0013-7227/91/1283-1352$03.00/0 Endocrinology Copyright © 1991 by The Endocrine Society
Vol. 128, No. 3 Printed in U.S.A.
Enhancement of in Vivo Humoral Immunity by Estrogen: Permissive Effect of a Thymic Factor* G. T. ERBACH AND JANICE M. BAHR Departments of Physiology (G.T.E., J.M.B.) and Animal Sciences (J.M.B.), University of Illinois, Urbana, Illinois 61801
ABSTRACT. Physiological levels of estrogen enhance humoral immune responses. Several in vitro studies indicate the hormone to have a direct effect on immune cells, and other studies show that estrogen may affect humoral immunity indirectly through the thymus. Therefore, we have conducted experiments to investigate the requirement of the thymus in the enhancement of humoral immune responsiveness by estrogen. In Exp I, adult ovariectomized Lewis rats were thymectomized or sham thymectomized and given estradiol (E2; 0.25 fig E2 in sesame oil, sc, once every 4 days) or the oil vehicle in a 2 X 2 factorial design, and their antifluorescein responses were followed by enzyme-linked immunosorbant assay across 21 days. Only animals that were thymus intact and given estrogen replacement showed significantly (P < 0.05) greater serum anti-
fluorescein titers than all other treatments. In Exp II, ovariectomized thymectomized rats were submitted to a 2 x 3 factorial design of oil vehicle or E2 replacement and saline, gelatin, or thymus replacement (thymosin fraction 5; 1 mg/kg in saline, sc). As described above, only the animals receiving both thymosin fraction 5 and E2 replacement displayed antifluorescein titers that were significantly (P < 0.03) increased over titers of all other treatment groups. These results indicate that the enhancement of in vivo humoral immunity by estrogen requires the thymus, and that a constitutive thymic factor, found in thymosin fraction 5, exerts a permissive influence on the action of E2 outside the thymus to increase a specific humoral immune response. {Endocrinology 128: 1352-1358, 1991)
E
STROGEN has been shown to enhance humoral immune responsiveness in several species and with various antigenic systems (1-5). A number of mechanisms have been proposed for this phenomenon, and several targets for estrogen have been identified within the immune system. In vitro studies indicate the hormone to have a direct influence on lymphocyte function. Kenney et al. (6) incubated sheep red blood cell (SRBC)primed murine spleen cells with various concentrations of estradiol (E2) and found that physiological levels of E2 caused an increase in the number of antibody-secreting cells in a hemolytic plaque assay. Myers and Petersen (3), using a similar experimental system, found E2 to increase not the number of antibody-producing cells, but the amount of antibody produced per plaque-forming colony (PFC). Incubation with E2 also caused an increase in the number of cells secreting immunoglobulin M (IgM) (7) and IgG (8) among human peripheral blood lymphocytes. The detection of receptor for estrogen in cytosol of mouse spleen (9), human spleen (10, 11), and human Received August 6. 1990. Address all correspondence and requests for reprints to: Dr. Janice Bahr, Animal Genetics Laboratory, University of Illinois, 1301 West Lorado Taft, Urbana, Illinois 61801. * Present address: Pathology Research Laboratory, Massachusetts General Hospital, 149 Thirteenth Street, Charlestown, Massachusetts 02129.
peripheral blood lymphocytes (10,12) and the measurement of significant estrogen binding in human peripheral CD8+ cells (13) and murine B cells (14) further support the idea that estrogen may act directly on lymphoid cells to alter their activities and effect enhanced antibody responses. Receptors for estrogen have been characterized and measured also in cells of the epithelial matrix of the thymus (15-22). The thymus orchestrates the functioning of the immune system through its control of T-cell maturation, differentiation, and activity, which it exerts by the production of soluble factors and thymic hormones (23-25). T-Lymphocytes play an important modulatory role in the humoral immune response. Several in vivo and in vitro studies show estrogen to influence only indirectly certain proliferative responses and functional characteristics of lymphoid cells that may affect humoral immunity (7, 26-31), raising the possibility that estrogen may affect humoral immune responsiveness via a thymic route of action. We report in this paper the results of two experiments, the first to determine that estrogen enhancement of in vivo humoral immunity requires the thymus, and the second to demonstrate that a constitutive soluble product of the thymus exerts a permissive influence on the action of estrogen outside the thymus to increase a specific humoral immune response.
1352
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 March 2015. at 05:35 For personal use only. No other uses without permission. . All rights reserved.
ESTROGEN ENHANCEMENT OF HUMORAL IMMUNITY
Materials and Methods Animals Lewis inbred female rats were purchased from Harlan-Sprague Dawley (Indianapolis, IN). They were housed in NIHapproved facilities (14 h of light, 10 h of darkness) and were provided Purina rat chow and water ad libitum. Surgeries Surgeries were performed under aseptic conditions. At 70 days of age, animals were ovariectomized through bilateral incisions under ketamine-acepromazine anaesthesia 21-23 days before immunization. Two weeks after ovariectomy (OVX), rats were again anesthetized, and thymi were removed through a single midline sternal incision. The sham operation consisted of the identical procedure, except that the thymus was not excised. Antigen preparation and immunization Fluorescein (FL) as fluorescein isothiocyanate was covalently conjugated to keyhole limpet hemocyanin (FL-KLH; Sigma Chemical Co., St. Louis, MO) as previously described (4, 32). FL-KLH, a hapten carrier, T-dependent immunogen was chosen for its proven immunogenicity in other systems (33) and for the relatively limited population of antibodies the hapten induces. Rats were immunized against 0.21 mg FLKLH in water in a total volume of 0.25 ml, sc. Determination of antibody titers Blood samples were obtained from the tail and allowed to clot overnight at 4 C. Sera were then frozen and stored at -20 C until assayed. Anti-FL activity in serum samples was determined by specific, solid phase enzyme-linked immunosorbent assay (ELISA). Microtiter plate wells were coated with 1 ng FL conjugated to BSA (Sigma) in 100 /x\ phosphate buffer, pH 8.0, and incubated overnight at 4 C. Wells were washed, and an overcoat of 300 n\ 1.0% BSA, 0.9% NaCl, and 100 Mg/ml thimerosol in 0.01 M Tris, pH 7.5, was applied and incubated for 30 min at 37 C. Wells were then aspirated dry. Serum samples, internal standards, and second antibodies were diluted in overcoat solution plus 0.05% Tween-20. The diluted serum samples containing anti-FL activity, interplate controls, complete titrations of internal standards, and normal sera were then added in 100-/ul volumes to duplicate wells and incubated for 2 h at 37 C. After this incubation and a subsequent wash, 100 /xl horseradish peroxidase-conjugated rabbit antirat Ig (heavy and light chain specific; Accurate Chemical Co., Westbury, NY) diluted 1:1000 were added to each well. The plates were again incubated for 2 h at 37 C. After a final wash, 1.5 mM H2O2 was added to each well with 0.4 mM 2,2'-azinobis-(3ethylbenzthiazoline sulfonic acid) in 0.05 M citrate buffer in a total of 100 n\. 2,2'-Azinobis-(3-ethylbenzthiazoline sulfonic acid develops color in the presence of free oxygen; this reaction was allowed to proceed for 1 h at room temperature, then was stopped by the addition of 100 /xl 0.15 M hydrogen fluoride plus 1.0 mM EDTA. Color development in plate wells was read at 406 nm. Controls for nonspecific binding were included in every trial.
1353
Before the assay of all samples, serial dilutions were carried out on representative antiserum samples from each experimental group to determine the dilution at which the assay end point (optical density) fell on the linear portion of all titration curves. This dilution was found to be 1:400, and in the assay of samples only this dilution was tested. Relative titers of sample antisera were determined by conversion to multiples of normal activity (MONA) by the method of Greenfield et al. (34). The conversion is based on the assumption, borne out in preliminary trials, that all titration curves within an assay are parallel. It follows, then, that from a single point a titration curve can be derived for a tested sample by comparison to a complete titration performed for a pooled sample (internal standard; Fig. 1). The relative sample titer, expressed as MONA, is defined as the ratio of dilutions (3/or. the ratio of /3, the extrapolated dilution of sample antiserum which yields the same optical density as that of the standard antiserum at 1:400, to a, the sample dilution (1:400). Titrations of the internal standard were run in every plate. A different pool of antiserum was used to measure the samples of each experiment; therefore, MONA values can be compared only within an experiment. The intraplate variation was less than 0.1%; the interplate variation was less than 0.5%. E2 replacement and measurements of serum E2 Subcutaneous injections of 0.25 ml of a 1.0 ng/m\ solution of E2 in sesame oil or the oil vehicle alone were administered the day before immunization and every fourth day, as characterized and described previously (5). E2 concentrations were measured by RIA, as previously described (35). The sensitivity of this assay is 8 pg. The E2 antiserum 244 was obtained from G. D. Niswender, Colorado
>•
H CO
z UJ Q
O I-
o.
o.
a
(3
DILUTION FIG. 1. Extrapolation of dilution 0 from a titration curve derived from a single test point. • , Measured data points; O, extrapolated point; , calculated from OD of diluted standard serum; , from test point. Adapted from Greenfield et al. (34).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 March 2015. at 05:35 For personal use only. No other uses without permission. . All rights reserved.
ESTROGEN ENHANCEMENT OF HUMORAL IMMUNITY
1354
State University (Fort Collins, CO). It was prepared against 6-hemisuccinate-17/3-estradiol and cross-reacts 0.44% with estrone. Thymic factor replacement
The epithelial cells of the thymus produce a number of peptides that have been shown to participate in the development of the T-cell system and in the regulation of immune cell function. Many of the activities of the endocrine thymus are present in thymosin fraction 5 (TF5), a semipurified aqueous extract of calf thymus consisting of between 20-30 acidic peptides ranging in mol wt from 3000-7000. As reviewed by Goldstein et al. (25), TF5 has been demonstrated to induce differentiation markers on immature T-cells and to restore immune competence to nude and neonatally thymectomized and adult thymectomized mice. TF5 has been shown also to alter the in vitro responsiveness of lymphoid cells from normal and immune-compromised mice and rats. This hormone preparation provides the best available broad replacement of possible active thymic factors. It was administered to animals by sc injection in saline at a dose of 1 mg/kg, as recommended by Dr. A. L. Goldstein (personal communication). This is generally the therapeutic dose for restoration of immune responsiveness in humans and rodents. Animals were injected with TF5 the day of and the day after E2 administration to assure the presence of thymic factors concurrent with the influence of estrogen on responsive cells. As TF5 is derived from a bovine source and was injected at potentially immunizing doses, controls injected with bovine gelatin (Bio-Rad Laboratories, Richmond, CA) at 1 mg/kg were included to test against general increases in anti-FL responses induced by repeated injections of a foreign protein. TF5 was graciously provided by Dr. Alan L. Goldstein, George Washington University (Washington, DC). Experimental design Exp I. This experiment was designed in a 2 X 2 factorial fashion, testing thymectomy (THYMX) against E2 replacement by injection in ovariectomized rats. Animals were randomly assigned to four treatment groups: 1) OVX, THYMX, and oil injections; 2) OVX, THYMX, and E2 injections; 3) OVX, sham THYMX, and oil injections; and 4) OVX, sham THYMX, and E2 injections. All animals were ovariectomized 2 weeks before THYMX or sham THYMX (Fig. 2). Seven to 10 days lapsed between THYMX and immunization against FL-KLH. Estrogen re-
Endo • 1991 Voll28«No3
placement commenced the day before immunization; sc injections of 0.25 fig E2 or the oil vehicle were given every fourth day. Blood samples (20-40 n\) were taken from the tail 3, 6, 9, 12,15,18, and 21 days after immunization; serum was harvested and assayed for anti-FL activity. On day 23 postimmunization, terminal blood was collected by decapitation 1.75 h after E2 or oil injection for measurement of peak serum E2 concentration. At that time, gross examination was made for ovarian and thymic remnants. Animals with incomplete organ extirpation were removed from the study. Exp II. This experiment was designed in a 3 X 2 factorial fashion, testing TF5 and E2 replacement in ovariectomizedthymectomized rats. Animals were randomly assigned to six treatment groups: 1) OVX, THYMX, oil injections and saline injections; 2) OVX, THYMX, E2 injections and saline injections; 3) OVX, THYMX, oil injections and TF5 injections; 4) OVX, THYMX, E2 injections and TF5 injections; 5) OVX, THYMX, oil injections and gelatin injections; and 6) OVX, THYMX, E2 injections and gelatin injections. All animals were ovariectomized 2 weeks before THYMX (Fig. 3). As in Exp I, 7-10 days lapsed between thymectomy and immunization, and estrogen replacement commenced the day before immunization. Injections of TF5, gelatin, or saline were administered in a volume of 0.2 ml the day of and the day after each E2 or oil injection. Blood samples were taken and assayed as described for Exp I. In a parallel study the dose response of ovariectomized and ovary-intact rats to TF5 was determined for anti-FL titers using doses of 1 and 10 mg/kg. TF5 was again prepared in saline and injected in a volume of 0.2 ml, sc, according to the schedule of treatments described by Fig. 3. Statistics Differences among means were determined by repeated measures analysis, factorial analysis, and analysis of variance using the SAS program (36).
Results Exp I The purpose of Exp I was to determine if the thymus is required for the enhancement of in vivo humoral EXPERIMENT2 SCHEDULE OF TREATMENTS
EXPERIMENT 1
» TF5 OR SALINE OR GELATIN INJECTION
SCHEDULE OF TREATMENTS
| E 2 OR OIL INJECTION
\ E:2 OR OIL INJECTION • OVX •
THYMX 1
BLOOD SAMPLING
•
OVX
THYMX
11 MM | ) | \ \ \ 1 I I I I I I I I I I I I I I i i i I i i i t i I
7-10 DAYS
llMM -I 0
I
I I I I I I t I I
(9 QU
1 14 DAYS
14 DAYS
BLOOD SAMPLING
7-10 DAYS
DAYS DAYS
FIG. 2. Exp I: schedule of treatments. Arrow indicates injection; box indicates blood sample. Immunization on day 0.
FIG. 3. Exp II: schedule of treatments. Arrow indicates E2 or control injection; asterisk indicates TF5 or control injection; box indicates blood sample. Immunization on day 0.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 March 2015. at 05:35 For personal use only. No other uses without permission. . All rights reserved.
ESTROGEN ENHANCEMENT OF HUMORAL IMMUNITY
immunity. E2 replacement in ovariectomized rats was fashioned to mimic the level and pattern of E2 surges of the 4-day estrous cycle. Shown in Table 1 are the E2 concentrations measured in terminal serum samples 1.75 h after the final hormone or vehicle injection. Groups receiving oil injections had serum E2 concentrations of approximately 7 pg/ml; groups receiving E2 injections experienced E2 surges of approximately 40 pg/ml, according to these measurements and previous studies (5), for 3 h once every four days. This means of E2 replacement is a potent enhancer of humoral immunity (5). The anti-FL responses for each of the treatment groups is shown in Fig. 4. The profiles of the responses of all of the groups were similar; all had titers equivalent to normal serum on day 3, then showed a gradual increase across 21 days. Thymectomized and oil, thymectomized and E2, and sham thymectomized and oil groups did not differ in anti-FL responses on any day. Only the group that was thymus intact and receiving E2 replacement showed a significantly increased titer (P < 0.05) on day 21. Thymic and uterine wet weights were also observed at the conclusion of the experiment. Four-day injections of 0.25 ng E2 did not cause involution of the thymus, but did significantly increase (P < 0.001) the growth of the uterus compared to that caused by injection of the vehicle. These data support previous observations made regarding the trophic responses of these tissues to the 4day injection regimen (5).
1355
THYMX + OIL, n = 6 9.0
THYMX + E 2 , n=ll SHAM THYMX+OIL, n = 9 SHAM THYMX + E 2 , n = 8
8.0
7.0
f
6.0
1
f
4.0
3.0
2.0
1.0
ExpII This experiment was designed to test thymus replacement against physiological estrogen, administered by injection once every 4 days. Table 1 presents the concenTABLE 1. E2 concentrations (picograms per ml) in terminal serum samples 1.75 h after injection of 0.25 /xg E2 or vehicle Treatment
Thymectomized
Thymus-intact
8.2 0.6 6
6.2 0.5 9
Expl OVX + oil Mean SE
n OVX + E2 Mean SE
n ExpII OVX + oil Mean SE
n OVX + E2 Mean SE
n
34.8 7.2 11 Saline
TF5
46.6 15.4 8 Gelatin
17.9 2.4 5
8.2 0.1 6
8.0
72.7 16.0 5
79.2 18.1 4
135.5 23.4 4
4
6
9
12
15
18
21
Days Post Immunization FlG. 4. Exp I: anti-FL titers (MONA values) vs. time after immunization. Each point represents the mean ± SEM of 6-11 rats. *, Significant difference from other treatments (P < 0.05, by repeated measures analysis).
trations of E2 measured in terminal serum samples 1.75 h after injection of a 0.25-jug bolus of E2 or the oil vehicle. Groups receiving vehicle injections after OVX had low serum E2 levels. Most of the values for the oil plus TF5 and oil plus gel groups were extrapolated from below the sensitivity of the assay. The groups receiving E2 injections had elevated serum levels of E2. As stated for Exp I, these levels are known from previous studies to remain elevated for 3 h (5). Although some measured serum E2 concentrations were slightly superphysiological, these E2 levels were still within the range shown in previous studies to be immunoenhancing (4). The development of anti-FL titers over time for each of the treatment groups is shown in Fig. 5. The profiles of the responses are similar to each other and to those of previous studies, rising exponentially between days 6
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 March 2015. at 05:35 For personal use only. No other uses without permission. . All rights reserved.
ESTROGEN ENHANCEMENT OF HUMORAL IMMUNITY
1356
2.4 2.2 2.0
—O— --O— —•— •— -•— •—
2.6
OIL, SALINE E 2 , SALINE OIL, TF5 E 2 , TF5 OIL, GELATIN E 2 , GELATIN
2.4
O • D •
Endo • 1991 Voll28-No3
INTACT; I mg/kg TF5 INTACT; 10 mg/kg TF5 OVARIECTOMIZED; I mg/kg TF5 OVARIECTOMIZED; 10 mg/kg TF5
2.2 2.0
•"I* 1.8 ^
1.6
£
1.4
1.8
1.6 1.4
5 '-2
1.2 1.0
I
0.8 Q8 0.6 0.6 0.4 0.4 0.2 0.2 12
15
18
21
Days Post Immunization
FIG. 5. Exp II: anti-FL titers (M0NA values) vs. time after immunization. Each point represents the mean ± SEM of four to six rats. *, Significant difference from other treatments (P < 0.03).
and 9 and reaching their peaks between days 12 and 21. The oil plus TF5 group and the saline- and gelatininjected groups did not differ from one another in antiFL titers on any day. Only the group receiving both TF5 and E2 replacement displayed significantly greater antiFL titers than all other treatment groups (P < 0.03) by day 18. These data are supported by the results of the test of dose response to TF5; the TF5 dose of 1 mg/kg was adequate to increase anti-FL responses in the ovaryintact group alone (Fig. 6). The dose of 10 mg/kg led to no improvement in anti-FL titers over the lower dose, but to an equivalent increase in antibody titers in ovaryintact animals compared to ovariectomized animals. At both levels of TF5 administration, the immunoenhancing effect of the intact ovary was apparent by factorial analysis by day 12 (P < 0.03). In both sets of results, as in Exp I, the presence of both E2 and thymus replacements led to increases in the anti-FL response, which amounted to a doubling in titer over all other treatments.
Discussion The results of the first experiment demonstrate that the thymus is required for the enhancement of in vivo humoral immune responsiveness by estrogen. Previous studies (5) show that the replacement of E2 in ovariectomized rats employed in this experiment significantly increases the titer of antibodies raised in the primary response against FL; the present study shows that the
0
6
9
12
15
18
21
Days Post Immunization
FlG. 6. Exp II: anti-FL titers (MONA values) us. time after immunization. Each point represents the mean ± SEM of three to six rats. Titers of intact groups were greater than those of ovariectomized groups (P < 0.05).
E2 replacement does so only when animals possess an intact thymus. The nature of the requisite thymic function is made clearer in the results of the second experiment, which demonstrate that a component of TF5 exerts a permissive influence on the action of E2 outside the thymus to increase the anti-FL response. Previous literature addressing the question of the necessity of the thymus in the adult are in agreement that THYMX in the adult, in the short term, does not diminish the responsiveness of the humoral branch of the immune system. Miller (37) found no differences in the number of spleen cells secreting antibody against SRBC between mice immunized 1 month after THYMX or sham THYMX. Simpson and Cantor (38) followed the anti-BSA response of thymectomized and sham-thymectomized CBA mice for 4 weeks and found that THYMX did not depress the production of serum antibody to that T-dependent immunogen. However, in the present study THYMX did depress humoral immunity in ovariectomized E2-treated rats relative to that in sham THYMX. These results, then, yield new insight into the necessity of the thymus for humoral immune responsiveness in the adult, i.e. the estrogen environment should be considered when assessing responses with thymic involvement. It is likely that the site of the direct influences of estrogen and the thymic factor is the peripheral immune
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 March 2015. at 05:35 For personal use only. No other uses without permission. . All rights reserved.
ESTROGEN ENHANCEMENT OF HUMORAL IMMUNITY
cell. Other investigators have shown in in vitro experiments that E2 can act directly on spleen cells to increase antibody production. Paavonen et al. (7) and Weetman et al. (8) have reported increases in the number of human peripheral blood lymphocytes secreting subclasses of Ig after incubation with physiological concentrations of E2. Kenney et al. (6) found that incubation of SRBC-primed murine spleen cells with physiological E2 increased the number of antibody-secreting cells in a hemolytic plaque assay. Myers and Petersen (3) observed a similar in vitro treatment increase in the amount of antibody secreted per PFC. In light of the results of the present study, the effects observed on these lymphocytes might well have been due to the joint actions of the steroid and its permissive factor, as these cultures of lymphocytes were conducted in the presence of a possible source of thymic hormones, fetal calf serum. The opportunity for the direct influence of estrogen on lymphocytes is apparent in the discovery of specific estrogen binding in rat spleen (39), mouse B-cells (14), and human peripheral lymphocytes (40) and CD8+ cells (13) and of a characterized estrogen receptor in mouse spleen (9), human spleen (10,11), and human peripheral lymphocytes (10, 12). The subpopulations of lymphocytes bearing receptors for estrogen and the direction of the cells' responsiveness to the steroid may determine the ultimate influence of estrogen on humoral immunity. Stimson (11) measured estrogen receptor in cells of the suppressor/cytotoxic phenotype from human spleen. If estrogen were to depress the activity of suppressor Tcells, that action would contribute to the hormone's effect on antibody production by plasma cells. This influence was first proposed by Paavonen et al. (7). Although firm evidence for this proposal has yet to be obtained, it is a possibility that should still be pursued, as TF5 contains thymosin a7, a peptide that has been shown to affect suppressor activity (25, 41). TF5 is a product of bovine calf thymus, collected when the animal is prepubescent (42). The peptides contained in this fraction are, therefore, unaffected by sex steroid hormones, but are produced by the thymus constitutively. Stimson and Hunter (27, 28) have suggested that estrogen indirectly influences immune functions through the induction of an element from the thymus. They found that sera from estrogen-treated (40 /u.g daily for 2 weeks) thymus-intact male rats decreased the percentage of functional T-cells and increased the percentage of lymphocytes bearing IgM among human peripheral blood lymphocytes in culture. A fraction of the active sera of less than 1000 mol wt when added to human lymphocyte cultures yielded the above activities as well as increased the number of PFCs against SRBCs. These activities were lost in sera from thymectomized rats. Estrogeninduced thymic factors have also been observed for their
1357
effects on mitogenic responses of thymocytes and spleen cells. Grossman et al. (43) found that the serum of castrated rats increased in vitro mitogenic responses of thymocytes to phytohemagglutinin compared to serum from intact rats; they also found that this trend was reversed by serum from estrogen-treated (15 ng daily for 3 days) rats, but not by the addition of E2 (15 pg/ml) directly to the cultures. The depression of phytohemagglutinin responsiveness in estrogen-complete animals was obviated by THYMX. Similar results were reported by Luster et al. (14) for mitogen-stimulated spleen cells of mice. These results again implicate a product of the thymus that is induced by estrogen and affects the responses of lymphoid cells. No such estrogen-inducible element has yet been purified or chemically identified. The results provided by the present study do not deny that in the models in which a thymic estrogen receptor has been observed, a factor may be evolved from the thymus in response to the steroid, and the postulated factor may have an immune activity. However, it is clear that in the model employed in our experiments, an estrogen-induced thymus product has no essential involvement in increasing specific serum antibody titers, as the effect of increased titer caused by the constitutive factor is indiscernible from that of an intact thymus. We conclude from the results of the present study that a constitutive thymic factor, found in TF5, must be present with estrogen in the course of the specific antibody response for the phenomenon of increased in vivo antibody titer to occur. Identification of the thymic factor, its target, and its mechanism of permissive action will further elucidate the role of the thymus in the enhancement of humoral immunity by estrogen.
Acknowledgments The authors are grateful to Dr. Alan L. Goldstein for the generous provision of TF5, and to Dr. David R. Trawick for technical assistance.
References 1. Stern K, Davidsohn I 1955 Effect of estrogen and cortisone on immune haemoantibodies in mice of inbred strains. J Immunol 74:479-484 2. Kenney JF, Gray JA 1971 Sex difference in immunologic response: studies of antibody production by individual spleen cells after stimulus with E. coli antigen. Pediatr Res 5:246-255 3. Myers MJ, Petersen BH 1985 Estradiol induced alterations on the immune system. I. Enhancement of IgM production. Int J Immunopharmacol 7:207-213 4. Trawick DR, Bahr JM 1986 Modulation of the primary and secondary antifluorescyl antibody response in rats by 17/?-estradiol. Endocrinology 118:2324-2330 5. Erbach GT, Bahr JM 1988 Effect of chronic or cyclic exposure to estradiol on the humoral immune response and the thymus. Immunopharmacology 16:45-51 6. Kenney JF, Pangburn PC, Trail G 1976 Effect of estradiol on immune competence: in vivo and in vitro studies. Infect Immunol
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 March 2015. at 05:35 For personal use only. No other uses without permission. . All rights reserved.
1358
ESTROGEN ENHANCEMENT OF HUMORAL IMMUNITY
13:448-456 7. Paavonen T, Andersson LC, Adlercreutz H 1981 Sex hormone regulation of in vitro immune response. Estradiol enhances human B cell maturation via inhibition of suppressor T cells in pokeweed mitogen-stimulated cultures. J Exp Med 154:1935-1945 8. Weetman AP, McGragor AM, Smith BR, Hall R 1981 Sex hormones enhance immunoglobulin synthesis by human peripheral blood lymphocytes. Immunol Lett 3:343-346 9. Detlefsen MA, Smith BC, Dickerman HW 1977 A high-affinity, low-capacity receptor for estradiol in normal and anemic mouse spleen cytosols. Biochem Biophys Res Commun 76:1151-1158 10. Danel L, Souweine G, Monier JC, Saez S 1983 Specific estrogen binding sites in human lymphoid cells and thymic cells. J Steroid Biochem 18:559-563 11. Stimson WH 1988 Oestrogen and human T lymphocytes: presence of specific receptors in the T suppressor/cytotoxic subset. Scand J Immunol 28:345-350 12. Weustein JJAM, Blankenstein MA, Gmeylig-Meyling FHJ, Schuurman HJ, Kater I, Thijssen JHH 1986 Presence of oestrogen receptors in human blood mononuclear cells and thymocytes. Acta Endocrinol (Copenh) 112:409-414 13. Cohen JHM, Danel L, Cordier G, Saez S, Reveillard J-P 1983 Sex steroid receptors in peripheral T cells: absence of androgen receptors and restriction of estrogen receptors to OKT8-positive cells. J Immunol 131:2767-2771 14. Luster MI, Hayes HT, Korach K, Tucker AN, Dean JH, Greenlee WF, Boorman GA 1984 Estrogen immunosuppression is regulated through estrogenic responses in the thymus. J Immunol 133:110116 15. Reichman ME, Villee CA 1978 Estradiol binding by rat thymus cytosol. J Steroid Biochem 9:637-641 16. Grossman CJ, Sholiton LJ, Blaha GC, Nathan P 1979a Rat thymic estrogen receptor-II. Physiological properties. J Steroid Biochem 11:1241-1246 17. Grossman CJ, Sholiton LJ, Nathan P 1979b Rat thymic estrogen receptor-I. Preparation, location and physiochemical properties. J Steroid Biochem 11:1233-1240 18. Gillette S, Gillette RW 1979 Changes in thymic estrogen receptor expression following orchidectomy. Cell Immunol 42:194-196 19. Thompson EA 1981 The effects of estradiol upon the thymus of the sexually immature female mouse. J Steroid Biochem 14:167174 20. Screpanti I, Gulino A, Pasqualini JR 1982 The fetal thymus of guinea pig as an estrogen target organ. Endocrinology 111:15521560 21. Haruki Y, Seiki K, Enomoto T, Fujii H, Sakabe K 1983 Estrogen receptor in the ''non-lymphocytes" in the thymus of the ovariectomized rat. Tokai J Exp Clin Med 8:31-39 22. Nilsson B, Carlsson S, Damber M-G, Lindblom D, Sodergaed R, von Schoulz B 1984 Specific binding of 17/3-estradiol in the human thymus. Am J Obstet Gynecol 149:544-547 23. Jordan RK, Robinson JH 1981 T lymphocyte diferentiation. In: Marion Kendall (ed) The Thymus Gland. Academic Press, London, pp 151-177 24. Dardenne M, Bach J-F 1981 Thymic hormones. In: Marion Kendall (ed) The Thymus Gland. Academic Press, London, pp 113-131
Endo • 1991 Voll28«No3
25. Goldstein AL, Low TLK, Thurman GB, Zatz MM, Hall N, Chen J, Hu S-K, Naylor PB, McClure JE 1981 Current status of thymosin and other hormones of the thymus gland. Recent Prog Horm Res 37:369-412 26. Barnes EW, MacCuish AC, Loudon NB, Jordan J, Irvine WJ 1974 Phytohaemagglutinin-induced lymphocyte transformation and circulating autoantibodies in women taking oral contraceptives. Lancet 1:898-900 27. Stimson WH, Hunter IC 1976 An investigation into the immunosuppressive properties of oestrogen. J Endocrinol 69:42P-43P 28. Stimson WH, Hunter IC 1980 Oestrogen-induced imunoregulation mediated through the thymus. J Clin Lab Immunol 4:27-33 29. Holdstock G, Chastanay BF, Krawitt EL 1982 Effect of testosterone, oestradiol and progesterone on immune responsiveness. Clin Exp Immunol 47:449-456 30. Staples LD, Binns RM, Heap RB 1983 Influence of certain steroids on lymphocyte transformation in sheep and goats studied in vitro. J Endocrinol 98:55-69 31. Myers MJ, Butler LD, Petersen BH 1986 Estradiol-induced alteration in the immune system. II. Suppression of cellular immunity in the rat is not the result of direct estrogenic action. Immunopharmacology 11:47-55 32. Lopatin DE, Voss EW 1971 Fluorescein. Hapten and antibody active site probe. Biochemistry 10:208-213 33. Voss EW 1984 Fluorescein Hapten: An Immunologic Probe. CRC Press, Boca Raton 34. Greenfield RA, Stephens JL, Bussey MJ, Lones JM 1983 Quantitation of antibody to Candidan mamman by enzyme-linked immunosorbent assay. J Lab Clin Med 101:758-771 35. Bahr JM, Ben-Jonathan N 1981 Preovulatory depletion of ovarian catecholamines in the rat. Endocrinology 108:1815-1820 36. SAS Institute Inc 1985 SAS User's Guide: Statistics, ed. 5. SAS Institute, Cary 37. Miller JFAP 1965 Effect of thymectomy in adult mice on immunological responses. Nature 208:1337-1338 38. Simpson E, Cantor H 1975 Regulation of the immune response by subclasses of T lymphocytes II. The effect of adult thymectomy upon humoral and cellular responses in mice. Eur J Immunol 5:337-343 39. Myers MJ, Heim MC, Hirsch KS, Queener SF, Petersen BH 1986 Translocatable estrogen receptors in rat splenocytes. Life Sci 39:313-320 40. Souweine G, Danel L, Costa O, Tubania N, Martin P, Monier JC, Saez S 1980 Effect of physiological concentrations of sex hormones on the formation of "early" sheep red blood cell rosettes by human lymphocytes. Possible relations with the presence of sex hormone cytosol receptors in lymphocytes. Biomedicine 33:150-152 41. Horowitz S, Borcherding W, Moorthy AV, Chesney R, SchulteWisserman H, Hong R 1977 Induction of suppressor T cells in systemic lupus erythematosis by thymosin and cultured thymic epithelium. Science 197:999-1001 42. Hooper JA, McDaniel MC, Thurman GB, Cohen GH, Schulof RS, Goldstein AL 1975 Purification and properties of bovine thymosin. Ann NY Acad Sci 249:125-153 43. Grossman CJ, Sholiton LJ, Roselle GA 1982 Estradiol regulation of thymic lymphocyte function in the rat: mediation by serum thymic factors. J Steroid Biochem 16:683-690
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 March 2015. at 05:35 For personal use only. No other uses without permission. . All rights reserved.
Vol. 128, No. 3 Printed in U.S.A.
Enhancement of in Vivo Humoral Immunity by Estrogen: Permissive Effect of a Thymic Factor* G. T. ERBACH AND JANICE M. BAHR Departments of Physiology (G.T.E., J.M.B.) and Animal Sciences (J.M.B.), University of Illinois, Urbana, Illinois 61801
ABSTRACT. Physiological levels of estrogen enhance humoral immune responses. Several in vitro studies indicate the hormone to have a direct effect on immune cells, and other studies show that estrogen may affect humoral immunity indirectly through the thymus. Therefore, we have conducted experiments to investigate the requirement of the thymus in the enhancement of humoral immune responsiveness by estrogen. In Exp I, adult ovariectomized Lewis rats were thymectomized or sham thymectomized and given estradiol (E2; 0.25 fig E2 in sesame oil, sc, once every 4 days) or the oil vehicle in a 2 X 2 factorial design, and their antifluorescein responses were followed by enzyme-linked immunosorbant assay across 21 days. Only animals that were thymus intact and given estrogen replacement showed significantly (P < 0.05) greater serum anti-
fluorescein titers than all other treatments. In Exp II, ovariectomized thymectomized rats were submitted to a 2 x 3 factorial design of oil vehicle or E2 replacement and saline, gelatin, or thymus replacement (thymosin fraction 5; 1 mg/kg in saline, sc). As described above, only the animals receiving both thymosin fraction 5 and E2 replacement displayed antifluorescein titers that were significantly (P < 0.03) increased over titers of all other treatment groups. These results indicate that the enhancement of in vivo humoral immunity by estrogen requires the thymus, and that a constitutive thymic factor, found in thymosin fraction 5, exerts a permissive influence on the action of E2 outside the thymus to increase a specific humoral immune response. {Endocrinology 128: 1352-1358, 1991)
E
STROGEN has been shown to enhance humoral immune responsiveness in several species and with various antigenic systems (1-5). A number of mechanisms have been proposed for this phenomenon, and several targets for estrogen have been identified within the immune system. In vitro studies indicate the hormone to have a direct influence on lymphocyte function. Kenney et al. (6) incubated sheep red blood cell (SRBC)primed murine spleen cells with various concentrations of estradiol (E2) and found that physiological levels of E2 caused an increase in the number of antibody-secreting cells in a hemolytic plaque assay. Myers and Petersen (3), using a similar experimental system, found E2 to increase not the number of antibody-producing cells, but the amount of antibody produced per plaque-forming colony (PFC). Incubation with E2 also caused an increase in the number of cells secreting immunoglobulin M (IgM) (7) and IgG (8) among human peripheral blood lymphocytes. The detection of receptor for estrogen in cytosol of mouse spleen (9), human spleen (10, 11), and human Received August 6. 1990. Address all correspondence and requests for reprints to: Dr. Janice Bahr, Animal Genetics Laboratory, University of Illinois, 1301 West Lorado Taft, Urbana, Illinois 61801. * Present address: Pathology Research Laboratory, Massachusetts General Hospital, 149 Thirteenth Street, Charlestown, Massachusetts 02129.
peripheral blood lymphocytes (10,12) and the measurement of significant estrogen binding in human peripheral CD8+ cells (13) and murine B cells (14) further support the idea that estrogen may act directly on lymphoid cells to alter their activities and effect enhanced antibody responses. Receptors for estrogen have been characterized and measured also in cells of the epithelial matrix of the thymus (15-22). The thymus orchestrates the functioning of the immune system through its control of T-cell maturation, differentiation, and activity, which it exerts by the production of soluble factors and thymic hormones (23-25). T-Lymphocytes play an important modulatory role in the humoral immune response. Several in vivo and in vitro studies show estrogen to influence only indirectly certain proliferative responses and functional characteristics of lymphoid cells that may affect humoral immunity (7, 26-31), raising the possibility that estrogen may affect humoral immune responsiveness via a thymic route of action. We report in this paper the results of two experiments, the first to determine that estrogen enhancement of in vivo humoral immunity requires the thymus, and the second to demonstrate that a constitutive soluble product of the thymus exerts a permissive influence on the action of estrogen outside the thymus to increase a specific humoral immune response.
1352
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 March 2015. at 05:35 For personal use only. No other uses without permission. . All rights reserved.
ESTROGEN ENHANCEMENT OF HUMORAL IMMUNITY
Materials and Methods Animals Lewis inbred female rats were purchased from Harlan-Sprague Dawley (Indianapolis, IN). They were housed in NIHapproved facilities (14 h of light, 10 h of darkness) and were provided Purina rat chow and water ad libitum. Surgeries Surgeries were performed under aseptic conditions. At 70 days of age, animals were ovariectomized through bilateral incisions under ketamine-acepromazine anaesthesia 21-23 days before immunization. Two weeks after ovariectomy (OVX), rats were again anesthetized, and thymi were removed through a single midline sternal incision. The sham operation consisted of the identical procedure, except that the thymus was not excised. Antigen preparation and immunization Fluorescein (FL) as fluorescein isothiocyanate was covalently conjugated to keyhole limpet hemocyanin (FL-KLH; Sigma Chemical Co., St. Louis, MO) as previously described (4, 32). FL-KLH, a hapten carrier, T-dependent immunogen was chosen for its proven immunogenicity in other systems (33) and for the relatively limited population of antibodies the hapten induces. Rats were immunized against 0.21 mg FLKLH in water in a total volume of 0.25 ml, sc. Determination of antibody titers Blood samples were obtained from the tail and allowed to clot overnight at 4 C. Sera were then frozen and stored at -20 C until assayed. Anti-FL activity in serum samples was determined by specific, solid phase enzyme-linked immunosorbent assay (ELISA). Microtiter plate wells were coated with 1 ng FL conjugated to BSA (Sigma) in 100 /x\ phosphate buffer, pH 8.0, and incubated overnight at 4 C. Wells were washed, and an overcoat of 300 n\ 1.0% BSA, 0.9% NaCl, and 100 Mg/ml thimerosol in 0.01 M Tris, pH 7.5, was applied and incubated for 30 min at 37 C. Wells were then aspirated dry. Serum samples, internal standards, and second antibodies were diluted in overcoat solution plus 0.05% Tween-20. The diluted serum samples containing anti-FL activity, interplate controls, complete titrations of internal standards, and normal sera were then added in 100-/ul volumes to duplicate wells and incubated for 2 h at 37 C. After this incubation and a subsequent wash, 100 /xl horseradish peroxidase-conjugated rabbit antirat Ig (heavy and light chain specific; Accurate Chemical Co., Westbury, NY) diluted 1:1000 were added to each well. The plates were again incubated for 2 h at 37 C. After a final wash, 1.5 mM H2O2 was added to each well with 0.4 mM 2,2'-azinobis-(3ethylbenzthiazoline sulfonic acid) in 0.05 M citrate buffer in a total of 100 n\. 2,2'-Azinobis-(3-ethylbenzthiazoline sulfonic acid develops color in the presence of free oxygen; this reaction was allowed to proceed for 1 h at room temperature, then was stopped by the addition of 100 /xl 0.15 M hydrogen fluoride plus 1.0 mM EDTA. Color development in plate wells was read at 406 nm. Controls for nonspecific binding were included in every trial.
1353
Before the assay of all samples, serial dilutions were carried out on representative antiserum samples from each experimental group to determine the dilution at which the assay end point (optical density) fell on the linear portion of all titration curves. This dilution was found to be 1:400, and in the assay of samples only this dilution was tested. Relative titers of sample antisera were determined by conversion to multiples of normal activity (MONA) by the method of Greenfield et al. (34). The conversion is based on the assumption, borne out in preliminary trials, that all titration curves within an assay are parallel. It follows, then, that from a single point a titration curve can be derived for a tested sample by comparison to a complete titration performed for a pooled sample (internal standard; Fig. 1). The relative sample titer, expressed as MONA, is defined as the ratio of dilutions (3/or. the ratio of /3, the extrapolated dilution of sample antiserum which yields the same optical density as that of the standard antiserum at 1:400, to a, the sample dilution (1:400). Titrations of the internal standard were run in every plate. A different pool of antiserum was used to measure the samples of each experiment; therefore, MONA values can be compared only within an experiment. The intraplate variation was less than 0.1%; the interplate variation was less than 0.5%. E2 replacement and measurements of serum E2 Subcutaneous injections of 0.25 ml of a 1.0 ng/m\ solution of E2 in sesame oil or the oil vehicle alone were administered the day before immunization and every fourth day, as characterized and described previously (5). E2 concentrations were measured by RIA, as previously described (35). The sensitivity of this assay is 8 pg. The E2 antiserum 244 was obtained from G. D. Niswender, Colorado
>•
H CO
z UJ Q
O I-
o.
o.
a
(3
DILUTION FIG. 1. Extrapolation of dilution 0 from a titration curve derived from a single test point. • , Measured data points; O, extrapolated point; , calculated from OD of diluted standard serum; , from test point. Adapted from Greenfield et al. (34).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 March 2015. at 05:35 For personal use only. No other uses without permission. . All rights reserved.
ESTROGEN ENHANCEMENT OF HUMORAL IMMUNITY
1354
State University (Fort Collins, CO). It was prepared against 6-hemisuccinate-17/3-estradiol and cross-reacts 0.44% with estrone. Thymic factor replacement
The epithelial cells of the thymus produce a number of peptides that have been shown to participate in the development of the T-cell system and in the regulation of immune cell function. Many of the activities of the endocrine thymus are present in thymosin fraction 5 (TF5), a semipurified aqueous extract of calf thymus consisting of between 20-30 acidic peptides ranging in mol wt from 3000-7000. As reviewed by Goldstein et al. (25), TF5 has been demonstrated to induce differentiation markers on immature T-cells and to restore immune competence to nude and neonatally thymectomized and adult thymectomized mice. TF5 has been shown also to alter the in vitro responsiveness of lymphoid cells from normal and immune-compromised mice and rats. This hormone preparation provides the best available broad replacement of possible active thymic factors. It was administered to animals by sc injection in saline at a dose of 1 mg/kg, as recommended by Dr. A. L. Goldstein (personal communication). This is generally the therapeutic dose for restoration of immune responsiveness in humans and rodents. Animals were injected with TF5 the day of and the day after E2 administration to assure the presence of thymic factors concurrent with the influence of estrogen on responsive cells. As TF5 is derived from a bovine source and was injected at potentially immunizing doses, controls injected with bovine gelatin (Bio-Rad Laboratories, Richmond, CA) at 1 mg/kg were included to test against general increases in anti-FL responses induced by repeated injections of a foreign protein. TF5 was graciously provided by Dr. Alan L. Goldstein, George Washington University (Washington, DC). Experimental design Exp I. This experiment was designed in a 2 X 2 factorial fashion, testing thymectomy (THYMX) against E2 replacement by injection in ovariectomized rats. Animals were randomly assigned to four treatment groups: 1) OVX, THYMX, and oil injections; 2) OVX, THYMX, and E2 injections; 3) OVX, sham THYMX, and oil injections; and 4) OVX, sham THYMX, and E2 injections. All animals were ovariectomized 2 weeks before THYMX or sham THYMX (Fig. 2). Seven to 10 days lapsed between THYMX and immunization against FL-KLH. Estrogen re-
Endo • 1991 Voll28«No3
placement commenced the day before immunization; sc injections of 0.25 fig E2 or the oil vehicle were given every fourth day. Blood samples (20-40 n\) were taken from the tail 3, 6, 9, 12,15,18, and 21 days after immunization; serum was harvested and assayed for anti-FL activity. On day 23 postimmunization, terminal blood was collected by decapitation 1.75 h after E2 or oil injection for measurement of peak serum E2 concentration. At that time, gross examination was made for ovarian and thymic remnants. Animals with incomplete organ extirpation were removed from the study. Exp II. This experiment was designed in a 3 X 2 factorial fashion, testing TF5 and E2 replacement in ovariectomizedthymectomized rats. Animals were randomly assigned to six treatment groups: 1) OVX, THYMX, oil injections and saline injections; 2) OVX, THYMX, E2 injections and saline injections; 3) OVX, THYMX, oil injections and TF5 injections; 4) OVX, THYMX, E2 injections and TF5 injections; 5) OVX, THYMX, oil injections and gelatin injections; and 6) OVX, THYMX, E2 injections and gelatin injections. All animals were ovariectomized 2 weeks before THYMX (Fig. 3). As in Exp I, 7-10 days lapsed between thymectomy and immunization, and estrogen replacement commenced the day before immunization. Injections of TF5, gelatin, or saline were administered in a volume of 0.2 ml the day of and the day after each E2 or oil injection. Blood samples were taken and assayed as described for Exp I. In a parallel study the dose response of ovariectomized and ovary-intact rats to TF5 was determined for anti-FL titers using doses of 1 and 10 mg/kg. TF5 was again prepared in saline and injected in a volume of 0.2 ml, sc, according to the schedule of treatments described by Fig. 3. Statistics Differences among means were determined by repeated measures analysis, factorial analysis, and analysis of variance using the SAS program (36).
Results Exp I The purpose of Exp I was to determine if the thymus is required for the enhancement of in vivo humoral EXPERIMENT2 SCHEDULE OF TREATMENTS
EXPERIMENT 1
» TF5 OR SALINE OR GELATIN INJECTION
SCHEDULE OF TREATMENTS
| E 2 OR OIL INJECTION
\ E:2 OR OIL INJECTION • OVX •
THYMX 1
BLOOD SAMPLING
•
OVX
THYMX
11 MM | ) | \ \ \ 1 I I I I I I I I I I I I I I i i i I i i i t i I
7-10 DAYS
llMM -I 0
I
I I I I I I t I I
(9 QU
1 14 DAYS
14 DAYS
BLOOD SAMPLING
7-10 DAYS
DAYS DAYS
FIG. 2. Exp I: schedule of treatments. Arrow indicates injection; box indicates blood sample. Immunization on day 0.
FIG. 3. Exp II: schedule of treatments. Arrow indicates E2 or control injection; asterisk indicates TF5 or control injection; box indicates blood sample. Immunization on day 0.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 March 2015. at 05:35 For personal use only. No other uses without permission. . All rights reserved.
ESTROGEN ENHANCEMENT OF HUMORAL IMMUNITY
immunity. E2 replacement in ovariectomized rats was fashioned to mimic the level and pattern of E2 surges of the 4-day estrous cycle. Shown in Table 1 are the E2 concentrations measured in terminal serum samples 1.75 h after the final hormone or vehicle injection. Groups receiving oil injections had serum E2 concentrations of approximately 7 pg/ml; groups receiving E2 injections experienced E2 surges of approximately 40 pg/ml, according to these measurements and previous studies (5), for 3 h once every four days. This means of E2 replacement is a potent enhancer of humoral immunity (5). The anti-FL responses for each of the treatment groups is shown in Fig. 4. The profiles of the responses of all of the groups were similar; all had titers equivalent to normal serum on day 3, then showed a gradual increase across 21 days. Thymectomized and oil, thymectomized and E2, and sham thymectomized and oil groups did not differ in anti-FL responses on any day. Only the group that was thymus intact and receiving E2 replacement showed a significantly increased titer (P < 0.05) on day 21. Thymic and uterine wet weights were also observed at the conclusion of the experiment. Four-day injections of 0.25 ng E2 did not cause involution of the thymus, but did significantly increase (P < 0.001) the growth of the uterus compared to that caused by injection of the vehicle. These data support previous observations made regarding the trophic responses of these tissues to the 4day injection regimen (5).
1355
THYMX + OIL, n = 6 9.0
THYMX + E 2 , n=ll SHAM THYMX+OIL, n = 9 SHAM THYMX + E 2 , n = 8
8.0
7.0
f
6.0
1
f
4.0
3.0
2.0
1.0
ExpII This experiment was designed to test thymus replacement against physiological estrogen, administered by injection once every 4 days. Table 1 presents the concenTABLE 1. E2 concentrations (picograms per ml) in terminal serum samples 1.75 h after injection of 0.25 /xg E2 or vehicle Treatment
Thymectomized
Thymus-intact
8.2 0.6 6
6.2 0.5 9
Expl OVX + oil Mean SE
n OVX + E2 Mean SE
n ExpII OVX + oil Mean SE
n OVX + E2 Mean SE
n
34.8 7.2 11 Saline
TF5
46.6 15.4 8 Gelatin
17.9 2.4 5
8.2 0.1 6
8.0
72.7 16.0 5
79.2 18.1 4
135.5 23.4 4
4
6
9
12
15
18
21
Days Post Immunization FlG. 4. Exp I: anti-FL titers (MONA values) vs. time after immunization. Each point represents the mean ± SEM of 6-11 rats. *, Significant difference from other treatments (P < 0.05, by repeated measures analysis).
trations of E2 measured in terminal serum samples 1.75 h after injection of a 0.25-jug bolus of E2 or the oil vehicle. Groups receiving vehicle injections after OVX had low serum E2 levels. Most of the values for the oil plus TF5 and oil plus gel groups were extrapolated from below the sensitivity of the assay. The groups receiving E2 injections had elevated serum levels of E2. As stated for Exp I, these levels are known from previous studies to remain elevated for 3 h (5). Although some measured serum E2 concentrations were slightly superphysiological, these E2 levels were still within the range shown in previous studies to be immunoenhancing (4). The development of anti-FL titers over time for each of the treatment groups is shown in Fig. 5. The profiles of the responses are similar to each other and to those of previous studies, rising exponentially between days 6
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 March 2015. at 05:35 For personal use only. No other uses without permission. . All rights reserved.
ESTROGEN ENHANCEMENT OF HUMORAL IMMUNITY
1356
2.4 2.2 2.0
—O— --O— —•— •— -•— •—
2.6
OIL, SALINE E 2 , SALINE OIL, TF5 E 2 , TF5 OIL, GELATIN E 2 , GELATIN
2.4
O • D •
Endo • 1991 Voll28-No3
INTACT; I mg/kg TF5 INTACT; 10 mg/kg TF5 OVARIECTOMIZED; I mg/kg TF5 OVARIECTOMIZED; 10 mg/kg TF5
2.2 2.0
•"I* 1.8 ^
1.6
£
1.4
1.8
1.6 1.4
5 '-2
1.2 1.0
I
0.8 Q8 0.6 0.6 0.4 0.4 0.2 0.2 12
15
18
21
Days Post Immunization
FIG. 5. Exp II: anti-FL titers (M0NA values) vs. time after immunization. Each point represents the mean ± SEM of four to six rats. *, Significant difference from other treatments (P < 0.03).
and 9 and reaching their peaks between days 12 and 21. The oil plus TF5 group and the saline- and gelatininjected groups did not differ from one another in antiFL titers on any day. Only the group receiving both TF5 and E2 replacement displayed significantly greater antiFL titers than all other treatment groups (P < 0.03) by day 18. These data are supported by the results of the test of dose response to TF5; the TF5 dose of 1 mg/kg was adequate to increase anti-FL responses in the ovaryintact group alone (Fig. 6). The dose of 10 mg/kg led to no improvement in anti-FL titers over the lower dose, but to an equivalent increase in antibody titers in ovaryintact animals compared to ovariectomized animals. At both levels of TF5 administration, the immunoenhancing effect of the intact ovary was apparent by factorial analysis by day 12 (P < 0.03). In both sets of results, as in Exp I, the presence of both E2 and thymus replacements led to increases in the anti-FL response, which amounted to a doubling in titer over all other treatments.
Discussion The results of the first experiment demonstrate that the thymus is required for the enhancement of in vivo humoral immune responsiveness by estrogen. Previous studies (5) show that the replacement of E2 in ovariectomized rats employed in this experiment significantly increases the titer of antibodies raised in the primary response against FL; the present study shows that the
0
6
9
12
15
18
21
Days Post Immunization
FlG. 6. Exp II: anti-FL titers (MONA values) us. time after immunization. Each point represents the mean ± SEM of three to six rats. Titers of intact groups were greater than those of ovariectomized groups (P < 0.05).
E2 replacement does so only when animals possess an intact thymus. The nature of the requisite thymic function is made clearer in the results of the second experiment, which demonstrate that a component of TF5 exerts a permissive influence on the action of E2 outside the thymus to increase the anti-FL response. Previous literature addressing the question of the necessity of the thymus in the adult are in agreement that THYMX in the adult, in the short term, does not diminish the responsiveness of the humoral branch of the immune system. Miller (37) found no differences in the number of spleen cells secreting antibody against SRBC between mice immunized 1 month after THYMX or sham THYMX. Simpson and Cantor (38) followed the anti-BSA response of thymectomized and sham-thymectomized CBA mice for 4 weeks and found that THYMX did not depress the production of serum antibody to that T-dependent immunogen. However, in the present study THYMX did depress humoral immunity in ovariectomized E2-treated rats relative to that in sham THYMX. These results, then, yield new insight into the necessity of the thymus for humoral immune responsiveness in the adult, i.e. the estrogen environment should be considered when assessing responses with thymic involvement. It is likely that the site of the direct influences of estrogen and the thymic factor is the peripheral immune
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 March 2015. at 05:35 For personal use only. No other uses without permission. . All rights reserved.
ESTROGEN ENHANCEMENT OF HUMORAL IMMUNITY
cell. Other investigators have shown in in vitro experiments that E2 can act directly on spleen cells to increase antibody production. Paavonen et al. (7) and Weetman et al. (8) have reported increases in the number of human peripheral blood lymphocytes secreting subclasses of Ig after incubation with physiological concentrations of E2. Kenney et al. (6) found that incubation of SRBC-primed murine spleen cells with physiological E2 increased the number of antibody-secreting cells in a hemolytic plaque assay. Myers and Petersen (3) observed a similar in vitro treatment increase in the amount of antibody secreted per PFC. In light of the results of the present study, the effects observed on these lymphocytes might well have been due to the joint actions of the steroid and its permissive factor, as these cultures of lymphocytes were conducted in the presence of a possible source of thymic hormones, fetal calf serum. The opportunity for the direct influence of estrogen on lymphocytes is apparent in the discovery of specific estrogen binding in rat spleen (39), mouse B-cells (14), and human peripheral lymphocytes (40) and CD8+ cells (13) and of a characterized estrogen receptor in mouse spleen (9), human spleen (10,11), and human peripheral lymphocytes (10, 12). The subpopulations of lymphocytes bearing receptors for estrogen and the direction of the cells' responsiveness to the steroid may determine the ultimate influence of estrogen on humoral immunity. Stimson (11) measured estrogen receptor in cells of the suppressor/cytotoxic phenotype from human spleen. If estrogen were to depress the activity of suppressor Tcells, that action would contribute to the hormone's effect on antibody production by plasma cells. This influence was first proposed by Paavonen et al. (7). Although firm evidence for this proposal has yet to be obtained, it is a possibility that should still be pursued, as TF5 contains thymosin a7, a peptide that has been shown to affect suppressor activity (25, 41). TF5 is a product of bovine calf thymus, collected when the animal is prepubescent (42). The peptides contained in this fraction are, therefore, unaffected by sex steroid hormones, but are produced by the thymus constitutively. Stimson and Hunter (27, 28) have suggested that estrogen indirectly influences immune functions through the induction of an element from the thymus. They found that sera from estrogen-treated (40 /u.g daily for 2 weeks) thymus-intact male rats decreased the percentage of functional T-cells and increased the percentage of lymphocytes bearing IgM among human peripheral blood lymphocytes in culture. A fraction of the active sera of less than 1000 mol wt when added to human lymphocyte cultures yielded the above activities as well as increased the number of PFCs against SRBCs. These activities were lost in sera from thymectomized rats. Estrogeninduced thymic factors have also been observed for their
1357
effects on mitogenic responses of thymocytes and spleen cells. Grossman et al. (43) found that the serum of castrated rats increased in vitro mitogenic responses of thymocytes to phytohemagglutinin compared to serum from intact rats; they also found that this trend was reversed by serum from estrogen-treated (15 ng daily for 3 days) rats, but not by the addition of E2 (15 pg/ml) directly to the cultures. The depression of phytohemagglutinin responsiveness in estrogen-complete animals was obviated by THYMX. Similar results were reported by Luster et al. (14) for mitogen-stimulated spleen cells of mice. These results again implicate a product of the thymus that is induced by estrogen and affects the responses of lymphoid cells. No such estrogen-inducible element has yet been purified or chemically identified. The results provided by the present study do not deny that in the models in which a thymic estrogen receptor has been observed, a factor may be evolved from the thymus in response to the steroid, and the postulated factor may have an immune activity. However, it is clear that in the model employed in our experiments, an estrogen-induced thymus product has no essential involvement in increasing specific serum antibody titers, as the effect of increased titer caused by the constitutive factor is indiscernible from that of an intact thymus. We conclude from the results of the present study that a constitutive thymic factor, found in TF5, must be present with estrogen in the course of the specific antibody response for the phenomenon of increased in vivo antibody titer to occur. Identification of the thymic factor, its target, and its mechanism of permissive action will further elucidate the role of the thymus in the enhancement of humoral immunity by estrogen.
Acknowledgments The authors are grateful to Dr. Alan L. Goldstein for the generous provision of TF5, and to Dr. David R. Trawick for technical assistance.
References 1. Stern K, Davidsohn I 1955 Effect of estrogen and cortisone on immune haemoantibodies in mice of inbred strains. J Immunol 74:479-484 2. Kenney JF, Gray JA 1971 Sex difference in immunologic response: studies of antibody production by individual spleen cells after stimulus with E. coli antigen. Pediatr Res 5:246-255 3. Myers MJ, Petersen BH 1985 Estradiol induced alterations on the immune system. I. Enhancement of IgM production. Int J Immunopharmacol 7:207-213 4. Trawick DR, Bahr JM 1986 Modulation of the primary and secondary antifluorescyl antibody response in rats by 17/?-estradiol. Endocrinology 118:2324-2330 5. Erbach GT, Bahr JM 1988 Effect of chronic or cyclic exposure to estradiol on the humoral immune response and the thymus. Immunopharmacology 16:45-51 6. Kenney JF, Pangburn PC, Trail G 1976 Effect of estradiol on immune competence: in vivo and in vitro studies. Infect Immunol
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 March 2015. at 05:35 For personal use only. No other uses without permission. . All rights reserved.
1358
ESTROGEN ENHANCEMENT OF HUMORAL IMMUNITY
13:448-456 7. Paavonen T, Andersson LC, Adlercreutz H 1981 Sex hormone regulation of in vitro immune response. Estradiol enhances human B cell maturation via inhibition of suppressor T cells in pokeweed mitogen-stimulated cultures. J Exp Med 154:1935-1945 8. Weetman AP, McGragor AM, Smith BR, Hall R 1981 Sex hormones enhance immunoglobulin synthesis by human peripheral blood lymphocytes. Immunol Lett 3:343-346 9. Detlefsen MA, Smith BC, Dickerman HW 1977 A high-affinity, low-capacity receptor for estradiol in normal and anemic mouse spleen cytosols. Biochem Biophys Res Commun 76:1151-1158 10. Danel L, Souweine G, Monier JC, Saez S 1983 Specific estrogen binding sites in human lymphoid cells and thymic cells. J Steroid Biochem 18:559-563 11. Stimson WH 1988 Oestrogen and human T lymphocytes: presence of specific receptors in the T suppressor/cytotoxic subset. Scand J Immunol 28:345-350 12. Weustein JJAM, Blankenstein MA, Gmeylig-Meyling FHJ, Schuurman HJ, Kater I, Thijssen JHH 1986 Presence of oestrogen receptors in human blood mononuclear cells and thymocytes. Acta Endocrinol (Copenh) 112:409-414 13. Cohen JHM, Danel L, Cordier G, Saez S, Reveillard J-P 1983 Sex steroid receptors in peripheral T cells: absence of androgen receptors and restriction of estrogen receptors to OKT8-positive cells. J Immunol 131:2767-2771 14. Luster MI, Hayes HT, Korach K, Tucker AN, Dean JH, Greenlee WF, Boorman GA 1984 Estrogen immunosuppression is regulated through estrogenic responses in the thymus. J Immunol 133:110116 15. Reichman ME, Villee CA 1978 Estradiol binding by rat thymus cytosol. J Steroid Biochem 9:637-641 16. Grossman CJ, Sholiton LJ, Blaha GC, Nathan P 1979a Rat thymic estrogen receptor-II. Physiological properties. J Steroid Biochem 11:1241-1246 17. Grossman CJ, Sholiton LJ, Nathan P 1979b Rat thymic estrogen receptor-I. Preparation, location and physiochemical properties. J Steroid Biochem 11:1233-1240 18. Gillette S, Gillette RW 1979 Changes in thymic estrogen receptor expression following orchidectomy. Cell Immunol 42:194-196 19. Thompson EA 1981 The effects of estradiol upon the thymus of the sexually immature female mouse. J Steroid Biochem 14:167174 20. Screpanti I, Gulino A, Pasqualini JR 1982 The fetal thymus of guinea pig as an estrogen target organ. Endocrinology 111:15521560 21. Haruki Y, Seiki K, Enomoto T, Fujii H, Sakabe K 1983 Estrogen receptor in the ''non-lymphocytes" in the thymus of the ovariectomized rat. Tokai J Exp Clin Med 8:31-39 22. Nilsson B, Carlsson S, Damber M-G, Lindblom D, Sodergaed R, von Schoulz B 1984 Specific binding of 17/3-estradiol in the human thymus. Am J Obstet Gynecol 149:544-547 23. Jordan RK, Robinson JH 1981 T lymphocyte diferentiation. In: Marion Kendall (ed) The Thymus Gland. Academic Press, London, pp 151-177 24. Dardenne M, Bach J-F 1981 Thymic hormones. In: Marion Kendall (ed) The Thymus Gland. Academic Press, London, pp 113-131
Endo • 1991 Voll28«No3
25. Goldstein AL, Low TLK, Thurman GB, Zatz MM, Hall N, Chen J, Hu S-K, Naylor PB, McClure JE 1981 Current status of thymosin and other hormones of the thymus gland. Recent Prog Horm Res 37:369-412 26. Barnes EW, MacCuish AC, Loudon NB, Jordan J, Irvine WJ 1974 Phytohaemagglutinin-induced lymphocyte transformation and circulating autoantibodies in women taking oral contraceptives. Lancet 1:898-900 27. Stimson WH, Hunter IC 1976 An investigation into the immunosuppressive properties of oestrogen. J Endocrinol 69:42P-43P 28. Stimson WH, Hunter IC 1980 Oestrogen-induced imunoregulation mediated through the thymus. J Clin Lab Immunol 4:27-33 29. Holdstock G, Chastanay BF, Krawitt EL 1982 Effect of testosterone, oestradiol and progesterone on immune responsiveness. Clin Exp Immunol 47:449-456 30. Staples LD, Binns RM, Heap RB 1983 Influence of certain steroids on lymphocyte transformation in sheep and goats studied in vitro. J Endocrinol 98:55-69 31. Myers MJ, Butler LD, Petersen BH 1986 Estradiol-induced alteration in the immune system. II. Suppression of cellular immunity in the rat is not the result of direct estrogenic action. Immunopharmacology 11:47-55 32. Lopatin DE, Voss EW 1971 Fluorescein. Hapten and antibody active site probe. Biochemistry 10:208-213 33. Voss EW 1984 Fluorescein Hapten: An Immunologic Probe. CRC Press, Boca Raton 34. Greenfield RA, Stephens JL, Bussey MJ, Lones JM 1983 Quantitation of antibody to Candidan mamman by enzyme-linked immunosorbent assay. J Lab Clin Med 101:758-771 35. Bahr JM, Ben-Jonathan N 1981 Preovulatory depletion of ovarian catecholamines in the rat. Endocrinology 108:1815-1820 36. SAS Institute Inc 1985 SAS User's Guide: Statistics, ed. 5. SAS Institute, Cary 37. Miller JFAP 1965 Effect of thymectomy in adult mice on immunological responses. Nature 208:1337-1338 38. Simpson E, Cantor H 1975 Regulation of the immune response by subclasses of T lymphocytes II. The effect of adult thymectomy upon humoral and cellular responses in mice. Eur J Immunol 5:337-343 39. Myers MJ, Heim MC, Hirsch KS, Queener SF, Petersen BH 1986 Translocatable estrogen receptors in rat splenocytes. Life Sci 39:313-320 40. Souweine G, Danel L, Costa O, Tubania N, Martin P, Monier JC, Saez S 1980 Effect of physiological concentrations of sex hormones on the formation of "early" sheep red blood cell rosettes by human lymphocytes. Possible relations with the presence of sex hormone cytosol receptors in lymphocytes. Biomedicine 33:150-152 41. Horowitz S, Borcherding W, Moorthy AV, Chesney R, SchulteWisserman H, Hong R 1977 Induction of suppressor T cells in systemic lupus erythematosis by thymosin and cultured thymic epithelium. Science 197:999-1001 42. Hooper JA, McDaniel MC, Thurman GB, Cohen GH, Schulof RS, Goldstein AL 1975 Purification and properties of bovine thymosin. Ann NY Acad Sci 249:125-153 43. Grossman CJ, Sholiton LJ, Roselle GA 1982 Estradiol regulation of thymic lymphocyte function in the rat: mediation by serum thymic factors. J Steroid Biochem 16:683-690
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 07 March 2015. at 05:35 For personal use only. No other uses without permission. . All rights reserved.
