, New York University School of Medicine, New York
Exposure of female mice to type I1 collagen reduces susceptibility to collagen-induced arthritis in offspring* The effect of exposure of female DBA/1 mice to collagen11 (CII) prior to breeding on the susceptibility of their offspring to CII-induced arthritis (CIA) was investigated. It was found that female offspring, born within 3 months after exposure of the mothers to CII, had a significantly reduced incidence of CIA, following immunization with CII. Just prior to this immunization, no anti-CII could be detected in the offspring. Offspring born more than 3 months after exposure of the mothers t o CII showed no differences in susceptibility to induction of CIA, if optimal conditions for induction were used. However, when suboptimal conditions for induction of CIA were used, offspring of females that had been exposed to CII developed less severe arthritis and had a delayed onset of arthritis as compared with controls. It is concluded that exposure of female mice to CII prior to mating results in changes in the immune response to CII in the offspring, leading to a subtle decrease in susceptibility t o CIA.
1 Introduction Exposure of neonatal animals to antigen usually results in tolerance to the antigen [l-31. However, depending on various factors including genetic background, type of antigen, dosage and route of administration, such exposure can result in immunity [4-61. Similarly, administration of antibody or anti-idiotypic antibody during the neonatal period may have long-lasting effects on the immune response to the corresponding antigen. For example, increased production of antibody of the same idiotype as that administered neonatally has been reported [7, 81. Keamey et al. [8] identified a network of idiotypes and anti-idiotypes in neonatal BALB/c mice which was presumably encoded in the germ line, and which was readily modulated by exposure to idiotypic or anti-idiotypic antibody. Such factors were felt to play an important role in shaping the adult B cell repertoire. Thus, it has become clear that neonatal mice are exquisitely sensitive to administration of minute quantities of anti-idiotypic antibody, which, depending upon details of experimental design, can lead to augmentation [9, 101 or suppression [ll]of subsequent production, in response to antigen, of antibody bearing the corresponding idiotype.
[I 76641
*
This work was supported in part by grants from the National Institutes of Health U.S.P.H.S. AG 04980 and A1 11694. Recipient of a Research Fellowship the Charles H. Revson Foundation. Current address: Department of Medicine, Stanford University, Stanford, CA 94305, USA.
Correspondence: George W. Siskind, Cornell University Medical College, Department of Medicine, 1300 York Avenue, New York, NY 10021, USA Abbreviations: CIA: Collagen-induced arthritis CII: Type I1 collagen 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1990
Findings such as those described above can be regarded as reflecting an experimental equivalent of the role which the transfer of maternal antigen, antibody or anti-idiotypic antibody through the placenta or via colostrum might play in the normal ontogeny of the immune response. One would predict that previous immunologic experience of female animals might influence the immunologic responses of their progeny. Weiler et al. [12] showed that immunization of female SJL mice with a major idiotype of antidextran antibodies resulted in decreased immune responsiveness to dextran in offspring born up to 5 months after immunization. Similar results were obtained in a different system by BernabC et al. [13]. Gorczynski et al. [14] demonstrated decreased immune mediated rejection of allogeneic skin grafts in the progeny of animals previously tolerized to cells with the haplotype of the skin graft. Treatment of the mother so as to modify an immune response which is linked to the induction of an experimental autoimmune disease in progeny has not been reported. In DBA/1 mice, an immune response to collagen type I1 (CII) is linked to the development of arthritis [15]. In the present study, we demonstrate decreased incidence and severity of collagen-induced arthritis (CIA) and delayed onset of arthritis in offspring of female mice that had been exposed to CII.
2 Materials and methods 2.1 Animals
Six- to ten-week-old DBA/1 mice were purchased from Jackson Laboratories (Bar Harbor, ME). For breeding purposes, two or three female DBA/1 mice were housed with one DBA/1 male.When a female was visibly pregnant she was moved to an individual cage. Five weeks after birth the young were separated according to gender and the mother was again allowed to mate.
+
0014-2980/90/0909-2149$3.50 .25/0
2150
Eur. J. Immunol. 1990.20: 2149-2152
R.Van Vollenhoven, G. J. Thorbecke and G. W. Siskind
2.2 Collagen Chick collagen type I1 (CII) was purchased from Genzyme Corp. (Boston, MA) and stored at -70°C. It was dissolved at 1mg/ml in 0.01 M acetic acid with gentle stirring at 4 "C.
in PBS; the plates were incubated for 90 min at 37°C and washed. p-Nitrophenyl phosphate (Sigma) was added as enzyme substrate and after color development the absorbance at 405 nm was determined. Dilutions of affinitypurified anti-CII antibodies were included on each plate as a standard, allowing expression of the results in pg anti-CII antibodiedml of serum.
2.3 Induction and assessment of arthritis 2.5 Statistical analysis CFA was prepared by mixing pulverized, lyophilized, heat-killed Mycobacreria (strains C, DTand PN, Ministry of Agriculture, Fisheries and Food, Weybridge, Surrey, GB) into IFA (Difco Labs., Detroit, MI) at 8 mg/ml. CII at 1 mg/ml was emulsified with an equal volume of CFA, and 0.2 ml of emulsion was injected i.d. divided over four sites on the back. All mice were boosted on day 28 after primary immunization, with the exception of those females that were used for breeding after the primary immunization. Boosting was by i.p. injection of 100 pg CII in aqueous solution. All animals were inspected for arthritis 2 or 3 times each week. Distal joint swelling was quantitated by measuring the thickness of the foot and the width of the ankle with a constant tension caliper (Dyer Co., Lancaster, PA). Each paw was scored on a scale from 0 to 2: 0 = no sign of inflammation, 1= mild swelling and erythema, 2 = pronounced swelling and/or ankylosis. Mean arthritis severity for a group of animals was calculated as the sum of the highest scores observed for each animal (that is a maximum of 2 for each paw x 4 paws or 8 per animal) divided by the number of arthritic animals. The mean arthritis index used in Fig. 1was calculated as the sum of the scores at a given time point divided by the total number of animals in the group. 2.4 Anti-CII antibody levels Serum anti-CII levels were assayed using an ELISA. Immulon II plates (Dynatech Labs., Alexandria,VA) were coated with CII (0.01 mg/ml, 100 pVwel1) dissolved in PBS (0.15 M NaCl, 0.01 M phosphate buffer, pH 7.2), at 4°C overnight. After washing with 0.1% Tween 20 (Sigma Chemical Co., St. Louis, MO) in PBS (PBS-Tween), blocking with 2% BSA (Miles Scientific, Naperville, IL) in PBS for 90 min at 37 "C, and washing, serial dilutions of the sera were added. The plates were incubated for 90 min at 37°C and were washed. Alkaline phosphatase (Sigma) conjugated goat anti-mouse IgG (Cappel, Cochranville, PA) was added at a previously determined optimal dilution
Comparison of arthritis incidence was done by two-tailed Fisher exact test; means of arthritis severity and arthritis index were analyzed by the nonparametric Wilcoxon rank-sum test; day of onset of arthritis and serum anti-CII levels were analyzed by Student's r-test.
3 Results 3.1 Effect of exposure of female mice to CII on the incidence of CIA in their offspring in response to an optimal CIA induction procedure Groups of 6-8-week-old female DBA/1 mice were treated in one of three ways: (a) immunization with 100 pg CII in CFA i d . , a procedure shown previously to optimally induce CIA; (b) tolerization with 100 pg CII in aqueous solution i.v., a procedure shown previously to render DBA/1 mice unresponsive to CII and resistant to induction of CIA, followed after 1 week by immunization with 100 pg CII in CFA i.d.; and (c) solvent (0.01 M acetic acid) i.v. followed after 1 week by 0.01 M acetic acid emulsified with CFA i.d. These mice were then mated. Since sex-related differences have been described in CIA, male and female progeny were used simultaneously but analyzed separately. As expected the female parents which had been immunized with CII i.d. (without i.v. pretreatment) all developed arthritis but this did not appear to interfere with reproduction. Control mice or mice injected with CII i.v. before immunization with CII i.d. did not develop arthritis. Female mice born within 3 months after the mothers had been injected with CII i.d. with or without a prior i.v. injection had a significantly lower incidence of CIA than did offspring of control mothers (Table 1). In contrast, male off spring showed no significant influence of maternal exposure to CII, and neither male nor female mice, born more than three months after CII treatment of mothers, demonstrated a significantly reduced CIA incidence under these challenge conditions, although the offspring from
Table 1. Incidence of CIA in progeny born < 3 months after CII injections of female mice@ Treatment of
mothers Control CII i.d. CII i.v. i.d.
+
Arthritis indicence in pro- Serum anti-CII antibodies in geny (n) female progeny (pg/ml? SD Female (n) Male (n)
(8) 77% (13) 33%b) (9) 92% (12)
100%
40%C)(10)
80%
(5)
2615 k 1752 1894 f 1231 1255f 626
a) Female DBAll mice were either immunized with 100 pg CII in CFA i.d., or tolerized with 100 pg CII in acqueous solution i.v. followed after 1 week by CII in CFA i.c. Control mothers received CFA alone. After immunization mice were mated with untreated syngeneic mice and the progeny were injected with 1OOpg CII in CFA i.d. at age 8 to 12 weeks.The mice were observed for at least 90 days to assess the incidence of CIA. Normal untreated male DBA/1 mice exhibit an indicence of 82% (n = 16).
b) Statistically significantly different from control group by Fisher exact test (p < 0.01). c) Statistically significantly different from control group by Fisher exact test (p < 0.02).
Eur. J. Immunol. 1990.20: 2149-2152
Maternal effect on autoimmunity
mice only immunized with CII i.d. did show a slight reduction (data not shown). Anti-CII serum levels were determined by ELISA and are presented in Table 1.The offspring of CII-treated mothers tended to have somewhat lower anti-CII antibody titers although the differences were not significant. No anti-CII antibody was detected (< 10 pg/ml) in sera obtained from the offspring prior to challenge with CII.
x
2151
3.2 Effect of exposure of female mice to CII on the development of CIA in their offspring in response to a suboptimal CIA induction procedure
Male progeny, born 6 to 8 months after treatment of the mothers, were subjected to a regimen previously shown to yield an approximately 50% incidence of CIA. This treatment consisted of 25 pg CII in aqueous solution i.v., followed one week later by 100 pg CII in CFA i.d. As shown in Table 2, the control mice did, in fact, have a 50% incidence of CIA. Although the incidence of CIA was unaltered by exposure of the mothers to CII, the offspring from CII-immunized or -tolerized/challenged mice exhibited a significantly decreased average severity of arthritis and a delayed onset of arthritis. The time course of CIA severity is illustrated in Fig. 1. It is seen that offspring of CII-treated mice show a significantly reduced mean arthritis index as compared to control mice, following this suboptimal challenge procedure. As indicated in Table 2 the difference in seventy and onset of arthritis was not associated with any change in anti-CII antibody titer in the arthritic animals. Anti-CII antibody levels were also assessed in sera of the progeny before challenge with CII; no anti-CII antibody was detected (< 10 pg/ml).
4 Discussion TIME
AFTER
IMMUNIZATION
( DAYS )
Figure 1. Course of CIA following suboptimal induction in male progeny of female mice previously exposed to CII. Female DBAll mice were either immunized with 100 pg CII with CFA i.d., or tolerized with 100 pg CII in aqueous solution i.v. followed after 1 week by CII in CFA i.d. After immunization they were mated with untreated syngeneic mice. Male progeny born 6 to 8 months after treatment of the mothers were injected at age 8 to 12 weeks with 25 pg CII in aqueous solution i.v. followed after 1week by 100 pg CII in CFA i.d. Arthritis severity was expressed for each entire group of mice using the mean arthritis index (max. = 8) which takes into account both severity and incidence as described in Sect. 2.3.Treatment of maternal mice: (a) no treatment; (b) CII i.v.. i.d.; (c) CII i.d. Data presented as mean k SD. * Statistically significantly different from control group, evaluated by Wilcoxon rank-sum test (p < 0.05).
+
The present results show that injection of female mice with CII prior to conception modifies the susceptibility of their offspring to induction of CIA. Such offspring, after immunization with CII, have a reduced incidence of CIA, and/or reduced severity and delayed onset of arthritis. These effects were observed regardless of whether or not the mothers had arthritis. Based upon previous studies [ 161 we would expect females immunized with CII in CFA to have 500-1500 pg/ml of anti-CII antibodies, while mice given 100 pg CII i.v. prior to i.d. immunization have about one-third that level. It is thus reasonable to assume that offspring of such mice were exposed to anti-CII antibodies of maternal origin in urero and/or via colostrum. It is less likely that they were exposed to significant amounts of CII, since most of the litters were conceived > 2 weeks after the last injection of CII, and effects could still be demonstrated
Table 2. Incidence, seventy and time of onset of CIA following a suboptimal CIA induction procedure in male progeny of female mice which had been previously exposed to CIIa)
Treatment of mothers
n
Control CII i.d. CII i.v. i.d.
12 8
+
11
Incidence of CIA
50%
50% 45%
Mean arthritis
Average day of
severity
onset
Anti-CII antibodies in arthritic mice (kg/ml* SD)
32.0f 5.3 45.5 7.W) 38.0 f 3.3c)
1443f 970 1370 & 1227 1469 f 1476
5.7k 1.2 3.0 & 1.2b) 3.2f l.lb)
*
Female DBA/1 mice were either immunized with 100 pg CII in CFA i.d., or tolerized with 100 pg CII in acqueous solution i.v. followed after 1 week by CII in CFA i.d. After immunization mice were mated with untreated syngeneic mice. Male progeny born 6 to 8 months after treatment of the mothers were injected at age 8 to 12 weeks with 25 pg CII in aqueous solution i.v. followed after 1week by 100 pg CII in CFA i.d.The mice were observed for at least 90 days to assess the incidence and severity of CIA. Arthritis severity was quantified for those animals that developed arthritis using the mean arthritis severity index (max. = 8 or 2 per paw X 4 paws) as described in Sect. 2.3. Data are presented as mean f SD. Statistically significantly different from control group by Wilcoxon rank-sum test (p < 0.05). Statistically significantly different from control group by Student’s t-test 0,< 0.04).
2152
Eur. J. Immunol. 1990. 20: 2149-2152
R. Van Vollenhoven, G. J. Thorbecke and G. W. Siskind
in offspring conceived > 3 months after antigen injection. Offspring born early after exposure of the mothers to CII showed a resistance to induction of CIA even when an optimal CII immunization procedure was used. This effect was seen only in female mice. Although in our hands female animals are equally susceptible to CIA as males, some authors who used CII of a bovine source found a gender difference ,female animals being less susceptible than males [15].Thus, it is possible that a subtle difference exists, and that a very small decrease in susceptibility to CIA induction might be more readily detected in female mice. However, when a CIA induction procedure which yielded disease in only 50% of control mice was used, a reduced severity and delayed onset of arthritis could be demonstrated even in male offspring of pretreated females. Assuming that antibody transfer from mother to offspring is the basis for the effects observed, it is unlikely that antibody levels in the young are sufficiently high to directly block an immune response 6 t o 8 weeks after birth when mice were immunized with CII. Rather an effect of these antibodies on the developing idiotypic network of the newborn can be reasonably postulated. Influences on the B cell idiotype network resulting from exposure of neonates to idiotypes or anti-idiotypes have been documented [7-11,171; however, the effects reported in this report were not associated with signficiant changes in anti-CII antibody titers. An alternative hypothesis could be that changes inT cell-mediated immune responsiveness to CII are involved, but such responses were not assessed in these experiments. Acquisition of the immunological repertoire has generally been seen as taking place at the level of the individual. Observations such as those reported here suggest that longitudinal transmission of immunological information
may take place, and that such information may therefore be a property of the species rather than of merely the individual. Received May 5, 1989; in revised form May 2, 1990.
5 References 1 Wells, H. G. and Osborne,T. B., J. Infect. Dis. 1911. 8: 66. 2 Owen, R. D., Nature 1945. 102: 400. 3 Billingham, R. E., Brent, L. and Medewar, €? B., Nature 1953. 112: 602. 4 Thorbecke, G. J., Siskind, G. W. and Goldberger, N. J., Immunology 1961. 87: 147. 5 Mitchison, N. A., Immunology 1962. 5: 341. 6 Siskind, G. W., Paterson, P. Y. and Thomas, L., J. Immunol. 1963. 90: 929. 7 Rubinstein, L. J.,Yeh, M. and Bona, C. A., J. Exp. Med. 1982. 156: 506. 8 Keamey, J. F. and Vakil, M., Immunol. Rev. 1986. 94: 39. 9 Rubinstein, L. J., Goldberg, B., Hiernaux, J., Stein, K. E. and Bona, C. A., J. Exp. Med. 1983. 158: 1129. J. Exp. Med. 1981. 10 Hiernaux, J., Bona, C. A. and Baker, €?, 153: 1004. 11 Kelsoe, G., Reth, M. and Rajewsky, K., Immunol. Rev. 1980. 52: 75. 12 Weiler, I. J.,Weiler, E., Sprenger, R. and Cosenza, H., Eur. J. Immunol. 1977. 7: 591. 13 BemabC, R. R., Coutinho, A., Cazenave, €?-A.and Forni, L., Proc. Natl. Acad. Sci. USA 1981. 78: 6416. 14 Gorczynski, R. M., Kennedy, M., McRae, S. and Ciampi, A., J. lmmunol. 1983. 131: 1115. 15 Courtenay, J. S., Dallman, M. J., Dayan, A. D., Martin, A. and Mosedale, B., Nature 1980. 283: 666. 16 Van Vollenhoven, R. F., Nagler-Anderson, C., Soriano, A., Siskind, G.W. and Thorbecke, G. J., Cell. Immunol. 1988.115: 146. 17 Stein, K. E. and Soderstrom, T., J. Exp. Med. 1984. 160: 1001.
