Gonadotropin treatment restores in vitro interleukin-1β and tumour necrosis factor-α production by stimulated peripheral blood mononuclear cells from patients with idiopathic hypogonadotropic hypogonadism

Abstract

In the present study, we aimed to investigate the effects of testosterone deficiency and gonadotropin therapy on the in vitro production of tumour necrosis factor-α (TNF-α) and interleukin-1β (IL-1β) by peripheral blood mononuclear cells (PBMCs) from patients with idiopathic hypogonadotropic hypogonadism (IHH) in order to elucidate the modulatory role of androgen in cytokine production. Fifteen male patients with untreated IHH and 15 age-matched healthy male subjects were enrolled in the study. Serum follicle-stimulating hormone (FSH), luteinizing hormone (LH), free testosterone (FT), sex hormone binding globulin (SHBG), prolactin, and IL-2 and IL-4 levels were also measured. In unstimulated cultures, IL-1β and TNF-α secretion were not significantly different between patient and control groups. However, after stimulation with lipopolysaccharide (LPS), secretion of IL-1β and TNF-α was significantly higher in cultures from untreated patients with IHH than in control subjects. Mean FSH, LH and FT levels were significantly lower, whereas SHBG, IL-2 and IL-4 levels were significantly higher in patients with IHH compared than in controls. In patients with IHH, FT negatively affected the serum levels of IL-4 and in vitro secretion of IL-1β and TNF-α. In addition, IL-2 and IL-4 affected the in vitro secretion of IL-1β in a positive manner. Gonadotropin therapy decreased both TNF-α and IL-1β in PBMCs from patients with IHH. The levels of serum IL-2 and IL-4 were also decreased by therapy. In conclusion, in the present study, gonadotropin treatment restored the in vitro production of IL-1β and TNF-α by PBMCs from patients with IHH, suggesting that androgen modulates proinflammatory cytokine production, at least directly through its effects on PBMCs. It seems probable that this effect plays an important role in the immunosuppressive action of androgens.

INTRODUCTION

Clinical and experimental evidence supports the hypothesis that sex hormones play a major role in modulating the immune system and that alteration of these hormones can affect the immune response [1–6]. Gender biases in the susceptibility to, and severity of, autoimmune and allergic diseases are well recognized [5,7]. Females are more immunoreactive than males, exhibiting greater antibody formation, cell-mediated immunity and predisposition for autoimmune disorders such as systemic lupus erythematosus (SLE), Sjogren's syndrome, rheumatoid arthritis (RA), Hashimoto's thyroiditis and multiple sclerosis [4,6,8]. However, it is not fully understood how these gender biases in immune-system function affect the clinical expression of autoimmunity. Classically, androgens exert an inhibitory influence on both humoral and cell-mediated immune responses, whereas oestrogens present a dual role by suppressing cell-mediated, but enhancing humoral, immune responses [9]. As androgens play a modifier role in the initiation and continuation of many autoimmune diseases, hypogonadism offers an interesting opportunity to study this relationship. Men with idiopathic hypogonadotropic hypogonadism (IHH) fail to undergo puberty because of deficient secretion of endogenous gonadotropin-releasing hormone (GnRH) [10]. Patients with IHH also have low testosterone without elevation of gonadotrophins. We have recently reported that patients with IHH [1,2] show enhanced immune responses. Although the lack of testosterone action plays a role in this enhanced immune response in these hypogonadal patients, the biological mechanisms behind it need clarification.

Although clinically distinct, all autoimmune diseases have some similarities in their pathogenesis and involve the production of cytokines – important protein mediators that specifically regulate the inflammatory response and the tissue damage and repair mechanisms. There is growing evidence that sex hormones can modify cytokine production or action [11,12]. Sex hormones may alter the T helper 1 (Th1)/T helper 2 (Th2) balance of proinflammatory and anti-inflammatory cytokines during autoimmune disease states, thus influencing the susceptibility to and severity of disease [11–14]. Furthermore, sex hormones might induce B-cell hyperactivity by modifying the cytokine milieu. As cytokines mediate lymphocyte growth and function, a defect in the cytokine network could disrupt the development of self-tolerance. In vitro stimulation of lymphocytes by sex hormones affects cytokine production by T cells and macrophages [15,16]. Proinflammatory cytokines such as IL-1β and TNF-α, secreted by activated monocytes, have a broad range of inflammatory and immunomodulatory actions [17–19]. IL-1β is a potent activator of the humoral immune response. Meanwhile, mice deficient in IL-1β, IL-2, IL-8 and TNF-α are relatively resistant to the induction of autoimmune diseases [20]. However, little is known about the production of these proinflammatory cytokines by peripheral blood mononuclear cells (PBMCs) from patients with IHH and the effects of gonadotropin treatment on this production.

In an attempt to clarify how androgens modulate proinflammatory cytokine production, we investigated the effect of testosterone deficiency and gonadotropin therapy on lipopolysaccharide (LPS)-induced cytokine production by PBMCs from patients with IHH.

PATIENTS AND METHODS

Patients

Fifteen male patients with untreated IHH and 15 age-matched healthy male subjects were enrolled in the study. The diagnosis of IHH was based on failure to undergo spontaneous puberty before 18 years of age and was confirmed by low serum testosterone, normal or low gonadotropin levels, absence of a pituitary or hypothalamic mass lesion on computerized tomography (CT) or magnetic resonance imaging (MRI), presence of a gonadotropin response to repetitive doses of GnRH, and a normal karyotype (46, XY). None of the patients had hyposmia, anosmia, or a family history of IHH. Patients had no history of autoimmune or rheumatic disease and displayed no clinical stigmata indicative of it. All patients had scrotal testis. None of the patients had abnormal levels of serum creatinine, liver enzymes, white and red blood cells, or thrombocytes. All controls had a history of spontaneous puberty and their physical and biochemical findings were within the normal range. Strenuous physical activity was not allowed before collection of blood samples. Use of drugs or conditions affecting lymphokine production (fever, infection, or another inflammatory process) was excluded. Patients were treated with human chorionic gonadotropin (hCG) (Profasi HP 2000; Serona SA, Aubonne, Switzerland) (containing 2000 IU hCG) and human postmenopausal gonadotropin (Pergonal; Serona SA) [containing 75 IU follicle-stimulating hormone (FSH) and 75 IU luteinizing hormone (LH)] three times a week for 6 months. Blood samples were collected from patients and controls between 08·00 and 08·30 h after overnight fasting. Post-treatment blood samples from the patient group were drawn 2 days after injection of human menopausal gonadotropin/hCG. All patients and control subjects were informed about the aim and procedures of the study and gave their consent. The study was approved by the Ethical Committee of Gülhane School of Medicine.

Analyses

Serum FSH, LH and prolactin (PRL) levels were measured by an immunoradiometric assay with reagents from Radim Techland (Angleur, Belgium). The intra- and interassay coefficients of variation (CVs) were 4·4% and 6·0% for FSH, 4·8% and 5·4% for LH, and 4·6% and 6·0% for PRL, respectively. Serum free testosterone (FT) was determined by a solid-phase ¹²⁵I radioimmunoassay (RIA), using reagents from Diagnostic Products (Los Angles, CA). The intra- and interassay CVs were 3·8% and 4·2% for FT, respectively. Serum sex hormone binding globulin (SHBG) was measured by RIA with reagents from Radim Techland. The intra- and interassay CVs were 2·4% and 2·9% for SHBG, respectively. The normal ranges in our laboratory are < 15 IU/l for FSH, < 20 IU/l for LH, 15–45 pg/ml for FT, and 9–55 nmol/l for SHBG. The upper limit for PRL is 12 µg/l.

Monocyte culture

Isolation of peripheral blood monocytes

Peripheral heparinized blood was diluted with serum saline (at a ratio of 1 : 1), and mononuclear cells were obtained by standard Ficoll–Hypaque Separating Solution (Seromed®; Biochrom KG, Berlin, Germany) gradient centrifugation. Mononuclear cells were harvested from the interface, washed three times in HEPES-buffered Hank's balanced salt solution (HBSS) and resuspended in RPMI-1640 (Sigma Chemical Co., St Louis, MO) supplemented with 10% fetal calf serum (FCS) (Biological Industries, Kibbutz Beth Haemek, Israel). Cells were layered on Petri dishes (5 ml; 35 × 10 mm) (Costar®; Cambridge, MA) and incubated at 37°C in a 5% CO₂ humidified atmosphere for 1 h. Non-adherent cells were discarded from the Petri dishes, and the remaining adherent cells (monocytes) were collected by rinsing with cold HBSS and by mechanical scraping. Adherent monocytes were washed in HBSS and resuspended in RPMI-1640 (supplemented with 10% FCS, 100 U/ml penicillin and 100 µg/ml streptomycin; Sigma Chemical Co.) for cell culture. Cell viability, as determined by staining with acridine orange, was> 98% after isolation of the monocytes. The cell suspension contained 92% monocytes.

Identification of monocytes after isolation

Monocytes stained with monoclonal antibodies (MoAbs) specific for CD14 (Leu-M3) and CD45 (Anti-HLe-1) (Simultest™ LeucoGATE™; CD14, Clone MφP9; CD45, Clone 2D1; Becton-Dickinson, San Jose, CA) were identified by flow cytometry (FASCalibur; Becton-Dickinson).

Stimulation of peripheral blood monocytes and monocyte culture

Peripheral blood monocytes were cultured at a concentration of 10⁶ cells/ml in RPMI-1640 (supplemented with 100 U/ml penicillin, 100 µg/ml streptomycin and 10% FCS). Cultures were either unstimulated (media alone) or stimulated with LPS from Escherichia coli serotype O55:B5 (Sigma Chemical Co.), and established in 24-well plates (Costar®). LPS was used at 10 ng/ml for stimulation, as described previously [21]. The cultures were incubated for 24 h at 37°C in a 5% CO₂ humidified atmosphere. Then, the cultures were harvested by centrifugation at 500 g for 10 min and the supernatants were collected and stored at −70°C until used. Cell viability, as determined by staining with acridine orange, was> 90% at the end of the culture.

Cytokine assays

Supernatant concentrations of IL-1β and TNF-α were measured using an enzyme-linked immunosorbent assay (ELISA) (Cytimmune Sciences Inc., College Park, MD, for IL-1β; and Bender MedSystems™ MedSystems Diagnostics GmbH, Vienna, Austria, for TNF-α). Sensitivity levels were 0·87 pg/ml for IL-1β and 5 pg/ml for TNF-α. The intra- and interassay CVs were 7·9% and 11·4% for IL-1β, and 6·9% and 7·4% for TNF-α, respectively. Serum IL-2 and IL-4 levels were determined by ELISA, using reagents from BioSource Int., Inc. (Camarillo, CA). The intra- and interassay CVs were 3·9% and 5·7% for IL-2, and 2·5% and 1·4% for IL-4, respectively. These reagents are sufficiently sensitive to measure concentrations of at least 8·7 pg/ml for IL-2 and of < 2 pg/ml for IL-4.

Statistical analysis

Statistical calculations were performed by using a PC-compatible statistical program, spss 6·0 for Windows (SPSS Inc., Chicago, IL). For comparison of the ages between the patient and the control groups, the Student's t-test was used, Spearman's correlation was used to evaluate the relationship between parameters, and for the remaining calculations the Mann–Whitney U-test was used. The relationships between the variables were evaluated by correlation-regression analysis. All data are expressed as mean ± standard deviation (s.d.). P-values of < 0·05 were used to define the statistical significance.

RESULTS

The mean age of patients with IHH was 21·3 ± 1·5 years (range: 20–23) and 21·7 ± 1·6 years (range: 20–23) in controls. There was no significant difference between the patients and controls with respect to mean age. The mean levels of pre- and post-treatment hormonal and immunological parameters are given in Table 1. The mean levels of FSH, LH, and FT were significantly lower in patients with IHH than in controls, whereas levels of SHBG, IL-2, and IL-4 were significantly higher (Table 1). However, PRL levels were not significantly different between the groups. After treatment with gonadotropin, the mean FT levels were significantly increased. FSH, LH, SHBG and PRL levels did not change significantly. IL-2 and IL-4 levels in the patients were significantly decreased after gonadotropin treatment.

Table 1.

Pre- and post-treatment hormonal and immunological parameters

	Patients with IHH†
Parameters	Pretreatment	Post-treatment	controls	p1*	p2*
FSH (mIU/ml)	0·55 ± 0·29	0·71 ± 0·21	4·10 ± 1·85	<0·001	NS
LH (mIU/ml)	0·92 ± 1·10	1·24 ± 0·74	4·20 ± 1·66	<0·001	NS
FT (pg/ml)	1·10 ± 0·70	25·30 ± 5·30	33·20 ± 11·60	<0·001	<0·001
SHBG (nmol/l)	45·90 ± 15·6	39·20 ± 6·50	29·80 ± 6·30	0·007	NS
PRL (µG/l)	6·60 ± 5·30	7·10 ± 3·60	7·40 ± 4·70	NS	NS
IL-2 (pg/ml)	21·33 ± 8·73	12·29 ± 6·14	13·01 ± 4·62	<0·001	<0·001
IL-4 (pg/ml)	13·19 ± 9·14	4·76 ± 1·94	5·33 ± 2·01	<0·001	<0·001

^*p1, pretreatment versus controls; p2, pretreatment versus post-treatment.

^†n = 15.

Values represent mean ± standard deviation (s.d.).

IHH, idiopathic hypogonadotropic hypogonadism; FSH, follicle-stimulating hormone; FT, free testosterone; IL, interleukin; LH, luteinizing hormone; NS, non-significant; PRL, prolactin; SHBG, sex hormone binding globulin.

IL-1β and TNF-α levels were significantly elevated in the stimulated compared with unstimulated cultures in all study groups (Table 2). In unstimulated cultures, IL-1β and TNF-α levels were not significantly different between patient and control groups. However, after LPS stimulation, a significant increase of IL-1β and TNF-α was found in cultures from untreated patients with IHH compared to controls (Table 2,Fig. 1). After treatment with gonadotropin, IL-1β and TNF-α secretion in stimulated cultures significantly decreased compared with baseline levels (Table 2,Fig. 1).

Table 2.

Values of parameters in stimulated and unstimulated cultures of patient and control groups

	IHH (pretreatment)(n=15)		IHH (post-treatment)(n=15)		IHH (controls)(n=15)
Parameters	Stimulated	Unstimulated	Stimulated	Unstimulated	Stimulated	Unstimulated
IL-1β (pg/ml)	67·9 ± 27·5	922 ± 320*	70·8 ± 32·2	574·5 ± 311·7†	77·7 ± 33·3	418·6 ± 270·9‡
TNF-α (pg/ml)	105·8 ± 37·2	2149 ± 647§	101·8 ± 32·2	1356·2 ± 459·2¶	95·1 ± 49·1	1196·0 ± 720**

Values represent mean ± standard deviation (s.d.).

IHH, idiopathic hypogonadotropic hypogonadism; IL-¹βinterleukin-¹β; TNF-α, tumour necrosis factor-α.

P < 0·001 for comparisons of unstimulated and stimulated cultures in all groups.

P < 0·05 for comparisons of unstimulated cultures in all groups.

P = 0·008 for *versus †P < 0·001 for *versus‡ P = 0·191 for †versus‡.

P = 0·001 for §versus ¶ P = 0·001 for § versus** P = 0·052 ¶versus **.

Fig. 1.

(a) Interleukin-1β (IL-1β) levels of supernatants in unstimulated (medium only) and stimulated [lipopolysaccharide (LPS)] cultures of patients with idiopathic hypogonadotropic hypogonadism (IHH) pre- and post-treatment and controls. (b) Tumour necrosis factor-α (TNF-α) levels of supernatants in unstimulated (medium only) and stimulated (LPS) cultures of patients with IHH pre- and post-treatment and controls.

Results of linear regression analysis for IHH patients are shown in Table 3. FT negatively affected serum levels of IL-4 and in vitro secretion of IL-1β and TNF-α. However, no relationship was found between FT and serum levels of IL-2 (data not shown). In addition, serum levels of IL-2 and IL-4 positively affected the in vitro secretion of IL-1β. No relationship was found between immune and other hormonal parameters in the patients with IHH (data not shown).

Table 3.

Linear regression results for patients with idiopathic hypogonadotropic hypogonadism (IHH)

	IHH
Parameters	Formula	F	P-value
IL-1β	− 0·388.FT + 0·326.IL-2 + 0·298.IL-4	9·455	<0·000
IL-4	− 0·446.FT	10·240	<0·000
TNF-α	− 0·365.FT	8·865	<0·008

FT, free testosterone; IL-1β, interleukin-1β; IL-4, interleukin-4; TNF-α, tumour necrosis factor-α.

DISCUSSION

Increasing attention has focused on the influence of androgens on immune-cell functions and how these may relate to the susceptibility to, or clinical outcomes of, autoimmune diseases. In mice, androgens exert an inhibitory effect on the thymus, causing thymic involution [22] and increased CD8⁺ activity [23]. Castration of male mice also results in an increased number of B cells in the periphery [24], thymic enlargement [25,26] and a decrease in the number of thymic CD3⁺ cells [27]. In humans, reduced serum concentration of androgens in RA, Sjögren's syndrome and adjuvant-induced arthritis have been reported previously [28–30]. Androgen therapy has also been used to decrease various signs and symptoms of SLE, Sjögren's syndrome and RA in humans [28–32]. Klinefelter's syndrome, a hypergonadotropic hypogonadism state, is associated with an increased incidence of autoimmune disorders, including SLE [32–34], progressive systemic sclerosis [35,36] and mixed connective tissue disease [37]. Furthermore, in Klinefelter's syndrome, immunological abnormalities and SLE disease activity have been reported to improve with testosterone replacement [38,39]. Recently, we demonstrated that hypogonadal men exhibited enhanced cellular and humoral immunity, which was reversed by androgenic therapy [1,3].

In the current study, it was observed that LPS-induced production of IL-1β and TNF-α by PBMCs was significantly higher in patients with IHH than in controls. Moreover, because only FT negatively affected the in vitro secretion of both IL-1β and TNF-α, this increased proinflammatory cytokine production may be related to the lack of testosterone action. Similarly to our results, it has been suggested that preincubation with androgens leads to a reduced inflammatory response to stimulation of macrophages from mice and rats, with suppression of TNF, IL-1 and IL-6 [40–42]. It has also been reported that castrated mice exhibit heightened TNF responsiveness to injection of LPS, while testosterone replacement reduced this [43]. Additionally, in vivo administration of the adrenal androgen dehydroepiandrosterone (DHEA) to mice leads to a diminished rise in TNF in response to LPS [44,45]. Research in humans has further demonstrated the anti-cytokine effects of androgens. Testosterone has been shown to reduce IL-1β production in vitro by monocytes from healthy subjects, patients with SLE and patients with RA [13,14]. Danazol, a synthetic androgen, also suppresses IL-1β and TNF production by stimulated monocytes from healthy subjects [46]. Consistent with the above observations, we demonstrated in the present study that gonadotropin treatment significantly attenuated LPS-stimulated secretion of TNF-α and IL-1β by PBMCs from patients with IHH. These results imply that the increased in vitro secretion of proinflammatory cytokines by PBMCs in hypogonadal males is caused by a lack of testosterone action. On the other hand, in the present study, IL-2 and IL-4 levels before treatment were higher than control values and decreased after hormone-replacement therapy. These results are in agreement with our previous findings [1]. Likewise, as demonstrated in our recent study [2], a negative relationship was found between FT and IL-4, but not IL-2. Furthermore, IL-2 and IL-4 positively affected the in vitro secretion of IL-1β in the present study. There are studies, similar to our results, which show that the increases of IL-2 and IL-4 may be a consequence of the high level of IL-1β[17,19]. Our results are also consistent with previous experimental observations that the IL-2 level is four times greater in mice with testicular feminization than in normal mice [26], and castration leads to an increase in IL-2 levels in normal mice [8,47]. Additionally, a high level of IL-4 production is reported in androgen resistance [48]. It has also been shown that dihydrotestosterone exerts a depressive effect on the production of IL-4 by activated murine T cells [49]. IL-2 is a mediator produced chiefly by the Th1 subset of helper/inducer T cells and stimulates other T cells, while IL-4 is produced by Th2 cells and targeted to T and B cells. Finally, interleukin-induced cytotoxicity and antibody production, if subverted against the host, could contribute to an increased incidence of autoimmune disease in hypogonadal men.

In conclusion, in the present study, gonadotropin treatment restored the in vitro production of IL-1β and TNF-α by PBMCs from patients with IHH, suggesting that androgen modulates proinflammatory cytokine production, at least directly through its effects on PBMCs. It seems probable that this effect plays an important role in the immunosuppressive action of androgens.