Kniss et al.
28. Teitelbaum I. The epidermal growth factor receptor is coupled to a phospholipase A2-specific pertussis toxininhibitable guanine nucleotide-binding regulatory protein in cultured rat inner medullary collecting tubule cells. J Bioi Chern 1990;265:4218-22. 29. Zick Y, Sagi-Eisenberg R, Pines M, Gierschik P, and Spie-
December 1990 Am J Obstet Gynecol
gel AM. Multisite phosphorylation of the a subunit of transducin by the insulin receptor kinase and protein kinase C. Proc Nat! Acad Sci USA 1986;83:9294-7. 30. Crouch M, Lapetina E. A role for G j in control of thrombin receptor-phospholipase C coupling in human platelets. J Bioi Chern 1988;263:3363-71.
Lymphokine production during term human pregnancy: Differences between peripheral leukocytes and decidual cells Donald J. Dudley, MD, Murray D. Mitchell, DPhil, Kathryn Creighton, BS, and D. Ware Branch, MD Salt Lake City, Utah Although the fetus is considered to be an "allograft," there is little information concerning the role of Iymphokines in human pregnancy. Lymphokines are polypeptides secreted by stimulated lymphocytes that direct the immune response by enabling immune effector cells to communicate with each other. To characterize Iymphokine production during normal human pregnancy, we isolated peripheral leukocytes and decidual lymphocyte-like cells from women undergoing repeat cesarean section at term. After stimulation with mitogen and paternal antigen for 24 hours, culture supernatants were assayed for granulocyte-macrophage colony-stimulating factor and interleukin-2 by enzyme-linked immunosorbent assay. There was no difference in the amount of interleukin-2 produced by stimulated peripheral and decidual cells. However, granulocyte-macrophage colony-stimulating factor production by stimulated decidual lymphocyte-like cells was significantly greater than granulocyte-macrophage colony-stimulating factor produced by peripheral lymphocytes. Decidual lymphocyte-like cells produced granulocyte-macrophage colony-stimulating factor both spontaneously and after stimulation with mitogen or paternal antigen, whereas peripheral leukocytes did not. This suggests that the decidua constitutes a distinct immunologic microenvironment. (AM J OBSTET GVNECOL 1990;163:1890-3.)
Key words: Pregnancy, lymphokines, granulocyte-macrophage colony-stimulating factor, interleukin-2 Our knowledge of the immunologic relationship between the mother and conceptus is incomplete. The conceptus appears to elicit a specific maternal immune response consisting of antibody production' and changes in maternal cellular immunity. This is detected as diminished proliferation of maternal peripheral blood leukocytes to irradiated paternal peripheral blood leukocytes in a mixed leukocyte culture! However, some investigators believe that the important cellular immune responses to fetopaternal antigens occur in the decidua at the maternal-fetal interface.' It is cur-
From the Department of Obstetrics and Gynecology, University of Utah School of Medicine. Supported by a grant from the Willard L. Eccles Charitable Foundation. Presented at the Thirty-seventh Annual Meeting of the Society for Gynecologic Investigation, St. Louis, Missouri, March 21-24, 1990. Reprint requests: Donald J. Dudley, MD, Department of Obstetrics and Gynecology, University of Utah School of Medicine, 50 N. Medical Dr., Salt Lake City, UT 84132. 616124761
1890
rently thought that these decidual immune responses are beneficial to the survival of the conceptus" Cellular immune responses are characterized by the production of a variety of polypeptides known' as lymphokines. Little information concerning the role of lymphokines in protecting the fetal allograft is available. Moreover, few studies have addressed the function of lymphoid cells in the human decidua. The purpose of our study was to characterize lymphokine production during normal human pregnancy, comparing the production of granulocyte-macrophage colonystimulating factor and interleukine-2 by peripheral leukocytes with that of decidual lymphocyte-like cells. We hypothesized that the decidua constitutes a distinct immunologic environment that supports immune effector cells that produce lymphokines different from those produced by peripheral leukocytes.
Material and methods Placentas were obtained aseptically from normal human pregnancies at term repeat cesarean section. The
Volume 163 l\:umber 6. Pan 1
membranes were isolated and rinsed in Dulbecco's modified Eagle's medium. A modification of the method described by Delvin et aI.' was used to isolate decidual lymphocyte-like cells. The decidua adherent to the maternal side of the chorion laevae was removed by sharp dissection to yield a cell homogenate. To disperse the cells, the homogenate was mixed with 63 units of collagenase per milliliter of Dulbecco's modified Eagle's medium (100 ml total volume) at 37° C for 60 to 120 minutes. The cell homogenate was filtered through a gauze mesh to remove large debris. The cells then were centrifuged over Histopaque 1077 (Sigma, St. Louis) at 1000 g for 30 minutes to isolate the lymphocyte-like cell fraction. This fraction was aspirated and rinsed. The culture medium used throughout these experiments was RPMI with 1% NutridomaHV (Boehringer-Mannheim, Indianapolis) and antibiotics (100 Vlml penicillin G sodium, 100 Vlml streptomycin sulfate, and 0.25 f-tg/ml amphotericin B, Gibco Laboratories, Grand Island, N.Y.). Heparinized peripheral blood was collected from the prospective mother and father just before the cesarean section. Peripheral blood leukocytes were isolated by standard techniques (Sigma Diagnostics, procedure 1077). The blood was diluted 1 : 1 with culture medium and then centrifuged over Histopaque 1077 at 1000 g for 30 minutes. The buffy coat was aspirated and rinsed. Separate cell suspensions of maternal peripheral blood leukocytes and decidual cells were adjusted to 10 6 cells/ml and duplicate 1 ml suspensions of each were placed in 24-well tissue culture plates. The cells were stimulated with 10 f-tg/ml of concanavalin A and incubated for 24 hours at 37° C. Supernatants were collected and stored at - 20° C until assayed. To evaluate the maternal peripheral lymphocyte and the decidual lymphocyte-like cell response to paternal antigen, 5 x 10 5 maternal peripheral blood leukocytes or decidual cells were coincubated with a similar number of irradiated (2000 rads) paternal cells for 24 hours. V nstimulated control cultures were established for both cell populations in all experiments. Culture supernatants were collected and stored at - 20° C until assayed. Supernatants were assayed for granulocyte-macrophage colony-stimulating factor and interleukin-2 with commercially available enzyme-linked immunosorbent assay kits (Genzyme, Boston). The limits of sensitivity for the interleukin-2 enzyme-linked immunosorbent assay range from 0.3 ng/ml to 20.8 ng/ml. The intraassay coefficient of variation ranges from 2% to 6% and the interassay coefficient of variation is approximately 4%. For the granulocyte-macrophage colonystimulating factor enzyme-linked immunosorbent assay, the limits of sensitivity range from 0.008 ng/ml to 1.0 ngl ml. The intraassay coefficient of variation
Lymphokine production during term human pregnancy
1891
JfJ p-O.056
.45
I
.40
.35
I
~
.30 .25
.20 .16 .10
.06
P
D
CONTROL
PDP D CONCANAVAI,JII A PATERNAL 8'1M.l.A110N
Fig. 1. Granulocyte-macrophage colony-stimulating factor production by maternal peripheral leukocytes (P) and deciduallymphocyte-like cells (D) in unstimulated, concanavalin Astimulated, and paternal antigen-stimulated conditions (n = 8). The hatched line delineates the lower limit of sensitivity of the assay. All values are mean ± SEM.
ranges from 2% to 8%, whereas the interassay coefficient of variation ranges from 5% to 10%. The data were evaluated statistically with the MannWhitney V test. Results
Eight term pregnancies were evaluated. Granulocytemacrophage colony-stimulating factor production by maternal peripheral blood leukocytes and decidual lymphocyte-like cells is depicted in Fig. 1. Concanavalin A-stimulated maternal peripheral blood leukocytes produced low levels of granulocytemacrophage colony-stimulating factor (0.018 ± 0.005 ng/ml, mean ± SEM). However, decidual lymphocytelike cells produced much greater amounts of granulocyte-macrophage colony-stimulating factor after stimulation with concanavalin A (0.2 ± 0.059 ng/ml, p = 0.006). Stimulation with paternal peripheral blood leukocytes also resulted in greater production of granulocyte-macrophage colony-stimulating factor by decidual cells (0.441 ± 0.262 ng/ml, p = 0.055 when compared with peripheral cells stimulated with paternal peripheral blood leukocytes). Fig. 1 also suggests that the decidual cells produce granulocyte-macrophage colony-stimulating factor constitutively at low, but significant, levels (0.073 ± 0.038 ng/mI, p = 0.01 when compared with unstimulated maternal peripheral cells). To evaluate whether this might reflect nonspecific stimulation caused by cell processing or culture, we performed a time course experiment with concanavalin A-stimulated and unsti-
1892
Dudley et al.
December 1990
Am
1.0
+
0.9 0.8 0.7
~ ~
~
~
0.6 0.5 0.4 0.3
Coneanavaln A
0.2
/
0.1 0
lmtimuIated
L-J~~~~-",-~=-----------I-----------r--
0
12
6
18
24
Tme (hru's)
Fig. 2. Granulocyte-macrophage colony-stimulating factor production by unstimulated and concanavalin A-stimulated decidual lymphocyte-like cells over a 24-hour time period (n = 3, mean ± SE). (The standard error at the asterisk is 0.283.)
.90 p.O.083
.80
II
.70 .80
1!
~
.:. C\I
.50 .40
.30 .20
.10 P
0
CONTROL
P
0
CONCANAVAltl II
Fig. 3. Interieukin-2 production by unstimulated and concanavalin A-stimulated maternal peripheral leukocytes (P) and decidual lymphocyte-like cells (D) (n = 8). The hatched line delineates the lower limit of sensitivity of the assay. All values are mean ± SEM.
mulated decidual cells from three different placentas. Culture supernatants were obtained at the start of the experiment and then every 6 hours for 24 hours. The results are shown in Fig. 2. Granulocyte-macrophage colony-stimulating factor was undetectable in the cell suspensions before stimulation (time 0). However, low levels of granulocyte-macrophage colony-stimulating factor were produced by unstimulated decidual cells over 24 hours. This may represent nonspecific acti-
J Obstet Gynecol
vation of the cells. Stimulation with concanavalin A resulted in significant production of granulocytemacrophage colony-stimulating factor after 6 hours of incubation, indicating that the apparent nonspecific activation contributes minimally to mitogen-induced granulocyte-macrophage colony-stimulating factor production. There was no significant difference in interleukin-2 production by maternal peripheral leukocytes and maternal decidual cells after stimulation with concanavalin A (Fig. 3). However, the P value approaches significance (p = 0.083). This lack of significance may be a result of the small sample size of the study. No interleukin-2 was detected after maternal cells were stimulated with irradiated paternal leukocytes. Irradiated paternal cells did not produce detectable amounts of interleukin-2 or granulocyte-macrophage colonystimulating factor. Comment
Recent observations suggest that Iymphokines are important mediators of immune effector cells. Lymphokines are polypeptides secreted by activated lymphocytes with autocrine and paracrine functions to direct the immune response. They signal lymphocytes to proliferate,6 develop cytotoxicity,' and produce antibody." To date, more than 10 lymphokines have been described. The nature of the immune response may be dictated by the predominant lymphokine produced by stimulated lymphocytes in response to specific antigen." Lymphokines also may function in a "trophic" role in certain nonlymphoid tissues. 4 Our data indicate that immune effector cells from maternal peripheral blood and decidua differ markedly in their capacity to produce the Iymphokine granulocyte-macrophage colony-stimulating factor after stimulation. Our findings also imply that assays of maternal peripheral immune cells are not reflective of the immunologic relationship between the mother and fetus. This is the first study to show that human term decidual cells produce large amounts of granulocytemacrophage colony-stimulating factor after stimulation. Granulocyte-macrophage colony-stimulating factor is generally thought to be primarily a hematopoietic cytokine, inducing bone marrow stem cells to proliferate and differentiate. 10 However, Athanassakis et al. 11 have found that granulocyte-macrophage colonystimulating factor produced by murine decidual cells induces proliferation of murine trophoblast. II This finding forms the basis of Wegmann's "immunotrophism" hypothesis.· According to this hypothesis, the maternal immune response to fetopaternal antigens is characterized by the secretion of granulocytemacrophage colony-stimulating factor and other trophic cytokines. These cytokines stimulate trophoblast growth and thus maintain the fetal allograft.
Volume 163 Number 6, Part 1
Our finding that granulocyte-macrophage colonystimulating factor is produced in large amounts by human decidual cells in response to paternal antigen supports this hypothesis. Although it is possible that the granulocyte-macrophage colony-stimulating factor produced after stimulation with paternal peripheral blood leukocytes may reflect nonspecific antigenic stimulation, we believe that this is unlikely given the large quantities produced. Whether granulocytemacrophage colony-stimulating factor is essential to human pregnancy is unknown, but it is interesting that the heterogenous collection of decidual cells we tested maintained the capacity to produce large amounts of this lymphokine after stimulation. It seems likely that decidual granulocyte-macrophage colony-stimulating factor is of teleologic significance and it will be of scientific and academic interest to dissect the cell populations and determine the source of the granulocytemacrophage colony-stimulating factor. The marked differences in granulocyte-macrophage colony-stimulating factor production response between cells from the peripheral blood and the decidua suggest that the immune cells in these environments are under different regulatory control. This is consistent with our hypothesis that the maternal decidua constitutes a distinct immunologic microenvironment. In the murine immune system, cells from different lymphoid tissues produce different lymphokines. I2 An important regulatory factor is the steroid hormonal milieu found in each lymphoid tissue. I2 . 13 We have recently shown that corticosteroids, dehyroepiandrosterone, and 1,25dihydroxyvitamin D3 at physiologic concentrations can dramatically alter lymphokine production by stimulated lymphocytes. 12. 13 Therefore markedly different lymphokine production from immune effector cells isolated from the decidua, where the cells would be influenced by the dramatic changes in the decidual steroid hormone environment, would be anticipated. In conclusion, our data show that deciduallymphocyte-like cells differ from maternal peripheral blood leukocytes in their capacity to produce large amounts of granulocyte-macrophage colony-stimulating factor. Furthermore, we have shown a similar response to pa-
Lymphokine production during term human pregnancy
1893
ternal antigen and this may reflect an immunotrophic maternal response to pregnancy. Further characterization of the normal lymphokine response by decidual immune effector cells and the local immunoregulatory factors that controllymphokine production will provide new insights into the normal and abnormal maternal immune responses evoked by the fetal allograft. REFERENCES 1. Bell SC, Billington WD. Anti-fetal allo-antibody in the pregnant female. Immunol Rev 1983;75:5-30. 2. Nicholas NS, Panayi GS, Nouri AME. Human pregnancy serum inhibits interleukin-2 production. Clin Exp Immunol 1984;58:587-95. 3. Lala PK, Chatterjee-Hasrouni S, Kearns M, Montgomery B, Colavincenzo V. Immunobiology of the maternal-fetal interface. Immunol Rev 1983;75:87-116. 4. Wegmann TG. Placental immunotrophism: the idea and the evidence. In: Chaouat G, ed. The immunology of the fetus. Boca Raton, Florida: CRC Press, Inc, 1990: 179. 5. Delvin EE, Arabian A, Glorieux FH, Mamer OA. In vitro metabolism of 25-hydroxycholecakiferol by isolated cells from human decidua. J Clin Endocrinol Metabol 1985; 60:880-5. 6. Mosmann TR, Coffman RL Thl and TH2 cells: different patterns of lymphokine secretion lead to different functional properties. Ann Rev Immunol 1989;7:145-73. 7. Farrar JJ, Benjamin WR, Hilfiker ML, Howard M, Farrar WL, Fuller-Farrar J. The biochemistry, biology, and role of interleukin 2 in the induction of cytotoxic T cell and antibody-forming B cell responses. Immunol Rev 1982; 63: 129-66. 8. Coffman RL, Seymour BW, Lebman DA, et al. The role of helper T cell products in mouse B cell differentiation and isotype regulation. Immunol Rev 1988; 102:5-28. 9. Janeway CA, Carding S, Jones B, et al. CD4+ T cells: specificity and function. Immunol Rev 1988; 101 :39-79. 10. Tomomonga M, Golde DW, Gasson JC. Biosynthetic (recombinant) human granulocyte-macrophage colonystimulating factor: effect on normal bone marrow and leukemia cell lines. Blood 1986;67:31-6. II. Athanassakis I, Bleackley RC, Paetkau V, Guilbert L, Barr PJ, Wegmann TG. The immunostimulatOl'y effect of T cells and T celllymphokines on murine fetally-drived placental cells. J Immunol 1987;138:37-44. 12. Daynes RA, Araneo BA, Dowell TA, Huang K, Dudley DJ. Regulation of murine Iymphokine production in vivo: regulatory influences over T helper cell function. J Exp Med 1990; 171 :979-96. 13. Daynes RA, Dudley DJ, Araneo BA. Regulation of murine lymphokine production in vivo: II. Dehydroepiandrosterone is a natural enhancer of IL-2 synthesis by helper T cells. Eur J Immunol 1990;20:793-802.
28. Teitelbaum I. The epidermal growth factor receptor is coupled to a phospholipase A2-specific pertussis toxininhibitable guanine nucleotide-binding regulatory protein in cultured rat inner medullary collecting tubule cells. J Bioi Chern 1990;265:4218-22. 29. Zick Y, Sagi-Eisenberg R, Pines M, Gierschik P, and Spie-
December 1990 Am J Obstet Gynecol
gel AM. Multisite phosphorylation of the a subunit of transducin by the insulin receptor kinase and protein kinase C. Proc Nat! Acad Sci USA 1986;83:9294-7. 30. Crouch M, Lapetina E. A role for G j in control of thrombin receptor-phospholipase C coupling in human platelets. J Bioi Chern 1988;263:3363-71.
Lymphokine production during term human pregnancy: Differences between peripheral leukocytes and decidual cells Donald J. Dudley, MD, Murray D. Mitchell, DPhil, Kathryn Creighton, BS, and D. Ware Branch, MD Salt Lake City, Utah Although the fetus is considered to be an "allograft," there is little information concerning the role of Iymphokines in human pregnancy. Lymphokines are polypeptides secreted by stimulated lymphocytes that direct the immune response by enabling immune effector cells to communicate with each other. To characterize Iymphokine production during normal human pregnancy, we isolated peripheral leukocytes and decidual lymphocyte-like cells from women undergoing repeat cesarean section at term. After stimulation with mitogen and paternal antigen for 24 hours, culture supernatants were assayed for granulocyte-macrophage colony-stimulating factor and interleukin-2 by enzyme-linked immunosorbent assay. There was no difference in the amount of interleukin-2 produced by stimulated peripheral and decidual cells. However, granulocyte-macrophage colony-stimulating factor production by stimulated decidual lymphocyte-like cells was significantly greater than granulocyte-macrophage colony-stimulating factor produced by peripheral lymphocytes. Decidual lymphocyte-like cells produced granulocyte-macrophage colony-stimulating factor both spontaneously and after stimulation with mitogen or paternal antigen, whereas peripheral leukocytes did not. This suggests that the decidua constitutes a distinct immunologic microenvironment. (AM J OBSTET GVNECOL 1990;163:1890-3.)
Key words: Pregnancy, lymphokines, granulocyte-macrophage colony-stimulating factor, interleukin-2 Our knowledge of the immunologic relationship between the mother and conceptus is incomplete. The conceptus appears to elicit a specific maternal immune response consisting of antibody production' and changes in maternal cellular immunity. This is detected as diminished proliferation of maternal peripheral blood leukocytes to irradiated paternal peripheral blood leukocytes in a mixed leukocyte culture! However, some investigators believe that the important cellular immune responses to fetopaternal antigens occur in the decidua at the maternal-fetal interface.' It is cur-
From the Department of Obstetrics and Gynecology, University of Utah School of Medicine. Supported by a grant from the Willard L. Eccles Charitable Foundation. Presented at the Thirty-seventh Annual Meeting of the Society for Gynecologic Investigation, St. Louis, Missouri, March 21-24, 1990. Reprint requests: Donald J. Dudley, MD, Department of Obstetrics and Gynecology, University of Utah School of Medicine, 50 N. Medical Dr., Salt Lake City, UT 84132. 616124761
1890
rently thought that these decidual immune responses are beneficial to the survival of the conceptus" Cellular immune responses are characterized by the production of a variety of polypeptides known' as lymphokines. Little information concerning the role of lymphokines in protecting the fetal allograft is available. Moreover, few studies have addressed the function of lymphoid cells in the human decidua. The purpose of our study was to characterize lymphokine production during normal human pregnancy, comparing the production of granulocyte-macrophage colonystimulating factor and interleukine-2 by peripheral leukocytes with that of decidual lymphocyte-like cells. We hypothesized that the decidua constitutes a distinct immunologic environment that supports immune effector cells that produce lymphokines different from those produced by peripheral leukocytes.
Material and methods Placentas were obtained aseptically from normal human pregnancies at term repeat cesarean section. The
Volume 163 l\:umber 6. Pan 1
membranes were isolated and rinsed in Dulbecco's modified Eagle's medium. A modification of the method described by Delvin et aI.' was used to isolate decidual lymphocyte-like cells. The decidua adherent to the maternal side of the chorion laevae was removed by sharp dissection to yield a cell homogenate. To disperse the cells, the homogenate was mixed with 63 units of collagenase per milliliter of Dulbecco's modified Eagle's medium (100 ml total volume) at 37° C for 60 to 120 minutes. The cell homogenate was filtered through a gauze mesh to remove large debris. The cells then were centrifuged over Histopaque 1077 (Sigma, St. Louis) at 1000 g for 30 minutes to isolate the lymphocyte-like cell fraction. This fraction was aspirated and rinsed. The culture medium used throughout these experiments was RPMI with 1% NutridomaHV (Boehringer-Mannheim, Indianapolis) and antibiotics (100 Vlml penicillin G sodium, 100 Vlml streptomycin sulfate, and 0.25 f-tg/ml amphotericin B, Gibco Laboratories, Grand Island, N.Y.). Heparinized peripheral blood was collected from the prospective mother and father just before the cesarean section. Peripheral blood leukocytes were isolated by standard techniques (Sigma Diagnostics, procedure 1077). The blood was diluted 1 : 1 with culture medium and then centrifuged over Histopaque 1077 at 1000 g for 30 minutes. The buffy coat was aspirated and rinsed. Separate cell suspensions of maternal peripheral blood leukocytes and decidual cells were adjusted to 10 6 cells/ml and duplicate 1 ml suspensions of each were placed in 24-well tissue culture plates. The cells were stimulated with 10 f-tg/ml of concanavalin A and incubated for 24 hours at 37° C. Supernatants were collected and stored at - 20° C until assayed. To evaluate the maternal peripheral lymphocyte and the decidual lymphocyte-like cell response to paternal antigen, 5 x 10 5 maternal peripheral blood leukocytes or decidual cells were coincubated with a similar number of irradiated (2000 rads) paternal cells for 24 hours. V nstimulated control cultures were established for both cell populations in all experiments. Culture supernatants were collected and stored at - 20° C until assayed. Supernatants were assayed for granulocyte-macrophage colony-stimulating factor and interleukin-2 with commercially available enzyme-linked immunosorbent assay kits (Genzyme, Boston). The limits of sensitivity for the interleukin-2 enzyme-linked immunosorbent assay range from 0.3 ng/ml to 20.8 ng/ml. The intraassay coefficient of variation ranges from 2% to 6% and the interassay coefficient of variation is approximately 4%. For the granulocyte-macrophage colonystimulating factor enzyme-linked immunosorbent assay, the limits of sensitivity range from 0.008 ng/ml to 1.0 ngl ml. The intraassay coefficient of variation
Lymphokine production during term human pregnancy
1891
JfJ p-O.056
.45
I
.40
.35
I
~
.30 .25
.20 .16 .10
.06
P
D
CONTROL
PDP D CONCANAVAI,JII A PATERNAL 8'1M.l.A110N
Fig. 1. Granulocyte-macrophage colony-stimulating factor production by maternal peripheral leukocytes (P) and deciduallymphocyte-like cells (D) in unstimulated, concanavalin Astimulated, and paternal antigen-stimulated conditions (n = 8). The hatched line delineates the lower limit of sensitivity of the assay. All values are mean ± SEM.
ranges from 2% to 8%, whereas the interassay coefficient of variation ranges from 5% to 10%. The data were evaluated statistically with the MannWhitney V test. Results
Eight term pregnancies were evaluated. Granulocytemacrophage colony-stimulating factor production by maternal peripheral blood leukocytes and decidual lymphocyte-like cells is depicted in Fig. 1. Concanavalin A-stimulated maternal peripheral blood leukocytes produced low levels of granulocytemacrophage colony-stimulating factor (0.018 ± 0.005 ng/ml, mean ± SEM). However, decidual lymphocytelike cells produced much greater amounts of granulocyte-macrophage colony-stimulating factor after stimulation with concanavalin A (0.2 ± 0.059 ng/ml, p = 0.006). Stimulation with paternal peripheral blood leukocytes also resulted in greater production of granulocyte-macrophage colony-stimulating factor by decidual cells (0.441 ± 0.262 ng/ml, p = 0.055 when compared with peripheral cells stimulated with paternal peripheral blood leukocytes). Fig. 1 also suggests that the decidual cells produce granulocyte-macrophage colony-stimulating factor constitutively at low, but significant, levels (0.073 ± 0.038 ng/mI, p = 0.01 when compared with unstimulated maternal peripheral cells). To evaluate whether this might reflect nonspecific stimulation caused by cell processing or culture, we performed a time course experiment with concanavalin A-stimulated and unsti-
1892
Dudley et al.
December 1990
Am
1.0
+
0.9 0.8 0.7
~ ~
~
~
0.6 0.5 0.4 0.3
Coneanavaln A
0.2
/
0.1 0
lmtimuIated
L-J~~~~-",-~=-----------I-----------r--
0
12
6
18
24
Tme (hru's)
Fig. 2. Granulocyte-macrophage colony-stimulating factor production by unstimulated and concanavalin A-stimulated decidual lymphocyte-like cells over a 24-hour time period (n = 3, mean ± SE). (The standard error at the asterisk is 0.283.)
.90 p.O.083
.80
II
.70 .80
1!
~
.:. C\I
.50 .40
.30 .20
.10 P
0
CONTROL
P
0
CONCANAVAltl II
Fig. 3. Interieukin-2 production by unstimulated and concanavalin A-stimulated maternal peripheral leukocytes (P) and decidual lymphocyte-like cells (D) (n = 8). The hatched line delineates the lower limit of sensitivity of the assay. All values are mean ± SEM.
mulated decidual cells from three different placentas. Culture supernatants were obtained at the start of the experiment and then every 6 hours for 24 hours. The results are shown in Fig. 2. Granulocyte-macrophage colony-stimulating factor was undetectable in the cell suspensions before stimulation (time 0). However, low levels of granulocyte-macrophage colony-stimulating factor were produced by unstimulated decidual cells over 24 hours. This may represent nonspecific acti-
J Obstet Gynecol
vation of the cells. Stimulation with concanavalin A resulted in significant production of granulocytemacrophage colony-stimulating factor after 6 hours of incubation, indicating that the apparent nonspecific activation contributes minimally to mitogen-induced granulocyte-macrophage colony-stimulating factor production. There was no significant difference in interleukin-2 production by maternal peripheral leukocytes and maternal decidual cells after stimulation with concanavalin A (Fig. 3). However, the P value approaches significance (p = 0.083). This lack of significance may be a result of the small sample size of the study. No interleukin-2 was detected after maternal cells were stimulated with irradiated paternal leukocytes. Irradiated paternal cells did not produce detectable amounts of interleukin-2 or granulocyte-macrophage colonystimulating factor. Comment
Recent observations suggest that Iymphokines are important mediators of immune effector cells. Lymphokines are polypeptides secreted by activated lymphocytes with autocrine and paracrine functions to direct the immune response. They signal lymphocytes to proliferate,6 develop cytotoxicity,' and produce antibody." To date, more than 10 lymphokines have been described. The nature of the immune response may be dictated by the predominant lymphokine produced by stimulated lymphocytes in response to specific antigen." Lymphokines also may function in a "trophic" role in certain nonlymphoid tissues. 4 Our data indicate that immune effector cells from maternal peripheral blood and decidua differ markedly in their capacity to produce the Iymphokine granulocyte-macrophage colony-stimulating factor after stimulation. Our findings also imply that assays of maternal peripheral immune cells are not reflective of the immunologic relationship between the mother and fetus. This is the first study to show that human term decidual cells produce large amounts of granulocytemacrophage colony-stimulating factor after stimulation. Granulocyte-macrophage colony-stimulating factor is generally thought to be primarily a hematopoietic cytokine, inducing bone marrow stem cells to proliferate and differentiate. 10 However, Athanassakis et al. 11 have found that granulocyte-macrophage colonystimulating factor produced by murine decidual cells induces proliferation of murine trophoblast. II This finding forms the basis of Wegmann's "immunotrophism" hypothesis.· According to this hypothesis, the maternal immune response to fetopaternal antigens is characterized by the secretion of granulocytemacrophage colony-stimulating factor and other trophic cytokines. These cytokines stimulate trophoblast growth and thus maintain the fetal allograft.
Volume 163 Number 6, Part 1
Our finding that granulocyte-macrophage colonystimulating factor is produced in large amounts by human decidual cells in response to paternal antigen supports this hypothesis. Although it is possible that the granulocyte-macrophage colony-stimulating factor produced after stimulation with paternal peripheral blood leukocytes may reflect nonspecific antigenic stimulation, we believe that this is unlikely given the large quantities produced. Whether granulocytemacrophage colony-stimulating factor is essential to human pregnancy is unknown, but it is interesting that the heterogenous collection of decidual cells we tested maintained the capacity to produce large amounts of this lymphokine after stimulation. It seems likely that decidual granulocyte-macrophage colony-stimulating factor is of teleologic significance and it will be of scientific and academic interest to dissect the cell populations and determine the source of the granulocytemacrophage colony-stimulating factor. The marked differences in granulocyte-macrophage colony-stimulating factor production response between cells from the peripheral blood and the decidua suggest that the immune cells in these environments are under different regulatory control. This is consistent with our hypothesis that the maternal decidua constitutes a distinct immunologic microenvironment. In the murine immune system, cells from different lymphoid tissues produce different lymphokines. I2 An important regulatory factor is the steroid hormonal milieu found in each lymphoid tissue. I2 . 13 We have recently shown that corticosteroids, dehyroepiandrosterone, and 1,25dihydroxyvitamin D3 at physiologic concentrations can dramatically alter lymphokine production by stimulated lymphocytes. 12. 13 Therefore markedly different lymphokine production from immune effector cells isolated from the decidua, where the cells would be influenced by the dramatic changes in the decidual steroid hormone environment, would be anticipated. In conclusion, our data show that deciduallymphocyte-like cells differ from maternal peripheral blood leukocytes in their capacity to produce large amounts of granulocyte-macrophage colony-stimulating factor. Furthermore, we have shown a similar response to pa-
Lymphokine production during term human pregnancy
1893
ternal antigen and this may reflect an immunotrophic maternal response to pregnancy. Further characterization of the normal lymphokine response by decidual immune effector cells and the local immunoregulatory factors that controllymphokine production will provide new insights into the normal and abnormal maternal immune responses evoked by the fetal allograft. REFERENCES 1. Bell SC, Billington WD. Anti-fetal allo-antibody in the pregnant female. Immunol Rev 1983;75:5-30. 2. Nicholas NS, Panayi GS, Nouri AME. Human pregnancy serum inhibits interleukin-2 production. Clin Exp Immunol 1984;58:587-95. 3. Lala PK, Chatterjee-Hasrouni S, Kearns M, Montgomery B, Colavincenzo V. Immunobiology of the maternal-fetal interface. Immunol Rev 1983;75:87-116. 4. Wegmann TG. Placental immunotrophism: the idea and the evidence. In: Chaouat G, ed. The immunology of the fetus. Boca Raton, Florida: CRC Press, Inc, 1990: 179. 5. Delvin EE, Arabian A, Glorieux FH, Mamer OA. In vitro metabolism of 25-hydroxycholecakiferol by isolated cells from human decidua. J Clin Endocrinol Metabol 1985; 60:880-5. 6. Mosmann TR, Coffman RL Thl and TH2 cells: different patterns of lymphokine secretion lead to different functional properties. Ann Rev Immunol 1989;7:145-73. 7. Farrar JJ, Benjamin WR, Hilfiker ML, Howard M, Farrar WL, Fuller-Farrar J. The biochemistry, biology, and role of interleukin 2 in the induction of cytotoxic T cell and antibody-forming B cell responses. Immunol Rev 1982; 63: 129-66. 8. Coffman RL, Seymour BW, Lebman DA, et al. The role of helper T cell products in mouse B cell differentiation and isotype regulation. Immunol Rev 1988; 102:5-28. 9. Janeway CA, Carding S, Jones B, et al. CD4+ T cells: specificity and function. Immunol Rev 1988; 101 :39-79. 10. Tomomonga M, Golde DW, Gasson JC. Biosynthetic (recombinant) human granulocyte-macrophage colonystimulating factor: effect on normal bone marrow and leukemia cell lines. Blood 1986;67:31-6. II. Athanassakis I, Bleackley RC, Paetkau V, Guilbert L, Barr PJ, Wegmann TG. The immunostimulatOl'y effect of T cells and T celllymphokines on murine fetally-drived placental cells. J Immunol 1987;138:37-44. 12. Daynes RA, Araneo BA, Dowell TA, Huang K, Dudley DJ. Regulation of murine Iymphokine production in vivo: regulatory influences over T helper cell function. J Exp Med 1990; 171 :979-96. 13. Daynes RA, Dudley DJ, Araneo BA. Regulation of murine lymphokine production in vivo: II. Dehydroepiandrosterone is a natural enhancer of IL-2 synthesis by helper T cells. Eur J Immunol 1990;20:793-802.
