T lymphocytes are not the target for estradiol-mediated suppression of DTH in reconstituted female severe combined immunodeficient (SCID) mice

Abstract

Oestrogen has the capacity to suppress T cell-dependent DTH. To explore the mechanisms whereby oestrogen exerts its effects on the immune system we have used SCID mice which are largely devoid of functional T and B lymphocytes, hence being unable to raise DTH, but display intact antigen-presenting capacity. Transfer of lymphocytes to SCID mice restores the DTH capacity. In order to analyse if oestrogen down-regulates DTH by a direct action on T cells we reconstituted SCID mice with either splenocytes or thymocytes from congenic C.B-17 or allogeneic B6 donor mice. Either donor or recipient mice were exposed to estradiol before cell transfer. DTH response was registered in recipient SCID mice 1 and 3 weeks after challenge with oxazolone (OXA). SCID mice receiving estradiol-exposed spleen cells from congenic or allogeneic donor mice displayed lower DTH responses compared with control mice. In contrast, SCID mice receiving estradiol-exposed thymocytes from congenic donor mice showed no significant difference in DTH response compared with control mice. Estradiol-treated SCID mice, transferred with either spleen cells or thymocytes from congenic, hormonally non-treated donors, displayed a significantly lower DTH response compared with control mice. In contrast, estradiol-treated SCID mice receiving hormonally non-treated allogeneic spleen cells showed no difference in DTH response compared with control mice. The results show that T lymphocytes are not the target cell population for estradiol-mediated suppression of DTH in reconstituted female SCID mice.

INTRODUCTION

It has been demonstrated in numerous studies that oestrogen can alter immune responsiveness [1]. Oestrogen has been reported to increase the interferon-gamma (IFN-γ) promoter in lymphoid cells [2] and to increase serum levels of tumour necrosis factor (TNF) but decrease IL-6 levels [3]. Hughes et al. suggested in a newly published study that oestrogen may prevent excessive bone loss by limiting osteoclast life span through promotion of apoptosis and that this effect is mediated by transforming growth factor-beta (TGF-β) [4]. Conflicting results have been reported with respect to the ability of oestrogen to alter endothelial cell adhesion molecule expression [5,6]. We have previously shown that exposure to physiological doses of estradiol in susceptible mouse strains suppresses hapten-specific DTH, a T cell-dependent inflammation [7,8]. Further studies revealed that oestrogen-mediated suppression of DTH is inherited as a single dominant trait without linkage to the MHC (H-2) and is equally expressed in female and male mice [7,9]. In contrast, B cell responses are stimulated by exposure to estradiol [1,8]. The dichotomous effect of estradiol on T and B cell responses results in an altered development of certain experimental autoimmune diseases. Thus, administration of oestrogen ameliorates several T cell-mediated autoimmune diseases but aggravates immune complex-mediated diseases [10]. It is not known whether the effects of oestrogen on immune responsiveness result from direct action on T and B cells, or if they are mediated through other cell populations.

Mice homozygous for the SCID mutation are largely devoid of functional B and T lymphocytes and expression of CD4 and CD8 [11]. The mutation appears to impair the recombination of antigen receptor genes and thereby causes an arrest in the early development of B and T lineage committed cells. These mice are essentially unable to produce antibodies and to reject allogeneic grafts [12]. In contrast, macrophages with characteristic morphology, adherence and receptor expression are present in spleen, liver and peritoneal cavity [12]. In addition, their antigen-presenting cell (APC) function [13] and killer cell activity [14] are also unimpaired. Thus, the SCID mouse offers a unique in vivo environment to study APC function, in the absence of functional autologous T cells.

In a recently published study we have shown that transfer of non-fractionated spleen cells from congenic and non-congenic mice restored the capacity to induce DTH in C.B-17 scid/scid (SCID) mice [15]. In the present study we have used this transfer system in order to elucidate whether the inhibitory effect of estradiol on T cell-dependent inflammation is mediated via direct effect on the T lymphocyte population.

MATERIALS AND METHODS

Mice

Female, inbred C.B-17 scid/scid H-2^d (SCID), C.B-17 non-scid H-2^d and C57Bl/6 H-2^b(B6), originally obtained from Bomholtgård Ltd (Ry, Denmark), were bred and maintained under sterile conditions, using a specially adapted laminar flow cabinet and filter-top cages, in the animal facility of the Department of Rheumatology, University of Gothenburg. Three to 10 mice were housed in each cage under standard conditions of temperature and light. They were fed with sterile food and water. A small proportion of SCID mice are said to be ‘leaky’, producing low levels of immunoglobulin [16]. Such mice were identified by estimation of serum IgG and IgM (> 5 μg/ml) by ELISA (see below) prior to the initiation of the experiment and excluded from the study.

Castration

All mice included in the study were oophorectomized. Castrations were performed in young but sexually mature female mice (7–9 weeks of age). Ovaries were removed after a flank incision. The operations were carried out under pentobarbital anaesthesia.

Hormone treatment

Mice were treated with high or low doses of estradiol. High dose exposure was obtained by filling 5 mm long silastic tubes with 2.5 mg 17β-estradiol (Sigma, St Louis, MO). The tube was placed under the skin after a flank incision and thereafter closed with a metallic clip. Tubes were left under the skin for 1 week. Such treatment resulted in very high levels of 17β-estradiol [8]. Low dose oestrogen treatment was obtained by a single s.c. injection of 1, 3 or 15 μg estradiol dissolved in olive oil (Apoteksbolaget, Göteborg, Sweden). Control mice were implanted subcutaneously with empty silastic tubes or injected subcutaneously with olive oil.

In order to determine the serum levels of estradiol in mice we injected castrated female mice subcutaneously with 3 or 15 μg of estradiol and serum samples were collected after 1, 3 and 4 days. Serum levels of 17β-estradiol were measured by a standard radioimmunoassay (RIA) technique (Sorin Biomedica Diagnostics, Saluggia, Italy).

Reconstitution

Spleen cell suspensions were prepared from C.B-17 and B6 mice, as previously described [15]. Briefly, spleens were teased with forceps and passed through a nylon sieve. The cells were suspended in PBS and centrifuged at 515 g for 5 min. The pelleted cells were resuspended in Tris-buffered 0.83% ammonium chloride to lyse erythrocytes. After washing twice in PBS, the mononuclear cells were counted and viability was assessed by trypan blue dye exclusion. Various numbers of these cells were injected intraperitoneally into SCID mice. In a control experiment using FACS analysis, as previously described [15], the phenotypes of spleen cells from estradiol-exposed and control mice were analysed for expression of B cell, T cell and macrophage surface markers. The percentages of these markers were not altered in mice exposed to estradiol compared with controls.

Thymocyte suspensions were prepared from C.B-17 mice. Thymuses were teased with forceps and passed through a nylon sieve. Cell suspensions were kept in Petri dishes for 2 h, to allow monocytes adhere to the plastic surface. Then the non-adherent cells were passed twice through nylon wool columns, to deplete B cells. SCID mice were then injected intraperitoneally with thymic T cells.

The efficiency of the purification procedure was controlled by FACS analysis of CD4, CD8 and immunoglobulin expression. The purified thymus cell suspensions contained > 99% of T cells and did not contain any immunoglobulin-expressing cells. High doses of estradiol profoundly decreased the percentage of double-positive T cells. There was a concomitant increase of single-positive T cells, expressing approx. 60% CD4 and 40% CD8.

Using the magnetic cell separation system [17] T cells from spleen were enriched using mouse anti-Thy1.2 (CD90)-coupled magnetic beads (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany). Briefly, single-cell suspension was obtained by standard preparation from spleen. Cells were resuspended in PBS containing buffer supplemented with 5 mm EDTA and 0.5% bovine serum albumin (BSA) and incubated with mouse anti-Thy1.2 microbeads (Miltenyi Biotec) for 15 min at 8°C. After washing the cells were separated by filtration through a magnetized MACS Column (Miltenyi Biotec). The efficiency of the purification procedure was controlled by FACS analysis and the purified cell suspension contained > 90% T cells (data not shown). Each SCID recipient received i.p. injection with 5 × 10⁶ cells.

Immunization procedure and registration of DTH reaction

The day after cell transfer mice were sensitized by epicutaneous application of 150 μl of a mixture of absolute ethanol and acetone (3:1) containing 3% 4-ethoxymethylene-2-phenyloxazolone (OXA; Sigma) on shaved abdomen and thorax skin. Six days after sensitization all the mice were challenged by topical application of 25 μl 1% OXA dissolved in olive oil on both sides of the right ear [7]. The thickness of the ear was measured before and 24 h after challenge using an Oditest spring calliper (Kröplin, Hessen, Germany) [18]. All challenges and measurements were performed under light pentobarbital anaesthesia. The intensity of the DTH reaction was expressed as ear thickness 24 h after challenge − thickness before challenge × 10⁻³ cm units. Three weeks after sensitization, each mouse was rechallenged with OXA topically on the left ear and DTH reactivity was measured again.

Serological assays

Total serum levels of IgG and IgM were measured by an ELISA, as previously described [15]. Briefly, 96-well microplates were coated overnight at 4°C with goat anti-mouse IgG (5 μg/ml) and goat anti-mouse IgM (5 μg/ml) (Southern Biotechnology Inc., Birmingham, AL). After washing three times in PBS and blocking with PBS containing 1% BSA, serial dilutions of sera from immunized mice were incubated overnight at 4°C. After washing, plates were stepwise incubated with alkaline phosphatase-labelled goat anti-mouse IgG or IgM (1 μg/ml; Southern Biotechnology) diluted in PBS–Tween 20 for 2 h at 20°C and p-nitrophenyl-phosphate in 10% diethanolamine buffer pH 9.8. The absorbance (405 nm) was recorded after 20–30 min in a Titertek (Flow Labs, McLean, VA) scan. The optical density (OD) values registered were related to the OD values obtained from calibration curves with known concentrations of mouse IgG and IgM (Southern Biotechnology).

Statistical analysis

All statistical analyses were made by two-tailed Student's t-test. All values are presented as means ± s.d. P < 0.05 was considered statistically significant.

RESULTS

In order to evaluate if the C.B-17 strain is susceptible for oestrogen-mediated suppression of DTH we exposed healthy C.B-17 mice to high dose of estradiol and registered DTH reactivity. Figure 1 shows that mice exposed to estradiol displayed a statistically significantly lower DTH reaction compared with controls implanted with empty tubes after both 1 and 3 weeks. We have previously shown that B6 mice display suppressed DTH after exposure to estradiol [7].

Fig. 1.

DTH reactivity as measured by increase in ear thickness and registered 1 and 3 weeks after sensitization with oxazolone (OXA) in healthy C.B-17 mice continuously treated with estradiol-containing tubes (▪) compared with control mice with empty tubes (□). Bars represent mean ± s.d. Each group consisted of three to four animals. *P < 0.05; **P < 0.01.

Treatment of donor mice

(i) Congenic donor C.B-17 mice were injected with a single dose of 1, 3 or 15 μg estradiol 3 days before cell preparation. Control donor mice received a single injection with olive oil. Spleen cell suspensions were made from these mice and 10 × 10⁶ cells were injected intraperitoneally into 40 recipient SCID mice, 10 in each group. Twenty-four hours later the recipient mice were epicutaneously sensitized with OXA. One and 3 weeks later all the recipients were challenged with OXA on both sides of one ear. DTH responses were equally pronounced in all SCID recipients after both 1 and 3 weeks, irrespective of hormonal status of the donor (data not shown).

(ii), (iii) Congenic donor C.B-17 mice were treated with estradiol-containing tubes, implanted under the skin 1 week before cell preparation. Control mice were implanted with empty tubes. Splenocytes and thymocytes were then obtained and 5 × 10⁶ spleen cells were injected intraperitoneally into each of 20 recipient SCID mice and 7 × 10⁶ thymocytes were injected intraperitoneally into each of 37 recipient SCID mice. Twenty-four hours later the recipient mice were epicutaneously sensitized with OXA and after another 1 and 3 weeks challenged with OXA on both sides of one ear. Figure 2 shows that SCID mice receiving splenocytes from estradiol-exposed donors displayed 3 weeks later statistically significantly lower DTH compared with controls receiving splenocytes from donor mice implanted with empty tubes (P < 0.05). In contrast, transfer of congenic thymocytes resulted early in statistically significantly higher DTH reaction among the SCID recipients receiving estradiol-exposed thymocytes (P < 0.05). This difference was not seen after 3 weeks (Fig. 2).

Fig. 2.

DTH responses, measured 1 and 3 weeks after sensitization with oxazolone (OXA), in recipient SCID mice transferred with freshly isolated spleen cells, thymus-derived T cells or T cells from spleen of congenic C.B-17 mice, or freshly isolated spleen cells from allogeneic B6 mice. Donor mice were previously exposed to high-dose estradiol. Each group consisted of 10–20 animals. Bars represent mean ± s.d. *P < 0.05; ***P < 0.001.

(iv) Non-congenic B6 (H-2^b) mice were exposed to high-dose estradiol during 1 week before cell preparation. Control mice received empty tubes. Spleen cell suspensions were prepared and 15 × 10⁶ cells were injected intraperitoneally into 20 recipient SCID mice. Twenty-four hours later the recipient mice were epicutaneously sensitized with OXA and after another 1 and 3 weeks challenged with OXA on both sides of one ear. After 1 week we found a tendency to lower DTH reaction in SCID recipients that were transferred with cells from estradiol-exposed donors compared with controls. This difference was statistically highly significant after 3 weeks (P < 0.001) (Fig. 2).

(v) In order to analyse the possible difference in susceptibility to estradiol exposure between T cell originating from thymus and spleen, we treated C.B-17 donor mice with high-dose estradiol for 1 week before cell preparation. Control donor mice were given an empty tube. After treatment with estradiol T cells from spleens were enriched and 5 × 10⁶ cells were transferred into 11 recipient SCID mice. Sensitization took place 1 day later and the DTH reactions were measured 1 and 3 weeks later. In Fig. 2 it is shown that DTH reactivity was equally pronounced in the two groups.

Treatment of recipient SCID mice

(vi), (vii) In the following experiment we analysed if exposure to oestrogen of recipient SCID mice affected the DTH response. We prepared spleen cells and purified thymic T cells from C.B-17 donor mice. Congenic SCID mice were given a single s.c. injection of various doses of estradiol 3 days before cell transfer. Control mice were given instead a single injection of olive oil. Spleen cell suspensions were prepared from C.B-17 donor mice and 20 recipient SCID mice received 5 × 10⁶ spleen cells and 25 recipient SCID mice received 19 × 10⁶ thymocytes. Sensitization with OXA was performed 1 day later and DTH responses were measured 1 and 3 weeks later. Figure 3 shows that recipient mice exposed to 3 or 15 μg of estradiol displayed statistically significant (P < 0.01–0.0001) inhibition of DTH reactivity compared with controls. The inhibition was even more pronounced after 3 weeks and when thymocytes rather than splenocytes were transferred.

Fig. 3.

DTH responses, measured 1 and 3 weeks after sensitization with oxazolone (OXA), in recipient SCID mice transferred with freshly isolated spleen cells or thymus-derived T cells from congenic C.B-17 mice or freshly isolated spleen cells from allogeneic B6 mice. Recipient SCID mice administered with a single injection of estradiol at different doses, 1, 3 and 15 μg, 3 days before cell transfer. Recipient controls received olive oil injection. Each group consisted of 8–12 mice. Bars represent mean ± s.d. *P < 0.05; **P < 0.01; ***P < 0.001.

(viii) In this experiment we prepared splenocytes from H-2 incompatible B6 donor mice. Twenty-seven recipient SCID mice were given a single injection of various doses of estradiol 3 days before cell transfer and nine control mice received instead an injection of vehicle. Each SCID recipient received 15 × 10⁶ spleen cells. Mice were sensitized with OXA 1 day after cell transfer and DTH responses were measured 1 and 3 weeks later. DTH reactivity was equally pronounced in recipient SCID mice irrespective of previous estradiol exposure, after both 1 and 3 weeks (Fig. 3).

Serum levels of estradiol

In order to determine turnover of estradiol in SCID recipients we injected 20 castrated female SCID mice with 3 or 15 μg of estradiol and serum samples were collected after 1, 3 and 4 days. Serum samples were also collected from normal female SCID mice and from castrated female SCID mice. Serum levels of 17β-estradiol were measured by a standard RIA technique. As shown in Fig. 4, serum levels of estradiol were high the day after injection and slowly decreased to reach levels comparable with normal mice at the day cell transfer took place.

Fig. 4.

Serum levels of estradiol in castrated female SCID mice 1, 3 and 4 days after a single injection with 3 and 15 μg estradiol. Hatched bars represent serum levels of estradiol in non-treated SCID mice with or without castration.

DISCUSSION

Several reports published during the last two decades have demonstrated immunoregulatory effects of oestrogen [1,19]. Most of these reports document stimulatory effects on immunoglobulin production [10] and inhibitory effects on T cell-mediated reactions such as DTH [8]. The mechanism whereby oestrogen exerts its effects on the immune system is to a large extent unknown. The primary aim of the present study was to elucidate whether the inhibitory effect of estradiol on T cell-dependent inflammation is mediated via direct effects of this hormone on T lymphocytes. This assumption is reasonable, since early studies showed the presence of oestrogen receptors in T lymphocytes obtained from the thoracic duct. The receptors were found to be restricted to the OKT8⁺ cells bearing the ‘suppressor/cytotoxic’ phenotype [20]. This observation was confirmed by other studies [21–23] indicating that CD8⁺ T lymphocytes may be specific targets for oestrogen. In addition, Fox et al. showed that expression of the mRNA for IFN-γ, a T cell product, can be enhanced through a direct effect of physiological concentrations of oestrogenic steroids. The most pronounced oestrogen response was observed using T lymphoid cell lines that had been transfected with an oestrogen receptor expression plasmid to ensure higher constitutive expression of the receptor [2]. The in vivo relevance of these observations is, however, uncertain.

In the present study we used a recently described transfer system enabling performance of in vivo studies on the immunomodulatory effects of oestrogen on selected cell populations. Our results clearly show that T lymphocytes are not target cells for estradiol-mediated suppression of DTH in this system, since estradiol-exposed T cells transferred to SCID mice did not decrease DTH response in the recipients (experiments iii and v). It is well established that oestrogen has a major impact on thymus development [24]. For instance, treatment of mice for 5 days with estradiol profoundly decreases the number of thymocytes [25], and there is a disproportionate loss of CD4⁺CD8⁺ double-positive cells in parallel with a proportional increase in the percentage of mature CD4⁺ and CD8⁺ single-positive cells [26]. In the maturation process in the thymus, when double-positive T cells interact with MHC-expressing cortical epithelial cells which are abundant in oestrogen receptors [27,28], it is conceivable that, in mice exposed to oestrogen, naive CD4⁺CD8⁺ T cells undergo apoptosis due to lack of positive selection. Interestingly, we saw a slight increase in DTH response in SCID recipients receiving estradiol-exposed thymus T cells. Thus one attractive explanation for our finding is that in mice exposed to high doses of estradiol, the thymus displays a relative increase of CD4⁺ T cells which are the main effector cells in the DTH reaction. Judging from the present results combined with previous studies [24–26], the impact of oestrogen on T cells is limited to their proliferation and maturation rather than exerting effector functions.

In the following experiments we asked if instead APC constitute the target cell population for oestrogen-mediated suppression of DTH. Therefore we exposed recipient SCID mice to estradiol and then transferred thymocytes or non-fractionated spleen cells from congenic C.B-17 mice (experiments vi and vii). When we used this experimental approach the effector T cells were undoubtedly unexposed to pharmacological levels of estradiol. This assumption is based on our data showing that 3 days after a single estradiol injection serum levels of the hormone were at the level found in normal female mice (Fig. 4). Clearly, we demonstrated a highly significant inhibition of DTH response in SCID recipient mice exposed to estradiol even when as low dose as 1 μg was used. Transfer of unfractionated spleen cells from congenic donor mice exposed to high doses of estradiol significantly lowered the DTH response in recipient SCID mice (experiment ii), and this inhibitory effect was even more pronounced upon transfer of allogeneic spleen cells previously exposed to high doses of estradiol (experiment iv). The main difference between these two experimental approaches is that when congenic cells are used, APC originate from both donor and recipient, while in the case of transfer of allogeneic spleen cells the APC–T cell interaction is strictly restricted to donor cells. This could explain the more impressive inhibitory effect of DTH response when transferring allogeneic cells.

A final possibility is that estradiol affects expression of adhesion molecules on endothelial cells and/or their ligands on inflammatory cells and thus regulates the cell influx. This is a relevant issue, since we showed in an earlier study that estradiol has the capacity to down-regulate granulocyte-mediated and T cell-independent inflammation [29]. However, transfer of non-fractionated spleen cells from allogeneic donor mice to SCID recipient mice pretreated with estradiol did not inhibit the DTH response (experiment viii). Therefore, endothelial cells are unlikely to be responsible for estradiol-induced suppression of T cell-mediated inflammation.

Taken together, our findings exclude T cells and endothelial cells as targets for estradiol-mediated suppression of DTH and instead indicate the possibility of the APC/macrophage cell population as mediators for the inhibitory effect of estradiol on T cell-dependent inflammation. This assumption is further supported by the fact that macrophages express oestrogen receptors [30]. In addition, Wira & Rossoll showed that antigen presentation in the vagina of rats is decreased when estradiol levels in blood are elevated [31]. They also showed that the number of MHC class II-expressing cells varies with the stage of the reproductive cycle and that these cells were significantly fewer when the serum level of estradiol was high. Accordingly, in both thymus and female reproductive tract MHC-expressing cells display suppressed APC function upon estradiol exposure.

There has been considerable interest in the possible beneficial effects of oestrogen administration in patients with rheumatoid arthritis (RA). This has mainly stemmed from the observations of the improvement of RA during pregnancy [32,33]. Interestingly, post-menopausal RA patients treated with oestrogen replacement therapy displayed not only increased bone mineral density and well being, but also amelioration of arthritic symptoms [34]. The potential mechanisms behind these observations are not known. Synovial inflammation in RA can be regarded as a DTH-like reaction, initiated by autoreactive T cells and perpetuated by activated macrophages. Since oestrogen receptors have been detected in human T cells [22] and in synovial macrophages [21], both cell populations must be regarded as possible candidates as target cells for oestrogen-mediated suppression of arthritis. The results of the present study strongly favour the macrophages as the target cell population for oestrogen-mediated amelioration of arthritis.