Michèle Lieberherr

Functional testosterone receptors in plasma membranes of T cells

T cells are considered to be unresponsive to testosterone due to the absence of androgen receptors (AR). Here, we demonstrate the testosterone responsiveness of murine splenic T cells in vitro as well as the presence of unconventional cell surface receptors for testosterone and classical intracellular AR. Binding sites for testosterone on the surface of both CD4+ and CD8+ subsets of T cells are directly revealed with the impeded ligand testosterone‐BSA‐FITC by confocal laser scanning microscopy (CLSM) and flow cytometry, respectively. Binding of the plasma membrane impermeable testosterone‐BSA conjugate induces a rapid rise (<5s) in[Ca2+]i of Fura‐2‐loaded T cells. This rise reflects influx of extracellular Ca2+ through non‐voltage‐gated and Ni2+‐blockable Ca2+ channels of the plasma membrane. The testosterone‐BSA‐induced Ca2+ import is not affected by cyproterone, a blocker of the AR. In addition, AR are not detectable on the surface of intact T cells when using anti‐AR antibodies directed against the amino and carboxy terminus of the AR, although T cells contain AR, as revealed by reverse transcription‐polymerase chain reactions and Western blotting. AR can be visualized with the anti‐AR antibodies in the cytoplasm of permeabilized T cells by using CLSM, though AR are not detectable in cytosol fractions when using the charcoal binding assay with 3H‐R1881 as ligand. Cytoplasmic AR do not translocate to the nucleus of T cells in the presence of testosterone, in contrast to cytoplasmic AR in human cancer LNCaP cells. These findings suggest that the classical AR present in splenic T cells are not active in the genomic pathway. By contrast, the cell surface receptors for testosterone are in a functionally active state, enabling T cells a nongenomic response to testosterone.—Benten, W. P. M., Lieberherr, M., Giese, G., Wrehlke, C., Stamm, O., Sekeris, C. E., Mossmann, H., Wunderlich, F. Functional testosterone receptors in plasma membranes of T cells. FASEB J. 13, 123–133 (1999)

Estradiol binding to cell surface raises cytosolic free calcium in T cells

The Fura‐2 method is used to examine a possible action of 17β‐estradiol (E2) on [Ca2+]i of splenic T cells isolated from female C57BL/10 mice. E2 concentrations between 10 fM and 10 nM induce a rapid and dose‐dependent increase in [Ca2+]i due to Ca2+ influx and release of Ca2+ from intracellular stores. Ca2+ influx is mediated by Ca2+ channels which are completely blockable by Ni2+ and partly by nifedipine. The antiestrogen tamoxifen does not inhibit the E2‐induced rise in [Ca2+]i. Ca2+ influx and Ca2+ release from intracellular stores is also inducible by plasma membrane impermeable E2 conjugated to BSA. E2‐BSA‐FITC binds to the surface of T cells of both the CD4+ and CD8+ subset. Our data suggest a novel E2‐signalling pathway in T cells which is not mediated through the classical nuclear estrogen receptor response but rather through putative plasma membrane receptors for E2.

Testosterone induces Ca2+ influx via non-genomic surface receptors in activated T cells

Using the Fura‐2 method we investigated a possible direct action of testosterone on cytosolic free calcium of splenic T cells isolated from female C57BL/10 mice. Testosterone at physiological concentrations of 1–10 nM induces an increase in [Ca2+]i within seconds, which is due to Ca2+ influx and not to Ca2+ release from intracellular stores. In contrast, estradiol induces both Ca2+ influx and Ca2+ release. The testosterone‐induced Ca2+ influx is mediated by Ni2+‐blockable channels and is not inhibited by cyproterone, a blocker of the classical androgen receptor. Ca2+ influx can also be induced by testosterone conjugated to BSA which is impermeable to the plasma membrane. These data indicate a novel mode of direct action of testosterone on T cells which is not mediated through the classical androgen receptor response, but through unconventional plasma membrane receptors.

Testosterone signaling in T cells and macrophages

This review summarizes data about non-genomic actions of testosterone on murine malaria, T cells and macrophages produced by our group during the last 15 years. In C57BL/10 mice, testosterone induces a lethal outcome of blood stage infections with Plasmodium chabaudi which normally takes a self-healing course controlled by genes of the H-2 complex and the non-H-2 background. This suppressive effect of testosterone is mediated neither via the classic intracellular androgen receptor (AR) response nor, after conversion of testosterone to estradiol, via the estrogen receptor. Testosterone acts non-genomically, i.e. through surface receptors, on murine T cells and macrophages, which becomes evident as a rapid rise in the intracellular free Ca(2+) concentration ([Ca(2+)](i)). In T cells, this rise reflects predominantly influx of extracellular Ca(2+), while it is predominantly due to release of Ca(2+) from intracellular Ca(2+)-stores in macrophages. The testosterone-induced rise in [Ca(2+)](i) of both macrophages and T cells is not inhibited by the AR-blocker cyproterone, and it is also inducible by the plasma membrane impermeable ligand testosterone-BSA. The surface receptors initiate a transcription-independent signaling pathway of testosterone. Currently, we are trying to isolate testosterone surface receptors and to investigate a possible cross-talk of non-genomic testosterone signaling with other genotropic signaling pathways.

Down‐Regulation of Osteoclast Differentiation by Daidzein via Caspase 3

Phytoestrogens are plant‐derived compounds with estrogen‐like activity. Phytoestrogen‐rich diets may prevent postmenopausal osteoporosis and these molecules maintain bone mass in ovariectomized animals. We compared the effects of the isoflavone daidzein, which has no action on tyrosine kinases, and 17β‐estradiol on the development and activity of osteoclasts in vitro. Nonadherent porcine bone marrow cells were cultured on dentine slices or on culture slides in the presence of 10−8 M of 1,25‐dihydroxyvitamin D3 [1,25(OH)2D3], with or without 10−8 M of daidzein, 10−8 M of 17β‐estradiol for 9‐11 days. Multinucleated tartrate‐resistant acid phosphatase‐positive (TRAP+) cells that resorbed bone (osteoclasts) developed in the presence of 1,25(OH)2D3. The number of osteoclasts formed in response to 1,25(OH)2D3 was reduced by 58 ± 8% by daidzein and 52 ± 5% by estrogen (p < 0.01); these effects were reversed by 10−6 M of ICI 182,780. The area resorbed by mature osteoclasts was reduced by 39 ± 5% by daidzein and 42 ± 6% by estradiol (p < 0.01). Both compounds also inhibited the 1,25(OH)2D3‐induced differentiation of osteoclast progenitors (mononucleated TRAP+ cells), 53 ± 8% by daidzein and 50 ± 7% by estradiol (p < 0.05). Moreover, daidzein and estradiol promoted caspase‐8 and caspase‐3 cleavage and DNA fragmentation of monocytic bone marrow cells. Caspase‐3 cleavage was reversed by 10−8 M of ICI 182,780. Both compounds up‐regulated the expression of nuclear estrogen receptors ER‐α and ER‐β. Thus, daidzein, at the same concentration as 17β‐estradiol, inhibits osteoclast differentiation and activity. This may be caused by, at least in part, greater apoptosis of osteoclast progenitors mediated by ERs.

Rac-GTPase, osteoclast cytoskeleton and bone resorption.

The members of the Rho-GTPase subfamily, Rac1 and Rac2, are intimately involved in the organization of the cytoskeleton, and the p21-activated kinases or PAKs are targets of these proteins. Rac1 and Rac2 are also essential components of NADPH oxidase, the enzyme responsible for generating free radicals. The cytoskeleton modulates the adhesion of osteoclasts to bone and its subsequent resorption. These cells contain NADPH diaphorase activity, and free radicals influence bone resorption. The influence of Rac1, Rac2 and PAK1 on the cytoskeleton, resorbing activity and NADPH diaphorase activity of disaggregated rat osteoclasts was investigated by permeabilisation with saponin and introducing specific anti-Rac1, anti-Rac2 or anti-PAK1 antibodies. Rhodamine-phalloidin stain was used to identify actin in osteoclasts cultured on plastic slides, and the bone-slice method was used to measure resorption. Saponin permeabilisation did not affect the cytoskeletal organization or bone resorption. Anti-Rac antibodies caused dose- and time-dependent cytoskeletal changes. The osteoclasts rounded up and developed retraction fibers; actin rings were disrupted and large actin dots were seen at the periphery of the cells. Osteoclast resorptive activity was depressed after incubation with the antibodies. The total area resorbed by treated cells and the mean pit area were smaller than those of controls. Anti-PAK1 antibody caused similar changes. None of the antibodies altered the NADPH diaphorase activity. Thus, Rac-GTPases are present in rat osteoclasts and are involved in the organization of the actin cytoskeleton and in resorptive activity. These effects may be mediated by PAK1 kinase, but do not influence osteoclast NADPH diaphorase activity.

G?q/11 and G?? proteins and membrane signaling of calcitriol and estradiol

17β‐estradiol and 1,25‐dihydroxyvitamin D3(calcitriol) rapidly increase (< 5 sec) the concentration of intracellular calcium by mobilizing Ca2+ from the endoplasmic reticulum and forming inositol 1,4,5‐trisphosphate (InsP3) and diacylglycerol. Calcitriol increases InsP3 formation via activation of phospholipase C (PLC)‐β1 linked to a pertussis toxin (PTX)‐insensitive G‐protein, and estradiol via activation of PLC‐β2 linked to a PTX‐sensitive G‐protein. Since PLC are effectors of different subunits of various G‐proteins, we looked for and identified several G‐subunits (Gαq/11, Gαs, Gαi, Gβ and Gγ) in female rat osteoblasts using Western immunoblotting. The action of calcitriol on InsP3 formation and Ca2+ mobilization in Fura‐2‐loaded confluent osteoblasts involved Gαq/11. The membrane effects of estradiol involved Gβγ; subunits, and principally Gβ subunits, but not α‐subunits. These results may provide additional evidence for membrane receptors of steroid hormones. Since PLC‐β1 is the target effector of Gαq/11, whereas PLC‐β2 is only activated by βγ subunits, this specificity may help to generate membrane receptor‐specific responses in vivo. J. Cell. Biochem. 75:138–146, 1999.

Progesterone triggers rapid transmembrane calcium influx and/or calcium mobilization from endoplasmic reticulum, via a pertussis-insensitive G-protein in granulosa cells in relation to luteinization process

We investigated the early effects (5–60 s) of progesterone (1 pM–0.1 μM) on cytosolic free calcium concentration ([Ca2+]i) and inositol 1,4,5‐trisphosphate (InsP3) formation in nonluteinized and in vitro luteinized porcine granulosa cells (pGCs). Progesterone increased [Ca2+]i and InsP3 formation within 5 s in both cell types. Progesterone induced calcium mobilization from the endoplasmic reticulum via the activation of a phospholipase C linked to a pertussis‐insensitive G‐protein. This process was controlled by protein kinases C and A. In contrast, only nonluteinized pGCs showed a Ca2+ influx via dihydropyridine‐insensitive calcium channel. In both cell types, the nuclear progesterone receptor antagonist RU‐38486 did not inhibit the progesterone‐induced increase in [Ca2+]i; progesterone immobilized on bovine serum albumin, which did not enter the cell, increased [Ca2+]i within 5 s and was a full agonist, but less potent than the free progesterone; pertussis toxin did not inhibit progesterone effect on InsP3. In conclusion, progesterone may interact with membrane unconventional receptors that belong to the class of membrane receptors coupled to a phospholipase C via a pertussis toxin‐insensitive G‐protein. The source of the Ca2+ for the progesterone‐induced increase in [Ca2+]i also depends on the stage of cell luteinization.

Different mechanisms are involved in intracellular calcium increase by insulin‐like growth factors 1 and 2 in articular chondrocytes: Voltage‐gated calcium channels, and/or phospholipase C coupled to a pertussis‐sensitive G‐protein

This study describes the mechanisms involved in the IGF‐1 and IGF‐2‐induced increases in intracellular calcium concentration [Ca2+]i in cultured chondrocytes and the involvement of type 1 IGF receptors. It shows that IGF‐1, IGF‐2, and insulin increased the cytosolic free calcium concentration [Ca2+]i in a dose‐dependent manner, with a plateau from 25 to 100 ng/ml for both IGF‐1 and IGF‐2 and from 1 to 2 μg/ml for insulin. The effect of IGF‐1 was twice as great as the one of IGF‐2, and the effect of insulin was 40% lower than IGF‐1 effect. Two different mechanisms are involved in the intracellular [Ca2+]i increase. 1) IGF‐1 and insulin but not IGF‐2 involved a Ca2+ influx through voltage‐gated calcium channels: pretreatment of the cells by EGTA and verapamil diminished the IGF‐1 or insulin‐induced[Ca2+]i but did not block the effect of IGF‐2.2)IGF‐1, IGF‐2, and insulin also induced a Ca2+ mobilization from the endoplasmic reticulum: phospholipase C (PLC) inhihitors, neomycin, or U‐73122 partially blocked the intracellular [Ca2+]i increase induced by IGF‐1 and insulin and totally inhibited the effect of IGF‐2. This Ca2+ mobilization was pertussis toxin (PTX) dependent, suggesting an activation of a PLC coupled to a PTX‐sensitive G‐protein. Lastly, preincubation of the cells with IGF1 receptor antibodies diminished the IGF‐1‐induced Ca2+ spike and totally abolished the Ca2+ influx, but did not modify the effect of IGF‐2. These results suggest that IGF‐1 action on Ca2+ influx involves the IGF1 receptor, while part of IGF‐1 and all of IGF‐2 Ca2+ mobilization do not implicate this receptor. J. Cell. Biochem. 64:414–422.

Effects of extracellular calcium on the proliferation and differentiation of porcine osteoblasts in vitro

Abstract We examined the effects of various extracellular calcium concentrations on DNA content, procollagen type I carboxy-terminal propeptide (PICP) release (reflects type I collagen synthesis), and alkaline phosphatase activity of porcine osteoblasts. Osteoblasts seeded in control medium (2.2 mM calcium) were transferred to low (0.5 or 1 mM) calcium medium or to high (3, 5, 7, or 10 mM) calcium medium at different stages of the culture period and for different incubation times. When osteoblasts were transferred to low or high (3 or 5 mM) calcium medium 1 or 2 days after plating and kept in that medium until the end of the culture period, PICP release was inhibited, but DNA content and alkaline phosphatase activity were unchanged, except in 5 mM calcium, which inhibited alkaline phosphatase activity. Short-term culture of subconfluent and near-confluent osteoblasts in 7 or 10 mM calcium for 48 h inhibited DNA content. DNA content returned to normal levels when cells were transferred back to control medium, whereas alkaline phosphatase inhibition induced by 5, 7, or 10 mM calcium was not reversible. Short-term culture in high calcium media did not affect PICP release. Thus, in porcine osteoblasts, low and high extracellular calcium concentrations affect DNA content, PICP release, and the expression of osteoblastic phenotype markers (alkaline phosphatase activity). These effects are dependent on the duration of calcium treatment and the state of differentiation of the osteoblasts.