Endometrial Lines Research Articles

Diagnostic strategies in endometrial cancer

In patients with abnormal uterine bleeding a normal and thin endometrial line during ultrasound examination can reliably exclude endometrial cancer. Ultrasonography with color Doppler imaging is useful to detect the pedicle artery sign, which–if present–is suggestive of focal endometrial lesions, which need further investigation by operative hysteroscopy or biopsy. If the endometrium is thick or if the endometrial line cannot be clearly visualized both saline infusion sonohysterography and hysteroscopy are very helpful to assess the endometrial cavity and to exclude focal lesions. Histology can often reliably be obtained by office sampling and is always necessary to confirm the diagnosis and to define the grading of endometrial cancer. Once the diagnosis of endometrial cancer has been established, contrast-enhanced MRI is the most reliable technique for pre-operative assessment of the depth of myometrial invasion. It provides more accurate results than peroperative macroscopic inspection or frozen section. However, MRI is not always available or feasible in obese patients with endometrial cancer. Therefore transvaginal ultrasound is still the most commonly used pre-operative imaging method to help in selecting the optimal management. In experienced hands the subjective assessment of the depth of myometrial invasion performs best. Single morphological ultrasound parameters cannot reliably predict depth of myometrial invasion and Doppler ultrasound parameters do not contribute in this prediction.

Receptor mediated antiproliferative effects of the cytotoxic LHRH agonist AN-152 in human ovarian and endometrial cancer cell lines.

Eighty percent of human ovarian and endometrial cancers express receptors for luteinizing hormone-releasing hormone (LHRH). These receptors might be used for targeted chemotherapy with cytotoxic LHRH analogs such as AN-152, in which doxorubicin is linked to agonist carrier [D-Lys6]LHRH. The antiproliferative effects of doxorubicin and AN-152 were assessed in LHRH receptor-positive ovarian (EFO-21, EFO-27) and endometrial (HEC-1A, Ishikawa) cancer cell lines as well as in LHRH receptor negative ovarian SKOV-3 and endometrial MFE-296 lines. The mechanism of action of AN-152 was investigated by a blockage of receptors using an excess of the LHRH agonist [D-Trp6]LHRH. In some cases, confocal laser-scanning microscopy was used to visualize the accumulation of AN-152 or doxorubicin within the cells. In 3 of 4 LHRH receptor-positive cell lines (EFO-21, HEC-1A, Ishikawa) AN-152 was more effective than doxorubicin in inhibiting cell proliferation. The effect of AN-152 was shown to be receptor-mediated because it could be reduced by competitive blockade of the LHRH receptors with [D-Trp6]LHRH. In contrast, AN-152 was less active than doxorubicin in LHRH receptor-negative lines. Confocal laser-scanning microscopy showed an intranuclear accumulation of AN-152 and competitive inhibition thereof by [D-Trp6]LHRH in LHRH receptor-positive cell lines, but no intracellular accumulation of AN-152 could be detected in the receptor-negative SKOV-3 line. These results suggest a selective receptor-mediated action of AN-152 in receptor-positive cell lines.

The detection and monitoring of early pregnancy in the vervet monkey (Cercopithecus aethiops) with the use of ultrasound and correlation with reproductive steroid hormones

Twenty early pregnancies were diagnosed and monitored in vervet monkeys by ultrasonography. Non-gravid uteri became increasingly echogenic from cycle days 7 to 26. The first definite sign of pregnancy was a gestational cavity of 2 mm (+/- 0.80) at 33.0 (+/- 1.48) days menstrual age, which was also used to date all subsequent features. Earlier signs, such as an endometrial line swelling or endometrial 'pregnancy' ring, as reported for other non-human primate species, could not be reliably and consistently used to diagnose pregnancy in vervet monkeys. A rapid increase of the gestational cavity size from days 37 to 49 corresponded closely to a rapid increase in plasma progesterone concentration from day 39 to 49. The first yolk sac was recognizable at 38.0 days (+/- 3.10) and measured 3.3 mm (+/- 0.40) in diameter. A heart beat could be detected at 45.5 (+/- 1.73) days and the size of the first measurable embryo at 35 days was 2 mm. The dating of most features was within the range reported for other non-human primate species.

Compared Randomized Roles and Pain Factor by Vabra and Wallach Endocell in Endometrial Evaluation

This randomized study was carried out to compare pain factors of Vabra and Endocell by an ordered classification of one to ten to evaluate the respective pathological yield and to establish the clinical acceptability. Patients had endometrial evaluation either for abnormal bleeding, abnormal endometrial line as demonstrated by echography or abnormal endometrial cells in the Pap smear. Age, gravidity, parity and hysterometry were similar in both groups, with abnormal bleeding being the most frequent indication (250/370 patients, 67.5%). Both techniques yielded equal results for insufficient tissue retrieval at 46/184 (25.0%) for Vabra and at 50/183 (27.3%) for Endocell. The pain score differed significantly in favor of Wallach Endocell at 4.36 ± 2.38 versus Vabra at 5.65 ± 2.41 (p = 0.001). Vabra and Endocell give similar pathological results; however, Endocell is significantly less painful and easier to manipulate and carry because of its weight (1.69 g vs. 37.80 g).

Cyclic AMP and PMA amplify the effects of IGF-I in the mitogenic activity of endometrial adenocarcinoma cell line HEC-I-A by acting at the G1 phase of the cell cycle

cyclic AMP and PMA amplify the effects of IGF-I in the mitogenic activity of endometrial adenocarcinoma cell line HEC-I-A by acting at the G1 phase of the cell cycle

Differentiation in human endometrial cells in monolayer culture: Dependence on a factor in fetal bovine serum

Human epithelial cells of the Ishikawa endometrial line can be stimulated to differentiate and form multicellular structures in 4-5 day-old monolayer cultures by the addition of a protein factor from fetal bovine serum. Multicellular structures become obvious over an 18-30-h period as the cells enlarge, separate from the dish, and form domes. These structures are similar to those that result from polarization in other epithelial cell lines. Ishikawa dome formation appears to be a multistage process. The appearance of enlarged differentiated cells is detected within hours of adding fetal bovine serum; these enlarged cells lift off the surface of the dish within 6-8 more hours. Domes are observed about 24 h after the addition of fetal bovine serum. Sometimes dome cells migrate into a "bud-like" structure that extends out from the dome. Differentiation of the domes is dependent on a factor from fetal calf serum that behaves similarly to a very large protein or complex of proteins, greater than 300 kd. Progesterone appears to enhance the formation of domes but does not elicit dome formation in the absence of serum factor.

The isolation and long-term culture of normal human endometrial epithelium and stroma: Expression of mRNAs for angiogenic polypeptides basally and on oestrogen and progesterone challenges

A highly reproducible and technically straightforward technique for the isolation and long-term culture of normal human endometrial epithelial cells is described. The essential conditions for long-term culture are that the cells be seeded onto a gelatin matrix and that 'endothelial cell growth supplement' be present in the culture medium. Normal endometrial epithelial cells express cytokeratins and oestrogen receptors. They may be passaged five to six times without change in properties. Growth of normal endometrial epithelial cells was stimulated by 17-beta-oestradiol and epidermal growth factor. Expression of the mRNA coding for seven polypeptide angiogenic factors, by normal endometrial epithelial, stromal and three endometrial carcinoma lines, was examined. The endometrial epithelial and stromal cells express mRNA for the polypeptide angiogenic factors, basic fibroblast growth factor, vascular endothelial cell growth factor, transforming growth factor-beta 1 and pleiotrophin, as well as the cytokine midkine. Expression of the mRNA for both vascular endothelial growth factor and midkine by normal endometrial epithelial cells showed a 2-fold increase on treatment with a physiological dose of 17-beta-oestradiol (10(-10) M) while, in contrast, the mRNA of transforming growth factor-beta 1 decreased 4-fold on treatment with 17-beta-oestradiol (10(-10) M) and was abolished by exposure to progesterone (5 x 10(-9) M). Expression of the mRNAs for angiogenic polypeptides by the endometrial carcinoma lines was more restricted.

The opaH locus of Neisseria gonorrhoeae MS11A is involved in epithelial cell invasion.

In order to produce a successful infection, Neisseria gonorrhoeae (GC) must attach to and invade mucosal epithelial cells. To identify GC gene products involved in this early interaction with host cells we constructed a gene bank derived from a clinical isolate of GC, and isolated a clone which had the capacity to adhere to the human endometrial adenocarcinoma tissue-culture line HEC-1-B. The cloned sequence was identified as a member of the opa gene family whose protein products have been associated with virulence. The GC chromosome contains numerous variant opa genes which, in MS11, are designated opaA-K. Previous work showed that expression of opaC confers a highly invasive phenotype upon strain MS11. When our cloned opa gene was mutated and returned to the GC MS11A chromosome by transformation and homologous recombination, we isolated one transformant that was significantly reduced in its invasive capacity. The locus mutated in this transformant was identified as opaH. Our results indicate that invasiveness of GC for human epithelial cells can be determined by more than one opa gene in strain MS11A.

Effects of transforming growth factors and regulation of their mRNA levels in two human endometrial adenocarcinoma cell lines

The effects of the transforming growth factor-β1 (TGF-β1) and epidermal growth factor (EGF) on the growth of cells from 2 endometrial cancer lines, Ishikawa and HEC-50 were evaluated by measuring rates of DNA synthesis and changes in cell numbers during culture. EGF at 17 and 1.7 nM concentrations consistently enhanced HEC-50 cell proliferation. TGF-β1 inhibited Ishikawa cell proliferation but, unexpectedly for epithelium-derived cells, stimulated HEC-50 cell growth. This effect of interest as it indicates that endometrial cells can acquire an altered responsiveness to a growth inhibitor during the process of malignant transformation. Northern blot analyses showed expression of TGF-α, TGF-β1 and EGF receptors mRNA in both cell lines. Neither estradiol (E 2) nor 4-hydroxytamoxifen (OHTam) affected mRNA levels for either TGF-α or TGF-β in HEC-50 cells, a line unresponsive to E 2 proliferation. In Ishikawa cells, previously shown to respond to both E 2 and OHTam by increasing proliferation rates, E 2 increased TGF-α mRNA and reduced TGF-β mRNA levels. OHTam lowered the levels of both mRNA species, although the effect was greater on TGF-β than TGF-α mRNA. These data are consistent with, but do not prove, the existence of a possible autocrine regulation by TGF-α and TGF-β of human cancer cell proliferation, which might be under E 2 influence in Ishikawa cells.

Intrinsic estrogenicity of some progestagenic drugs

The intrinsic estrogenic activity of some progestins cannot be properly evaluated by using hormone responsive systems when the chosen end-points are also sensitive to progestagenic activity, usually antagonistic of estrogenic actions. We have therefore applied to the evaluation of some drugs commonly used in contraceptive and hormone replacement formulations a recently developed in vitro method to estimate estrogenic activities, which is based on measurements of the estrogen-stimulated alkaline phosphatase activity in cells of the Ishikawa-Var I human endometrial adenocarcinoma line, a response not influenced by progestins. Whereas progesterone, medroxyprogesterone acetate and danazol were found to be devoid of estrogenic activity in this assay, Org OD-14, norethynodrel, gestrinone (R 2323), norethindrone and dl-norgestrel provoked half maximal increases in alkaline phosphatase activity at concentrations (EC-50) of 7, 14, 140, 200 and 2900 nM, respectively, under conditions in which the corresponding value for estradiol was 8 pM. This intrinsic estrogenic activity can be inhibited by antiestrogens, as verified by reversing the effect of R 2323 with 4-hydroxytamoxifen. Since prostaglandin F 2α output by secretory endometrium is increased by estrogens and diminished by progestins, this end-point can serve to evaluate the net effect of drugs with intrinsic estrogenic and progestagenic activities. For instance, R 2323 showed estrogenic activity in this assay whereas Org OD-14 did not. The same in vitro system can be used to evaluate estrogen antagonistic activities of test compounds, using estradiol as the agonist. These in vitro systems are useful in establishing a profile of activities of a drug on a relevant human target tissue, in the screening of synthetic or natural compounds under investigation, and in studies on structure/action relationships.

The growth inhibition of human breast cancer cells by a novel synthetic progestin involves the induction of transforming growth factor beta.

Recent experimental work has identified a novel intracellular binding site for the synthetic progestin, Gestodene, that appears to be uniquely expressed in human breast cancer cells. Gestodene is shown here to inhibit the growth of human breast cancer cells in a dose-dependent fashion, but has no effect on endocrine-responsive human endometrial cancer cells. Gestodene induced a 90-fold increase in the secretion of transforming growth factor-beta (TGF-beta) by T47D human breast cancer cells. Other synthetic progestins had no effect, indicating that this induction is mediated by the novel Gestodene binding site and not by the conventional progesterone receptor. Furthermore, in four breast cancer cell lines, the extent of induction of TGF-beta correlated with intracellular levels of Gestodene binding site. No induction of TGF-beta was observed with the endometrial cancer line, HECl-B, which lacks the Gestodene binding site, but which expresses high levels of progesterone receptor. The inhibition of growth of T47D cells by Gestodene is partly reversible by a polyclonal antiserum to TGF-beta. These data indicate that the growth-inhibitory action of Gestodene may be mediated in part by an autocrine induction of TGF-beta.

Effects of substrata on the polarization of bovine endometrial epithelial cells in vitro

Epithelial-cell function requires cellular polarity in which apical membrane surfaces have unique characteristics and cellular organelles are stratified. Physiological investigations of endometrial epithelial cells would be enhanced greatly by the ability of a method to polarize cells in culture. This study investigates the effects of different substrata on polarization of cultured bovine endometrial epithelial cells. Fetal bovine endometrial epithelial-cell lines were developed from explant outgrowth. Epithelial monolayers were subcultured onto amniotic membranes, Millicell-HA membranes, or Millicell-CM membranes coated with rat-tail collagen, Matrigel, laminin, Vitrogen, or fibronectin. Cultures on these substrata were maintained at the air/liquid interface. Cells grown on either collagen-coated or uncoated Millicell membranes also were maintained submerged in medium. Excellent polarized morphology was attained in cultures grown at the air/liquid interface on amniotic membranes and rat-tail collagen-coated membranes. Lectin-binding patterns to apical membranes of polarized epithelial cell cultures paralleled patterns of binding to bovine endometrial surfaces in vivo. Cultures on rat-tail collagen were maintained for several weeks. These methods provide a valuable system for studying the endometrium in vitro.

Altered estrogen receptor system in estrogen-unresponsive human endometrial adenocarcinoma cells

Previous studies from our laboratories demonstrated that cells from a human endometrial adenocarcinoma cell line (Ishikawa) responded to estradiol whereas cells from another endometrial cancer line (HEC-50) did not. In an attempt to identify factors responsible for the observed estiogen insensitivity we compared the characteristics of the estradiol receptor (ER) systems in Ishikawa and HEC-50 cells. Saturation analyses of cytosolic estrogen binders were performed over a 0.1–70nM range of [ 3H]estradiol concentrations. Equilibrium dissociation constants and number of binding sites were determined by graphic analysis of Scatchard plots or computed by applying Fourier-derived affinity spectrum analysis (FASA) of the binding data. No significant differences were noted in the dissociation constants ( K d approx. 0.6 nM) or number of binding sites (approx. 6–10 fmol/mg protein) for the single binder that could be evaluated by the graphic method in cytosol from the two cell lines. However, 2 binders in Ishikawa cells ( K d approx. 0.2 and 6 nM) could be detected by the FASA method; the higher affinity binder in HEC-50 cells could not be clearly demonstrated. Structural differences in the specific estrogen binders which might distinguish HEC-50 from Ishikawa cells or normal endometrial tissue were investigated by using the anti-ER monoclonal antibody JS 34/32. Interaction of the antibody with [ 3H]estradiol binders ofestrogen-responsive cells and tissue was evident from the formation of labeled complexes that were shown to sediment faster in glycerol density gradients and could be immunoprecipitated with Protein A attached to Sepharose beads. In contrast, the antibody did not recognize labeled specific binders in the HEC-50 cells. Furthermore, [ 3H]estradiol receptors in Ishikawa cells could be transformed into a species that exhibited increased hydrophilicity, evident from its binding to DNA-cellulose, whereas binders from HEC-50 could not. These results indicate that the lack of responsiveness of HEC-50 cells to estrogens might be due to structural or functional alterations in the ER protein resulting in a loss of its capability to undergo estrogen-directed conformational changes required for biological activity.

A nonsialylated high-molecular-weight glycoprotein defined by a monoclonal antibody to adenocarcinoma of the uterine endometrium

A murine monoclonal antibody, MCA-97, was prepared against a human endometrial adenocarcinoma cell line. In a cellular enzyme-linked immunospecific assay, the MCA-97 antibody reacted with all adenocarcinoma cell lines tested, including four endometrial adenocarcinoma lines. The antigen defined by MCA-97 was estimated to be a nonsialylated high-molecular-weight glycoprotein containing galactose and N-acetylglucosamine in its determinant. In immunoperoxidase staining of formalin-fixed, paraffin-embedded sections, MCA-97 reacted with most endometrial adenocarcinomas and ovarian endometrioid-type adenocarcinomas, and also reacted with normal glandular epithelium of the female genital and gastrointestinal tracts. However, it was mostly unreactive with squamous cell carcinoma and ovarian serous adenocarcinoma tissues. By the reversed passive hemagglutination method, the antigen defined by MCA-97 was detected in the sera of one-third of patients with endometrial adenocarcinomas, and in half of those with ovarian endometrioid adenocarcinomas. The antigen was not demonstrable in the sera of most normal female volunteers. Thus, MCA-97 monoclonal antibody has potential clinical applications in diagnostic serology and patholgy.

Effects of serine and glycine on proliferation of an ishikawa cell variant

The proliferation of cells on an Ishikawa human endometrial adenocarcinoma line variant (Ishikawa-Var I) is markedly influenced by the medium used to culture them, viz. MEM vs BM (basal medium; DMEM/Ham's F12, 1/1, with additional glutamine and HETES), under serum-free conditions. Components of BM which are not present in MEM were systematically tested in order to identify those that might account for these differences. Cells were cultured for various periods of time, up to 8 days, in serum-free MEM to which the components to be tested were added. Cell population densities were evaluated using a fluorometric DNA assay when the cells were grown in multiwell plates, or by cell counting when the cells were cultured in plastic dishes. It was found that addition to MEM of a mixture of the amino acids that this medium lacks, significantly increased cell density. By testing individual amino acids at the concentrations present in BM, it could be demonstrated that addition of serine alone was sufficient to obtain the densities achieved with BM. Glycine, a metabolic precursor of serine, had a similar but smaller effect. None of the other missing compounds of BM was effective. Effects of serine on DNA synthesis were also estimated by measuring incorporation of [3H]thymidine for 1 h after a 24 h culture period in MEM. The effect of serine was similar and additive to that of 1% charcoal-treated fetal bovine serum. A serine concentration dependence studied either with this method or measuring DNA/well after 8 days in culture showed detectable effects at 0.005 mM concentration and maximal responses at about 0.025 mM. These findings are of potential importance in studies on regulatory mechanisms of cell proliferation. A possibility to be explored, for instance, is that serine added to the medium increases intracellular phosphatidylserine concentrations leading to increases in the activity of protein kinase C, a stimulator of cell proliferation in some systems.

Mullerian inhibiting substance reduction of colony growth of human gynecologic cancers in a stem cell assay

Mullerian Inhibiting Substance (MIS), a fetal testicular product that causes regression of the Mullerian duct in the male mammalian embryo, was evaluated for its antitumor effect on the premise that a substance active against this genital precursor in the fetus might also be active against tumors derived from these tissues. Increasingly pure fractions of biologically active MIS, prepared from newborn calf testes, were tested in the soft agar colony inhibition assay against single cell suspensions of fresh tumors derived in ascitic or solid form from patients with gynecologic malignancies. Twenty-eight tumor specimens placed in soft agar culture have provided sufficient growth to assess an MIS effect. Twentyfive of these 28 tumors showed significant colony inhibition after incubation with MIS. Increased antitumor response correlated with increased purification of MIS when the same tumor was treated with preparations of different purity. Samples obtained from the same patient at different times, from both ascites and solid tumor sources, produced nearly identical responses to MIS. MIS preparations, previously shown to be active in microcytotoxicity and colony inhibition assays against established human ovarian and endometrial carcinoma lines demonstrate consistent antitumor activity against fresh human gynecologic cancers removed at surgery.

Rapid changes in specific estrogen binding elicited by cGMP or cAMP in cytosol from human endometrial cells.

Addition of cGMP to cytosol of human endometrium or to cells of the endometrial cancer line HEC-1 produced severalfold increases in specific estrogen binding (EB) levels. This effect was maximal with 1 microM cGMP in the presence of 0.1 mM isobutylmethylxanthine (a phosphodiesterase inhibitor) during incubations with [3H]estradiol. In contrast, cAMP decreased EB levels under similar conditions. The effects of cyclic nucleotides on EB levels were complete in less than 15 min in the presence of Mg2+, Mn2+, or Ca2+. The EB sites generated by the addition of cGMP during labeling of cytosol with 10 nM [3H]estradiol were found to sediment in the 8S and 4S regions of low-salt glycerol gradients. No changes in EB levels were observed when cyclic nucleotides were added to cytosol depleted of ATP by preincubation at 4 degrees C for 3 hr, but responsiveness was restored by addition of exogenous ATP. The ATP requirement and the pattern of dependence of cyclic nucleotide actions on divalent cation concentrations suggest that cGMP and cAMP effects may be mediated by kinases and may involve phosphorylations. Another possibility is that the cyclic nucleotides interact allosterically with the binder in the presence of ATP. Addition of sodium molybdate, ATP, and GTP to homogenates of endometrial tissue or HEC-1 cells produces increases in EB levels similar to those obtained by the addition of cGMP. However, these compounds are much less active when added to cytoplasm or cytosol. On the basis of these and other observations, it is hypothesized that molybdate, ATP, and GTP affect EB levels primarily by increasing cGMP concentrations through processes involving a plasma membrane-bound guanylate cyclase.

Effects of cyclic nucleotides on estradiol binding in human endometrium.

The addition of molybdate to intact or homogenized cells of the endometrial adenocarcinoma line HEC-1 during incubation with [3H]estradiol ([3H]E2) at 4 C causes substantial increases in cytoplasmic E2 binding levels. A similar effect can be observed in homogenates of normal human endometrium. These effects of molybdate appear to involve activation of E2-binding sites. Fractionation of the homogenates and recombination of different fractions revealed that activation of specific E2-binding sites by by MoO4= requires cytosolic factors as well as factors associated with the cell membrane. In homogenates of neoplastic cells (HEC-1) and normal endometrium, the addition of ATP, GTP, or cGMP was also found to increase E2 binding to levels as high as those obtained by the addition of MoO4=. In contrast, the addition of cAMP was found to lower specific E2 binding levels and to counteract the effects of MoO4=, ATP, GTP, and cGMP. Levels of intracellular cAMP and cGMP can change rapidly in cells in culture. Since cGMP causes E2 binding levels to increase while cAMP causes them to decrease, changes in the levels of these two cyclic nucleotides may explain the fluctuation in concentrations of specific estrogen binders that we have previously reported to occur in cultured endometrial cells.

Hormone-related enzymatic activities in normal and cancer cells of human endometrium

Histochemical work has identified enzymes in human endometrium predominating in the follicular phase, likely under estrogen control, or in the luteal phase, under combined estrogen-progestin stimulation. We have measured the activity of several of these enzymes in endometrial adenocarcinoma and in normal endometrium during the menstrual cycle, and examined their distribution in glands and stroma. Estradiol dehydrogenase, found preferentially localized in secretory glandular epithelial cells, was induced in isolated glands in organ culture by medroxyprogesterone acetate. This finding indicates that progestins can act directly on epithelial cells in the absence of stroma. Alkaline phosphatase activity was highest in proliferative endometrium and was preferentially localized in glandular epithelial cells. The activity in those cells was lost in primary culture. Levels in adenocarcinoma fell within the range of normal endometrium. Patterns of enzyme inhibition by heat, phenylalanine or homoarginine indicated that endometrial alkaline phosphatase belongs to the group of non-placental organ specific isoenzymes. Specific activities of peroxidase were somewhat higher in secretory than in proliferative endometrium and were located mainly in the stroma. These results contrast with the estrogen inducibility and preferential glandular location of the enzyme in the rodent uterus. High peroxidase activities were found in the majority of cancer tissues examined. Ornithine decarboxylase activity in a human endometrial cancer line (HEC-1B) was found during the period of active proliferation but was not detectable at confluence. In contrast, ornithine decarboxylase activity was absent in cultures of stromal cells even during periods of growth. These findings suggest that elevated ornithine decarboxylase activity may not be a reliable indication of growth.

A method for the study of steady-state kinetics in cell suspensions. evaluation of estrogen metabolism in HEC-1B cells

A procedure has been developed for the superfusion of cells in suspension with pairs of metabolically related labeled compounds ([ 3H]-A and [ 14C]-B). Cells (approx. 6 × 10 6) are placed in a superfusion chamber limited by a cellulose ester membrane of 3μ pore size. The superfusion medium with the tracers is driven through capillary Teflon tubing into a chamber by a syringe pump (10 ml/h) and collected in test tubes immersed in an ice bath. The chamber is mounted on a Vortex mixer for agitation. The superfusion is conducted at 37°C inside a temperature controlled room. After a period of superfusion sufficiently long to produce an isotopic steady state (1–2 h). the concentrations of 3H- and 14C-labeled compounds A and B are measured in the superfusion medium, superfusate and superfused cells. Rates of entry, interconversion, exit and metabolism of A and B per cell are estimated from the data, using formulas derived on the basis of a suitable model. Application of the method to the study of the metabolism of estradiol and estrone in an endometrial adenocarcinoma eell line of human origin (HEC) demonstrated its feasibility. Constant flow rates were maintained without excessive pressure build-up. The cells appeared undamaged and vital at the end of the procedures. Superfusion of homogeneous cell suspensions offers the advantage of allowing uniform access of tracers, oxygen and nutrients to each cell, and provides an experimental design which closely approximates the model on which the calculations are based.