Phosphate transport by embryonic chick duodenum. Stimulation by vitamin D3.

Embryonic chick duodenum maintained in organ culture is a well-suited model for the study of vitamin D effects on inorganic phosphate (Pi) absorption. The system is sensitive to as little as 6.5 nM vitamin D3 (0.1.I.U./ml culture medium). Increased phosphate absorption is observed after 6--12 h of culture. Maximal response (133% of vitamin D-efficient control) is achieved at 24 h. Phosphate uptake by embryonic chick duodenum involves a saturable and a non-saturable component. The former displays characteristics of an active sodium-dependent transport mechanism and is also sensitive to vitamin D3. Presence of the sterol in culture medium raises the maximal velocity from 55 to 75 nmol Pi/min per g tissue. Km remains unchanged (0.5 mM Pi). Duodena cultured in presence of inhibitors of protein synthesis (actinomycin D, alpha-amanitin and cycloheximide) display reduced rates of phosphate absorption. This treatment also prevents vitamin D3 action on phosphate transport. It is concluded that the sterol affects phosphate transport by modulation of synthesis of proteins which are functional in the Pi absorptive process.

Membrane potential measurements in isolated rat liver plasma membrane vesicles: effect of transmembrane ion concentration gradients.

In isolated basolateral and canalicular rat liver plasma membrane vesicles the membrane potential (measured with DiS-C2 (5] varied with transmembrane concentration gradients of Na+, K+ and Cl- revealing the following ion permeabilities: basolateral vesicles: PNa/PK: 0.76, PCl/PK: 0.45 and canalicular vesicles: PNa/PK: 0.69, PCl/PK: 0.56. The data indicate a permselectivity of PK greater than PNa greater than PCl for both membranes.

Investigations on the sodium dependence of bile acid fluxes in the isolated perfused rat liver.

At [Na+]o = 118 mM the concentrative transfer of cholic and taurocholic acid from the perfusate into the isolated rat liver displays saturation kinetics (taurocholate: V = 299 nmol-min-1-g-1, Km = 61 muM; Cholate: V=327 nmol-min-1-g-1, Km = 436 muM). Perfusion with an isotonic sodium-free medium did not change the feature of a carrier-mediated transport but did markedly reduce V without affecting Km (taurocholate: V = 65 nmol-min-1-g-1, Km = 78 muM; cholate: V = 104 nmol-min-1-g-1, Km = 354 muM). It was experimentally assured that the observed reduction of bile salt uptake was not a consequence of regurgitation of bile salts or due to an excessive intracellular accumulation during cholestasis in the sodium-free state. The rate of taurocholate efflux is very low when compared with the rapid rate of the uptake. A stimulatory action of extracellular sodium on this pathway was also observed. Inhibition of the (Na+ + K+)-ATPase by 1 mM ouabain resulted in a decrease of bile salt uptake. Activation of the enzyme by potassium readmission to a K+-deprived liver enhanced bile salt uptake. The immediate response to alteration of the enzyme activity suggests a close association of a fraction of bile acid active transport with the sodium pump.

Stimulation of D-glucose transport. A novel effect of vitamin D on intestinal membrane transport.

Vitamin D stimulates absorption of D-glucose in chick jejunum and ileum by a specific action on the maximal velocity of Na+-gradient driven D-glucose transport across the brush-border membrane of intestinal cells. Induction of D-glucose transport by either vitamin D-3 or 1,25-dihydroxyvitamin D-3 in embryonic intestine can be blocked by inhibitors of RNA and protein synthesis.

On the role of cyclic AMP as a mediator of bone resorption: gamma-interferon completely inhibits cholera toxin- and forskolin-induced but only partially inhibits parathyroid hormone-stimulated 45Ca release from mouse calvarial bones.

The effects of gamma-interferon (gamma-IFN) on bone resorption and cyclic AMP formation stimulated by parathyroid hormone (PTH), forskolin, and cholera toxin have been studied in cultured neonatal mouse calvarial bones. Bone resorption was assessed by the release of 45Ca from prelabeled mouse calvarial bone fragments. Cyclic AMP formation was quantified by analyzing the amount of the nucleotide in calvarial bone tissue. gamma-IFN completely blocked the 45Ca release response to forskolin and cholera toxin in 96 h cultures. In contrast, the 45Ca release response to PTH was only partially inhibited, an effect that was seen over a wide range of PTH concentrations. The inhibitory effect of gamma-IFN was dose dependent, with a threshold for action at 10 U/ml. Forskolin-stimulated 45Ca release could only be inhibited when gamma-IFN was added simultaneously with forskolin; gamma-IFN added to bones prestimulated with forskolin had no effect. The inhibitory effect of gamma-IFN on PTH-stimulated 45Ca release was seen first after a time lag of 48 h. In contrast calcitonin caused an inhibition after only 3 h. PTH and cholera toxin stimulation of radioactive calcium release was also inhibited by gamma-IFN in bones treated with indomethacin. gamma-IFN inhibited forskolin-induced 45Ca release in bones treated with the mitotic inhibitor hydroxyurea. No effect of gamma-IFN on cyclic AMP formation induced by PTH, cholera toxin, or forskolin could be seen. These data show that gamma-IFN inhibits forskolin- and cholera toxin-induced bone resorption by a mechanism unrelated to prostaglandin production or mitotic activity.(ABSTRACT TRUNCATED AT 250 WORDS)

Interferon gamma and calcitonin induce differential changes in cellular kinetics and morphology of osteoclasts in cultured neonatal mouse calvaria.

The present study compares the effects of calcitonin (CT) and interferon gamma (IFN-gamma) on the size, distribution, and ultrastructure of osteoclasts in cultured neonatal mouse calvaria. The number and cross-sectional area of osteoclasts in cultured bones was increased by the addition of parathyroid hormone (PTH) to the culture medium for 24-48 h. Prolonged treatment (up to 72 h) with PTH led to extensive rarefication and formation of holes in the mineralized matrix. PTH-activated osteoclasts exhibited an elaborate ruffled border and showed a typical zonal arrangement of intracellular organelles with the outer cytoplasmic region containing numerous membrane-bound vesicles. CT (20 mU/ml) within 90 minutes caused a complete loss of the ruffled border in PTH-activated osteoclasts. The typical zonal architecture disappeared with intracellular vesicles spread through the entire cytoplasm. Prolongation of CT treatment to 24-48 h led to the appearance of vesicles with dark granular content in inactivated osteoclasts. This morphologically distinct vesicle population allowed us to identify the latter cells as "postosteoclasts" and thereby to distinguish them from osteoclasts precursors. Small active osteoclasts that had originated very likely from these precursors appeared in calvarial bones coincidentally with escape from CT inhibition of bone resorption and resulted in increased total number of osteoclasts. In contrast, IFN-gamma (500 U/ml) acting as a proliferation inhibitor, reduced the total number of osteoclasts. In the presence of PTH it caused no immediate (90 minutes) change in the ultrastructure of PTH-induced osteoclasts apart from an increase in the number of autophagic vacuoles. After prolonged exposure (24-48 h) in the presence of PTH, osteoclasts with low resorbing activity exhibited intermediate borders at their contact zone with the mineralized matrix. In the absence of PTH, the short-term effect of the immune interferon on osteoclast morphology was almost comparable to that of CT. After prolonged treatment in the absence of PTH, postosteoclasts comparable in vesicle population and size to those after CT treatment were found.

Bone-resorbing activity of thyroid hormones is related to prostaglandin production in cultured neonatal mouse calvaria.

The bone-resorbing activity of thyroid hormones was evaluated in neonatal mouse calvaria maintained in organ culture for 96 h. Thyroxine (T4) between 10(-8) and 10(-5) mol/liter and triiodothyronine (T3) between 10(-8) and 10(-7) mol/liter caused a dose-dependent release of calcium from cultured bone. The thyroid hormone effect was delayed in onset for at least 24 h, and after 96 h of culture amounted to 50-90% of the bone-resorbing activity of 10(-8) mol/liter parathyroid hormone (PTH). The bone-resorbing action of T4 as well as of T3 was completely blocked by 100 U/ml interferon-gamma (IF-gamma) or 20 mU/ml salmon calcitonin (CT). "Escape" from CT inhibition, which is a well-known phenomenon in the action of PTH, was not observed with thyroid hormone-mediated bone resorption. Thyroid hormone treatment of cultured calvaria resulted in a gradual increase between 48 and 96 h of medium concentrations of prostaglandin (PG) E2 and particularly of 6-keto-PGF1 alpha, the inactive metabolite of prostacyclin (PGI2). The release of PGF2 alpha in general was not significantly affected. Although the effect of thyroid hormones on PG release from cultured calvaria was completely abolished by 5 x 10(-7) mol/liter indomethacin, in some experiments indomethacin reduced thyroid hormone-mediated bone resorption by only 50%. This indicates that thyroid hormone action on bone is also mediated by a PG-independent mechanism.

17Beta-estradiol antagonizes effects of 1alpha,25-dihydroxyvitamin D3 on interleukin-6 production and osteoclast-like cell formation in mouse bone marrow primary cultures.

In mouse bone marrow primary cultures, the formation of osteoclast-like, i.e. tartrate-resistant acid phosphatase (TRAP)- and calcitonin receptor-positive multinucleated cells (MNC), when induced by 1alpha,25-dihydroxyvitamin D3 (1alpha,25(OH)2D3), can be suppressed by 17beta-estradiol (17beta-E2), whereas 17alpha-E2 is without any effect. 17beta-E2, above 10(-11) M, significantly reduced 1alpha,25(OH)2D3-mediated TRAP+ MNC formation in cultured bone marrow cells from both female and male mice. The estrogen at 10(-8) M suppressed the peak response to the vitamin D sterol by 50%. 17beta-E2 significantly suppressed basal and 1alpha,25(OH)2D3-stimulated cellular production of interleukin (IL)-6. IL-6 alone, although bone marrow cells in hormone-free culture produced appreciable amounts of the cytokine, did not induce any TRAP+ MNC. Therefore, the changes in IL-6 production induced by the hormones could not be the sole determinant for the extent of TRAP+ MNC formation. However, the stimulatory effect of 1alpha,25(OH)2D3 on osteoclastogenesis nevertheless can be significantly reduced by a neutralizing monoclonal anti-IL-6 antibody. In the presence of 10(-8) M 17beta-E2, the anti-IL-6 monoclonal antibody does not achieve any further suppression of 1alpha,25(OH)2D3-related osteoclast-like cell formation. Our data suggest that induction of osteoclastogenesis by 1alpha,25(OH)2D3 is partially dependent on IL-6 signaling and can be modulated by 17beta-E2 through interference with IL-6 receptor activation, in addition to inhibition of IL-6 production by marrow stromal cells.

Regulatory effects of 1alpha,25-dihydroxyvitamin D3 on the cytokine production of human peripheral blood lymphocytes.

We studied the possible regulatory effects of 1alpha,25-dihydroxyvitamin D3 [1alpha,25-(OH)2D3] on cytokine production and differentiation of subsets of CD4+ [T helper 1 (Th1) and Th2] and CD8+ [T cytotoxic 1 (Tc1) and Tc2] lymphocytes at the single cell level. PBMC from healthy donors were cultured with or without 1alpha,25-(OH)2D3 for up to 21 days. On days 0, 7, 14, and 21, the percentage of cytokine-producing T lymphocytes was analyzed by intracellular cytokine detection with mAb and flow cytometry. Simultaneous staining for cell surface markers allowed discrimination of CD4+ and CD8+ T cell subsets. After culture with 1alpha,25-(OH)2D3 (10(-8) mol/L), no significant effects on the proportion of interferon-gamma (IFNgamma)- or interleukin-4 (IL-4)-producing cells were detected, whereas reduced frequencies of IL-2-producing cells in the CD4+ as well as in the CD8+ population were found. An increase in the low percentage of CD4+ and CD8+ T cells producing the Th2 cytokine IL-13 was noticed. Most interestingly, IL-6-producing CD4+ and CD8+ T cells could only be detected in cultures with 1alpha,25-(OH)2D3, reaching a plateau after 14 days. The percentage of IL-6-producing T cells induced by 1alpha,25-(OH)2D3 after a given time period remained stable for at least 7 weeks. Studies of cytokine coexpression revealed that about 70% of IL-6-producing CD4+ and CD8+ cells were also positive for IL-2, but more than 90% were negative for IFNgamma, IL-4, or IL-13, respectively. This suggests that the IL-6-producing population does not match the Th1/Tc1-like (IFNgamma+) or Th2/Tc2-like (IL-4+ or IL-13+) subset. The influence of 1alpha,25-(OH)2D3 on cytokine production by lymphocytes is probably an important point of intersection between the endocrine and the immune system.

Vitamin D stimulates (Na+ + K+)-ATPase activity in chick small intestine.

Vitamin D3 and 1,25-dihydroxyvitamin D3 raise (Na+ + K+)-ATPase activity (ouabain-sensitive 86Rb+ uptake) in cultured embryonic and 4-week-old chick small intestine. Vitamin D stimulation of the sodium pump, which requires genomic action of the sterol, may lead to enhanced Ca2+ extrusion via a basolateral Na+/Ca2+ exchange mechanism, and, in addition, may provide a proliferative signal in undifferentiated enterocytes.

Recombinant gamma-interferon inhibits prostaglandin-mediated and parathyroid hormone-induced bone resorption in cultured neonatal mouse calvaria.

A role of gamma-interferon in the bone remodeling process can be implicated from its interference with bone resorptive processes in cultured neonatal mouse calvaria. The immune interferon is an efficient inhibitor of endogenous prostaglandin synthesis, particularly after stimulation by thrombin or arachidonic acid, and, in addition, has a calcitonin-like inhibitory effect on PTH-induced osteoclastic bone resorption.

Inhibitory effect of amylin on basal and parathyroid hormone-stimulated bone resorption in cultured neonatal mouse calvaria.

Amylin is a secretory product of pancreatic beta-cells which shares major sequence homology with calcitonin gene-related peptide. In neonatal mouse calvaria maintained in organ culture for 72 h, amylin inhibited basal (i.e., unstimulated) resorption in medium concentrations above 2.5 x 10(-9) M. In addition, amylin (> or = 1.0 x 10(-7) M) in a calcitonin-like fashion transiently inhibited bone resorption induced by 1.0 x 10(-8) M PTH ("escape phenomenon"). Pretreatment of calvarial bones with amylin (1.0 x 10(-8) - 1.0 x 10(-6) M) for 72 h attenuated the subsequent response to 1.0 x 10(-8) M PTH. Changes in location and appearance of osteoclasts in amylin-treated bones, as visualized by light microscopy, suggest that amylin inhibits bone resorption by causing a loss of specialized contact zones to the mineralized matrix in resorbing osteoclasts, and in addition, by preventing retraction of osteoblasts from the mineralized surface which impedes attachment of osteoclasts thereon.

Expression of the vitamin D receptor, of estrogen and thyroid hormone receptor alpha- and beta-isoforms, and of the androgen receptor in cultures of native mouse bone marrow and of stromal/osteoblastic cells.

Marrow stromal cells mediate the effect of 1alpha,25-dihydroxyvitamin D3 on formation of osteoclast-like cells from undifferentiated hematopoetic precursors in bone marrow. Induction by the vitamin D hormone of multinucleated, calcitonin receptor- and tartrate-resistant acid phosphatase-positive cells in primary mouse bone marrow culture can be modulated by other members of the steroid/thyroid hormone family, such as triiodothyronine, which has a positive effect, as well as 17beta-estradiol and 5alpha-dihydrotestosterone, which both act as inhibitors of osteoclastogenesis. In an attempt to relate these effects of the steroid/thyroid hormones to the presence of their respective nuclear receptors, we studied expression of the vitamin D receptor (VDR), estrogen receptor (ER)-alpha and -beta, thyroid hormone receptor (TR)-alpha and -beta, and androgen receptor (AR) in total bone marrow as well as primary marrow stromal cell cultures. By using reverse-transcriptase-polymerase chain reaction, in both cases amplification products were obtained, which were identified by multiple restriction fragment length analysis as transcripts from mRNA specific for the ligand-binding domains of the VDR, ER-alpha, ER-beta, TR-alpha, TR-beta, and AR. Specific immunostaining by indirect peroxidase labeling revealed that among the various cell types present in bone marrow, the steroid/ thyroid hormone receptors are abundant particularly in marrow stromal cells. In another series of experiments, we extended our survey on receptor expression also to stromal/osteoblastic cell lines. At the mRNA level, the complete repertoire of steroid/thyroid hormone receptors was present in preadipocytic ST2 cells as well as in osteoblastic MC3T3-E1 cells. By immunocytochemical staining of the latter, it became apparent that single cells exhibit wide variations in intensity of specific signals for all the receptors investigated, so that, notably in contrast to primary stromal cells and ST2 cells, MC3T3-E1 display a mosaic pattern of receptor protein expression.

Comparison between the effects of forskolin and calcitonin on bone resorption and osteoclast morphology in vitro.

The adenylate cyclase activator forskolin (1-10 mumol/L) inhibited 45Ca release from parathyroid hormone (PTH; 10 nmol/L) stimulated prelabeled neonatal mouse calvaria in short term culture (24 h). This effect of forskolin was potentiated by rolipram, Ro 20-1724, and isobutyl-methylxanthine, three structurally different inhibitors of cyclic AMP phosphodiesterase. Forskolin (10 mumol/L) and calcitonin (30 mU/mL) inhibited the mobilization of stable calcium and inorganic phosphate as well as the release of the lysomal enzymes beta-glucuronidase and beta-N-acetylglucosaminidase from PTH-stimulated unlabeled bones. Osteoclasts in PTH-stimulated calvaria showed active ruffled borders with numerous membrane infoldings. Treatment of PTH-stimulated bones with forskolin and calcitonin resulted in a rapid (2 h) loss of the active ruffled border. In addition, forskolin and calcitonin induced similar changes with respect to the number and size distribution of cytoplasmic vesicles in PTH-activated osteoclasts. After 24 h, all signs of osteoclast inactivation were still prominent, whereas after 48 h of treatment with forskolin or calcitonin, the reappearance of a ruffled border on a number of osteoclasts signaled an escape from the inhibitory action of both calcitonin or forskolin. These data indicate that forskolin inhibits bone resorption by a cyclic AMP dependent mechanism and that the effect of forskolin and calcitonin on bone resorption and osteoclast morphology are comparable. These observations lend further support to the view that cyclic AMP may be an intracellular mediator of the inhibitory action of calcitonin on multinucleated osteoclasts.

Interaction of triiodothyronine with 1alpha,25-dihydroxyvitamin D3 on interleukin-6-dependent osteoclast-like cell formation in mouse bone marrow cell cultures.

In mouse bone marrow cultures, the formation of osteoclast-like, that is, tartrate-resistant acid phosphatase-positive (TRAP+) and calcitonin (CT) receptor-positive multinucleated cells (MNCs), induced by 10(-10) to 10(-8) mol/L 1alpha,25-dihydroxyvitamin D3 [1alpha,25(OH)2D3], could be augmented by triiodothyronine (T3), which alone had no effect on osteoclast-like cell formation. The permissive effect of T3 increased the response to 1alpha,25(OH)2D3 by approximately one order of magnitude. Linear concentration dependence was observed between 10(-11) and 10(-8) mol/L T3. Importantly, inhibition of prostaglandin synthesis by indomethacin significantly impeded osteoclast-like cell formation by 1alpha,25(OH)2D3 and abrogated the effect of T3 thereon. Basal interleukin-6 (IL-6) production by cultured marrow cells was significantly stimulated by 1alpha,25(OH)2D3. However, even at an exceedingly high concentration of 20 ng/mL, IL-6 was ineffective in inducing osteoclast-like cell formation. Therefore, any hormonally induced rise in IL-6 release from bone marrow cells could not account for the observed changes in TRAP+ MNC numbers. Nevertheless, the stimulatory effect of 1alpha,25(OH)2D3 on osteoclastogenesis was partially dependent on IL-6 because it could be significantly blocked by a neutralizing monoclonal anti-IL-6 antibody, and to the same extent by a monoclonal anti-IL-6 receptor antibody. Unimpaired signaling through the IL-6/IL-6R system is also a prerequisite for the auxiliary effect of T3 on induction of osteoclast-like cells by 1alpha,25(OH)2D3. Our data provide evidence that 1alpha,25(OH)2D3 induces osteoclast-like cell formation, at least in part, in an IL-6-dependent mode of action, which is also subject to modulation by T3. The mechanism of interaction of the two hormones apparently involves joint stimulation of prostaglandin synthesis.

Growth regulation of human colon cancer cells by epidermal growth factor and 1,25-dihydroxyvitamin D3 is mediated by mutual modulation of receptor expression.

The human colon adenocarcinoma-derived cell line Caco-2 was used as a model system to study the interaction of epidermal growth factors (EGF) and 1,25-dihydroxyvitamin D3 (1,25(OH)2D3) in control of colorectal cancer cell growth. The mitogenic stimulus of EGF was rapidly transduced via apical and basal membrane receptors alike into elevation of c-myc expression, causing a shift of Caco-2 cells from the G0/G1 into the S phase of the cell cycle. The stimulatory effect of EGF on cell division was effectively counteracted by 1,25(OH)2D3: the presence of the steroid hormone prevents the negative effect of EGF on vitamin D receptor abundance and concurrently minimises ligand-occupied EGF receptor numbers on both sides of Caco-2 cell monolayers. Our data suggest that EGF and 1,25-(OH)2D3 actions on mutual receptor levels represent a specific feature of the potent antimitogenic effect of the steroid hormone on colon cancer cells.

Immunocytochemical localization of the extracellular calcium-sensing receptor in normal and malignant human large intestinal mucosa.

We identified the parathyroid type Ca(2+)-sensing receptor (CaR) in normal human colon mucosa and in cancerous lesions at the mRNA and protein level. Polymerase chain reaction produced an amplification product from reverse-transcribed large intestinal RNA which corresponded in size and length to a 537-bp sequence from exon 7 of the CaR gene. With a specific antiserum against its extracellular domain, the CaR could be detected by immunostaining in normal human colon mucosa in cells preferentially located at the crypt base. The CaR protein was also expressed in tumors of the large bowel in all 20 patients examined. However, the great majority of CaR-positive cells in the adenocarcinomas inspected were confined to more differentiated areas exhibiting glandular-tubular structures. Poorly or undifferentiated regions were either devoid of specific immunoreactivity or contained only isolated CaR-positive cells. In the normal mucosa and in glandular-tubular structures of cancerous lesions, the CaR was exclusively expressed in chromogranin A-positive enteroendocrine cells and in only a small fraction of PCNA-positive cells.

Aging, neuroendocrine function, and osteoporosis.

Osteoporosis may occur in both women and men. Osteoporosis is defined as age-related involutional bone loss that exceeds that observed in the normal population of the same age and sex. The hallmark of the disease is a reduction in mineralized bone matrix. Osteoporosis is a multifactorial disease and can be considered the result of impaired coordination of bone remodeling ("resorption/formation coupling"). This review describes the role of systemic hormones and local acting mediators, including neuropeptides and neurotransmitters, in the control of bone turnover. Furthermore the significance of age-related (neuro)endocrine dysfunction for bone remodeling and consequently for development of osteoporosis is discussed.

Surface markers and transendothelial migration of dendritic cells from elderly subjects.

Age-related changes of immune functions have been extensively investigated in both humans and animal models; nevertheless, the literature on potential alterations of dendritic cells, potent antigen presenting cells responsible for initiating immune responses, with aging is very scarce. We studied the immuno-phenotype of peripheral blood dendritic cells of elderly and young subjects by three-color flow cytometry. In addition, the capacity of transendothelial migration, an important step in inflammatory reactions, of peripheral blood dendritic cells of elderly subjects was investigated in an in vitro model. The expression of HLA-DR in the peripheral blood dendritic cells of the elderly subjects was significantly decreased when compared to the young control subjects. The expression of various other surface markers was similar in the young and elderly subjects. The ability of transendothelial migration of dendritic cells was found to be unimpaired in the elderly subjects. Both in the young and elderly subjects a significantly higher expression of CD29, CD86, HLA-DR, and HLA-DQ in the dendritic cells that had migrated through the endothelium in comparison to nonadherent, nonmigrating cells was found. In the migrating dendritic cells of the elderly subjects a significantly increased expression of CD11c was observed, whereas the expression of CD54 was significantly enhanced in the migrating dendritic cells of the young subjects only. In conclusion, our results demonstrate intact functions and a normal immunophenotype of dendritic cells derived from elderly subjects. Dendritic cells thus seem to be functional and therefore are not responsible for the well-known decline of T cell functions with aging.

Immune phenotype and intracellular cytokine production of peripheral blood mononuclear cells from postmenopausal patients with osteoporotic fractures.

A number of factors with known effects on bone turnover are also immune regulatory factors. Disturbances of bone remodeling thus may be a consequence of altered local immune reactivity. We therefore determined surface markers and intracellular cytokine production of peripheral blood mononuclear cells by four-color flow cytometry in 19 postmenopausal patients with established osteoporosis and a control group of 11 postmenopausal women without fragility fractures. No significant differences in bone mineral density as assessed by dual energy X-ray absorptiometry were observed between the two groups. The following surface markers and cytokines were studied: CD3, CD4, CD8, CD16, CD19, CD29, CD45RA, CD56, CD57, HLA-DR, interleukin (IL)-1beta, IL-2, IL-4, IL-6, IL-13, tumor necrosis factor (TNF)-alpha, interferon (IFN)-gamma and granulocyte macrophage colony stimulating factor. In the fracture patients, the percentage of CD8+ cells co-expressing CD57 was increased (14+/-2 vs. 8+/-1%; p=0.03). Moreover, the proportion of CD8+ cells co-expressing TNF-alpha (47+/-5 vs. 33+/-4; p=0.05) and both TNF-alpha and IFN-gamma was significantly higher in the patients than the controls (41+/-6 vs. 22+/-3%; p=0.04). IL-1beta expression tended to be increased in monocytes from patients with established osteoporosis. Distinct subsets of CD8+ cells thus appear to contribute to the development of osteoporotic fractures.