Alterations in Immunological Reactivity during Pregnancy in Mice Determined in vitro by Lymphoproliferation Tests

Immunobiol., vol. 175, pp. 236-244 (1987)

Department of Physiology and Immunology, Faculty of Medicine, University of Rijeka, Yugoslavia

Alterations in Immunological Reactivity during Pregnancy in Mice Determined in vitro by Lymphoproliferation Tests MILJENKO DORIC and DANIEL RUKAVINA Received February 24,1987' Accepted in Revised Form June 1, 1987

Abstract Primiparous and multiparous mice, either syngeneically or allogeneically pregnant, were sacrificed at various stages of pregnancy, and the immune reactivity of cells from different lymphatic organs was analyzed by mixed lymphocyte reaction (MLR) and concanavalin A (Con A)-induced lymphocyte proliferation. In the MLR, spleen cells and cells from the axillary lymph nodes of pregnant animals showed similar changes in their response to allogeneic cells during the course of gestation. In comparison to age-matched virgin controls they had an increased reactivity between the 7th and 11 th day of pregnancy. During the preimplantation period and the last week of pregnancy, their alloreactivity was comparable to that of the controls. Cells from the para-aortic lymph nodes, which drain the uterus, also showed an increased reactivity at mid-gestation, but in the preimplantation period and third week of pregnancy their alloreactivity was even decreased in comparison to the control animals. When T cell immunocompetence was measured by means of lymphocyte transformation induced with Con A, the pattern of reactivity was completely different. Lymphoid cells from all the above-mentioned tissues showed the highest response to Con A during thepreimplantation period, constantly weakening towards term.

Introduction Mating in an outbred population, or between two inbred strains, results in an intrauterine fetus that bears a variety of paternally derived antigens against which the maternal immune system can respond. However, the fetus develops most successfully throughout gestation showing no evidence of being immunologically prejudiced. Many theories have been advanced to explain the unusual acceptance of the fetal allograft by the maternal immune system (1, 2). Among these, in the last decade, evidence supporting a pregnancy-associated modulation of the maternal immune response has been demonstrated using the mixed lymphocyte reaction (3), macrophage migration inhibition assay (4), mitogen-induced transformation (5), local graft versus host reaction (6, 7) as well as by comparing the effects of Abbreviations: MLR

= Mixed lymphocyte reaction;

Con A

= concanavalin A.

Immune Response in Pregnant Mice . 237

pregnancy serum and normal serum on cellular reactivity monitored using similar techniques (8, 9). Although these studies support the concept of both specific and nonspecific suppression of maternal cellular activity, contradictory results are frequent, and it is impossible to conclude whether these observed results are of any relevance to fetal survival. The present investigation was designed to further examine whether alterations of the immune response exist in the pregnant mice and if possible, to explain some of the mentioned contradictory results. A major objection that concerns most of the above-mentioned papers is that the responsiveness of pregnant mice was analyzed during different stages of pregnancy and by using lymphoid cells from different lymphatic organs. Therefore, we considered it necessary to perform a kinetic study of cellular reactivities of both primiparous and multiparous females, in syngeneic as well as allogeneic pregnancies from fertilization to term. Furthermore, the cellular reactivities were analyzed at the local (draining lymph nodes) and systemic level (spleen and axillary lymph nodes).

Materials and Methods Mice CBA (H-2k) and AI] (H-2') mice were bred in our animal quarters. Female CBA mice were used at the age of 2-2.5 months (primiparous) or 4.5-5 months (multiparous). The stage of pregnancy was assessed by daily examination for vaginal plugs in mated females. Multiparous mice were sacrificed during their third pregnancy.

Lymphocyte preparations Spleen cell suspensions were obtained by gently passing the splenic pulp through a metallic sieve in RPM I medium. Para-aortic (draining lymph nodes) and axillary lymph nodes were excised, disrupted with two needles and sieved. Cell suspensions were counted on a hemacytometer and adjusted to the appropriate cell concentration in RPM I medium.

Reactivity of lymphocytes to allogeneic cells and Con A To determine the reactivity of lymphocytes to allogeneic cells,S X lOS responding lymphoid cells were incubated for 120 h with the same number of mitomycin C-treated stimulator cells in 0.25 ml RPMI medium supplemented with 10 % horse serum,S X 10- 5 M 2-mercaptoethanol, 25 X 10- 3 M HEPES, penicillin (100 IU/ml) and streptomycin (50 !-lg/ml). The incubation was performed at 37°C in a humidified atmosphere of 5 % CO 2 and 95 % air. Eighteen h before harvest, 1 !-lCi 3H-thymidine was added to each well and the incorporation measured in a liquid scintillation counter. The reactivity of lymphocytes to ConA was determined by incubating 5 X 105 lymphoid cells in 0.25 ml RPM I medium with 10 !-lg/ml of Con A (Pharmacia Fine Chemicals). The medium was supplemented with the same contents as for the MLR and incubated under the same conditions for 72 h. Eighteen h prior to harvest, 3H-thymidine was added.

Statistical analysis Values obtained in the experimental groups were compared to those of the control groups by the Student's t-test.

238 . MILJENKO DORIC and DANIEL RUKAVINA

Results

Mixed lymphocyte responses We first established both in the MLR and in the response to Con A that the kinetics were not changed by the origin of cells (from virgin or pregnant CBA mice). The mixed lymphocyte responses of spleen, axillary and para-aortic lymph node cells were followed in pregnant mice from fertilization to term. Three stages of pregnancy were found to be very interesting and are shown in the figures. They included 1-3 days, 7-11 days and 15-19 days of pregnancy. Between the 7th and 11th day of pregnancy, the response of lymphoid cells from all the analyzed lymphatic tissues was higher in comparison to other stages of pregnancy, as well as in comparison to age-matched virgin controls. During this period, the response of spleen cells (Fig. 1a) was significantly higher, in comparison to that of virgin mice, in the groups of syngeneically pregnant primiparous, syngeneically pregnant multiparous and allogeneically pregnant primiparous mice. The response of spleen cells during the preimplantation period as well as between 15-19 days of pregnancy was not significantly different in comparison to that of virgin mice. When the response of axillary lymph node cells (Fig. 1b) was analyzed, similar results were obtained. The MLR significantly differed in comparison to the control animals, only during the mid-pregnancy period in syngeneically pregnant primiparous and multiparous mice. Such a comparison made at the level of the para-aortic lymph nodes (Fig. 1c) also gave a significantly higher response between 7-11 days of pregnancy in syngeneically pregnant primiparous, allogeneically pregnant primiparous and allogeneically pregnant multiparous mice. However, besides the above-mentioned differences, it is also interesting to point out that the response of cells from the draining lymph nodes was significantly decreased in comparison to the control mice, in all the studied groups between 15-19 days of pregnancy. During the preimplantation period, a significantly decreased response was also obtained in primiparous versus .. . vIrgm mlCe.

Figure 1. a) Response of spleen cells from pregnant CBA mice mixed in culture with mitomycin C-treated AI] cells. b) Response of axillary lymph node cells from pregnant CBA mice mixed in culture with mitomycin C-treated AI] cells. c) Response of para-aortic lymph node cells from pregnant CBA mice mixed in culture with mitomycin C-treated AI] cells. Curves represent the modifications, in comparison to the reactivity of virgin mice, of syngeneically pregnant (CBA x CBA) primiparous ( - - ) and multiparous mice (---) and allogeneically pregnant (CBA x AI]) primiparous (_._._) and multiparous mice (.. "'). Each point represents a group of 6-9 animals. Significance versus virgin controls: x = p < 0.05, xx = P < 0.01, xxx = P < 0.001. ~

Immune Response in Pregnant Mice . 239

~!!; z

Q

!:i(

~

~

~ttto~~~tEttQ11ttttttb~~Jj~11~~~tt~~~~SE) 1

a

1-)

7_11 15-19 STAGE OF PREGNANCY (DAYS)

10

I

I II..

I

I

I

I

/

I

/

I

/

I

\

\

\

\

\

\

\

\

\

\

\

\ IRGIH (MEAN!SE)

b

1_)

7-11 15-19 STAGE OF PREGNANCY (DAYS)

10

c 1_)

7-11 15-19 STAGE OF PREGNANCY (DAYS)

. MILJENKO DORIC and DANIEL RUKAVINA

240

20

k \

15

>< 0

\

w

~

\.

:z:

0

10

~::> 1:

~

a I- 3

7 - tt

15-19

STAGE OF PREGNANCY (DAYS)

600

k\ \

y. \.

400

g

r··. .......\ \

.....\

~

\"\".......

~

!ii

~200th~tthHTtthhH~yy~++if~~-~-hW~~~+~~,RGI

to

b 1-3

7-tt 15-19 STAGE OF PREGNANCY (DAYS)

I-J

7-tt 15-19 STAGE OF PREGNANCY (DAYS)

100

><

UJ

~ 50

c

Immune Response in Pregnant Mice . 241

Response of lymphoid cells to Con A At the systemic level, a striking difference was seen when the response to Con A was compared to the above-described alterations in alloreactivity. Both spleen and axillary lymph node cells showed the highest response to Con A during the preimplantation period, and the response thereafter nearly linearly decreased towards term. During the first three days of pregnancy, the response of spleen cells (Fig. 2a) was significantly higher, in comparison to virgin mice, only in the group of allogeneically pregnant primiparous mice. In the mid-pregnancy period, the response was significantly decreased, in comparison to virgin mice, in all the analyzed groups of pregnant animals. Between 15-19 days of pregnancy, the response of spleen cells from pregnant mice was still lower. The difference was highly significant in all the experimental groups when compared to virgin controls. The axillary lymph node cells (Fig. 2b) showed similar changes in the response to Con A during the course of gestation as did spleen cells. However, at the level of the draining lymph nodes (Fig. 2c), the response was different, being already lower during the preimplantation period than in the control group. The decrease was statistically significant in syngeneically pregnant primiparous mice during the preimplantation period and in all the groups between 7-11 days and 15-19 days of pregnancy.

Discussion Several points have been demonstrated by this study of cellular immunocompetence in pregnant mice. First, pregnancy alters the in vitro responsiveness to allogeneic cells as well as to the T cell mitogen Con A. Second, the alterations strictly depend on the stage of pregnancy. During the preimplantation period, alloreactivity seems to be decreased at the level of the draining lymph nodes, while there seems to be no difference, in comparison to age-matched controls, at the systemic level. Between days 7-11 of pregnancy, alloreactivity is highly increased in all the studied groups, while between days 15-19 of pregnancy the reactivity towards allogeneic cells is again similar to that described during the preimplantation period. These changes in alloreactivity, during the course of gestation, are very similar to those found in our earlier experiments using the in vivo local Figure 2. a) Response of spleen cells from pregnant CBA mice to Con A. b) Response of axillary lymph node cells from pregnant CBA mice to ConA. c) Response of para-aortic lymph node cells from pregnant CBA. Curves represent the modifications, in comparison to the reactivity of virgin mice, of syngeneically pregnant (CBA x CBA) primiparous ( - - ) and multiparous mice (---) and allogeneic ally pregnant (CBA x AI]) primiparous (_._._) and multiparous mice (- .... ). Each point represents a group of 6-9 animals. Significance versus virgin controls: x= p
242 . MILJENKO DORIC and DANIEL RUKAVINA

GVHR assay. Third, the reactivity to Con A also depends on the stage of pregnancy, but shows a completely different pattern in comparison to changes in alloreactivity. During the preimplantation period, the responsiveness of spleen and axillary lymph node cells to Con A is higher than that which was obtained in virgin mice. However, the responsiveness of these cells is strongly decreased in the mid-pregnancy period and further decreases towards term. The responsiveness of para-aortic lymph node cells is already lower than in control mice 1-3 days afte fertilization but, similar as at the systemic level, continues to decrease during the course of gestation. Fourth, differences in the responsiveness of syngeneically and allogeneically pregnant mice were not statistically significant. Therefore, the mentioned alterations in alloreactivity should be considered non-specific, since the genotype of the paternal strain had no significant effect on the response of cells from pregnant mice. Finally, some authors have demonstrated specific depression of the response towards paternal strain alloantigens in multiparous mice (10, 11). Our results are not in accordance with these findings, and the changes in the responsiveness of multiparous mice were found to be non-specific as well. It is interesting to point out that data from literature concerning the immune responsiveness of pregnant mice are often contradictory. Some investigators found a markedly suppressed response in pregnant mice (6, 10, 11), others did not find significant alterations (8, 12, 13), while in some experiments an increased responsiveness was obtained (7, 14). However, even though these findings may seem confusing, this study, as well as our earlier results obtained in experiments in vivo (7), clearly shows that the immune responsiveness of pregnant mice primarily depends on the stage of pregnancy. If the stage is taken into account, then the majority of the results of the above-mentioned authors (stated as contradictory) are in agreement with our results demonstrating changes in reactivity during the course of gestation. Furthermore, data from literature seem to prove that the same or very similar alterations in the responsiveness during pregnancy exist in the rat as well (15). A possible explanation for the increased reactivity in the MLR experiments, between 7-11 days of pregnancy, could be the response towards embryonal, gametic and oncofetal antigens (16, 17). A specific response towards paternal strain alloantigens was absent or had only a minor effect on the responsiveness studied during pregnancy, since a significant difference did not occur between the response of cells obtained from syngeneically or allogeneically pregnant mice. Lymphoid cells from lS-19-day-pregnant mice responded to allogeneic cells much weaker than during the mid-pregnancy period. The response of draining lymph node cells was even significantly suppressed in comparison to the virgin controls. These results can be expained by the influence of many immunosuppressive factors released at this stage of pregnancy,

Immune Response in Pregnant Mice . 243

mainly from the feto-placental unit, therefore most markedly affecting local lymphoid tissue. Such candidates are: 1) corticosterone, whose highest level is found on the 16th day of pregnancy in the mouse (18), 2) pregnancyassociated and pregnancy-specific proteins (19, 20), 3) placental hormones (21), 4) immunosuppressive factors released by decidual cells (22), 5) immunosuppressive factors released by trophoblast cells (2) and 6) other placental extracts and fractions (23). Finally, it is essential to point out that the above-mentioned changes in alloreactivity were completely different to those found in the response to Con A. It seems that the cellular response, in general, decreases nearly linearly from fertilization to term, while only alloreactive T cells seem to be stimulated by undefined antigen(s), giving a peak response between 7-11 days of pregnancy. Acknowledgements The authors wish to acknowledge the excellent technical help from Mrs. M. KASTELA and Mrs. N. KRUZIC, and the financial support from the Republic of Croatia Research Fund (SIZ V), Zagreb.

References 1. BEER, A. E., and R. E. BILLINGHAM. 1980. Mechanisms of non-rejection of foeto-placental allografts. Fo!. Bio!. (Praha) 26: 225. 2. CHAOUAT, P., J. P. KOLB, and T. G. WEGMANN. 1983. The murine placenta as an immunological barrier between the mother and the fetus. Immuno!. Rev. 75: 31. 3. CHAOUAT, G., and G. A. VOISIN. 1979. Regulatory T cell subpopulations in pregnancy. I. Evidence for suppressive activity in the early phase of MLR. J. Immuno!. 122: 1383. 4. RABSON, A. R., M. C. BEY, J. E. KERRICH, and H. J. HOORNO. 1976. The blocking by autologous serum of maternal cell mediated immune reaction to placental antigens. s. A. Med. J. 50: 201. 5. BURRELLS, c., P. W. WELLS, and A. D. SUTHERLAND. 1978. Reactivity of ovine lymphocytes to phytohaemagglutinin and pokeweed mitogen during pregnancy and in the immediate post-parturient period. Clin. expo Immuno!. 33: 410. 6. NICKLIN,S., and W. D. BILLINGTON. 1982. Impairment of graft versus host reactivity in pregnant mice. Clin. expo Immuno!. 49: 135. 7. DORIC, M., R. G. KINSKY, and G. A. VOISIN. 1984. Allogeneic reactivity of maternal lymphoid cells during the course of gestation. Modifications and sex differences in a local GVH assay. J. Reprod. Immuno!. 6: 187. 8. HARRISON, M. R. 1976. Maternal immunocompetence. I. The graft-versus-host reactivity of lymphocytes from pregnant rats and the distribution pattern of 51Cr-labelled lymphocytes in pregnant mice. Scand. J. Immuno!. 5: 549. 9. SMITH, G. 1978. Inhibition of cell-mediated microcytotoxicity and stimulation of mixed lymphocyte reactivity by mouse pregnancy serum. Transplantation 26: 278. 10. CHAOUAT, G., and G. A. VOISIN. 1980. Regulatory T-cell subpopulations in pregnancy. II. Evidence for suppressive activity of the late phase of MLR. Immunology 39: 239. 11. CHAOUAT, G., P. MONNOT, M. HOFFMANN, and G. A. VOISIN. 1982. Regulatory T-cells in pregnancy. VI. Evidence for T-cell mediated suppression to CTL generation toward paternal alloantigens. Cel!. Immuno!. 68: 322.

244 . MILJENKO DORIC and DANIEL RUKAVINA 12. WEGMANN, T. G., C. A. WATERS, D. W. DRELL, and G. A. CARLSON. 1979. Pregnant mice are not primed but can be primed to fetal alloantigens. Proc. Nat!. Acad. Sci. USA 76: 2410. 13. PAVIA, C. S., and D. P. STITES. 1979. Humoral and cellular regulation of alloimmunity in pregnancy. J. Immunol. 123: 2194. 14. BONAPARTE, Y. P., L. COLOMBO, S. KLEIN, S. OISBOLD-DAGA, and L. KAPLANIKONICOFF. 1982. Modifications de la reaction greffon-contre-h6te (RGCH) des cellules lymphoides maternelle au cours de l'evolution de la gestation allogenique. C. R. Acad. Sci. Ser. D. 294: 577. 15. BAUMINGER, S., and S. PELEG. 1978. Changes in immunological activity of rat lymphoid organs during pregnancy. Clin. expo Immunol. 32: 179. 16. GHANI, A. M., T.]. III. GILL, H. W. KUNZ, and D. N. MISRA. 1984. Elicitation of the maternal antibody response to the fetus by a broadly shared MHC class I antigenic determinant. Transplantation 37: 187. 17. GHANI, A. M., H. W. KUNZ, and T.]. III. GILL. 1984. Pregnancy induced monoclonal antibody to a unique fetal antigen. Transplantation 37: 503. 18. BARLOW, S. M., P. J. MORRISON, and F. M. SULLIVAN. 1973. Plasma corticosterone levels during pregnancy in the mouse. Brit.]. Pharmacol. 48: 346. 19. BOHN, H., R. ]OHANNSEN, and W. KRAUS. 1980. Neues Plazentaprotein (PP15) mit immunsuppressiver Wirkung. Arch. Gynecol. 230: 167. 20. BOHN, H., W. KRAUS, and W. WINCKLER. 1982. New soluble placental tissue proteins: Their isolation, characterization, localization and quantification. Placenta 4 (suppl.): 67. 21. SUTERI, P. K., and D. P. STITES. 1982. Immunologic and endocrine interrelationships in pregnancy. BioI. Reprod. 26: 1. 22. SLAPSYS, R. M., and D. A. CLARK. 1982. Active suppression of host vs graft reaction in pregnant mice: IV. Local suppressor cells in decidua and uterine blood. ]. Reprod. Immunol. 4: 355. 23. BOBE, P., M. DORIC, R. G. KINSKY, and G. A. VOISIN. 1984. Modulation of mouse antiSRBC antibody response by placental extracts. Cell. Immunol. 89: 355. Dr. M. DORIC, Department of Physiology and Immunology, Faculty of Medicine, University of Rijeka, Olge Ban 22, 51000 Rijeka, Yugoslavia