Genomic structure and promoter analysis of the ecto-phosphodiesterase I gene (PDNP3) expressed in glial cells.

PDNP (phosphodiesterase I/nucleotide pyrophosphatase) is one of a series of ectoenzymes that are involved in hydrolysis of extracellular nucleotides. PDNP possesses ATPase (EC 3.6.1.3) and ATP pyrophosphatase (EC 3.6.1.8) activities. Mammalian PDNP consists of three closely related family proteins (PDNP1, -2, and -3), and they are expressed in different cell types and at different developmental stages. Rat PDNP3 is expressed in a subset of immature glial cells and in the alimentary tract. Human PDNP3 is expressed in glioma cells, prostate, and uterus, but not in the alimentary tract. We have cloned genomic DNA containing the whole coding region of the human PDNP3 gene and determined its exon-intron structure. The human PDNP3 gene spans over 60 kb and is organized into 25 exons and 24 introns. We determined the nucleotide sequence of the 5'-flanking region of human and rat PDNP3 genes. The upstream region of both species lacks a canonical TATA box and contains a putative binding site for CCAAT enhancer-binding proteins near the transcription start site. Promoter activity analysis of the 5'-flanking region revealed that the sequence around the CCAAT box is required for its transcriptional activity in 9L rat glioma cells. A gel shift assay demonstrated that 9L nuclear extract contains proteins that bind to this region.

A comparative study on transforming growth factor-beta and activin A for preantral follicles from adult, immature, and diethylstilbestrol-primed immature mice.

Both transformation growth factor-beta (TGFbeta) and activin belong to the TGFbeta superfamily, and each receptor is structurally related. We have shown that the action of activin A on folliculogenesis is different in immature and adult mice, so it is of interest to study whether TGFbeta has such an action on follicular development. The effect of TGFbeta on folliculogenesis was studied in isolated preantral follicles from immature, adult, and diethylstilbestrol (DES)-primed immature mice and was compared with that of activin A. TGFbeta caused a significant increase in follicular diameter and estradiol and immunoreactive inhibin secretion in adult mice in a dose-related manner, but did not affect the size of preantral follicles from immature mice. Activin A, on the other hand, caused a significant increase in the size of follicles from immature mice, but did not change the size of preantral follicles from adult mice. TGFbeta enhanced the effect of FSH, whereas activin A completely blocked the action of FSH on preantral follicles from adult mice. Such a specific action of TGFbeta and activin A was age dependent because preantral follicles obtained from 28-day-old mice, compared with those from 11- and 56-day-old mice, showed an intermediate reaction to TGFbeta and activin A. DES pretreatment of 11- and 28-day-old mice caused an enhanced response to FSH, but this response was completely inhibited by TGFbeta. These results indicate that both TGFbeta and activin A have proliferative action and cytodifferentiative action on granulosa cells, but the action of each is age dependent and opposite in direction. In conclusion, although both TGFbeta and activin A belong to the same family, and each receptor is structurally related, both share a specific role in early folliculogenesis before and after puberty.

Effects of growth hormone, activin, and follistatin on the development of preantral follicle from immature female mice.

The aim of this study was to investigate whether GH and insulin-like growth factor I (IGF-I) are involved in preantral folliculogenesis and, if so, to clarify the relationship between GH/IGF-I and activin/follistatin (FS) systems in immature female mice. Ovaries were obtained from 11-day-old mice, and preantral follicles, 100-105 microm in diameter, were mechanically isolated and selected for culture. Ten preantral follicles per well were cultured with different quantities and combinations of additives as follows: no additives (control), recombinant human FSH (rhFSH), IGF-I, recombinant human GH (rhGH), activin A, and recombinant human FS (rhFS). Mean diameters of the follicles were measured daily, and estradiol and immunoreactive inhibin levels in the cultured medium were assayed by RIA on day 4. rhGH showed stimulatory effects on the follicular diameter and the secretion of estradiol and immunoreactive inhibin. These effects were augmented by the presence of IGF-I and activin A. IGF-I alone did not show any stimulatory effect. The addition of rhFSH to activin A or to rhGH and activin A promoted preantral follicular growth and hormone production. On the other hand, GH- or activin-stimulated follicular growth was suppressed by rhFS in a dose-dependent manner. These results indicate that activin A and rhGH may play an important role in controlling earlier phases of follicular development during the infantile period, which is considered to be gonadotropin independent.

Effect of adrenal and ovarian androgens on type 4 follicles unresponsive to FSH in immature mice.

The present study investigates the physiological significance of dehydroepiandrosterone, dehydroepiandrosterone sulfate, T, androstenedione (Delta(4)), dihydrotestosterone (DHT), estrone (E1), and E2 on recombinant human FSH- (rhFSH) resistant type 4 follicles obtained from immature mice. Type 4 follicles of a diameter of 100-120 microm with one or two granulosa cell layers around the oocyte and an intact basal lamina with theca cells were isolated from the ovaries of 11-d-old BDF-1 mice and cultured with medium alone (control) or with dehydroepiandrosterone, dehydroepiandrosterone sulfate, T, Delta(4), DHT, E1, or E2 at concentrations ranging from 1 x 10(-11) to 1 x 10(-7) M for 4 d. We examined the mean diameters of type 4 follicles, levels of immunoreactive (IR)-inhibin, and E2 and progesterone in the culture media on day 4. In addition, we evaluated follicular cell proliferation by immunofluorescence staining with 5-bromo-2'-deoxyuridine. All tested androgens significantly increased the diameter of type 4 follicles in a dose-dependent manner without the production of IR-inhibin and E2. The nuclei of granulosa cells in type 4 follicles cultured with all tested androgens exhibited intense 5-bromo-2'-deoxyuridine-positive staining, compared with those of controls. In contrast, neither E1 nor E2 had any stimulatory effects. The stimulatory effects of T, Delta(4), or DHT were inhibited by an AR antagonist in a dose-related fashion but not by an aromatase inhibitor. Furthermore, all tested androgens had a synergistic effect with rhFSH on follicular growth and the production of IR-inhibin and E2. These results demonstrated that neither adrenal nor ovarian androgens are arteriogenic but that they stimulate type 4 follicles unresponsive to rhFSH and augment the responsiveness of these follicles to rhFSH.

Activin from secondary follicles causes small preantral follicles to remain dormant at the resting stage.

The purpose of the present study was to investigate 1) whether activin A can cause primary follicles to become dormant at the resting stage, and 2) the role of the secondary follicle on follicular growth of primary follicles. Preantral follicles (100-120 microm in diameter) harvested from adult mice and cultured in in vitro follicle culture system showed a significant increase in size and estrogen and inhibin secretion in response to FSH, but the administration of activin A blocked the effect of FSH. Withdrawal of activin A not only restored the follicular response to FSH but also enhanced the effect of FSH, indicating that the action of activin A is to cause small preantral follicles to become dormant at the preantral stage. To investigate the role of secondary follicles in early folliculogenesis, small preantral follicles were cocultured with secondary follicle (300-350 microm in diameter) in the presence of FSH. The secondary follicle showed a significant increase in follicular diameter as a result of stimulation by FSH, but the small preantral follicles did not increase in size. After removal of the secondary follicle, however, the small preantral follicles commenced follicular growth, indicating that the growth of small preantral follicles is suppressed by the secondary follicle. Administration of the activin binding protein follistatin caused a significant increase in follicular diameter of both small preantral and secondary follicles as a result of stimulation by FSH. These results have suggested that secondary follicles cause primary follicles to become dormant at the resting stage by secreting activin.

Analysis of prognostic factors in stage IIB-IVA cervical carcinoma treated with radiation therapy: value of computed tomography.

PURPOSE: To define the influence of the tumor size measured by computed tomography (CT) and lymph node involvement detected by CT in patients treated with radiation therapy for Stage IIB-IVA carcinoma of intact uterine cervix. METHODS AND MATERIALS: This was a retrospective analysis of 233 patients with uterine cervical cancer managed with both external irradiation and high-dose-rate intracavitary brachytherapy (HDR-ICR) at Kanagawa Cancer Center. The results were analyzed for the end points of absolute survival (AS), disease-free survival (DFS), pelvic control (PC), and central control (CC). The parameters of stage, CT-measured anterior-posterior (AP) cervix size, and CT-detected lymph node metastases were evaluated using univariate and multivariate analysis. RESULTS: The stage, AP cervix size, and lymph node involvement were significant pretreatment factors in univariate analysis with respect to AS, DFS, PC, and CC. Multivariate analysis confirmed that significant risk was associated with certain prognostic parameters. Those in terms of AS, in order of decreasing significance, were lymph node involvement, AP cervix size, age, and total HDR-ICR dose. When DFS was studied, lymph node involvement and AP cervix size were demonstrated to have a significant effect. Stage and lymph node involvement significantly affected PC. CONCLUSION: Because the International Federation of Gynecological Obstetrics staging system fails to incorporate important prognostic information about tumor volume and lymph node involvement, CT-detected lymph node metastases as well as CT-measured cervix size should be determined as complementary additional prognostic measures.

Observations on the cell biology of tissue factor in endothelial cells.

Human umbilical vein endothelial cells (HUVEC) are inducible for tissue factor (TF) activity in culture. Based on experiments using ECGF (4-20 micrograms/ml) with heparin (90 micrograms/ml), we obtained the following results: 1) In confluent HUVEC cultures, ECGF had essentially no influence on the levels of inducible TF. 2) In growing HUVEC cultures, ECGF reduced the TF response shortly after seeding but full response was regained when cells were kept confluent for 2-3 days. 3) Although secondary cultures responded best to TF induction in the absence of ECGF, the response was essentially equal over at least 8 passages in the presence of ECGF. 4) Of total cellular TF induced in HUVEC, about 25% was available on the surface, and less than 4% was released with the shed plasma membrane vesicles. The proportion of total TF activity available on the surface of intact cells was not influenced by the presence of ECGF. 5) T1/2 for the decay of TF activity induced was 8.3-9.5 h, whereas in HUVEC when protein synthesis was blocked after TF induction a T1/2 of about 30 h was found.

Tissue factor released from leukemic cells.

Using clotting assay and radioimmunoassay (RIA), tissue factor activity (TFA) and TF related antigen (TFR:AG) were determined in an extracellular culture medium of HL-60 cells. After 12 h incubation, TFA and TFR:AG in the medium with endotoxin (EDX: 1 microgram/ml) reached maximums which were 1.8 and 2.1 times greater than those in the medium without EDX, respectively. In the leukemic cells of 10 patients with acute nonlymphoid leukemia (ANLL), TFR:AG showed a significant correlation with TFA (p less than 0.01). On day 1 of the induction chemotherapy, TFR:AG in the 7 patients with DIC significantly increased to 288.9 +/- 153.1 ng/ml (p less than 0.01), whereas no increase in TFR:AG was recognized in the 3 patients without DIC. These results suggest that TF may be released from leukemic cells into the culture medium or blood stream, and that this may correlate with the development of DIC.

Induction of tissue factor synthesis in human umbilical vein endothelial cells involves protein kinase C.

Incubation of human umbilical vein endothelial cells with one of the following compounds: endotoxin, recombinant interleukin-1 beta, recombinant tumor necrosis factor alpha, allogenic lymphocyte subpopulations or phorbol ester resulted in significant induction of tissue factor synthesis. Diacylglycerol had the same effect and also enhanced synergistically the induction caused by endotoxin and interleukin-1 beta. Two different inhibitors of protein kinase C, H7 and sphingosine, inhibited tissue factor synthesis at concentrations which did not depress protein synthesis in general, suggesting that protein kinase C is involved in the processes leading to tissue factor synthesis. Cells down-regulated for the tissue factor response to TPA responded essentially normally to endotoxin and interleukin-1 with regard to tissue factor synthesis.

One-stage method for assay of tissue factor activity of leukemic cell with special reference to disseminated intravascular coagulation.

Tissue factor activity (TFA) of leukemic cells (1 x 10(8) cells/mL) was measured in 44 patients with acute nonlymphoid leukemia (ANLL) by the one-stage assay using factor-IX deficient plasma (OSA-dIX) and two-stage assay (TSA). According to the preventative heparin dose schedule based on the TFA measured by the TSA, all disseminated intravascular coagulation (DIC) was controlled successfully. The procedure of the TSA was too complicated for clinical use, and its minimal measurable value was 125 units (U)/L of TFA. The OSA-dIX was simpler in its procedure and sensitive enough to measure accurately a TFA quantity as small as 30 U/L with high reproducibility. In 20 ANLL patients with 125 U/L or more of TFA measured by both assays, there was a significant relationship between their logarithms of TFA (r = 0.93, P less than 0.01). These results suggested that DIC complication in ANLL patients would be controlled successfully by the administration of heparin dosage based on the TFA measured by the OSA-dIX.

Blood coagulation factor Xa interacts with a linear sequence of the kringle 2 domain of prothrombin.

Prothrombin is a vitamin K-dependent plasma protein composed of several functional domains, which is proteolytically activated into thrombin by factor Xa in the presence of factor Va, Ca2+, and phospholipids. During the activation, prothrombin is cleaved into three fragments: fragment 1, containing a domain rich in gamma-carboxyglutamic acid residues and kringle 1 domain; fragment 2, containing the kringle 2 domain; and a protease catalytic domain, thrombin. Here we studied the interaction site for factor Xa in human prothrombin during the activation. The isolated fragment 2 inhibited the activation of prothrombin by either prothrombinase complex or factor Xa alone in a dose-dependent manner, whereas fragment 1 and diisopropylphosphate (DIP)-thrombin did not. Factor Xa directly bound to fragment 2 immobilized to microwell plates with a Kd of 9.0 x 10(-8) M, but not to fragment 1 or DIP-thrombin. Factor Xa also bound to immobilized prothrombin and prethrombin 1 with Kds of 2.0 x 10(-7) and 1.5 x 10(-7) M, respectively, suggesting that factor Xa interacts with the kringle 2 domain in these molecules. The binding of factor Xa to immobilized fragment 2 was Ca(2+)-dependent with an optimal concentration at 6 mM. In the presence of Ca2+, the interaction was enhanced by phospholipids in a concentration-dependent manner. To localize the factor Xa-binding site in the kringle 2 domain, fragment 2 was digested with lysyl endopeptidase and then trypsin after reduction and S-carboxymethylation. The resulting peptides were immobilized to microwell plates and assayed for factor Xa binding ability. The amino acid sequence of the peptide positive in the assay was determined to be residues His205 to Arg220. Factor Xa bound to a synthetic peptide corresponding to the residues His205 to Arg220 immobilized to microwell plates. The peptide inhibited factor Xa-catalyzed activation of prothrombin, but a peptide with the reversed sequence of His205 to Arg220 did not. These findings indicate that factor Xa interacts with at least a linear sequence, His205 to Arg220, in the kringle 2 domain of prothrombin during its activation into thrombin.

Endometrial dating in the conception cycle.

OBJECTIVE: To investigate endometrial dating during the conception cycle. DESIGN: Endometrial biopsies of the last half of the luteal phase of conception cycles were dated based on urinary luteinizing hormone (LH) surges. SETTING: Endometrial samples were obtained from women attending the fertility clinic of the Department of Obstetrics and Gynecology at Gunma University Hospital. PATIENTS: One hundred eighty-five women, of whom 15 (8.1%) conceived during this study. MAIN OUTCOME MEASURES: Urinary LH surge, serum progesterone (P) levels, and endometrial dating. RESULTS: All 10 women who conceived showed 2 or within 2 days in-phase endometrial biopsies on days 7 to 11 after the LH surge. In 4 of 5 women biopsied on day 12, an unequivocally delay in the stroma was found, i.e., a persistence of edema and poor development of predecidual reaction. Since serum P levels in conception cycles were significantly higher than in nonconception cycles on days 10, 11, and 12, we interpreted this delay in the stroma as a consequence of conception. CONCLUSION: Endometrial specimens during the last half of the luteal phase of conception cycles are in-phase until day 12. On day 12, gestational hyperplasia causes apparent out-of-phase.

Prediction and prevention of ovarian hyperstimulation by monitoring endogenous luteinizing hormone release during purified follicle-stimulating hormone therapy.

OBJECTIVE: To elucidate whether early detection of premature luteinizing hormone (LH) release can be useful for prediction and prevention of ovarian hyperstimulation during purified follicle-stimulating hormone (FSH) therapy. DESIGN: Retrospective and prospective study. PATIENTS: Infertile women with polycystic ovarian syndrome. MAIN OUTCOME MEASURES: Correlation between rate of endogenous LH release and incidence of excessive ovarian enlargement. In the prospective study, LH was measured by fluorometric enzyme immunoassay to obtain real-time concentration. Maximal ovarian diameter by ultrasonography. RESULTS: The rate of excessive ovarian enlargement (greater than or equal to 60 mm) in cycles that were treated by a daily administration of purified FSH and accompanied by premature LH release was 83.3%. This rate was significantly higher than that in cycles without premature LH release (24.1%, P less than 0.001). Interruption of FSH administration or alternate day FSH administration in cycles with premature LH release reduced the rate of ovarian enlargement to 45.5% (P less than 0.05). CONCLUSION: Early detection of premature LH release is useful for prediction of ovarian hyperstimulation. Ovarian hyperstimulation can be reduced by modulating the dose of FSH and the interval of administration in cycles with premature LH release.

Ovulation induction by step-down administration of purified urinary follicle-stimulating hormone in patients with polycystic ovarian syndrome.

The step-down method can be an alternative method of ovulation induction in women with PCOS. This protocol can induce ovulation with a smaller dose of FSH and with lower incidence of excessive ovarian enlargement in comparison with the traditional fixed-dose administration method.

The role of endogenous gonadotropin release in the etiology of ovarian enlargement during purified urinary follicle-stimulating hormone therapy.

Patients with polycystic ovarian disease were grouped into three groups according to their maximum ovarian diameter (maxD) after ovulation induction by purified urinary follicle-stimulating hormone (FSH). Serum luteinizing hormone (LH) and FSH were measured daily by radioimmunoassay and size and number of follicles were assessed by ultrasonography. Follicle-stimulating hormone in group A (80 mm less than or equal to maxD) was significantly higher than those of group B (60 mm less than or equal to maxD less than 80 mm) and C (maxD less than 60 mm) for the last 4 days of the treatment. This FSH rise in group A was not accounted for by FSH accumulation by the study of pharmacodynamics, and so was thought to be of endogenous origin. Luteinizing hormone was also elevated 3 days before the administration of human chorionic gonadotropin (hCG) in both groups A and B. The number of follicles in group A at hCG administration was significantly greater than that of group C. Any significant differences were not found in either total amount of purified urinary FSH or in the basal FSH and the LH levels before treatment of the three groups. These results suggest that excessive ovarian enlargement during gonadotropin therapy is caused by multiple follicular development primarily stimulated by endogenous FSH. Endogenous LH release further enhances ovarian enlargement.

Effects of recombinant human FSH (rhFSH), urinary purified FSH (uFSH), and hMG on small preantral follicles and tertiary follicles from normal adult and androgen-sterilized female mice.

OBJECTIVE: To examine the stage-specific follicular response to recombinant human FSH (rhFSH), urinary FSH (uFSH), and hMG preparations. SETTING: In vitro follicle culture. INTERVENTION(S): Small preantral and tertiary follicles isolated from adult normal BDF-1 mice and androgen-sterilized mice were cultured with rhFSH, uFSH, and hMG for 4 days. MAIN OUTCOME MEASURE(S): Follicular diameter. Immunoreactive inhibin, E2, and progesterone concentrations in cultured medium. RESULT(S): The minimal effective dose of rhFSH, uFSH, and hMG for the follicular growth of small preantral follicles from normal mice was 10 mIU/mL, 1 mIU/mL, and 0.1 mIU/mL, respectively. For tertiary follicles from normal mice, the minimal effective dose of rhFSH, uFSH, and hMG was 10 mIU/mL, 10 mIU/mL, and 1 mIU/mL, respectively. The minimal effective dose of hMG for the follicular growth of small preantral follicles from androgen-sterilized mice was 0.01 mIU/mL, and that of rhFSH and uFSH on tertiary follicles from androgen-sterilized mice was 1 mIU/mL and 10 mIU/mL, respectively. No significant increase was found in the follicular diameter of the tertiary follicles from androgen-sterilized mice as a result of stimulation by hMG, but an hMG dose of >10 mIU/mL produced a significant increase in progesterone secretion. CONCLUSION(S): Human menopausal gonadotropin preparation acts detrimentally on follicles from androgen-sterilized mice by increasing the sensitivity of small preantral follicles to FSH and by inducing the luteinization of tertiary follicles.

Assessment of the implantation site by transvaginal ultrasonography.

OBJECTIVE: To investigate the implantation site in a singleton pregnancy. DESIGN: Transvaginal ultrasonography (US) was performed at the mid or late follicular phase and/or at very early gestation before 6 weeks. SETTING, PATIENTS: Ultrasound monitoring was performed on 21 women with spontaneous cycles or treated by administration of 2.5 mg/d bromocriptine mesylate (Parlodel; Sando Co., Tokyo, Japan) for occulted hyperprolactinemia at the Fertility Clinic of the Department of Obstetrics and Gynecology at Gunma University Hospital. MAIN OUTCOME MEASURES: The site of the ovary with a growing follicle and/or a corpus luteum of pregnancy, and the location of a gestational sac in longitudinal and transverse views were observed. RESULTS: When ovulation occurred in the right ovary, 12 of the 14 gestational sacs were located on the right wall, 1 on the midwall, and 1 on the left wall. Ovulation in the left ovary resulted in 5 gestational sacs located in the left wall and 2 on the midwall. That is, of 21 gestational sacs, 17 were located on the ipsilateral uterine wall to the ovulating ovary, 3 on the midwall, and only 1 on the contralateral. CONCLUSION: Implantation occurs on the ipsilateral uterine wall to the ovulating ovary.

A comparative study of fixed-dose, step-down, and low-dose step-up regimens of human menopausal gonadotropin for patients with polycystic ovary syndrome.

OBJECTIVE: To compare the efficacy and safety of the fixed-dose, the step-down, and the low-dose step-up regimens of hMG for women with polycystic ovary syndrome (PCOS). DESIGN: Prospective randomized study. SETTING: Gunma University School of Medicine, Maebashi, Japan. PATIENT(S): Thirty-seven women with PCOS. INTERVENTION(S): The fixed-dose, the step-down. and the low-dose step-up regimens were administered. MAIN OUTCOME MEASURE(S): The number of growing follicles and serum hormone levels. RESULT(S): Serum FSH levels on the day of hCG administration were significantly higher in the fixed-dose regimen group than in the step-down and the low-dose step-up regimen groups, and the number of growing follicles (> or =11 mm) in the low-dose step-up regimen group was significantly smaller than in the fixed-dose regimen group. On the 7th day after hCG administration, the maximal diameter of the ovaries in the low-dose step-up regimen group was significantly smaller than in the fixed-dose and the step-down regimen groups, and the risk of excessive ovarian enlargement (> or =70 mm) was significantly lower in the low-dose step-up regimen group than in the fixed-dose regimen group. CONCLUSION: The low-dose step-up regimen of hMG for patients with PCOS may be the safest protocol among the three stimulation regimens for reducing multiple follicular development.

Radioimmunoassay of human tissue factor.

The apoprotein (AP) of tissue factor (TF) has been purified 72,000-fold to homogeneity from human placenta using acetone delipidation, sodium deoxycholate (DOC) extraction, Sephacryl S-300 column chromatography, preparative polyacrylamide-gel electrophoresis (PAGE) in DOC and tryptic digestion. The purified AP had an apparent molecular weight of 54,000 by sodium dodecyl sulfate/PAGE. A radioimmunoassay (RIA) for quantitation of the TF-AP using an antibody against this purified AP of TF was devised which was sensitive enough to measure as small a quantity as 100 pg/ml of TF-AP accurately with high reproducibility. In addition to TF clotting activity (TFA), the immunoreactive TF-AP (IR-TFR) in the homogenates of leukemic leukocytes from patients with acute non-lymphoid leukemia (ANLL) was determined using this RIA. In 30 patients with ANLL, the mean IR-TFR with standard deviation (SD) of 21 cases with DIC was 157.9 +/- 188.1 ng/10(8) cells, which was significantly higher than that (37.1 +/- 29.9 ng/10(8) cells) of 9 cases with no DIC during remission induction chemotherapy (p less than 0.01).

Studies on leukemic cell tissue factor.

Apoprotein part of tissue factor of human placenta was purified 871 fold from the starting material with 4.2% yield by concanavalin A-Sepharose affinity chromatography and SDS-PAGE. The molecular weight of purified apoprotein was 45,000 in non-reduced condition and 49,000 in reduced condition. Tissue factor of human leukemia cells (FAB classification:M2 and M3) and cultured leukemia cell lines (HL-60 and Molt-4) was analyzed using specific rabbit anti-tissue factor IgG raised against purified material. Endotoxin stimulated HL-60 and Molt-4 also expressed procoagulant activity which was inhibited by tissue factor immune IgG. By immunostaining of the purified material, the lysate of leukemia cells (M2 and M3) and cultured leukemia cells (HL-60 and MOLT-4) revealed a major band of the same apparent molecular weight. Immuno-electron microscopic study on tissue factor of HL-60 cells produced the following findings: stimulation by endotoxin resulted in the formation of pseudopods of the cell membrane, and immunogold particles accumulated mainly on these pseudopods and cisternal spaces of rough endoplasmic reticulum, indicating exposure of the tissue factor to the surface of perturbed cell membrane with concurrent increase in tissue factor synthesis.