Skip to main content
NCBI home page
Immunology logoLink to Immunology
. 2003 Sep;110(1):149–157. doi: 10.1046/j.1365-2567.2003.01704.x

Influence of oestrogen receptor α and β on the immune system in aged female mice

U Islander *, M C Erlandsson *, B Hasséus †, C A Jonsson *, C Ohlsson ‡, J-Å Gustafsson §, U Dahlgren *,†, H Carlsten *
PMCID: PMC1783017  PMID: 12941152

Abstract

Oestrogen has a dichotomous effect on the immune system. T and B lymphopoiesis in thymus and bone marrow is suppressed, whereas antibody production is stimulated by oestrogen. In this study the importance of the oestrogen receptors (ER) ER-α and ER-β in the aged immune system was investigated in 18 months old-wild type (WT), ER-α (ERKO), ER-β (BERKO) and double ER-α and ER-β (DERKO) knock-out mice, and compared with 4 months old WT mice. Cell phenotypes in bone marrow, spleen and thymus, and the frequency of immunoglobulin (Ig) spot forming cells (SFC) were determined. We show here that the 17-β-oestradiol (E2)-induced downregulation of B lymphopoietic cells in bone marrow of young ovariectomized mice can be mediated through both ER-α and ER-β. However, only ER-α is required for the age-related increased frequency of immunoglobulin M (IgM) SFC in the bone marrow, as well as for the increased production of interleukin-10 (IL-10) from cultured splenocytes in aged mice. Furthermore, increased age in WT mice resulted in lower levels of both pro- and pre-B cells but increased frequency of IgM SFC in the bone marrow, as well as increased frequency of both IgM and IgA SFC in the spleen. Results from this study provide valuable information regarding the specific functions of ER-α and ER-β in the aged immune system.

Introduction

It is well established that oestrogens affect the development and regulation of the immune system. For instance B lymphopoiesis is suppressed during pregnancy1 and elevated in oestrogen-deficient mice.2 Furthermore, both T and B lymphopoiesis is suppressed by treatment with 17β-oestradiol (E2).3,4 Oestrogen has a dichotomous effect on the immune system. Hence, it has been shown that exposure to oestrogens stimulates antibody production5 but decreases T-cell mediated delayed-type hypersensitivity (DTH)5,6 granulocyte-mediated inflammation7 and natural killer (NK)-cell mediated cytotoxicity.8,9

It is still not fully understood how the biological functions of oestrogens are mediated. However the oestrogen molecule enters the target cell where it binds to oestrogen receptors (ER). There are two known ER subtypes termed ER-α and ER-β.10,11 They act as transcription factors on different gene promoters and have different tissue distribution. ER knock-out mice, ER-α knock-out (ERKO)12 ER-β knock-out (BERKO)13 and the double ER knock-out (DERKO)14 are valuable tools to use when studying the specific functions of ER-α and ER-β, respectively.

We have previously shown that ER-α is important for full development of thymus and spleen in 4-month-old male ER knock-out mice. Furthermore, E2 treatment of female WT and BERKO mice revealed that ER-β is required for the E2 mediated thymic cortex atrophy.15

Both ER-α and ER-β are important for the maturation of single positive CD4 and CD8 cells from double positive thymocytes. Staples et al. showed that no suppression of CD4/CD8 double positive cells could be seen in male ERKO (ER-α−) mice upon E2 exposure.16 Also, E2 treatment of female WT and BERKO mice revealed that ER-β is required for the E2 mediated thymocyte shift towards more mature phenotypes.15 Taken together, the data suggest that both ER-α and ER-β are needed to achieve full E2 mediated thymic atrophy.

Ageing affects the immune system by a general suppression of activity. The production of T cells from the thymus and B cells from the bone marrow is decreased at higher age. However, the number of peripheral B cells is maintained during ageing.17 This is suggested to be mainly due to increased survival time when the competition with newly formed B cells from the bone marrow decrease, as well as to increased self-renewal. Old age also induces a change in the quality of the antibody response. There is a shift from antibodies directed against foreign antigens towards more autoantibodies, and also the affinity of antibodies produced from newly formed B cells is lower in aged individuals. However, this does not reflect a generalized impairment in immunoglobulin secretion. In fact, the serum concentration of immunoglobulin M (IgM) increases with age (reviewed in 18).

Following these data, the aim of this study was to investigate the importance of ER-α and ER-β on the aged immune system. Eighteen-month-old female wild type (WT) and ER knock-out ERKO, BERKO and DERKO mice in comparison to 4-month-old female WT mice were used. Also, 4-month-old WT, ERKO, BERKO and DERKO mice were ovariectomized and treated with E2.

Materials and methods

Mice

This study was approved by the ethical committee for animal experiments at Göteborg University.

Generation and identification of WT, ERKO, BERKO and DERKO mice

Male double heterozygous (ER-α+/− β+/−) mice were mated with female double heterozygous (ER-α+/− β+/−) mice on a mixed C57Bl/6J/129 background resulting in WT (ER-α+/+ β+/+), ERKO (ER-α−/− β+/+), BERKO (ER-α+/+ β−/−) and DERKO (ER-α−/− β−/−) offspring.19

Genotyping of tail DNA was performed using polymerase chain reaction (PCR). The ER-α gene was analysed with the primer pairs 5′-AACTCGCCGGCTGCCACTTACCAT-3′ and 5′-CATCAGCGGGCTAGGCGACACG-3′, resulting in a 320-bp fragment for the normal ER-α gene. The primer pairs 5′-TGTGGCCGGCTGGGTGTG-3′ and 5′-GGCGCTGGGCTCGTTCTC-3′ resulted in a 700-bp fragment for the mutant ER-α gene. The ER-β gene was analysed with primers βNHD4-25 (5′-AGAATGTTGCACTGCCCCTGCTGCT-3′), Clwt-27 (5′-GGAGTAGAAACAAGCAATCCAGACATC-3′) and Neo-25 (5′-GCAGCCTCTGTTCCACATACACTTC-3′). A 650-bp fragment (primer pairs βNHD4-25 and Clwt-27) was amplified for the normal ER-β gene, and a 450-bp fragment (primer pairs βNHD4-25 and Neo-25) was amplified for the mutant ER-β gene.

Housing conditions

Mice were kept in the animal facility at Göteborg University under standard conditions of temperature and light, and fed standard laboratory chow ad libitum.

Ovariectomy and oestrogen treatment

4-month-old female WT, ERKO, BERKO and DERKO mice were ovariectomized under Ketalar®/Dormitor® (Apoteksbolaget, Göteborg, Sweden) anaesthesia and left to rest for 2 weeks before the start of experiments. Mice were injected subcutaneously (s.c) with 3·2 µg 17β-oestradiol-3-benzoate (Sigma, St Louis, MO) per mouse, dissolved in olive oil (Apoteksbolaget), five times a week for 3 weeks. Control mice received injections of vehicle olive oil (Apoteksbolaget).

Ageing mice

Untreated WT, ERKO, BERKO, and DERKO mice were kept in the animal facility at Göteborg University for 18 months before start of experiments. 4-month-old untreated WT mice were used as young control.

Tissue collection and single-cell preparation

Mice were anaesthetized by Ketalar®/Dormitor® (Apoteksbolaget), bled and killed by cervical dislocation. Thymuses and spleens were removed and weighed. Single-cell suspensions were prepared by mashing the organs through a nylon wool sieve. Bone marrow cells were harvested from the right femur by a syringe containing 2 ml Phosphate-buffered saline (PBS). Cells were kept in a 50/50 mixture of complete medium (RPMI-medium enriched with 50 µg/ml gentamicin (Sigma), 1%l-glutamine (Sigma), 5 × 10−5 m mercaptoethanol (Sigma), 0.25 µg/ml amphotericin B (Roche Diagnostics Scandinavia AB, Bromma, Sweden) and 5% fetal calf serum (FCS; Biological Ind., Beit Haemek, Israel) and PBS until use. Thymus, spleen and bone marrow cells were centrifuged at 515 g for 5 min. Pelleted spleen cells were re-suspended in Tris-buffered 0·83% NH4Cl solution (pH 7·29) for 5 min to lyse erythrocytes. After washing in PBS the total number of leucocytes from the organs was calculated using an automated cell counter (Sysmex, Kobe, Japan). Cells were re-suspended in complete medium before use. The total number of nucleated cells in the bone marrow was not used since it varies too much to be reliable.

Flow cytometry

Cells from thymus, spleen and bone marrow were subjected to fluorescence-activated cell sorting (FACS) analysis. Isolated thymocytes were stained with phycoerythrin (PE)-conjugated antibodies to CD4 (clone H129·19, BD PharMingen, Franklin Lakes, NJ), and fluoroscein isothiocyanate (FITC)-labelled antibodies to CD8 (clone 53-6·7, BD PharMingen). Spleen cells were labelled with anti-CD4–PE, anti-CD8–FITC and anti-CD45R/B220-FITC (clone RA3-6B2, BD PharMingen) antibodies. Bone marrow cells were stained with anti–CD45R/B220–FITC and R-phycoerythrin (R-PE) labelled anti-µ (cat. nr. 1021–09, Southern Biotech, Birmingham, AL) antibodies. Flow cytometry was performed on a FACSCalibur and analysed using Paint-A-Gate software (Beckton-Dickinson, Franklin Lakes, NJ). (No differences in the results were detected when comparing frequencies of cells, with the absolute number of cells.) Thymus and spleen data are presented as percentage of all gated lymphocytes, whereas bone marrow data is presented as percentage of all nucleated cells.

Concanavalin A-induced proliferation

Freshly isolated spleen cells suspended in complete medium (1 × 106 cells/ml), were cultured in round bottom 96-well plates (Nunc, Roskilde, Denmark) in 37°, 5% CO2 and 95% humidity. The T-cell mitogen concanavalin A (Con A; Sigma) was added to the medium in a final concentration of 2·5 µg/ml. Cells in complete medium alone served as control. After 72 hr of culture 1 µCi [3H]thymidine (Amersham Pharmacia Biotech, Uppsala, Sweden) was added for 24 hr. Cells were harvested onto glass-fibre filters and counted in a β-counter. Cell cultures were set up in triplicates, and results are presented as proliferation index (mean of counts per minute (c.p.m.) in cultures with Con A/mean of c.p.m. in cultures without Con A).

Interleukin-10 enzyme-linked immunosorbent assay (ELISA)

Interleukin (IL)-10 in supernatants from the cell cultures was measured with a mouse IL-10 ELISA kit (R & D Systems Europe Ltd, Abingdon, UK). Culture supernatants were used undiluted and the procedure was performed according to the manufacturers instructions. Plates were read in a spectrophotometer (SPECTRAmax PLUS) and Delta Soft 3 software (Biometallics, Inc., Princeton, NJ) was used for analysing the results.

Immunoglobulin production

The enzyme-linked immunospot (ELISPOT)20 assay was used for enumeration of IgM IgG and IgA spot forming cells (SFC) in freshly prepared spleen and bone marrow cell suspensions. Briefly, 96-well nitrocellulose plates (Millipore MultiScreen, Bedford, MA) were coated with 100 µl of affinity-purified F(ab′)2 fragments to goat anti mouse IgM, IgG or IgA (Cappel, Organon Teknika, Turnhout, Belgium) diluted to 5 µg/ml in PBS, and then incubated over night at 4° in a moist atmosphere. After washing with PBS and blocking with 5% FCS in PBS, 50 µl of complete medium containing 1 × 106 or 1 × 105 spleen or bone marrow cells per ml were added to each well, and the plates were incubated for 3·5 hr at 37°, 5% CO2 and 95% humidity. Plates were washed in PBS and then 100 µl of alkaline phosphatase (AP) conjugated goat anti-mouse IgM, IgG or IgA antibodies (Southern Biotechnology Associates Inc.) diluted 1 : 750 in PBS, were added to the wells. Finally, concentrated 5-bromo-4-chloro-3-indolyl phosphate solution (Sigma) was added. All cell cultures were done in triplicates. Results are presented as the number of immunoglobulin SFC per 1 × 103 B cells (B220+ cells).

Serum E2 levels

Blood was collected at the termination of the experiments and sera were individually stored at −20° until use. Serum E2 levels were measured using an E2 radioimmunoassay (RIA) scaled down 1 : 4 (Diagnostic Systems Laboratories Inc., Webster, TX). The sensitivity of the assay was less than 1 pg/ml.

Statistical analysis

With only a minor number of exceptions, all data proved to be normally distributed using the Shapiro–Wilk test. Student's unpaired t-test was used to compare data from E2-treated mice with control mice. This test was also used to find statistical differences between 4- and 18-month-old WT mice. Two-way analysis of variance (anova) was used to find statistical differences with respect to the expression of either ER-α or ER-β. Fisher's test was used to find statistical differences between aged mice of the different genotypes. Horizontal lines in the dot-plots represent mean values. Results are presented as mean ± standard deviation. P < 0·05 was considered statistically significant.

Results

Influence of ER expression on E2-mediated effects on B lymphopoiesis

Ovariectomized 4-month-old female WT, BERKO, ERKO and DERKO mice were injected s.c. with E2 (3·2 µg/mouse per day, 5 days per week for 3 weeks) or olive oil as control. The dose of E2 chosen gave a serum concentration of 140 ± 50 pg/ml.21

The frequency of B220 expressing bone marrow cells were analysed by flow cytometry. In WT, BERKO and ERKO mice, exposure to E2 dramatically reduced the frequency of B220+ cells whereas no such reduction was found in DERKO mice (Fig. 1). A two-way anova revealed that both ER-α and ER-β are involved in E2 induced reduction of B lymphopoietic cells.

Figure 1.

Figure 1

Open in a new tab

Both ER-α and ER-β can mediate the inhibitory effect on B lymphopoietic cells after oestrogen treatment. The percentage of B220+ cells in bone marrow of 4-month-old ovariectomized mice of all ER genotypes are presented. Mice were injected s.c. with 3·2 µg E2/mouse per day five times a week for 3 weeks (shaded bars), or with olive oil as control (open bars), Students unpaired t-test: control versus E2 treatment **P < 0·01, ***P < 0·001. A two-way anova reveals that both ER-α and ER-β contribute to the E2 mediated inhibitory effect, ER-α: P < 0·001, ER-β: P < 0·01.

Influence of age and ER expression on B lymphopoiesis in the bone marrow

Untreated 4-months-old-female WT and 18-month-old female WT, ERKO, BERKO and DERKO mice were compared in order to investigate the role of age and expression of ER-α and ER-β on B-cell phenotype and immunoglobulin-producing cells in the bone marrow. Flow cytometry analysis of B220+ cells in the bone marrow revealed a non-significant tendency towards lower frequency of B lymphopoietic (B220+) cells in aged WT mice compared to young (Fig. 2a). However, ELISPOT analysis on bone marrow cells showed that the frequency of IgM SFC clearly increased with age in WT mice (Fig. 2b).

Figure 2.

Figure 2

Open in a new tab

ER-α is needed for the age-induced elevation of IgM producing B cells in the bone marrow. The figure shows percentage of B220+ cells (a) and number of immunoglobulin spot forming cells (SFC) per 1 × 103 B220+ cells in the bone marrow (b). In (a) aged WT mice show a non-significant tendency towards lower frequency of B220+ cells. No significant differences in frequency of B220+ cells could be found between aged mice of the different genotypes. Horizontal lines represent mean values. In (b), aged WT mice display a significantly higher frequency of IgM SFC compared to young WT mice, Students unpaired t-test: *P < 0·05. A two-way anova reveals that ER-α is needed for the age-induced elevation of IgM SFC, ER-α: P < 0·001, ER-β: P = ns. ERKO mice show significantly higher frequency of IgA-producing bone marrow B cells compared to aged mice of the other genotypes, Fisher's test: *P < 0·05.

No statistically significant differences could be detected between aged mice of the different genotypes when comparing the frequency of B220+ cells in the bone marrow (Fig. 2a). However, mice lacking ER-α (ERKO and DERKO) had significantly lower frequency of IgM SFC in the bone marrow compared to ER-α+ mice (Fig. 2b). Interestingly, ERKO mice (ER-α−) showed a significantly higher frequency of IgA-producing bone marrow B cells compared to aged mice of the other genotypes (Fig. 2b).

Flow cytometry was also used to analyse B lineage cells at different developmental stages in the bone marrow. B220 and µ chains are expressed with various intensities on different stages of B lymphopoietic cells. Four fractions of B lymphopoietic cells were identified using anti-B220/CD45R and anti-µ antibodies. Figure 3 shows representative FACS plots of the four fractions of B lymphopoietic cells in the bone marrow from young and old WT mice. The earliest fraction is B220lo and µ chain negative cells (fraction 1), which includes intermediary and late pro-B cells. The fraction of cells that is B220lo and µ chain positive (fraction 2), includes large and small pre-B cells as well as the fraction of almost mature B cells (i.e. immature B cells). The B220hi and µ chain positive fraction (fraction 3), represents mature B cells. These three stages correspond to the late part of B lymphopoiesis: from the time of rearrangement of the heavy chain immunoglobulin genes to newly formed mature B cells. Finally, the B220hi and µ chain negative fraction (fraction 4), represents immunoglobulin switched activated B cells as well as memory B cells. The results in Fig. 3 and Table 1 show that aged WT mice have significantly lower frequencies of pro- and pre-B cells (fraction 1–2) compared to young WT mice. No statistically significant difference could be detected between young and aged WT mice in the fractions containing mature- and immunoglobulin-swiched activated B cells (fractions 3–4). Eighteen months old WT, BERKO, ERKO and DERKO mice showed no differences in any of the four fractions of B lymphopoietic cells (data not shown).

Figure 3.

Figure 3

Open in a new tab

Aged WT mice have less early B lymphopoietic cells (fraction 1 and 2) in the bone marrow compared to young WT mice. Representative FACS plots from bone marrow of young and aged WT mice is shown. The numbers in the figure indicate four different fractions of B lymphopoietic cells. See text for explanation of fractions, and Table 1 for data values.

Table 1.

Increased age in WT mice results in less early B lymphopoietic cells in the bone marrow

% pro-B cells (fraction 1) % pre-B cells (fraction 2) % mature B cells (fraction 3) % immunoglobulin switched activated B cells (fraction 4)
WT n = 4 4·2 ± 3·1 1·5 ± 0·8 1·0 ± 0·3 0·3 ± 0·04
ER α+, β+
4 months
WT n = 9 1·6 ± 1·1 0·7 ± 0·5 2·1 ± 1·2 0·6 ± 0·3
ER α+, β+
18 months
Student's unpaired t-test * * P = 0·10 P = 0·13
Open in a new tab

Fraction 1, 2, 3 and 4 represent four different stages in B cell development. These fractions are also indicated in Fig. 3. This shows that aged WT mice display less early B lymphopoietic cells in the bone marrow compared to young WT mice, Student's unpaired t-test:

*

P < 0·05.

These results clearly show that ER-α alone is necessary for the age-induced high frequency of IgM SFC in the bone marrow. Furthermore, increased age in WT mice results in lower frequencies of pro- and pre-B cells as well as increased frequency of IgM SFC in the bone marrow.

Influence of age and ER expression on B-cell phenotype and function in the spleen

Flow cytometry analysis of B220+ cells in the spleen revealed that, as expected, there was no difference in B-cell frequencies between young and aged WT mice (Fig. 4a). However, the frequencies of IgM and IgA SFC clearly increased with age in WT mice (Fig. 4b). Also, aged mice lacking both ER-α and ER-β (DERKO) showed a statistically significant higher frequency of B220+ cells compared to aged BERKO and ERKO mice (Fig. 4a).

Figure 4.

Figure 4

Open in a new tab

The age induced increase in IgM and IgA SFC is not related to the frequency of B cells in the spleen. The figure shows the percentage of B220+ cells (a), and the number of immunoglobulin SFC per 1 × 103 B220+ cells in the spleen (b). Despite the non-changing frequency of B220+ cells in young compared to aged WT mice, the frequency of IgM and IgA SFC increase significantly with age, Students unpaired t-test: *P < 0·05. No significant differences could be found in either frequency of B220+ cells or immunoglobulin-producing cells, regarding the expression of ER-α and ER-β using the two-way anova. However, DERKO mice show significantly higher frequencies of B220+ cells compared to aged BERKO and ERKO mice, Fisher's test: **P < 0·01. Horizontal lines in the dot-plot represent mean values.

These results demonstrate that the age-induced increase in IgM and IgA SFC is not related to the total frequency of B cells in the spleen.

Influence of age and ER expression on T-cell phenotype and function in the spleen

ER-α is required for full development of spleen in young male mice.15 In this study the relative spleen weights (percent of total body weight) of 18-month-old female mice were: WT 0·45 ± 0·18 (n = 9), BERKO 0·52 ± 0·29 (n = 8), ERKO 0·38 ± 0·09 (n = 7) and DERKO 0·24 ± 0·07 (n = 7). A two-way anova revealed that the ER-α-mediated effect on spleen weight is also preserved in very aged female mice, ER-α: P < 0·05, ER-β: P = ns.

Flow cytometry analysis of spleen cells showed that the total frequency of T cells did not differ between young and old WT mice (data not shown). However, aged mice had significantly lower frequency of CD8+ T cells compared to young mice (Fig. 5a). A two-way anova revealed a statistically significant lower frequency of CD4+ T cells in aged ER-α− mice compared to ER-α+ mice (Fig. 5a). Although no differences in the total frequency of splenic T cells could be found between young and aged WT mice, spleen cells from aged WT mice displayed a significantly lower proliferative response to Con A compared to cells from young mice (Fig. 5b).

Figure 5.

Figure 5

Open in a new tab

Lack of ER-α results in lower frequency of CD4+ T cells in the spleen. (a) The percentage of CD4+ and CD8+ T cells in spleen. Aged WT mice have significantly lower frequency of CD8+ T cells compared to young, Students unpaired t-test: **P < 0·01. A two-way anova reveals that mice lacking ER-α have lower frequency of CD4+ T cells compared to ER-α+ mice, ER-α: P < 0·05, ER-β: P = ns. (b) Proliferation index of Con A-stimulated splenic T cells. Spleen cells from aged WT mice display significantly lower proliferative response to Con A compared to cells from young WT mice, Students unpaired t-test: *P < 0·05. No statistically significant differences could be found when comparing aged mice of the different genotypes. Horizontal lines represent mean values.

Supernatants from Con A-stimulated spleen cell cultures were analysed for the presence of the T helper 2 (Th2) cytokine IL-10. No difference between young and aged WT mice could be detected; however, a two-way anova revealed that mice lacking ER-α had significantly lower levels of IL-10 in supernatants compared to mice expressing ER-α (Fig. 6). Supernatants from Con A-stimulated spleen cell cultures were also analysed for the presence of the Th1 cytokine interferon-γ but no differences could be detected (data not shown).

Figure 6.

Figure 6

Open in a new tab

Lack of ER-α expression induces a change in IL-10 production. The figure shows the amount of IL-10 in supernatants from Con A-stimulated spleen cells. No differences in IL-10 production, between young and aged WT mice, were detected using the Student's unpaired t-test. However, a two-way anova reveals significantly higher levels of IL-10 in supernatant from mice expressing ER-α compared to ER-α− mice, ER-α: P < 0·05, ER-β: P = ns. Horizontal lines represent mean values.

Taken together, these results show that lack of ER-α results in lower frequency of CD4+ T cells in the spleen. Also, ER-α is involved in the changed levels of IL-10 produced from Con A-stimulated cultured spleen cells, and spleen cells from aged WT mice have significantly lower proliferative response to Con A compared to young WT mice.

Influence of age and ER expression on T-cell phenotype in the thymus

In 18-month-old mice the thymuses were atrophic and difficult to separate from the surrounding tissue, making correct measurement of weight and cellularity uncertain. However, thymuses in old mice were clearly smaller compared to that of young mice.

In comparison to young WT mice, aged WT mice had significantly lower frequency of CD4/CD8 double positive cells (Fig. 7). No difference could be found in the frequencies of CD4 and CD8 single positive cells in the thymus of aged WT mice compared to young (data not shown).

Figure 7.

Figure 7

Open in a new tab

The ER-α dependent decreased frequency of double positive thymocytes, is present also at a very high age. The figure shows percentage of CD4/CD8 double positive T cells in the thymus. Aged WT mice have lower frequency of CD4/CD8 double positive T cells compared to young WT mice, Student's unpaired T-test: **P < 0·01. A two-way anova reveals that aged mice lacking ER-α display a higher frequency of CD4/CD8 double positive cells compared to ER-α+ mice, ER-α: P < 0·01, ER-β: P = ns. Horizontal lines represent mean values.

We have previously shown that young male mice lacking ER-α display higher frequencies of CD4/CD8 double positive cells compared to mice expressing ER-α.15 Interestingly, we now show that this effect is preserved in very aged (18 months old) female mice (Fig. 7). We have also previously shown that ER-α− male mice display less CD4 and CD8 single positive cells in the thymus compared to ER-α+ male mice.15 A similar pattern was found in aged female mice, however, not reaching statistical significance (data not shown).

Taken together these results demonstrate that the ER-α dependent decreased frequency of double positive thymocytes, is present also at a very high age.

17β-oestradiol levels in serum of aged mice

The serum levels of E2 in 18-month-old mice were: WT 21 ± 11 pg/ml (n = 4), BERKO 26 ± 6 pg/ml (n = 4), ERKO 52 ± 53 pg/ml (n = 4) and DERKO 113 ± 107 pg/ml (n = 4). A two-way anova revealed that mice lacking ER-α had higher serum levels of E2 compared to ER-α+ mice, however, not reaching statistical significance, ER-α: P = 0·07, ER-β: P = ns.

Discussion

The immune system is less effective in old age because of a generalized dysregulation of its component parts. It has been suggested that thymic involution indirectly leads to decreased production of B cells since products of activated T lymphocytes regulate the development of B cells in the bone marrow. Supporting this, it has been reported that the number of B lineage cells in the bone marrow decrease during ageing at a rate similar to but somewhat later than the thymic involution (reviewed in 22). Furthermore, Stephan et al. suggested that the gradual reduction in B lymphopoiesis observed in ageing mice may be attributable to impaired release of IL-7 from bone marrow stromal cells.23 In this study, we show lower levels of early B lymphopoietic cells in aged WT bone marrow compared to young (Fig. 3, Table 1). Previous reports have shown that ageing reduces the levels of pre-B cells by 60–90% without any reduction in the pro-B-cell fraction (reviewed in 22). Interestingly, our analyses of B lymphopoietic cells in old WT mice show lower frequencies of both pro- (fraction 1) and pre-B cells (fraction 2) when compared to young WT mice (Table 1). However, since we did not perform triple staining using anti-B220, anti-µ and anti-CD43, we can not exclude that some cells in fraction 1 could consist of transient B220lo and µ− plasma cells. An indication of the higher ratio of differentiated B cells in the bone marrow of age mice, was that the frequency of IgM SFC in the bone marrow was significantly higher in aged WT mice (Fig. 2b).

Kline et al. reported that despite the decreased rate of B lymphocyte production in the bone marrow of aged mice, the overall number of peripheral B cells remain constant during ageing. This is suggested to be mainly because of the increased lifespan of peripheral B cells when competition with newly formed B cells from the bone marrow is decreased, as well as to increased self-renewal.17 The serum concentration of IgM increases with age and the frequency of immunoglobulin secreting B cells increase two- to 10-fold during ageing. The basis for this B-cell activation is not clear but increased production of IL-4 and IL-6 by T cells in old mice has been claimed to play a role (reviewed in 22). Results from this study agree with previous findings and showed no significant difference between young and aged WT mice regarding the frequency of B cells in the spleen (Fig. 4a). Also the frequency of IgM and IgA SFC was higher in aged mice compared to young (Fig. 4b).

Age-induced involution of the thymus leads to decreased production of naive T cells. In this study we confirm this and show that aged WT mice have lower frequency of CD4/CD8 double positive T lymphocytes compared to young WT mice (Fig. 7). In spleen there was no difference in the total frequency of T cells in young and aged mice, even though aged mice showed significantly lower frequency of CD8+ T cells (Fig. 5a). Furthermore, ConA stimulated T-cell proliferation was significantly lower in aged WT mice compared to young mice (Fig. 5b), confirming a lower activity in the ageing immune system.

It has previously been shown that E2 treatment suppresses B-cell development in the bone marrow24 and stimulates antibody production.5 However it is not known through which one of the two oestrogen receptors this effect is mediated. ER-α has been found in peripheral B cells, and both ER-α and ER-β have been found in bone marrow stromal cells.25–27 Therefore we hypothesize that the oestrogen-induced suppression of B lymphopoiesis, requiring both ER-α and ER-β, is mediated by action of the hormone on stromal cells. E2 treatment of ovariectomized 4-month-old female mice revealed that both ER-α and ER-β can mediate the E2-induced reduction of B lymphopoietic cells in the bone marrow (Fig. 1). This finding correlates well with our previously reported findings in male mice.28 The oestrogen treatment resulted in a mean serum concentration of 140 pg/ml21 Normal serum level of E2 in mice varies between 25 and 50 pg/ml during dioestrus, while it is between 150 and 200 pg/ml in the oestrus phase.29 Thus, our E2 treatment resulted in serum levels that were similar to those normally seen during oestrus in female mice.

In this study we show that ER-α, but not ER-β, is necessary for the age-induced stimulation of IgM SFC in bone marrow (Fig. 2b). This finding correlates well with our recent discovery that ER-α, but not ER-β, is necessary for the E2-mediated elevation of immunoglobulin producing B cells in young mice.28 Furthermore in Fig. 2b, ERKO mice show a higher frequency of IgA-producing B cells in the bone marrow compared to aged mice of the other genotypes. Interestingly, Macpherson et al.30 have shown that mice with a developmental block at the pro-B cell stage (µ-MT mice) selectively expresses IgA, and the pathway for development of IgA-expressing B cells is dependent on the presence of Peyer's patches and peripheral lymph nodes. We speculate that the elevated frequency of IgA-producing B cells in bone marrow of ERKO mice could be related to peripheral B-cell development in the Peyer's patches.

In this study we show that the previously reported ER-α mediated oestrogenic effects on both spleen weight15 as well as cell phenotype in the thymus, remain at high age. Flow cytometry analysis of spleen cells revealed that aged mice lacking ER-α displayed a lower frequency of CD4+ T cells, with reduced IL-10 production as a consequence. Dudley et al.31 have previously shown a shift in CD4+ T-cell cytokine production towards the B lymphocyte-supporting Th2 profile during pregnancy. This study supports that finding and shows that lack of ER-α reduces IL-10 production from Con A-stimulated cultured spleen cells.

The serum levels of E2 in untreated 18-month-old female WT mice were similar to those of 2–3-month-old WT mice32 indicating that 18-month-old mice still have endogenous E2 production. However, aged mice lacking ER-α had higher serum levels of E2 compared to aged mice expressing ER-α, a phenomenon also demonstrated in young mice.33

With increasing age the variability between individual mice increases. In this study, the age related increased variability is one important reason why there in many cases are trends but no statistically significant changes. In future studies of aged mice, more animals in each group are needed to provide more clear-cut results.

This report demonstrates, among other things, that oestrogen via both ER-α and ER-β has important effects on the immune system and that these effects are significant also in very old mice. Whether or not these findings have any relevance in humans are still not known. One difference between female mice and women is that mice still have high serum levels of E2 at high age whereas elderly women display very reduced concentrations. Further studies in postmenopausal women with and without oestrogen replacement therapy could cast light on this issue.

Acknowledgments

We thank Mrs Lena Svensson for excellent technical assistance. This study was supported by grants from the Swedish cancer foundation, the Börje Dahlin foundation, the Göteborg Medical Society, Association against Rheumatism, the King Gustav V's 80 years foundation, the Anna-Greta Crafoord foundation, Reumaforskningsfond Margareta, the Medical Faculty of Göteborg University (LUA), the Swedish Foundation for Strategic Research, the Torsten and Ragnar Söderbergs Foundation, Petrus and Augusta Hedlunds Foundation and the Swedish Medical Research Council.

References

  • 1.Medina KL, Smithson G, Kincade PW. Suppression of B lymphopoiesis during normal pregnancy. J Exp Med. 1993;178:1507. doi: 10.1084/jem.178.5.1507. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Smithson G, Beamer WG, Shultz KL, Christianson SW, Shultz LD, Kincade PW. Increased B lymphopoiesis in genetically sex steroid-deficient hypogonadal (hpg) mice. J Exp Med. 1994;180:717. doi: 10.1084/jem.180.2.717. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Medina KL, Kincade PW. Pregnancy-related steroids are potential negative regulators of B lymphopoiesis. Proc Natl Acad Sci U S A. 1994;91:5382. doi: 10.1073/pnas.91.12.5382. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Rijhsinghani AG, Thompson K, Bhatia SK, Waldschmidt TJ. Estrogen blocks early T cell development in the thymus. Am J Reprod Immunol. 1996;36:269. doi: 10.1111/j.1600-0897.1996.tb00176.x. [DOI] [PubMed] [Google Scholar]
  • 5.Carlsten H, Holmdahl R, Tarkowski A, Nilsson LA. Oestradiol- and testosterone-mediated effects on the immune system in normal and autoimmune mice are genetically linked and inherited as dominant traits. Immunology. 1989;68:209. [PMC free article] [PubMed] [Google Scholar]
  • 6.Carlsten H, Holmdahl R, Tarkowski A, Nilsson LA. Oestradiol suppression of delayed-type hypersensitivity in autoimmune (NZB/NZW) F1 mice is a trait inherited from the healthy NZW parental strain. Immunology. 1989;67:205. [PMC free article] [PubMed] [Google Scholar]
  • 7.Josefsson E, Tarkowski A, Carlsten H. Anti-inflammatory properties of estrogen. I. In vivo suppression of leukocyte production in bone marrow and redistribution of peripheral blood neutrophils. Cell Immunol. 1992;142:67. doi: 10.1016/0008-8749(92)90269-u. [DOI] [PubMed] [Google Scholar]
  • 8.Hanna N, Schneider M. Enhancement of tumor metastasis and suppression of natural killer cell activity by beta-estradiol treatment. J Immunol. 1983;130:974. [PubMed] [Google Scholar]
  • 9.Nilsson N, Carlsten H. Estrogen induces suppression of natural killer cell cytotoxicity and augmentation of polyclonal B cell activation. Cell Immunol. 1994;158:131. doi: 10.1006/cimm.1994.1262. [DOI] [PubMed] [Google Scholar]
  • 10.Green S, Walter P, Greene G, et al. Cloning of the human oestrogen receptor cDNA. J Steroid Biochem. 1986;24:77. doi: 10.1016/0022-4731(86)90035-x. [DOI] [PubMed] [Google Scholar]
  • 11.Kuiper GG, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson JA. Cloning of a novel receptor expressed in rat prostate and ovary. Proc Natl Acad Sci U S A. 1996;93:5925. doi: 10.1073/pnas.93.12.5925. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Korach KS, Couse JF, Curtis SW, et al. Estrogen receptor gene disruption: molecular characterization and experimental and clinical phenotypes. Recent Prog Horm Res. 1996;51:159. [PubMed] [Google Scholar]
  • 13.Krege JH, Hodgin JB, Couse JF, et al. Generation and reproductive phenotypes of mice lacking estrogen receptor beta. Proc Natl Acad Sci U S A. 1998;95:15677. doi: 10.1073/pnas.95.26.15677. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Couse JF, Hewitt SC, Bunch DO, Sar M, Walker VR, Davis BJ, Korach KS. Postnatal sex reversal of the ovaries in mice lacking estrogen receptors alpha and beta. Science. 1999;286:2328. doi: 10.1126/science.286.5448.2328. [DOI] [PubMed] [Google Scholar]
  • 15.Erlandsson MC, Ohlsson C, Gustafsson JA, Carlsten H. Role of oestrogen receptors alpha and beta in immune organ development and in oestrogen-mediated effects on thymus. Immunology. 2001;103:17. doi: 10.1046/j.1365-2567.2001.01212.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Staples JE, Gasiewicz TA, Fiore NC, Lubahn DB, Korach KS, Silverstone AE. Estrogen receptor alpha is necessary in thymic development and estradiol-induced thymic alterations. J Immunol. 1999;163:4168. [PubMed] [Google Scholar]
  • 17.Kline GH, Hayden TA, Klinman NR. B cell maintenance in aged mice reflects both increased B cell longevity and decreased B cell generation. J Immunol. 1999;162:3342. [PubMed] [Google Scholar]
  • 18.LeMaoult J, Szabo P, Weksler ME. Effect of age on humoral immunity, selection of the B-cell repertoire and B-cell development. Immunol Rev. 1997;160:115. doi: 10.1111/j.1600-065x.1997.tb01032.x. [DOI] [PubMed] [Google Scholar]
  • 19.Vidal O, Lindberg MK, Hollberg K, et al. Estrogen receptor specificity in the regulation of skeletal growth and maturation in male mice. Proc Natl Acad Sci U S A. 2000;97:5474. doi: 10.1073/pnas.97.10.5474. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Czerkinsky CC, Nilsson LA, Nygren H, Ouchterlony O, Tarkowski A. A solid-phase enzyme-linked immunospot (ELISPOT) assay for enumeration of specific antibody-secreting cells. J Immunol Methods. 1983;65:109. doi: 10.1016/0022-1759(83)90308-3. [DOI] [PubMed] [Google Scholar]
  • 21.Lindberg MK, Weihua Z, Andersson N, et al. Estrogen receptor specificity for the effects of estrogen in ovariectomized mice. J Endocrinol. 2002;174:167. doi: 10.1677/joe.0.1740167. [DOI] [PubMed] [Google Scholar]
  • 22.Weksler ME, Szabo P. The effect of age on the B-cell repertoire. J Clin Immunol. 2000;20:240. doi: 10.1023/a:1006659401385. [DOI] [PubMed] [Google Scholar]
  • 23.Stephan RP, Reilly CR, Witte PL. Impaired ability of bone marrow stromal cells to support B- lymphopoiesis with age. Blood. 1998;91:75. [PubMed] [Google Scholar]
  • 24.Medina KL, Strasser A, Kincade PW. Estrogen influences the differentiation, proliferation, and survival of early B-lineage precursors. Blood. 2000;95:2059. [PubMed] [Google Scholar]
  • 25.Benten WP, Stephan C, Wunderlich F. B cells express intracellular but not surface receptors for testosterone and estradiol. Steroids. 2002;67:647. doi: 10.1016/s0039-128x(02)00013-2. [DOI] [PubMed] [Google Scholar]
  • 26.Smithson G, Medina K, Ponting I, Kincade PW. Estrogen suppresses stromal cell-dependent lymphopoiesis in culture. J Immunol. 1995;155:3409. [PubMed] [Google Scholar]
  • 27.Smithson G, Couse JF, Lubahn DB, Korach KS, Kincade PW. The role of estrogen receptors and androgen receptors in sex steroid regulation of B lymphopoiesis. J Immunol. 1998;161:27. [PubMed] [Google Scholar]
  • 28.Erlandsson MC, Jonsson CA, Islander U, Ohlsson C, Carlsten H. Oestrogen receptor specificity in oestradiol-mediated effects on B lymphopoiesis and immunoglobulin production in male mice. Immunology. 2003;108:346. doi: 10.1046/j.1365-2567.2003.01599.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Offner H, Adlard K, Zamora A, Vandenbark AA. Estrogen potentiates treatment with T-cell receptor protein of female mice with experimental encephalomyelitis. J Clin Invest. 2000;105:1465. doi: 10.1172/JCI9213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Macpherson AJ, Lamarre A, McCoy K, Harriman GR, Odermatt B, Dougan G, Hengartner H, Zinkernagel RM. IgA production without mu or delta chain expression in developing B cells. Nat Immunol. 2001;2:625. doi: 10.1038/89775. [DOI] [PubMed] [Google Scholar]
  • 31.Dudley DJ, Chen CL, Mitchell MD, Daynes RA, Araneo BA. Adaptive immune responses during murine pregnancy: pregnancy-induced regulation of lymphokine production by activated T lymphocytes. Am J Obstet Gynecol. 1993;168:1155. doi: 10.1016/0002-9378(93)90361-l. [DOI] [PubMed] [Google Scholar]
  • 32.Erlandsson MC, Jonsson CA, Lindberg MK, Ohlsson C, Carlsten H. Raloxifene- and estradiol-mediated effects on uterus, bone and B lymphocytes in mice. J Endocrinol. 2002;175:319. doi: 10.1677/joe.0.1750319. [DOI] [PubMed] [Google Scholar]
  • 33.Lindberg MK, Alatalo SL, Halleen JM, Mohan S, Gustafsson JA, Ohlsson C. Estrogen receptor specificity in the regulation of the skeleton in female mice. J Endocrinol. 2001;171:229. doi: 10.1677/joe.0.1710229. [DOI] [PubMed] [Google Scholar]

Articles from Immunology are provided here courtesy of British Society for Immunology

RESOURCES