Both IgG and IgM anti-beta2 glycoprotein I antibodies assays are clinically useful to the antiphospholipid syndrome diagnosis.

OBJECTIVES: The aim of our study was to evaluate the clinical values of anti-beta2 glycoprotein I antibodies (anti-beta2GPI) IgG and IgM comparing with lupus anticoagulant (LA), anticardiolipin antibodies (aCL) in the two clinical groups of antiphospholipid syndrome (APS), vascular thrombosis (VT) and pregnancy morbidity (PM). METHODS: Eighty patients who fulfilled the APS clinical criteria, VT n = 34; PM n = 40, both VT and PM n = 6 were included. LA, aCL and three anti-beta2GPI ELISA kits were tested. RESULTS: Sensitivities of LA, aCL and anti-beta2GPI assays were found respectively 62, 26 and 41% in VT, and 28, 28 and 30% in PM. The sensitivity for the APS diagnosis could reach to 63% using triple tests. The presence of LA (P<0·01, OR = 4·3) or anti-beta2GPI IgG alone (P<0·05, OR = 8·4) was significantly associated with VT. IgM isotype was found more frequent in PM (92%) than in VT (57%) among all positive anti-beta2GPI cases. CONCLUSION: Both IgG and IgM anti-beta2GPI assays were useful when clinical features of APS presented, even its standardization is ongoing. A decreased by half sensitivity of LA in PM compared with that in VT underlines the importance of adding anti-beta2GPI in PM of APS, especially IgM isotype although recent review questioned its significance.

Estrogen-like activity of a subpopulation of natural antiestrogen receptor autoantibodies in man.

We recently reported that a subpopulation of immunoglobulin G (IgG) in man interacts with the hormone-binding site of estrogen receptors (ER), competes with [3H]estradiol (E2) uptake, and decreases effective ER concentrations in cell cultures. The present work further characterizes the immunological properties of these antibodies and defines their biological activity. Using sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blotting techniques, enriched preparations of the natural anti-ER IgG subpopulation (IgGs) were found to specifically immunoprecipitate ER extracted from MCF-7 mammary carcinoma cells and to compete with [3H]tamoxifen-aziridine for ER binding. During 18-h incubations IgGs decreased [3H]E2 binding capacity of MCF-7 cells in a dose-dependent manner similar to E2. Like E2 but unlike antiestrogens, this biological effect corresponded to down-regulation of the receptor protein and depended on a mechanism specifically inhibited by actinomycin D. Moreover, IgGs antagonized the decrease of [3H]E2 binding capacity produced by the strong antiestrogen methyl-hydroxytamoxifen; this antagonism was additive to that of E2. On the other hand, IgGs like estrogens increased progesterone receptor concentrations and cathepsin D secretion. The biological activity of IgGs was neutralized by anti-IgG antibodies and by ICI 164,384, a "pure" steroid antagonist of E2, confirming that immunoglobulins G were responsible for this activity and acted at the E2-binding site. These observations indicate that some natural antibodies in man can function like potent estrogens on ER and mammary cells.

Estrogenic and antiestrogenic down-regulation of estrogen receptor levels: evidence for two different mechanisms.

Preincubation of MCF-7 cells with estradiol (E2) produces a decrease of 3H-E2 binding capacity ("processing"); the strong antiestrogen methylhydroxytamoxifen (MHT) is also effective but with a approximately 100 fold lower efficiency. Parallel immunological measurement of estrogen receptor contents of the cells (ER-EIA from Abbott) revealed that the mechanisms by which these ligands operate are not of the same nature. Thus, while E2 produced a loss of the ER peptide, MHT increased it; indicating an accumulation of a non-binding form of the receptor under its treatment. Measurement of the binding capacity of the cells for 3H-ORG 2058 showed a decrease of PgR concentration after pre-incubation with MHT which contrasted with the classical E2-induced increase of the receptor. MHT at relatively low concentrations also antagonised the E2-induced decrease of 3H-E2 binding capacity; this property did not result from a difference in chemical structure between the ligands since bisphenol a weak estrogenic analogue of MHT failed to show a similar antagonistic activity. This property confers to MHT the ability to reduce the efficiency of E2 to induce PgR. Finally, actinomycin D a known antagonist of the E2-induced processing was found to be totally ineffective towards the MHT processing. This clearly confirmed that the term "processing" covers at least two distinct mechanisms.

Estrogen and antiestrogen interaction with estrogen receptor of MCF-7 cells--relationship between processing and estrogenicity.

Overnight preincubation of MCF-7 cells with 2 x 10(-10) M estradiol (E2) produces a dramatic reduction of their specific [3H]E2 binding capacity. Scatchard plot analysis revealed that this loss of estrogen receptor (ER) concentration, usually termed "processing", occurs without any significant modification of binding properties of the unprocessed receptors. Direct measurement of ER (ER-EIA from Abbott) gave residual receptor concentrations close to those established by binding assay indicating that processing involves the loss of at least one epitope other than the steroid binding site. Incubation with increasing amounts of E2 (0.1 to 5 x 10(-10) M) resulted in an increasing reduction of binding capacity indicating that the extent of processing is associated with the hormone concentration. Steroidal estrogens other than E2 as well as antiestrogens of the triphenylethylene category behaved similarly in this regard although the latter compounds usually acted only when at higher concentrations. The processing capacity of a large series of ligands was compared with the corresponding binding affinity for ER as assessed by classical competitive inhibition of [3H]E2 binding in both cytosol and whole cells. For steroidal estrogens, a large spectrum of concordant values was found which correlated with the known uterotrophic activity of the compounds. On the contrary, weak estrogen and antiestrogens of the triphenylethylene category displayed low processing capacities which were in the order of magnitude of the binding affinities established in whole cells; these values were considerably lower than the corresponding values measured in the cytosol. These observations are consistent with the concept that the capacity of a ligand to process ER is related to its agonistic activity. They also support our hypothesis (J. steroid Biochem. 25 (1986) 677-682) that assessment of the ability of a ligand to inhibit the binding of [3H]E2 in whole cells provides an estimate of its agonistic activity, an estimate which can not be established in the corresponding cytosol assay.

A subpopulation of immunoglobulin G in man selectively interacts with the hormone-binding sites of estrogen receptors.

Since the binding sites of hormone receptors may be similar to those of antihormone antibodies, we wondered whether the former might not be recognized by the idiotypic network. To test this hypothesis we investigated the interaction of plasma immunoglobulin G (IgG) with the binding sites of estrogen receptors (ER) from uterine or mammary tissue. Using ER isolated from uterine cytosol we found that IgG from normal subjects shifted the position of purified receptor in sucrose gradients and displaced [3H]estradiol (E2) from its receptor-binding sites. Equilibrium studies revealed competitive inhibition by IgG of E2 binding to the ER. IgG isolated by adsorption on a rat uterine cytosol-Blue B matrix gel column also bound to the ER, and this binding was inhibited by an excess of E2. After an 18-h exposure of MCF-7 mammary carcinoma cells in monolayer culture to IgG (2 mg/ml), Scatchard analysis of [3H]E2 binding revealed a reproducible decrease in the available receptor sites from 2.52 +/- 0.56 (+/- SEM) to 0.68 +/- 0.48 fmol/microgram DNA (n = 10). This effect was selective, since enriched anti-ER IgG obtained by adsorption on purified receptor was 20 times more potent than total IgG, whereas IgG identically prepared but not retained by affinity chromatography had no activity. Exposure of the cells to the IgG for 45 min also revealed, as with isolated ER, specific competition of the IgG with E2 for the E2-binding sites; the Kd increased from 10.5 +/- 1.6 to 27.5 +/- 7.2 X 10(-11) M (n = 7). Enriched antireceptor IgG was a 20 times more effective competitor, and the IgG not retained by affinity chromatography had no activity. In conclusion, our observations indicate the presence of ER on the cell surface, interaction of ER with IgG from plasma of normal subjects, and competitive antagonism of these IgG with E2 uptake leading to a decrease in effective ER concentrations.

Human chorionic gonadotropin-like substance in plasma of normal nonpregnant subjects and women with breast cancer.

To understand ectopic hormone secretion in cancer we compared the plasma concentrations of the hCG-like substance in normal nonpregnant subjects and in women with breast cancer. In 45 normal men the plasma concentrations did not vary with age (median, 5 pg/ml; range, less than 3-169) whereas in 45 normal women they increased after menopause (median, 48 pg/ml; range, less than 5 -569, n = 20, P less than 0.0001). In 56 women with breast cancer the plasma concentrations of the hCG-like substance after menopause were much higher than in normal women (median, 202 pg/ml; range, 14-1561; n = 35; P less than 0.001), with no abnormally high pituitary gonadotropin values and no relationship with the tumor burden (same median after mastectomy, 198 pg/ml; n = 21). This hCG-like substance was glycosylated and similar to standard hCG according to molecular size, ionic strength, and immunoreactivity. Our data are compatible with the following conclusions: 1) the plasma concentration of the hCG-like substance is normally very low but dependent on gonadal function in women. Its source might be the pituitary gland or peripheral tissues. 2) Its concentration is much increased in postmenopausal women with breast carcinoma. This increase is found with normal pituitary gonadotropin values and is independent of the tumor burden, suggesting it is of extrapituitary and nontumoral origin.