In mammals, estrogens have dose- and cell-type-specific effects on immune cells and may act as pro- and anti-inflammatory stimuli, depending on the setting. In the bivalve mollusc Mytilus, the natural estrogen 17β-estradiol (E2) has been shown to affect neuroimmune functions. We have investigated the immunomodulatory role of E2 in Mytilus hemocytes, the cells responsible for the innate immune response. E2 at 5–25 nM rapidly stimulated phagocytosis and oxyradical production in vitro; higher concentrations of E2 inhibited phagocytosis. E2-induced oxidative burst was prevented by the nitric oxide (NO) synthase inhibitor NG-monomethyl-l-arginine and superoxide dismutase, indicating involvement of NO and O2−; NO production was confirmed by nitrite accumulation. The effects of E2 were prevented by the antiestrogen tamoxifen and by specific kinase inhibitors, indicating a receptor-mediated mechanism and involvement of p38 MAPK and PKC. E2 induced rapid and transient increases in the phosphorylation state of PKC, as well as of a aCREB-like (cAMP responsive element binding protein) transcription factor, as indicated by Western blot analysis with specific anti-phospho-antibodies. Localization of estrogen receptor-α- and -β-like proteins in hemocytes was investigated by immunofluorescence confocal microscopy. The effects of E2 on immune function were also investigated in vivo at 6 and 24 h in hemocytes of E2-injected mussels. E2 significantly affected hemocyte lysosomal membrane stability, phagocytosis, and extracellular release of hydrolytic enzymes: lower concentrations of E2 resulted in immunostimulation, and higher concentrations were inhibitory. Our data indicate that the physiological role of E2 in immunomodulation is conserved from invertebrates to mammals.
- innate immunity
- kinase-mediated cell signaling
in mammals, estrogens exert a broad spectrum of activities on a wide variety of cells and tissues, including the immune system. Estrogens have dose- and cell-type-specific effects on immune cells and may act as pro- and anti-inflammatory stimuli, depending on the setting (5, 22, 24, 43). 17β-Estradiol (E2) can modulate the function of neutrophil granulocytes, monocytes, and macrophages; these cell types have been reported to express intracellular and membrane estrogen receptors (ERs), and their response to E2 can be mediated by nuclear, classical or “genomic” pathways, as well as by rapid, “nongenomic” mechanisms of action (2, 3, 24, 45). These latter mechanisms may be initiated at membrane or cytosolic locations (29, 39, 49) and can result in direct local effects (e.g., modification or ion fluxes) and regulation of gene transcription secondary to activation of cytosolic kinase cascades (5, 22, 24, 29, 43).
Among invertebrates, estrogens have been identified in bivalve molluscs (25, 34, 40, 43), in which their role has been mainly investigated in the control of gametogenesis (20, 21, 30, 33, 34). However, there is evidence that estrogen can represent an important signaling molecule that is involved in roles other than reproduction in these organisms. In neural tissues of the marine mussel Mytilus spp., E2 has been shown to downregulate ganglionic microglial cells after surgical insult, which normally stimulates their egress from the tissue (46), as well as after activation of N-formylmethionyl-leucyl-phenylalanine (48). The effects of E2 on microglial cells were mediated by rapid induction of nitric oxide (NO) release by nervous tissue and were antagonized by classical antiestrogens. These data indicated a receptor-mediated event in Mytilus pedal ganglia (46), where a fragment of ER-β with 100% sequence identity to the human receptor was identified (48).
In Mytilus, E2 was also shown to affect the digestive cells and circulating hemocytes, in particular at the level of lysosomal function (6, 32). In bivalves, hemocytes are responsible for cell-mediated immunity through phagocytosis and various cytotoxic reactions (e.g., lysosomal enzyme and antimicrobial peptide release and production of reactive oxygen intermediates) (13, 41). We previously showed that addition of E2 (in the low nanomolar range) to hemocyte monolayers induced a moderate increase in cytosolic Ca2+ concentration ([Ca2+]), destabilization of lysosomal membranes, morphological changes, hydrolytic enzyme release, and stimulated the bactericidal activity to Escherichia coli (12); all these effects were rapid, occurring from seconds to minutes from E2 addition, and were prevented by the antiestrogen tamoxifen. The effects of E2 on mussel hemocytes were affected by components of tyrosine kinase-mediated cell signaling (12), in particular, the stress-activated p38 MAPK and STAT-like (signal transducers and activators of transcription) proteins, which play a key role in activation of these cells (8–10).
To clarify the effects and mechanisms of action of E2 in the immune response of mussel hemocytes, the effects of the hormone on phagocytosis and oxyradical production were evaluated in vitro. To investigate estrogen signaling, we used specific kinase inhibitors and evaluated the phosphorylation state of PKC and the transcription factor cAMP response element binding protein (CREB) by electrophoresis and Western blot analysis with specific anti-phospho-antibodies. Intracellular localization of ER-like proteins was evaluated by immunofluorescence confocal microscopy. Finally, the possible immunomodulatory role of E2 in vivo was investigated in hemocytes collected from mussels injected with different concentrations of the hormone.
MATERIALS AND METHODS
Mussels (Mytilus galloprovincialis Lam.; 4–5 cm long) were obtained from SEA (Gabicce Mare, Italy) and kept for 1–3 days in static tanks containing artificial seawater (ASW; 1 l/mussel) at 16°C. The seawater was changed daily.
Hemolymph collection, preparation of hemocyte monolayers, and hemocyte treatments.
A sterile 1-ml syringe with an 18-gauge 0.5-in. needle was used to extract hemolymph from the posterior adductor muscle of 10–12 mussels for each experiment. The needle was removed, and hemolymph was filtered through sterile gauze and pooled in 50-ml Falcon tubes at 4°C. Hemolymph serum was obtained by centrifugation of whole hemolymph at 200 g for 10 min, and the supernatant was sterilized through a 0.22-μm-pore filter. Hemocyte monolayers were prepared as previously described (10). Briefly, aliquots of 0.5 ml of hemolymph, corresponding to ∼1–2 × 106 cells, were seeded onto glass coverslips (40 × 22 mm), placed in plastic culture dishes, and incubated at 16°C for 30 min to allow for cell attachment. Nonadherent hemocytes were subsequently removed by gentle washing of the preparations with sterilized ASW. Hemocyte monolayers were added with 1.5 ml of hemolymph serum and kept before use at 16°C.
Hemocytes were incubated at 16°C with E2 (from 10 mM stock solutions in ethanol suitably diluted in ASW). Untreated and control vehicle hemocyte samples were run in parallel. In experiments with antiestrogens, tamoxifen (100 nM final concentration, from a 10 mM stock solution in ethanol) was added 10 min before addition of E2. In experiments with kinase inhibitors, before addition of E2, hemocyte monolayers were pretreated for 20 min with 20 μM SB-203580 (for p38 MAPK) or 2.5 μM GF-109203X (for PKC) as previously described (10, 15). For each experiment, control hemocyte samples were run in parallel. Triplicate preparations were made for each sample. All incubations were carried out at 16°C.
Phagocytosis of neutral red-stained zymosan by hemocyte monolayers was used to assess the phagocytic ability of hemocytes according to the method of Pipe et al. (36) with slight modifications (4). Neutral red-stained zymosan in 0.05 M Tris·HCl buffer [Tris-buffered saline (TBS), pH 7.8, with 2% NaCl (to maintain osmotic conditions)] was added to each monolayer at a concentration of ∼1:50 hemocytes-zymosan diluted in ASW in the presence or absence of E2 and allowed to incubate for 60 min. Monolayers were then washed three times with TBS, fixed with Baker's formol calcium [4% (vol/vol) formaldehyde, 2% NaCl, and 1% calcium acetate] for 30 min, and mounted in Kaiser's medium for microscopic examination with an Olympus Vanox optical microscope. For each slide, the percentage of phagocytic hemocytes was calculated from ≥200 cells. The effect of E2 on phagocytosis was also compared with that induced by 200 nM human recombinant tumor necrosis factor-α (4).
Extracellular oxyradical production by mussel hemocytes was measured by the reduction of cytochrome c (36) with slight modifications. Hemolymph was extracted into an equal volume of TBS (0.05 M Tris·HCl buffer, pH 7.6, containing 2% NaCl). Aliquots (500 μl) of hemocyte suspension in triplicate were incubated with 500 μl of cytochrome c solution (75 μM ferricytochrome c in TBS), with or without E2. Cytochrome c in TBS was used as a blank. The samples were read at 550 nm from 0 to 60 min, and the results are expressed as change in optical density (OD) per milligram of protein. Intracellular oxyradicals were also detected by nitro blue tetrazolium (NBT) assay (36) as described elsewhere (1). Aliquots (500 μl) of hemocyte suspension in triplicate were diluted 1:1 with TBS added with 2% NaCl and incubated for 60 min with an equal volume of 0.1% NBT solution in the presence or absence of E2 at room temperature in the dark. The samples were centrifuged (180 g, 5 min), and the pellet was resuspended in 70% methanol for fixation. The samples were centrifuged again, the supernatant was discarded, and the pellet was air-dried. The pellet was resuspended in 1.1 ml of extraction medium (6 ml of 2 M KOH + 7 ml DMSO) to dissolve the insoluble formazan that had formed, mixed by vortexing, and centrifuged at 3,500 g for 20 min at 10°C. Supernatants were read at 620 nm, and the results were expressed as OD at 620 nm per milligram of protein. Experiments were carried out also in the presence of 1 mM NG-monomethyl-l-arginine (l-NMMA) or superoxide dismutase (SOD) or after preincubation of the sample with tamoxifen or kinase inhibitors. The effects of E2 on extra- and intracellular oxyradical production were compared with those induced by 10 μg/ml phorbol 12-myristate 13-acetate (PMA) in the same experimental conditions.
NO production by mussel hemocytes was evaluated as described previously (51) by the Griess reaction, which quantifies the nitrite (NO2−) content of supernatants. Aliquots of hemocyte suspensions (1.5 ml) were incubated at 16°C with 25 nM E2 or vehicle (ethanol) for 0–5 h in the presence and absence of tamoxifen or inhibitors (l-NMMA, SB-203580, and GF-109203X). Every 30 min, the samples were immediately frozen and stored at −80°C until use. Before analysis, the samples were thawed and centrifuged (12,000 g for 30 min at 4°C), and the supernatants were analyzed for NO2− content. Aliquots (200 μl) in triplicate were incubated for 10 min in the dark with 200 μl of 1% (wt/vol) sulfanilamide in 5% H3PO4 and 200 μl of 0.1% (wt/vol) N-(1-naphthyl)-ethylenediamine dihydrochloride. The samples were read at 540 nm, and the molar concentration of NO2− in the sample was calculated from standard curves generated using known concentrations of sodium nitrite. The effect of E2 on NO2− accumulation was compared with that induced by PMA (50 μg/ml) in the same experimental conditions.
Lysosomal membrane stability.
Lysosomal membrane stability in mussel hemocytes was evaluated by the neutral red retention time assay as previously described in in vitro (9, 12, 14, 15) and in vivo (11) experiments by Lowe et al. (31). Hemocyte monolayers on glass slides were incubated with 30 μl of a neutral red solution (final concentration 40 μg/ml from a stock solution of neutral red at 20 mg/ml DMSO); after 15 min the excess dye was washed out, 30 μl of ASW were added, and the slides were sealed with a coverslip. In in vitro experiments, hemocyte monolayers were incubated with E2 for 30 min before addition of neutral red (12). Every 15 min, hemocytes were examined under an optical microscope, and the percentage of cells showing loss of the dye from lysosomes in each field was evaluated. For each time point, 10 fields, each containing 8–10 cells, were randomly observed. The end point of the assay was defined as the time at which 50% of the cells showed lysosomal leaking (i.e., the cytosol stains red and the cells are rounded). The effect of E2 on lysosomal membrane stability was compared with that induced by hemocyte incubation with E. coli (strain MG155) under the experimental conditions previously described (9).
Immunolocalization of ER-like receptors.
Immunolocalization of ER-like receptors in hemocytes was evaluated by confocal laser scanning microscopy (CLSM). Hemocyte monolayers, seeded on polylysine-coated glass coverslips, were fixed for 30 min at room temperature in 4% paraformaldheyde dissolved in 5 mM PBS, pH 7.4, with 2% NaCl. All subsequent steps were carried out at room temperature in the presence of PBS with NaCl. The cells were washed three times with PBS, permeabilized for 10 min with 0.2% Triton X-100 in PBS, and treated for 30 min with 2% BSA in PBS (PBS-BSA) to minimize possible unspecific binding of the antiserum. Hemocytes were then incubated with human polyclonal anti-ER-α (anti-rabbit) or anti-ER-β (anti-goat) antibodies (1:50 dilution in PBS-BSA) for 60 min. The cells were washed three times in PBS-BSA and incubated for 60 min with donkey anti-rabbit IgG-rhodamine-labeled (Santa Cruz Biotechnology) or rabbit anti-goat IgG-Cy3-labeled (Santa Cruz Biotechnology) as secondary antibodies (1:250 dilution in PBS-BSA), respectively, following the manufacturer's instructions. The cells were washed three times in PBS, and coverslips were mounted on glass slides using 5% glycerol-1,4-diazabicyclo[2.2.2]octane in a 90% glycerol-10% PBS (10×) antifading mixture.
The slides were mounted on the stage of a Leica inverted microscope, which is part of a CLSM system (model TCS SL, Leica). The hemocytes were observed using a ×63 immersion objective (1.32 NA). Illumination was provided by an argon ion laser filter. A tetramethylrhodamine isothiocyanate wide filter (excitation at 543 nm) and a Cy3 filter (excitation at 520 nm) were used to visualize ER-α- and ER-β-like immunoreactivity. Neutral density filters and reduced laser powers were utilized to reduce photobleaching nearly to zero. To focus hemocytes, samples were viewed under reduced transmitted light illumination; then, in confocal fluorescence mode, 10 video frames, from 500-nm adjacent sections, were averaged. The confocal images (1,024 × 1,024 bit resolution) were viewed on a high-resolution monitor and saved to disk.
Electrophoresis and Western blot analysis.
Levels of phosphorylated PKC and CREB in whole cell extracts from hemocyte monolayers were determined using phosphospecific antibodies as previously described (9, 11, 14). Supernatants from each culture dish were discarded, and each hemocyte monolayer was lysed with 1 ml of ice-cold lysis buffer [50 mM Tris·HCl, pH 7.8, 0.25 M sucrose, 1% (wt/vol) SDS, 1 μg/ml pepstatin, 10 μg/ml leupeptin, 2 mM sodium orthovanadate, 10 mM NaF, 5 mM EDTA, 5 mM N-ethylmaleimide, 40 μg/ml PMSF, and 0.1% Nonidet P-40] and sonicated for 45 s at 50 W. The samples were boiled for 4 min and then centrifuged for 10 min at 14,000 g to remove insoluble debris. Supernatants were mixed 1:1 (vol/vol) with sample buffer (0.5 M Tris·HCl, pH 6.8, 2% SDS, 10% glycerol, 4% 2-mercaptoethanol, and 0.05% bromophenol blue), and the samples (normalized for protein content before they were loaded to 30 μg of protein) were resolved by 10% (for PKC) or 12% (for CREB) SDS-PAGE (27). Prestained molecular mass markers were run on adjacent lanes. The gels were electroblotted and stained with Coomassie blue (53). The blots were probed with human recombinant anti-pan-phospho-PKC (1:1,000 dilution), anti-phospho-PKC-α/βII (1:1,000 dilution), or anti-phospho-CREB (1:1,000 dilution) as primary antibodies and horseradish-peroxidase-conjugated goat anti-rabbit IgG (1:3,000 dilution) as secondary antibody. Nitrocellulose membranes were stripped for 30 min at 50°C with stripping buffer (62.5 mM Tris·HCl, pH 6.7, containing 10 mM β-mercaptoethanol and 2% SDS) and reprobed with anti-actin antibodies (1:1,000 dilution) as loading control (9, 14) or with anti-CREB (1:1,000 dilution) (11). Immune complexes were visualized using an enhanced chemiluminescence Western blot analysis system (Amersham Pharmacia) following the manufacturer's specifications. Western blot films were digitized (Chemidoc; Bio-Rad), and band ODs were quantified using a computerized imaging system (QuantityOne). Relative ODs (arbitrary units) were normalized for the control band in each series.
In vivo E2 exposure experiments.
Mussels were injected with different concentrations of E2: 50 μl of 0.1, 0.5, and 2 μM E2 solutions containing 5, 25, and 100 pmol, respectively, of E2 (from a 10 mM stock solution in ethanol diluted in ASW) were injected into the posterior adductor muscle of groups of 10–12 mussels with use of a sterile 0.1-ml syringe as previously described (11). This range of concentrations was chosen because the average volume of hemolymph that can be withdrawn from mussels of this size (4–5 cm) is 0.6–1 ml. Experiments were repeated three times. For each experiment, a parallel set of control mussels were injected with 50 μl of a solution of ASW containing an equal amount of ethanol (≤0.005%). The mussels were then placed in plastic tanks containing ASW at 16°C (0.5 l/mussel). At 6 and 24 h after injection, hemolymph was withdrawn using a sterile syringe, and hemocytes were utilized for determination of lysosomal membrane stability, phagocytosis, and hydrolytic enzyme release. Lysosomal enzyme release was evaluated by measurement of lysozyme activity in the extracellular medium (12) as described previously (18), and data were expressed as lysozyme equivalents per milligram of protein per milliliter with hen egg white lysozyme as a standard.
Values are means ± SD of at least three experiments in triplicate. Data from densitometric analysis of Western blots are means ± SD of three independent experiments. Statistical analysis was performed by using Mann-Whitney's U-test, with significance at P ≤ 0.05.
All reagents were of analytic grade. E2 and anti-β-actin (rabbit polyclonal) antibody were obtained from Sigma (St. Louis, MO); rabbit polyclonal anti-pan-phospho-PKC (Ser660), anti-phospho-PKC-α/βII (Thr638/641), anti-phospho-CREB (Ser133), and anti-CREB from New England Biolabs (Beverly, MA); anti-ER-α (catalog no. H-184) and anti-ER-β (catalog no. N-19) antibodies from Santa Cruz Biotechnology (Santa Cruz, CA); and prestained low- and high-molecular-mass markers from Bio-Rad (Hercules, CA). All other reagents were purchased from Sigma.
Effects of E2 on hemocyte functional parameters.
The effects of E2 on hemocyte phagocytosis are reported in Fig. 1. A significant stimulation was observed at 5 and 25 nM E2 (+26 and +37%, respectively, P ≤ 0.05), whereas higher concentrations of E2 (50 nM) inhibited the phagocytic process (−32%, P ≤ 0.05; Fig. 1A). E2-stimulated phagocytosis was prevented by cell pretreatment with the antiestrogen tamoxifen and with the specific p38 MAPK inhibitor SB-302580; a similar effect was observed with the PKC inhibitor GF-109203X (Fig. 1B). Neither inhibitor, at the concentration tested, showed cytotoxic effects in mussel hemocytes or affected lysosomal membrane stability (10, 15) and phagocytosis (4) in control cells.
The effects of E2 on oxyradical production by mussel hemocytes are reported in Fig. 2. As shown in Fig. 2A, 25 nM E2 induced a rapid and time-dependent increase in extracellular oxyradical production, evaluated spectrophotometrically as cytochrome c reduction, reaching a maximal twofold increase with respect to control at 60 min (P ≤ 0.05); the effect of E2 was prevented by hemocyte pretreatment with tamoxifen. A significant increase was also observed with 5 nM E2 (+50%, P ≤ 0.05). As shown in Fig. 2B, addition of the O2− scavenger SOD, which significantly decreased the basal level of oxyradical production in control hemocytes, did not affect E2-induced cytochrome c reduction; on the other hand, the NO synthase (NOS) inhibitor l-NMMA prevented the effect of E2. Similar results were obtained with SB-302580 and GF-109203X (Fig. 2B).
As shown in Fig. 2C, 25 nM E2 induced a twofold increase in intracellular oxyradical production compared with control (P ≤ 0.05), as evaluated by the NBT reduction assay. The effect was abolished by SOD and l-NMMA. Moreover, E2-induced NBT reduction was prevented by tamoxifen, as well as by SB-302580 and GF-109203X.
NO production in hemocyte supernatants was evaluated by the Griess reaction, which measures the concentration of the stable NO2− product. In control hemocytes, little NO2− production was observed over 0–6 h; addition of 25 nM E2 induced a significant increase in NO2− concentration from 30 min to 4 h (not shown); although a large variability in the time course of E2-induced NO production was observed among different hemocyte pools, in all experiments, maximal NO2− concentration was observed at 4 h (Fig. 2D). The effect was prevented by cell pretreatment with l-NMMA, as well as with tamoxifen and p38 MAPK and PKC inhibitors. Higher concentrations of E2 (50 nM) did not affect oxyradical production (data not shown).
We previously showed that E2-induced destabilization of lysosomal membranes in mussel hemocytes was mediated by p38 MAPK activation (12). The effect of E2 on lysosomal membrane stability was prevented by cell pretreatment with the PKC inhibitor GF-109203X (Fig. 3).
Immunofluorescence microscopy of ER-like receptors.
Intracellular localization of ER-like receptors in hemocyte monolayers was investigated by immunofluorescence labeling and CLSM utilizing anti-ER-α and anti-ER-β antibodies (Fig. 4). Prominent ER-α immunoreactivity was observed in the nuclei; cytoplasmic staining was also observed, except in large intracellular vacuoles (Fig. 4A). ER-β immunoreactivity was extranuclear; a strong signal was observed in the perinuclear region, and punctate cytoplasmic staining indicated association with small intracellular organelles (Fig. 4B).
Effect on PKC phosphorylation.
The effect of E2 on hemocyte PKC phosphorylation was first evaluated by Western blotting with anti-pan-phospho-PKC antibodies as previously described (9, 14) (Fig. 5). The antibody, which detects α-, βI-, βII-, δ-, ε-, and η- isoforms of mammalian PKC only when phosphorylated at a COOH-terminal residue homologous to Ser660 of human PKC-βII, recognizes two phosphorylated protein bands of ∼70 and 75 kDa, respectively, in mussel hemocytes, as in protein extracts from mammalian cells (9). As shown in Fig. 5A, 25 nM E2 induced a rapid and transient increase in phosphorylation of the 75-kDa protein band; densitometric band analysis (inset) revealed that the effect was maximal at 5 min (∼5-fold compared with control, P ≤ 0.05). On the other hand, phosphorylation of the 70-kDa protein band was decreased. The 75-kDa PKC isoform was previously shown to correspond to PKC-α and -β isoforms, as evaluated with a specific anti-phospho-PKC-α/βII antibody directed toward phosphorylated Thr638/641 residues (14). As shown in Fig. 5B, E2 induced an increase in phosphorylation of the PKC-α/βII protein band (up to 3-fold compared with control at 5 min, P ≤ 0.05). An anti-actin blot is shown as loading control (Fig. 5C). As previously demonstrated (9), no significant changes in PKC phosphorylation were observed in control hemocytes (not shown).
Effect on CREB phosphorylation.
The effect of E2 on CREB phosphorylation was evaluated as previously described (11) using specific antibodies directed toward the phosphorylated Ser133 of CREB (Fig. 6). E2 induced a rapid increase in CREB phosphorylation compared with control (Fig. 6A), whereas the level of total, unphosphorylated CREB did not change (Fig. 6B). Densitometric band analysis revealed a maximal effect at 5 min (4-fold compared with control, P ≤ 0.05).
In vivo effects of E2 on hemocyte function.
The possible in vivo effects of exogenous E2 on hemocyte function were investigated in mussels injected with different amounts of E2. Hemocyte parameters were evaluated in cells collected from mussels injected with 5, 25, and 100 pmol of E2 after 6 and 24 h, and the results were compared with those obtained in hemocytes from vehicle-injected mussels (Fig. 7). E2 induced a significant and concentration-dependent decrease in lysosomal membrane stability at 6 and 24 h of exposure (Fig. 7A). A particularly large lysosomal membrane destabilization (≥50%) was observed with 25 and 100 pmol of E2 (nominal concentration in hemolymph). A small but significant increase in phagocytic activity was observed only after 6 h with the lowest E2 concentration (+16% compared with control, P ≤ 0.05; Fig. 7B). On the other hand, higher concentrations inhibited phagocytosis; with 100 pmol of E2, the effect was observed after only 6 h (−23% compared with control, P ≤ 0.05), and with 25 and 100 pmol of E2 the effect was observed after 24 h (−17 and −37%, respectively, P ≤ 0.05).
E2 also stimulated extracellular enzyme release, evaluated as lysozyme activity in the extracellular medium (Fig. 7C). The lowest concentration induced a large increase in lysozyme release 24 h after injection (+231% compared with control, P ≤ 0.05); higher concentrations induced smaller but significant increases at 6 and 24 h (+70 and +31% at 25 pmol of E2 and +35 and +23% at 100 pmol of E2, respectively, P ≤ 0.05).
The results demonstrate that E2 in vitro can rapidly affect phagocytosis and oxyradical production in Mytilus hemocytes; the effects were prevented by the antiestrogen tamoxifen. The results confirm and expand previous observations indicating that E2, in the same experimental conditions, induced tamoxifen-sensitive lysosomal membrane destabilization, morphological changes, degranulation of hydrolytic enzymes, and stimulation of bactericidal activity (12).
E2 significantly stimulated the phagocytic process in a narrow concentration range (5–25 nM), whereas higher concentrations (50 nM) were inhibitory. Along with increased phagocytosis, stimulation of the oxidative burst was observed, as indicated by measurement of extra- and intracellular oxyradical production by standard methods for bivalve hemocytes (54). The extent of stimulation was similar to that observed after challenge of Mytilus hemocytes with PMA (1; present study) or laminarin (1). Addition of SOD prevented only E2-stimulated intracellular oxyradical production, indicating an increase in O2− generation associated with increased phagocytic activity. Moreover, extra- and intracellular oxyradical production was prevented by addition of the NOS inhibitor l-NMMA, which prevents NO synthesis in molluscan hemocytes (44). A similar effect was induced by nitro-l-arginine methyl ester (data not shown). NO production was confirmed by the significant increase in NO2− concentration, with a maximum after 4 h of E2 addition. In the same experimental conditions, a similar, although smaller, effect was induced by PMA (50 μg/ml). The effect was prevented by tamoxifen and l-NMMA, suggesting that, in Mytilus hemocytes, E2 induces NO synthesis through a receptor-mediated event, as previously demonstrated in ganglionic tissue (46). Because in bivalve molluscs phagocytosis and oxyradical production through the action of NADPH oxidase and NOS represent important nonspecific defense mechanisms (1, 23, 35, 44, 50), the results demonstrate that E2 can act as an immunostimulant on mussel hemocytes. In invertebrates, NO can be produced rapidly and after a latent period through constitutive (cNOS) and inducible (iNOS) NOS isoforms, with cNOS modulating basal NO actions that are further enhanced by iNOS-derived NO (44). If we consider that the NO2− determination usually does not include cNOS-derived NO release and the time course of NO production in response to E2, our data support the hypothesis that rapid E2 signaling subsequently leads to increased NO production, probably through disinhibition of an iNOS form in the hemocyte (44).
In vertebrate phagocytes, the oxidative burst is mediated by several intracellular pathways, including MAPKs, and Ca2+-dependent and Ca2+-independent PKC isoforms, leading to NADPH oxidase activation (52). E2-stimulated phagocytosis and oxyradical (including NO) production were prevented by SB-302580, confirming that, as previously demonstrated, the stress-activated p38 MAPK represents a target for the action of E2 in these cells (12). In particular, 5 and 25 nM E2 induced a rapid (from 5 min) and transient increase in the level of phosphorylated p38 (12). The p38 MAPK inhibitor prevented the effect of E2 on lysosomal membrane stability (12), a sensitive parameter of in vitro exposure to E2 in Mytilus cells (6, 32). In mussel hemocytes, the effects of E2 on all the functional parameters tested, including lysosomal membrane stability, were prevented by GF-109203X, and PKC and, in particular, a PKC-α/βII-like isoform (14) were rapidly and transiently phosphorylated in response to E2. In the same conditions, E2 induced a small but significant increase in cytosolic [Ca2+], indicating that [Ca2+]-mediated cell signaling may participate in activation of PKC-like classical isoforms (12). PKC is involved in rapid estrogen signaling through ER-dependent and -independent mechanisms (19, 50). Because activation of p38 MAPK and PKC has been shown to play a key role in the response of Mytilus hemocytes to bacterial challenge (9, 10), overall, the results demonstrate that estrogen signaling through p38 MAPK and PKC occurs in invertebrate cells and participates in immunostimulation.
E2 also induced a transient increase in the phosphorylation of a CREB-like protein previously identified in mussel hemocytes (11). CREB, a transcription factor that is activated by serine/threonine kinases, is involved in the cross talk between estrogen, ERs, and cytosolic kinase cascades, including MAPKs (17, 28). CREB-like proteins have been identified in the nervous system of gastropod molluscs, where they play a role in neuronal plasticity and learning (42). In mussel hemocytes, in vivo exposure to the xenoestrogen bisphenol A significantly decreased phosphorylation of CREB at Ser133 (11); on the other hand, large CREB phosphorylation is rapidly induced by bacterial challenge with E. coli (unpublished results). The present results demonstrate that E2 signaling in mussel hemocytes, as in mammalian cells, involves CREB-like proteins.
Overall, the results further support the hypothesis that rapid activation of kinase pathways represents a significant mode of action by the natural estrogen E2 in Mytilus hemocytes in vitro. In Mytilus ganglia, the effects of the membrane-impermeant E2-BSA conjugate were similar to those of free E2, suggesting an E2-mediated signaling pathway at the cell surface coupled to NO production (46). In hemocytes, E2-BSA, similar to E2, induced lysosomal destabilization and MAPK and STAT activation; however, part of the E2-BSA conjugate was rapidly internalized by the cells and, therefore, seemed unsuitable to demonstrate the extracellular action of the hormone (12). E2-BSA stimulated the oxidative burst, but it induced a slight decrease (10–15%), rather than an increase, in phagocytosis of zymosan particles (data not shown).
The results obtained with the antiestrogen tamoxifen indicate a possible role for ER-like receptors in mediating the rapid effects of E2. We previously demonstrated the presence of ∼70- and 49-kDa immunoreactive ER-α- and ER-β-like proteins, respectively, by electrophoresis and Western blot analysis of soluble hemocyte protein extracts with use of antibodies directed against a specific region in the NH2-terminal domain of ER-α and ER-β, respectively, which is the highly divergent domain between the two mammalian forms (12). The results of confocal microscopy utilizing the same antibodies confirm the presence of ER-like proteins in control hemocytes. ER-α immunoreactivity showed a prominent nuclear localization and a diffuse signal in the cytoplasm; on the other hand, ER-β-like proteins were observed at extranuclear sites, with most cells showing a strong signal in the perinuclear region. A distinct immunostaining was also observed in cytoplasmic organelles: large vacuoles showed a characteristic absence of immunoreactivity to ER-α antibodies, whereas punctate ER-β immunostaining indicated association with smaller vacuoles. In Mytilus spp., granular hemocytes, the dominant cell type in the hemolymph, are characterized by a low nucleus-to-cytoplasm ratio, high phagocytic activity, and capacity for oxyradical production (54). The distinct association of the two ER-like proteins with intracellular granules may reflect the heterogeneity of these organelles in terms of enzyme composition (lysosomal hydrolytic enzymes and oxidases) (16, 37) and the possible role of lysosomal compartments in ER degradation (39). E2-induced localization of ER in perinuclear lysosomes was first demonstrated by Szego (49) and others (Ref. 39 and references quoted therein), supporting a role for lysosomes in nucleocytoplasmic communication in the response to estradiol. Moreover, recent studies indicated that, in different cell types, perinuclear localization of ER-β isoforms was due to association of the ER with the mitochondria (Ref. 7 and references quoted therein). Our results, although preliminary, suggest that, in mussel hemocytes, ER localization in lysosomal compartments is highly likely; however, experiments are needed to identify the nature of ER-positive vacuoles utilizing lysosomal and mitochondrial markers in control and E2-treated hemocytes.
The possibility that E2 may affect the hemocyte function in vivo at longer exposure times was investigated in mussels injected with different concentrations of E2. In hemocytes from control and E2-injected mussels, functional parameters that have been shown to be affected by E2 in vitro, i.e., lysosomal membrane stability, extracellular lysozyme release, and phagocytosis (12; present study), were evaluated. In mussels injected with E2, a concentration-dependent decrease in lysosomal membrane stability was observed compared with control 6 and 24 h after injection. These results demonstrate that in vivo the lysosomal function represents a sensitive target for the action of the hormone in mussel cells, as previously shown in vitro (6, 12, 32). At the lowest concentration tested (5 pmol), E2 induced a large increase in extracellular release of lysosomal hydrolytic enzymes, comparable to that induced by E. coli (38). The smaller effect observed at higher concentrations may be due to unspecific degranulation associated with cell damage, as indicated by the large decrease in lysosomal membrane stability observed in these conditions. Similarly, the lowest concentration of E2 stimulated phagocytosis only 6 h after injection, whereas at longer exposure times, and with higher concentrations, a decrease in phagocytosis was observed.
The effects and mechanisms of action of E2 at longer incubation times in vivo may be more complex than those involved in the rapid effects on cell signaling observed in vitro: they may be mediated by membrane and intracellular signaling pathways, with or without involvement of ER-like receptors, or reflect the integrations of actions at different receptor pools, as in mammalian cells (29). In addition to the direct effect on the hemocytes, exposure to E2 can also affect the activity of steroid-metabolizing enzymes in other mussel tissues (gonad and digestive gland) that metabolize exogenous E2 to an esterified form (25), thereby affecting the concentration of free E2 in the hemolymph; E2 exposure may also affect the levels of endogenous modulators that have been shown to affect the immune function in bivalves (4, 26, 47).
Steroid concentrations reported in the literature in bivalve tissues widely differed (from picograms to nanograms) depending on the species, tissue, sexual maturation stage, and analytic technique (21). The only available data on estrogen content in the hemolymph are those reported by Stefano et al. (46) and Zhu et al. (55) in Mytilus edulis. Although the concentration of free E2 was not evaluated in the present study at different times after injection, preliminary data, obtained utilizing a commercial competitive chemiluminescent enzyme immunoassay kit (Immulite 2000 Estradiol), indicate that a only a small fraction of the injected E2 was retained in hemolymph serum (unpublished observations). Further research utilizing E2 dissolved in seawater and measurements of E2 isoforms in the hemolymph with appropriate analytic techniques are needed to clarify the effects of exogenous E2 on mussel immune function in vivo.
In mammalian immunocytes, estrogens induce dose- and cell-type-specific responses, thus resulting in pro- or anti-inflammatory effects (2, 5, 22, 24, 43, 45). Overall, the results indicate that, in Mytilus hemolymph, changes in the concentrations of E2 in a narrow range may elicit immunostimulation through activation of the signaling pathways involved in the immune response. Higher concentrations can downregulate the hemocyte function, probably through impairment of the mechanism involved in controlled membrane fusion events during the phagocytic process leading to lysosomal membrane destabilization (present study; 6), lysosomal enlargement, and increased protein degradation (6). In bivalve immunocytes, the lysosomal vacuolar system, which plays a key role in different aspects of the immune function, seems to represent a significant target for the action of E2.
Synthetic estrogens, such as 17α-ethynylestradiol, diethylstilbestrol, and mestranol, showed similar effects, although at much higher concentrations (15; unpublished data). In mussels, the effects of the natural estrogen E2 on the immune function may be relevant in physiological conditions. Overall, our data demonstrate that, in bivalve molluscs, E2 can modulate the immune function in vitro and in vivo and support the hypothesis that the role of estrogens in nonreproductive functions is conserved in evolution.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 the American Physiological Society