European collaborative study of LH assay: 3. relationship of immunological reactivity, biological activity and charge of human luteinizing hormone

This report describes the results of the third part of the collaborative study organized by a working group sponsored by the Community Bureau of Reference of the European Community Commission. The aim of the present work was to establish the link between immunoreactivity and biological activity of human LH, thus allowing to determine the antigenic domains of the molecule involved in the induction of the biological effect. The relationship between immunoreactivity and electric charge of hLH was also studied. This work allowed to further apprehend hLH isomorphism and its role in discrepancies observed among hLH assays and clinical status. It also made the feasibility of measuring biologically active isoforms by an immunological method to be assessed. The effect of 36 mAb with known epitopic specificity, was evaluated on both hLH binding to rat membrane receptor and hLH induced production of testosterone by porcine Leydig cells. All the epitopes located on the beta subunit were found to be strongly involved in the biological activity whereas 4/9 and 10/18 epitopes present on the alpha subunit or specific for the holomolecule respectively appeared weakly involved. Assaying biological hLH using immunological method would require that mAb specific for all the epitopes involved in the receptor activation be tested, and thus appears presently unsuitable for routine clinical evaluation. In the previous work some LH immunoassays were found to underestimate LH concentrations (J. Endocrinol. Invest 1994, 17: 397-406 and 407-416). The mAb used in liquid phase in these kits were found in the present work to be directed against the domains of LH weakly involved in the activation of the receptor and would suggest that bioactive LH would be misevaluated by these kits. The immunoreactivity of hLH isoforms separated by isoelectric focusing (IEF) in liquid phase was also determined. IEF allowed to separate three groups of hLH isoforms but none of them exhibited a specific discriminating pattern of immunoreactivity when they were tested against a panel of mAb. It suggests that, in our experimental conditions, the electric charge and the immunoreactivity of hLH were not closely linked.     

Isolation, characterization and radioimmunoassay of luteinizing hormone in the brushtail possum

Luteinizing hormone (LH) was purified from brushtail possum (Trichosurus vulpecula) pituitary glands. The purification procedure consisted of ammonium sulfate precipitation followed by triazinyl-dye chromatography, hydrophobic interaction chromatography and gel filtration. A yield of 10 microg LH g-1 pituitary with a recovery of 20% was obtained from 1400 pituitary glands (20.3 g). Contamination with possum follicle-stimulating hormone (FSH) was < or =0.05%. The amino acid analysis and the N-terminal sequencing for 10 cycles revealed close homology with LH from other mammals. Minor amounts of LH that had been truncated near the N-terminal were also detected. No contaminating proteins were found by amino acid sequencing. The potency of possum LH was 20% that of ovine LH in a receptor assay using possum testicular receptors and 4% that of ovine LH when bovine corpora lutea receptors were used. Possum LH was able to stimulate production of cyclic adenosine 3',5'-monophosphate by bovine granulosa cells. A radioimmunoassay (RIA) for possum LH using 125I-possum LH and an antiserum raised against ovine LH was developed. The RIA has a sensitivity of 0.15 ng mL-1, a 50% displacement of 1.9 ng mL-1 and a cross-reactivity of <0.02% against possum FSH. Plasma concentrations were 0.24+/-0.04 ng mL-1 (n = 8) and 0.39+/-0.12 ng mL-1 (n = 8) in female and male possums respectively. Administration of mammalian gonadotrophin-releasing hormone (GnRH) and chicken GnRH II stimulated increases in plasma LH concentrations in male and female possums. When comparing LH responses with administration of mammalian GnRH or chicken GnRH II, plasma LH concentrations remained elevated for a longer period of time in males than in females (P < 0.01); plasma LH concentrations also remained elevated for longer after mammalian GnRH than after chicken GnRH II (P < 0.01). Gonadectomy stimulated an increase in plasma concentrations of LH in both male (P < 0.01) and female (P < 0.05) possums. The rate of increase in plasma LH concentrations in males was faster than that in females. In summary, we have purified, partially characterized, and developed a RIA for possum LH.     

Recombinant uracil phosphoribosyltransferase from the thermophile Bacillus caldolyticus: expression, purification, and partial characterization

The hydrophobic surfactant protein C (SP-C) is known to modulate the biophysical properties of surfactant phospholipid. Although SP-C mRNA has been demonstrated in human fetal lung, there is limited information regarding developmental expression and processing of proSP-C protein. Two epitope-specific human proSP-C antisera, anti-hCPROSP-C (His59-Ser72) and anti-hCTERMSP-C (Gly162-Gly175), were generated to complement previously produced anti-NPROSP-C (Met10-Gln23) for the study of proSP-C expression in human fetal lung. Western blotting and immunocytochemistry detected expression of proSP-C protein by 12-16 wk of gestation. ProSP-C immunoreactivity of preculture lung, limited to expression of proSP-C21 in airway epithelial cells, was markedly enhanced by culture of lung explants in dexamethasone. To examine synthesis of proSP-C, homogenates from explants were labeled with 35S-Met/Cys for 0.5-4 h. Immunoprecipitation with anti-NPROSP-C detected 35S-proSP-C21 by 30 min and, after 2 h of labeling, there was a 15-fold increase in 35S-proSP-C21 in dexamethasone-treated lungs versus controls. Synthesis of proSP-C21 was followed by the appearance of a 24-kD form and smaller processing intermediates including 6-10-kD forms. Posttranslational processing of proSP-C21 was not observed in control explants. SP-C(6-10) were not recognized by either anti-CPROSP-C or anti-hCTERMSP-C. These results indicate that low level expression of proSP-C protein first occurs in epithelial cells early in the second trimester and that expression can be enhanced by dexamethasone. Initial posttranslational processing of human proSP-C involves modification of proSP-C21 to SP-C24 and subsequent proteolysis of C-terminal propeptide domains. We speculate that absence of low Mr intermediates in unstimulated second trimester fetal lung tissue reflects developmental and glucocorticoid dependent regulation of proSP-C21 synthesis and posttranslational processing.     

Purification, physicochemical characterization and biological properties of a lectin from Erythrina velutina forma aurantiaca seeds

A lectin was purified from seeds of Erythrina velutina forma aurantiaca by affinity chromatography on cross-linked guar gum. The lectin is a potent agglutinin for human (minimal concentration of protein able to cause visible agglutination of a 2% erythrocyte suspension varying from 1 to 4 micrograms/ml), rabbit (4 micrograms/ml) and chicken erythrocytes (8 micrograms/ml) but presented low activity against cow (250 micrograms/ml) or sheep (333 micrograms/ml) blood cells. Hemagglutination of human O+ erythrocytes was inhibited by D-lactose (0.2 mM) > D-galactose (0.8 mM) > D-raffinose (2.1 mM). At pH 7.5, chromatography on a Superose 12 HR 10/30 column showed that the lectin was primarily a dimer (56.0 kDa) composed of two identical subunits (31.6 kDa each). A small amount of a tetrameric form was also apparently present. The lectin is a glycoprotein (7.3% carbohydrate), has a pI of 4.5, contains high levels of acidic (Asp and Glu, 64.2 and 51.6 residues/mol, respectively) and hydroxy amino acids (Ser and Thr, 42.9 and 38.5 residues/mol, respectively) but relatively low amounts of sulfur amino acids (Cys and Met, 1.0 and 5.0 residues/mol, respectively) and has an N-terminal sequence of Val-Glu-Thr-Ile/Leu-Pro-Phe-Ser. Its hemagglutinating activity was abolished by heating at 70 degrees C for 10 min. The activation energy (delta G') required for denaturation measured by loss of hemagglutination activity was 24.87 kcal/mol. In rats, the purified lectin (100 micrograms) induced neutrophil migration into the peritoneal cavity (3.7 +/- 0.6 x 10(6) neutrophils/ml) or into the air pouch (2.75 +/- 0.25 x 10(6) neutrophils/ml), 8 and 10 times greater than the negative control, respectively.     

Recombinant human phenylethanolamine N-methyltransferase: overproduction in Escherichia coli, purification, and characterization

We studied major malformations in 5,581 infants with Down syndrome (DS) from three registers of congenital malformations. THe prevalence at birth of 23 different malformations was compared with the program-specific rates for each malformation in non-DS infants. An about 300 times risk increase was seen for annular pancreas, cataracts and duodenal atresia and an about 100 times risk increase for megacolon and small choanal atresia. Esophageal, anal and small bowel atresia, preaxial polydactyly, and omphalocele all showed risk increases between 10 and 30 times. Statistically significantly elevated risk ratios around 3-5 were seen for cleft palate, cleft lip/palate, and limb deficiencies. No increased risk was seen for neural tube defects, hydrocephaly, microtia, renal agenesis or severe dysgenesis, hypospadias or polydactyly other than preaxial. Oral clefts were more often present in DS in the Swedish material than in the other two materials. Cardiac defects were registered in 26% of all cases (varying between programs) but 28% of the cardiac defects were unspecified. DS infants born to women younger than 25 years had a significantly increased risk for megacolon and there was a trend increasing risk for esophageal or anal atresia with maternal age. A decreased risk for cardiac defect in DS infants born to teenage mothers was found, quite pronounced for endocardial cushion defects and ventricular septum defects. There were no statistically significant differences in the sex distribution of specific malformations in infants with DS and in non-DS infants.     

Molecular cloning, expression, and partial characterization of a second human tissue-factor-pathway inhibitor

Previous studies have shown that tissue-factor-pathway inhibitor (TFPI) is an important regulator of the extrinsic pathway of blood coagulation through its ability to inhibit factor Xa and factor VIIa-tissue factor activity. We describe the molecular cloning and expression of a full-length cDNA that encodes a molecule, designated TFPI-2, that has a similar overall domain organization and considerable primary amino acid sequence homology to TFPI. After a 22-residue signal peptide, the mature protein contains 213 amino acids with 18 cysteines and two canonical N-linked glycosylation sites. The deduced sequence of mature TFPI-2 revealed a short acidic amino-terminal region, three tandem Kunitz-type domains, and a carboxyl-terminal tail highly enriched in basic amino acids. Northern analysis indicates that TFPI-2 is transcribed in umbilical vein endothelial cells, liver, and placenta. TFPI-2 was expressed in baby hamster kidney cells and purified from the serum-free conditioned medium by a combination of heparin-agarose chromatography, Mono Q FPLC, Mono S FPLC, and Superose 12 FPLC. Purified TFPI-2 migrated as a single band in SDS/PAGE and exhibited a molecular mass of 32 kDa in the presence and absence of reducing agent. The amino-terminal sequence of recombinant TFPI-2 was identical to that predicted from the cDNA. Despite its structural similarity to TFPI, the purified recombinant TFPI-2 failed to react with polyclonal anti-TFPI IgG. Preliminary studies indicated that purified recombinant TFPI-2 strongly inhibited the amidolytic activities of trypsin and the factor VIIa-tissue factor complex. In addition, the inhibition of factor VIIa-tissue factor amidolytic activity by recombinant TFPI-2 was markedly enhanced in the presence of heparin. TFPI-2 at high concentrations weakly inhibited the amidolytic activity of human factor Xa, but had no measurable effect on the amidolytic activity of human thrombin.     

Chemical, metabolic and immunological characterization of gangliosides of human glioma cells

The patterns of ganglioside profiles were studied in 10 human glioma and one melanoma cell lines. Ganglio-series gangliosides, GM3 (NeuAc alpha2-3Gal beta1-4Glc beta1-Cer) and GM2 (GalNAc beta 1-4 (NeuAc alpha2-3)Gal beta1-4Glc beta 1-1Cer), and a neolacto-series ganglioside, sialylparagloboside (SPG) (NeuAc alpha 2-3Gal beta1-4GlcNAc beta1-3Gal beta1-4Glc beta1-1Cer), were the predominant constituents. The activities of the two key enzymes, GM3 synthetase and lactotriaosyl ceramide (Lc3Cer) synthetase, alone did not account for the ganglioside profile. Metabolic labeling with the use of [3H]glucosamine-HCl showed more pronounced difference in the synthetic rate of each ganglioside type, in which GM2 was the most strongly labeled in 7 out of the 10 glioma cell lines. On quantifying the chemical content of GM3 and GM2, the GM3/GM2 molar ratio of above 2.0 was arbitrarily classified into GM3 dominant type (KG-1C and Mewo); the ratio below 0.5 was designated as GM2 dominant type (H4, U138MG, U373MG, T98G and A172); and the ratio between 0.5 and 2.0 was regarded as GM3 and GM2-co-dominant type (U87MG, Hs683, SW1088 and U118MG). Subsequently, the capabilities of the antibody binding to these gangliosides were examined in native forms in the cell membrane and in chemically-isolated forms. The intensity of reaction against chemically isolated GM3 and GM2 gangliosides was dependent on the quantity, and GM2 was more reactive than GM3; however, the reactivities on the cell surface did not correlate with the chemical content indicating other factors to influence their immunoreactivities.     

Heterologous radioimmunoassay for llama and alpaca luteinizing hormone with a monoclonal antibody, an equine standard and a human tracer

A radioimmunoassay for llama and alpaca LH was developed using a human I125LH tracer from a commercial kit, equine LH diluted in human LH free serum as standard, and a monoclonal antibody (518B7) specific for LH but with low species specificity. A 60-min delay in the addition of the tracer and overnight incubation gave a sensitivity of 0.8 microgram L-1. The intra-assay coefficient of variation was 37% at 1 microgram L-1, declined to 15% at 4 micrograms L-1 and was below 6% for concentrations up to 32 micrograms L-1. The inter-assay coefficients of variation for 3 control samples were 20% (2.8 micrograms L-1), 16% (7.1 micrograms L-1) and 9.8% (19 micrograms L-1). In an attempt to increase sensitivity, all tubes were preincubated for 4 h at room temperature before adding the tracer, and the sample volume was increased from 50 microL to 100 microL (in the standard curve the increased volume was compensated for by human LH free serum). With this protocol, the assay sensitivity was 0.5 microgram L-1. The assay was validated clinically and demonstrated increased concentrations of LH after mating in llamas and alpacas. Furthermore, the assay was used to monitor LH responses to a single dose of GnRH in llamas (adult males and females at different ages).     

Synthesis, physicochemical characterization, and crystallization of a putative retro-coiled coil

OBJECTIVES: The purposes of this study were to investigate the use of radial artery applanation tonometry and a generalized transfer function for the assessment of central aortic pressure augmentation in subjects taking commonly used antihypertensive agents (angiotensin-converting enzyme inhibitors, beta-adrenergic blockers, Ca2+ antagonists, diuretic therapy). BACKGROUND: Applanation tonometry of the radial artery with a generalized transfer function has been proposed as a means of assessing central aortic blood pressure. Recently, a commercial apparatus based on this technique has become available; we therefore examined the effect of a generalized transfer function on derived central aortic pressure compared with measured brachial blood pressures and also investigated the potential of this technique to assess the influence of differing drug therapy. METHODS: Two hundred and sixty-two hypertensive patients on stable medication were studied using the PWV Medical Blood Pressure Analysis System (version 2, DAT-1). RESULTS: In univariate analysis, augmentation index showed association with age, sex, height and heart rate. In multivariate analysis, diastolic blood pressure and age (positively), height and heart rate (negatively) and sex were significantly associated. After adjustment for these variables, pressure augmentation was not associated with any antihypertensive treatment investigated. Linear relationships were demonstrated between brachial blood pressures and corresponding central pressures derived by transfer function methods. CONCLUSIONS: Our findings suggest that if adjustment for central-peripheral pressure difference is necessary, simple linear relationships may be sufficient. Age, heart rate and height but not the class of antihypertensive medication affected the degree of pressure augmentation observed using this technique.     

Twenty-four-hour profiles of luteinizing hormone, follicle-stimulating hormone, testosterone, and estradiol levels: a semilongitudinal study throughout puberty in healthy boys

Monocytes and endothelial cells interact at sites of vascular injury during inflammatory response, thrombosis, and development of atherosclerotic lesions. Such interactions result in modulation of several biological functions of the two cell types. Because both cells, on appropriate stimulation, synthesize tissue factor (TF), we examined the effect of human umbilical vein endothelial cell (HUVEC)/monocyte coculture on the expression of TF. We found that the coincubation resulted in TF generation, which was maximal at 4 hours, increased with increasing numbers of monocytes, and required mRNA and protein synthesis. Supernatant from HUVEC/monocyte coculture induced TF activity in HUVECs, but not in monocytes, indicating that HUVEC were the cells responsible for the activity, and that soluble mediators were involved. Interleukin-1 beta (IL-1 beta) and tumor necrosis factor-alpha (TNF-alpha), well-known inducers of TF in HUVECs, were found in the supernatant from the coculture, and specific antibodies directed against either cytokine inhibited TF generation. The need of IL-1 beta and TNF-alpha synthesis in order to elicit TF expression was also suggested by the delay observed in TF mRNA formation and TF activity generation when monocytes were incubated with HUVECs. IL-1 beta and TNF-alpha antigen levels in the coculture supernatant, and, consequently, HUVEC TF expression, were inhibited in the presence of anti-CD18 monoclonal antibody. These findings emphasize the role of cell-cell contact and cross-talk in the procoagulant activity, which could be responsible for the thromboembolic complications observed in those vascular disorders in which monocyte infiltration is a common feature.     

Recombinant growth hormone: a new cardiovascular drug therapy

GH has an important role in normal cardiovascular physiologic functioning, working indirectly through effects on IGF-1. An excess or deficiency of GH causes an increased rate of cardiovascular disease, including cardiomyopathy. A relative GH deficiency in older subjects may also increase cardiovascular morbidity and mortality risk. In replacement doses, GH can enhance myocardial contractility; can decrease peripheral vascular resistance; and can reduce total cholesterol and LDL-cholesterol values and fibrinogen and PAI levels. These effects of GH, coupled with the ability to improve skeletal muscle function and reduce adiposity, make it an attractive treatment for patients with CHF and a potential maintenance drug for elderly people. Clinical trials, including studies with GHRH that may reduce the adverse effects of GH therapy, such as hyperglycemia and hypertension, are now in progress.     

Embryonic origin of preimplantation factor (PIF): biological activity and partial characterization

Preimplantation factor (PIF) is detected in the serum of women shortly after fertilization; its origin, however, has not been established. In this study, the embryonal origin of PIF was investigated and partial characterization of the factor was carried out. Culture media from viable human 2-8-cell stage embryos and mouse 2-cell-blastocyst stage embryos were analysed using the lymphocyte/platelet binding assay (LPBA). The assay was performed by combining culture media with donor O+ type blood-derived lymphocytes/platelets, complement and an antibody against CD2. Increased autorosette formation between lymphocytes and platelets (> 9%) was an indication for the presence of PIF. In addition, the effect of platelet-activating factor (PAF) and chaperonin 10 on PIF activity was determined. Partial purification of PIF was carried out using gel filtration and reverse-phase high purification liquid chromatography (HPLC), followed by mass spectrometry. Culture media of single human viable fertilized oocytes were negative for PIF; however, the 10-fold concentrated medium was positive for PIF. In medium in which five or more mouse embryos were cultured, PIF activity was observed starting at the morula stage and was higher by the blastocyst stage. Addition of PAF or chaperonin 10 to the PIF assay did not elicit a specific effect on PIF activity. Chromatographic data suggest that PIF activity is due to low molecular weight proteins. PIF appears to be a low molecular weight protein which is derived from viable preimplantation embryos. It is different from PAF or chaperonin 10. Its final characterization will be valuable for better understanding of maternal recognition of pregnancy and implantation.     

Regulation of the activin-binding protein follistatin cultured human luteinizing granulosa cells: characterization of the effects of follicle stimulating hormone, prostaglandin E2, and different growth factors

Regulation of the activin-binding protein follistatin (FS) by recombinant human (rh) FSH, prostaglandin E2 (PGE2), and several polypeptide growth factors was examined in cultures of human granulosa-luteal (GL) cells obtained from in-vitro fertilization patients at oocyte retrieval. Northern and dot blot hybridization analyses demonstrated that both rhFSH and PGE2 caused stimulatory effects on FS mRNA levels in a culture stage-, time-, and concentration-dependent manner. An 8-h stimulation with rhFSH (100 ng/ml) significantly increased FS mRNA levels on Days 5 and 7 of culture and PGE2 (10(-6)M) on Days 2, 4, and 7. The stimulatory effect of rhFSH and PGE2 on FS mRNA levels were rapid and transient. Maximal inductions occurred 8 h after stimulation, whereas weak or no stimulatory effects were seen at 24 or 48 h. PGF2 alpha did not affect FS mRNA levels at any time point studied. Treatment of the cells with the protein synthesis inhibitor cycloheximide prior to rhFSH stimulation did not inhibit the rapid induction of FS mRNAs, but it prevented the decline at 24 h. Both rhFSH and PGE2 clearly also increased the levels of secreted FS proteins are detected by immunoprecipitation studies with a specific antibody. The effects of the polypeptide growth factors epidermal growth factor (EGF); transforming growth factor beta 1 (TGF beta 1), and activin A on FS mRNA levels were also examined. TGF beta 1 and activin A had no effect on basal FS expression at any concentration or time point studied. An 8-h stimulation with EGF increased FS mRNA levels, but the effect was weaker than those caused by rhFSH and PGE2. We conclude that rhFSH and PGE2 induce FS mRNA and protein in human cultured GL cells. EGF is able to induce FS mRNA to a lesser extent than are rhFSH and PGE2, whereas PGF2 alpha, TGF beta 1, and activin A do not affect basal FS mRNA levels in human cultured GL cells. This study together with our previous report on the stimulatory effect of hCG on FS levels suggest that in the luteal phase of the human menstrual cycle, FS expression in granulosa cells is likely to be positively controlled by luteotropic factors such as gonadotropins and PGE2. Consequently, elevated FS levels may support the survival of the human CL since FS is known to prevent the antisteroidogenic effects of activin in human GL cells.     

Recombinant rat follicle-stimulating hormone; production by Chinese hamster ovary cells, purification and functional characterization

In S. cerevisiae, the chromatin structure of DNA replication origins changes as cells become competent for DNA replication, suggesting that G1 phase-specific association of replication factors with origin DNA regulates entry into S phase. We demonstrate that ORC, Cdc45p, and MCM proteins are components of prereplication complexes (pre-RC). The MCM-origin association is dependent upon ORC and Cdc6p. During S phase, MCM proteins and Cdc45p dissociate from origin DNA and associate with nonorigin DNA with similar kinetics as DNA Polymerase epsilon, which is present at DNA replication forks. Our results identify protein components of the pre-RC and a novel replication complex appearing at the G1/S transition (the RC), and suggest that after initiation MCM proteins and Cdc45p move with eukaryotic replication forks.     

The unique exon 10 of the human luteinizing hormone receptor is necessary for expression of the receptor protein at the plasma membrane in the human luteinizing hormone receptor, but deleterious when inserted into the human follicle-stimulating hormone receptor

The LH receptor (LHR) is a member of the family of G protein-coupled seven-times plasma membrane transversing receptors. Its gene consists of 11 exons, the last one encoding the transmembrane and intracellular domains of the receptor. The FSHR, and its gene, resemble structurally those of the LHR, with the exception that the sequences corresponding to exon 10 in LHR are missing in FSHR, which is thus encoded by a total of ten exons. Our recent studies on the marmoset monkey testis LHR cDNA indicated that an 81 bp nucleotide sequence, encoding the complete exon 10 of the LHR gene in other mammalian species, is absent in this species without affecting the LHR function. To study further the role of the exon 10 encoded sequences of the LHR in the gonadotropin receptor function, a deletion of exon 10 from the human LHR (hLHdeltaexon10R), and a chimeric hFSHR with exon 10 from hLHR inserted (hFSHLHexon10R), were constructed in expression vectors. The results presented here demonstrate that 293 cells transfected with the hLHdeltaexon10R display a decrease in the proportion of the receptor binding at the cell surface, compared with cells transfected with wild-type hLHR. However, the cells expressing hLHdeltaexon10R showed similar high affinity binding of [125I]iodo-hCG as those transfected with wild-type hLHR, in either intact cells or their detergent extracts. In addition, cells expressing the hLHdeltaexon10R and wild-type hLHR displayed similar dose-response of cAMP production to hCG stimulation. Cells transfected with chimeric hFSHLHexon10R showed barely detectable [125I]iodo-FSH binding in intact cells compared with those transfected with wild-type hFSHR. The FSH binding detected in cellular detergent extracts displayed 10-fold lower binding activity than wild-type receptors, in spite of similar level of immunoreactive FSHR protein expression in the transfected cells. The hFSHLHexon10R had a modest 5-fold lower binding affinity for FSH as compared with wild-type hFSHR. In conclusion, the present study indicates that the sequences encoding exon 10 of the hLHR are essential for the LHR expression at the plasma membrane, but deleterious for function if inserted into the hFSHR.     

Purification and partial characterization of a highly immunogenic human leukocyte protein, the L1 antigen

L1 is a major and highly immunogenic protein component of human granulocytes. It was purified from random samples of blood-donor leukocytes. An antiserum to L1 was initially raised in rabbits by immunization with protein fractions from preparative agarose gel electrophoresis. The main problem in purifying L1 was its poor stability when carried through multiple steps. Preparative isoelectric focussing was therefore adopted as an efficient one-step method. The isoelectric focussing pattern of L1 was greatly influence by the presence of EDTA or calcium ions in the sample buffer. With low EDTA concentrations or calcium excess, L1 focussed as seven protein bands in two regions, pH 6.1-6.5 and pH 7.6-8.4. Conversely, L1 was found only in the pH-6.1-6.5 region when excess EDTA was added to the sample. Irrespective of conditions, the bulk of L1 focussed at pH 6.3 and pH 6.5. Also the electrophoretic mobility of L1 was strongly influenced by calcium ions, suggesting uptake of calcium by the protein. In the presence of calcium, L1 adhered to glass surfaces and filters used for concentration of protein solutions. The latter problem could be prevented by addition of EDTA. The molecular mass of L1 was determined to be about 36.5 kDa. The molecule was shown to consist of three non-covalently linked 12.5-kDa subunits.     

Purification and characterization of a cell growth factor from a human leukemia cell line: immunological identity with ferritin

We have succeeded in long-term cultivation of a human erythroleukemia cell line, K-562-T1 (T. Okabe, M. Fujisawa, and F. Takaku, Proc. Natl. Acad. Sci. USA, 81: 453-455, 1984). The cells grown in a protein-free chemically defined medium have been shown to produce cell growth factors (A. Mihara et al., In Vitro Cell. Dev. Biol., 23: 317-322, 1987). In this study, we have purified a cell growth factor from the conditioned medium that stimulates the proliferation of human leukemia cells, HL-60. In the purified factor, two major protein bands of 24 kDa and 22 kDa were identified on a sodium dodecyl sulfate-polyacrylamide gel. The 22 kDa protein was stained with a monoclonal antibody to the light chain of ferritin. The growth-promoting activity of the purified factor was coprecipitated with a monoclonal antibody to the light chain or heavy chain of human ferritin. These results suggest that K-562-T1 cells produce a cell growth factor that is related to ferritin.