Proposed method for dimensionality reduction based on framework in gene expression domain.

The excessive use of attributes may affect the search for patterns and extraction of useful knowledge, because they harm the learning performance of algorithms in both speed and success rate. The use of dimensionality reduction methods is therefore an important alternative; however, these methods do not deal with the reduction of attributes in a specific area. This article presents a method based on framework concepts of domain for reducing attributes in a domain. The input method is a set of databases related to a domain, and the main process is the identification of common and variable attributes, plus the reduction of attributes in the original database. The proposed method was applied in the gene expression domain, using databases. The method can be used to analyze the most relevant attributes in a specific domain, granting greater confidence for models created for the application of a data mining task, thus, a previously known method in data mining. Attribute selection was also applied in the three databases for the comparison of the results. Analyses of the results using the criterion of cross-validation revealed that the employment of the methods resulted in the improvement of success rates compared to the databases containing the full range of attributes.

Reirradiation for recurrent malignant brain tumor with radiotherapy or proton beam therapy. Technical considerations based on experience at a single institution.

BACKGROUND AND PURPOSE: Radiotherapy for recurrent malignant brain tumors is usually limited because of the dose tolerance of the normal brain tissue. The goal of the study was to evaluate the efficacy and feasibility of reirradiation for patients with recurrent malignant brain tumors. PATIENTS AND METHODS: The subjects comprised 26 patients with recurrent malignant brain tumors treated with conventional radiotherapy (RT, n = 8), stereotactic radiotherapy (SRT, n = 10), and proton beam therapy (PBT, n = 8) at our institute. Fifteen patients had glioblastoma, 6 had WHO grade 3 glioma, and 5 had other tumors. The dose of initial radiotherapy was 34.5-94.4 Gy. Different radiation schedules were compared using the equivalent dose in 2-Gy fractions. RESULTS: Reirradiation was completed in all patients without a severe acute reaction. The reirradiation doses were 30-60 Gy (median, 42.3 Gy) and the total doses for the initial and second treatments were 64.5-150.4 Gy (median, 100.0 Gy). Currently, 11 patients are alive (median follow-up period, 19.4 months) and 15 are dead. The median survival and local control periods after reirradiation of the 26 patients were 18.3 and 9.3 months, respectively. For the 15 patients with glioblastoma, these periods were 13.1 and 11.0 months, respectively. Two patients showed radiation necrosis that was treated by surgery or conservative therapy. CONCLUSION: Reirradiation for recurrent malignant brain tumor using conventional RT, SRT, or PBT was feasible and effective in selected cases. Further investigation is needed for treatment optimization for a given patient and tumor condition.

Mandibular and femoral growth alteration after sex hormone disruption in growing mice.

OBJECTIVES: To investigate how mandibular and femoral growth is affected when sex hormone- specific receptor antagonist is administered in growing mice. MATERIALS AND METHODS: Forty C57BL/6J mice were used in this experiment. At 5 days of age, the mice received daily injection of estrogen receptor alpha (ERα), beta (ERß), or androgen receptor (AR) antagonists, and their body weight was assessed every 4 days. One, four and eight weeks after the initial injection, radiographs of the mandible and femur were taken and measured. Analyses of variance and pairwise comparisons (Fisher) were performed to examine the differences in values measured among the groups. RESULTS: Mandibular growth was affected by ERß antagonist injection in male mice at 4 and 8 weeks. In female mice, the growth was affected during all the experimental period, when ERß was administered. Moreover, at 8 weeks, mandibular growth was also affected in male and female mice injected with ERα antagonist and in male mice injected with AR antagonist. Femoral growth was affected during all the experimental period in male and female mice injected with ERß antagonist. Moreover, at 8 weeks, the growth was affected in male and female mice injected with ERα antagonist and in male mice injected with AR antagonist. CONCLUSIONS: Growth of the mandible and femur in mice, in part, is induced in response to the stimulation of ERß in chondrocytes before and during early puberty. In late and after puberty, the growth is induced by the stimulation of ERα in male and female mice and that of AR in male mice.

An fMRI Study of the Brain Network Involved in Teeth Tapping in Elderly Adults.

Cortical activity during jaw movement has been analyzed using various non-invasive brain imaging methods, but the contribution of orofacial sensory input to voluntary jaw movements remains unclear. In this study, we used functional magnetic resonance imaging (fMRI) to observe brain activities during a simple teeth tapping task in adult dentulous (AD), older dentulous (OD), and older edentulous subjects who wore dentures (OEd) or did not wear dentures (OE) to analyze their functional network connections. (1) To assess the effect of age on natural activation patterns during teeth tapping, a comparison of groups with natural dentition-AD and OD-was undertaken. A general linear model analysis indicated that the major activated site in the AD group was the primary sensory cortex (SI) and motor cortex (MI) (p < 0.05, family wise error corrected). In the OD group, teeth tapping induced brain activity at various foci (p < 0.05, family wise error corrected), including the SI, MI, insula cortex, supplementary motor cortex (SMC)/premotor cortex (PMA), cerebellum, thalamus, and basal ganglia in each group. (2) Group comparisons between the OD and OEd subjects showed decreased activity in the SI, MI, Brodmann's area 6 (BA6), thalamus (ventral posteromedial nucleus, VPM), basal ganglia, and insular cortex (p ¡ 0.005, uncorrected). This suggested that the decreased S1/M1 activity in the OEd group was related to missing teeth, which led to reduced periodontal afferents. (3) A conjunction analysis in the OD and OEd/OE groups revealed that commonly activated areas were the MI, SI, cerebellum, BA6, thalamus (VPM), and basal ganglia (putamen; p < 0.05, FWE corrected). These areas have been associated with voluntary movements. (4) Psychophysiological interaction analysis (OEd vs OE) showed that subcortical and cortical structures, such as the MI, SI, DLPFC, SMC/PMA, insula cortex, basal ganglia, and cerebellum, likely function as hubs and form an integrated network that participates in the control of teeth tapping. These results suggest that oral sensory inputs are involved in the control of teeth tapping through feedforward control of intended movements, as well as feedback control of ongoing movements.

Bacteriological characteristics of Staphylococcus aureus isolates from humans and bulk milk.

The aim of this study was to clarify the epidemiological association and bacteriological characteristics of human and animal Staphylococcus aureus isolates. Pulsed-field gel electrophoresis showed that pulsotypes (PT) of isolates from bulk milk differed from PT from human isolates, suggesting that there is no epidemiological association between isolates from these 2 sources. The absence of a common PT could result from the lack of contact between the sources. Methicillin-resistant S. aureus from human secretions and S. aureus from bulk milk in Japan consisted of 1 and 2 dominant clusters, respectively, whereas methicillin-susceptible S. aureus from humans consisted of assorted clusters. Isolates belonging to the dominant clusters showed the coagulase serotype, the capsule serotype, detection of exotoxin genes, and antimicrobial susceptibility. Isolates from bulk milk did not show the penicillin-binding protein 2a gene, and 252 of 275 isolates belonging to the 2 dominant clusters of bulk milk were susceptible to ampicillin, cefazolin, erythromycin, chloramphenicol, oxacillin, and vancomycin. Moreover, the LukM/LukF'-PV leukotoxin gene was detected in 233 of 275 isolates belonging to the dominant clusters in bulk milk isolates. These results support the hypothesis that a number of factors play a role in the adaptation of S. aureus isolates to specific hosts.

Comparative study of plasminogen activators in cancers and normal mucosae of human urinary bladder.

We have developed a highly sensitive sandwich enzyme immunoassay for determination of urokinase-type plasminogen activator (u-PA) and tissue-type plasminogen activator (t-PA) antigen levels in extracts of human tissues. We determined antigen levels of PAs in extracts of 31 primary cancers and 15 normal mucosal tissues of the urinary bladder using this method. U-PA antigen levels in extracts of bladder cancers were significantly higher than those in normal tissues (p less than 0.005). U-PA antigen levels significantly increased as histological grading of malignancy advanced. There was no correlation between t-PA antigen level and malignancy. These results indicate that an increase of u-PA antigen level may be a parameter of malignant transformation and may play an important role in invasiveness of cancer.

Nucleotide binding to the chaperonin GroEL: non-cooperative binding of ATP analogs and ADP, and cooperative effect of ATP.

Chaperonin-assisted protein folding proceeds through cycles of ATP binding and hydrolysis by GroEL, which undergoes a large structural change by the ATP binding or hydrolysis. One of the main concerns of GroEL is the mechanism of the productive and cooperative structural change of GroEL induced by the nucleotide. We studied the cooperative nature of GroEL by nucleotide titration using isothermal titration calorimetry and fluorescence spectroscopy. Our results indicated that the binding of ADP and ATP analogs to a single ring mutant (SR1), as well as that to GroEL, was non-cooperative. Only ATP induces an apparently cooperative conformational change in both proteins. Furthermore, the fluorescence changes of pyrene-labeled GroEL indicated that GroEL has two kinds of nucleotide binding sites. The fluorescence titration result fits well with a model in which two kinds of binding sites are both non-cooperative and independent of each other. These results suggest that the binding and hydrolysis of ATP may be necessary for the cooperative transition of GroEL.

Nucleotide-induced transition of GroEL from the high-affinity to the low-affinity state for a target protein: effects of ATP and ADP on the GroEL-affected refolding of alpha-lactalbumin.

We studied the refolding kinetics of alpha-lactalbumin in the presence of wild-type GroEL and its ATPase-deficient mutant D398A at various concentrations of nucleotides (ATP and ADP). We evaluated the apparent binding constant between GroEL and the alpha-lactalbumin refolding intermediate quantitatively by numerical simulation analysis of the alpha-lactalbumin refolding curves in the presence and absence of GroEL. The binding constant showed a co-operative decrease with an increase in ATP concentration, whereas the binding constant decreased in a non-co-operative manner with respect to ADP concentration. For the D398A mutant, the ATP-induced decrease in affinity occurred much faster than the steady-state ATP hydrolysis by this mutant, suggesting that ATP binding to GroEL rather than ATP hydrolysis, was responsible for the co-operative decrease in the affinity for the target protein. We thus analyzed the nucleotide-concentration dependence of affinity of GroEL for the target protein using an allosteric Monod-Wyman-Changeux model in which GroEL underwent an ATP-induced co-operative conformational transition between the high-affinity and low-affinity states of the target protein. The transition midpoint of the ATP-induced transition of GroEL has been found to be around 30 microM, in good agreement with the midpoint evaluated in other structural studies of GroEL. The results show that the observed difference between ATP and ADP-induced transitions of GroEL are brought about by a small difference in an allosteric parameter (the ratio of the nucleotide affinities of GroEL in the high-affinity and the low-affinity states), i.e. 4.1 for ATP and 2.6 for ADP.

Usefulness of subtraction of 3D T2WI-DRIVE from contrast-enhanced 3D T1WI: preoperative evaluations of the neurovascular anatomy of patients with neurovascular compression syndrome.

BACKGROUND AND PURPOSE: High-resolution 3D MR cisternography techniques such as 3D T2WI-driven equilibrium radiofrequency reset pulse (DRIVE) are used preoperatively to assess neurovascular anatomy in patients with neurovascular compression syndrome, but contrast between vessels and cranial nerves at the point of neurovascular contact is limited. The postprocessing technique subtraction of 3D T2WI-driven equilibrium radiofrequency reset pulse from contrast-enhanced 3D T1WI (sDRICE) provides both high spatial resolution and excellent contrast in depicting the neurovascular contact. We evaluated the usefulness of sDRICE compared with 3D T2WI-DRIVE. MATERIALS AND METHODS: Twelve patients who underwent microvascular decompression for hemifacial spasm or trigeminal neuralgia were examined preoperatively with 3D T2WI-DRIVE and sDRICE. Two neuroradiologists retrospectively analyzed and scored lesion conspicuity, defined as the ease of discrimination between offending vessels and compressed nerves or the brain stem at the neurovascular contact. They also quantitatively analyzed the contrast and contrast-to-noise ratio at the neurovascular contact. RESULTS: The lesion conspicuity scores of sDRICE images were significantly higher than those of 3D T2WI-DRIVE for all 12 patients (P = .006) and the 6 cases of hemifacial spasm (P = .023) but were not significantly higher in the 6 trigeminal neuralgia cases alone (P = .102). For all 12 patients, the contrast-to-noise ratio between the offending vessels and the brain stem and between the vessels and nerves on sDRICE images was significantly higher than that on 3D T2WI-DRIVE (P = .003 and P = .007, respectively). Among these structures, the contrast values were also significantly higher on the sDRICE than on the 3D T2WI-DRIVE (P < .001) images. CONCLUSIONS: The postprocessing technique sDRICE is useful to evaluate neurovascular anatomy and to improve contrast and the contrast-to-noise ratio in patients with neurovascular compression syndrome.

Multilocus sequence typing of bifidobacterial strains from infant's faeces and human milk: are bifidobacteria being sustainably shared during breastfeeding?

Bifidobacteria are considered to be one of the most important beneficial intestinal bacteria for infants, contributing to the priming of the mucosal immune system. These microbes can also be detected in mother's milk, suggesting a potential role of human milk in the colonisation of infant's gut. However, little is known about the timing of bacteria appearance in human milk, and whether human milk is the first source of inoculation. Here, we investigated whether specific strains are shared sustainably between maternal milk and infant's gut. Faecal samples and human milk were collected from 102 healthy mother-infant pairs (infant's faeces: meconium, 7, 30 days of age; mother's milk: once before delivery, colostrum, 7, 30 days after delivery). Bifidobacterial strains were isolated from these samples, and were discriminated by means of multilocus sequencing typing. No bifidobacteria were detected from human milk collected before delivery, or colostrum. Strains were isolated only from human milk samples obtained 7 days after birth or later. On the other hand, bifidobacterial strains were obtained from infant's faeces throughout the study period, sometimes as early as the first day of life (meconium). We have found that bifidobacterial species belonging to Bifidobacterium bifidum, Bifidobacterium breve, and Bifidobacterium longum subsp. longum could be identified as monophyletic between infant's faeces and their mother's milk. These strains were confirmed to be sustainably shared between maternal milk and infant's gut. Moreover, monophyletic strains were isolated at the same time point or earlier from infant's faeces than from human milk, and none were isolated earlier from human milk than from infant's faeces. Although it remains unclear whether human milk is the first source of microbes for infants, our results confirm that human milk is a reservoir of bifidobacteria, and specific strains are shared between infant's intestine and human milk during breastfeeding.

Enzyme immunoassay of angiotensin I.

A very sensitive and specific enzyme immunoassay has been developed for angiotensin I. Angiotensin I was coupled to beta-D-galactosidase by a novel cross-linking reagent, N-(meta-maleimidobenzoyloxy)succinimide. No decrease in the enzyme activity was observed during the coupling procedure. In the angiotensin I-beta-D-galactosidase conjugate, 0.39 mol immunoreactive angiotensin I/mol enzyme were present. A competitive assay with the enzyme-labeled angiotensin I was performed. Antibody-bound and free labeled antigen were separated from each other by the second antibody method, and the enzyme activity of the former was estimated. Using this assay, angiotensin I could be detected in the range of 1.2--50 pg. The sensitivity was 4.5-fold higher than that of the usual RIA. This assay distinguished clearly angiotensin I from angiotensin II, angiotensin III, and (Sar1, Ile8)-angiotensin II. The present method was applied to measure PRA in dogs; the results correlated fairly well with those obtained by the RIA (r = 0.94).

Urinary growth hormone levels measured by ultrasensitive enzyme immunoassay in patients with renal insufficiency.

An ultrasensitive enzyme immunoassay was used to measure urinary GH levels in patients with renal insufficiency and normal subjects. Urinary GH excretion varied widely, but was significantly higher (P less than 0.01) in patients with renal insufficiency (median, 339; range, 2-17,000 ng/day) than in normal subjects (5.4; 1.2-15 ng/day). Urinary GH excretion correlated positively with urinary beta 2-microglobulin excretion (r = 0.79; P less than 0.001) and negatively with creatinine clearance (r = -0.83; P less than 0.001). Gel chromatography of urine from patients with renal insufficiency revealed a major peak of urinary GH corresponding to a mol wt of 22K, that of pituitary GH. These findings suggest that the kidneys play an important role in the catabolism of GH and that urinary GH may reflect, at least in part, renal function as well as hypothalamo-pituitary function.

Urinary growth hormone (GH) measurements are useful for evaluating endogenous GH secretion.

Daily (24-h) urinary GH excretion was measured using a highly sensitive sandwich enzyme immunoassay in 10 normal adults, 6 patients with hypopituitarism, 25 normal but short children who had normal plasma GH responses (peak plasma GH level, greater than 10 micrograms/L) to provocative tests, and 8 patients with acromegaly. The mean urinary GH values in the normal adults, patients with acromegaly, and patients with hypopituitarism were 13.8 +/- 4.0 (+/- SE) and 431.1 +/- 149.1 ng/g creatinine (Cr) (1.56 +/- 0.45 and 48.77 +/- 16.87 ng/mmol Cr) and undetectable, respectively; these mean values were significantly different from each other. In the normal but short children the urinary values ranged from undetectable to 55.8 ng/g Cr (6.31 ng/mmol Cr). All of the normal but short children and 4 patients with hypopituitarism participated in a 24-h endogenous GH secretion study. The urinary GH values correlated significantly with the mean 24-h plasma GH concentrations as an index of endogenous GH secretion (r = 0.81; P less than 0.001) and plasma somatomedin-C levels (r = 0.67; P less than 0.001), respectively. In 6 patients with acromegaly whose plasma GH levels were constant throughout a 4-h period, the urinary GH values also significantly correlated with the mean plasma GH levels (r = 0.95; P less than 0.01). These data indicate that urinary GH measurements reflect endogenous GH secretion and that measurement of urinary GH excretion is a useful, simple, and practical method for evaluating endogenous GH secretion.

Two novel mutations in the coding region for neurophysin-II associated with familial central diabetes insipidus.

Familial central diabetes insipidus is an autosomal dominant disease caused by a deficiency of arginine vasopressin (AVP). We previously reported three distinct mutations in the AVP gene in Japanese familial central diabetes insipidus pedigrees that result in a substitution of Ser for Gly57 in the neurophysin-II (NPII) moiety of the AVP precursor, a substitution of Thr for Ala at the COOH-terminus of the signal peptide, and a deletion of Glu47 in the NPII moiety. In this study, we analyzed the AVP gene in two pedigrees by direct sequencing of the polymerase chain reaction-amplified DNA and found two novel mutations in exon 2, which encodes the central part of the NPII moiety of the precursor. The mutation in one pedigree was a C to A transition at nucleotide position 1891, which replaces Cys67 (TGC) with stop codon (TGA). As the premature termination eliminates part of the COOH domain of the NPII moiety and the glycoprotein moiety, the conformation of the truncated protein is likely to be markedly different from that of normal precursor. In another pedigree, a G to T transversion was detected at nucleotide position 1874, which substitutes polar Trp (TGG) for hydrophobic Gly62 (GGG). It is possible that mutated NPII molecules, as a consequence of a conformational change, cannot bind AVP or self-associate to form higher oligomer complexes. Interestingly, all mutations we have identified to date, with the exception of the signal peptide mutation, are located in exon 2, suggesting the importance of the highly conserved central part of the NPII molecules and/or the NPII moiety in the precursor for AVP synthesis.

TMP-153, a novel ACAT inhibitor, lowers plasma cholesterol through its hepatic action in golden hamsters.

The mechanism of the hypocholesterolemic action of N-[4-(2-chlorophenyl)-6,7-dimethyl-3-quinolyl]-N'-(2, 4-difluorophenyl) urea (TMP-153), a potent acyl-CoA:cholesterol acyltransferase (ACAT) inhibitor, was studies in Golden hamsters. TMP-153 (0.5-1.5 mg/kg) dose-dependently reduced plasma total- and low density lipoprotein (LDL)-cholesterol without affecting high density lipoprotein (HDL)-cholesterol. TMP-153 markedly reduced the cholesterol influx into the plasma upon intravenous injection of Triton WR-1339. The compound also decreased cholesterol absorption calculated from dietary intake, biliary secretion and the absorption co-efficient. Hepatic cholesterol secretion was calculated by substracting the cholesterol absorption from the cholesterol influx. In hamsters, the liver accounted for 92% of the cholesterol influx with the remaining 8% coming from the intestine, and both were markedly decreased by TMP-153. Thus, it is likely that TMP-153 lowers plasma cholesterol through its hepatic action. In the liver, the compound significantly reduced the unesterified cholesterol content in addition to markedly reducing the content of esterified cholesterol. In accordance with this reduction, the half-life time of [125I]-LDL was significantly shortened by the compound, suggesting an increase in LDL receptors. However, the hepatic cholesterogenesis from [14C]acetate was decreased by TMP-153 treatment. This effect seems to be secondary, since the compound did not inhibit cholesterogenesis from [14C]acetate in HepG2 cells. From the data described above, the contribution of hepatic secretion and intestinal absorption of cholesterol to the plasma cholesterol level in Golden hamsters are discussed.

TMP-153, a novel ACAT inhibitor, inhibits cholesterol absorption and lowers plasma cholesterol in rats and hamsters.

Effects of TMP-153, N-[4-(2-chlorophenyl)-6,7-dimethyl-3-quinolyl]-N'-(2,4-difluorophe nyl)urea, on intestinal and hepatic acyl-CoA:cholesterol acyltransferase (ACAT) activities, cholesterol absorption and plasma cholesterol level in rats and hamsters were studied. TMP-153 has IC50 values of around 5-10 nM for the hepatic and intestinal ACAT from various animals. The most potent inhibition was observed in the intestinal ACAT from Golden hamsters (IC50 = 2.3 nM). The inhibition mode of TMP-153 was non-competitive for rat intestinal ACAT. TMP-153 inhibited cholesterol esterification both in human colonic adenocarcinoma cells, LS180, and in human hepatoma cells, HepG2 (IC50 = 150 nM and 330 nM, respectively). [14C]cholesterol and cold cholesterol absorption from the small intestine was markedly inhibited by oral administration of TMP-153 (1 mg/kg) without affecting lymph flow and triglyceride absorption. When the compound was given as a dietary admixture, plasma cholesterol was reduced in rats fed a cholesterol diet (ED50 = 0.25 mg/kg/day), but not in those fed a stock diet. On the other hand, TMP-153 showed more prominent hypocholesterolemic effect in Golden hamsters fed the stock diet (ED50 = 0.81 mg/kg/day) than in those fed the cholesterol diet (ED50 = 8.01 mg/kg/day). In hamsters fed the stock diet, TMP-153 markedly decreased the hepatic unesterified cholesterol in addition to esterified cholesterol content, but did not affect bile flow and the biliary secretion of bile acid and lipids. Different mechanisms for plasma cholesterol lowering by TMP-153 between rats and hamsters was discussed.

A general and mild procedure for the purification of rabbit Fab' antibodies.

A mild procedure for the purification of rabbit Fab' antibodies is described. A small column of normal goat IgG-Sepharose 4B (3 cm x 5mm) was saturated with F(ab')2 prepared from rabbit anti-goat IgG, and anti-goat IgG Fab was eluted from the column by splitting the disulfide bond in the hinge of the F(ab')2 molecule using 10 or 12 mM 2-mercaptoethylamine at pH 7. The recovery of anti-goat IgG Fab' in the eluate was about 35% of anti-goat IgG F(ab')2 bound to the column, and the purity of anti-goat IgG Fab' eluted was more than 90%. This procedure may be applicable to most kinds of rabbit IgG antibodies and useful for various immunoassays.

Conjugation of recombinant reverse transcriptase of HIV-1 to beta-D-galactosidase from Escherichia coli for ultrasensitive enzyme immunoassay (immune complex transfer enzyme immunoassay) of anti-HIV-1 IgG.

Recombinant reverse transcriptase (RT) of HIV-1 was conjugated to beta-D-galactosidase from Escherichia coli in three different ways. Maleimide groups were introduced into beta-D-galactosidase molecules using N,N'-o-phenylenedimaleimide in the absence (method I) or presence (method II) of N-ethylmaleimide or into beta-D-galactosidase molecules, which had been treated with excess of 4,4'-dithiodipyridine to block thiol groups, using N-succinimidyl-6-maleimidohexanoate (method III). Subsequently, the maleimide groups were reacted with thiol groups introduced into recombinant RT molecules using N-succinimidyl-S-acetylmercaptoacetate. The conjugates were tested by a sensitive enzyme immunoassay (immune complex transfer enzyme immunoassay). The immune complex consisting of 2,4-dinitrophenyl-bovine serum albumin-recombinant RT conjugate, anti-HIV-1 IgG and recombinant RT-beta-D-galactosidase conjugate was captured by polystyrene beads coated with (anti-2,4-dinitrophenyl group) IgG, eluted with N epsilon-2,4-dinitrophenyl-L-lysine and transferred to polystyrene beads with (anti-human IgG gamma chain) IgG. The conjugate prepared by method III, which showed the least polymerization, the least loss of the specific enzyme activity and the lowest nonspecific binding, improved the sensitivity of the enzyme immunoassay for anti-HIV-1 IgG approximately 30-fold compared with RT-horseradish peroxidase conjugate.

Development of highly sensitive enzyme-linked immunosorbent assays for hepatocyte growth factor/scatter factor (HGF/SF): determination of HGF/SF in serum and urine from normal human subjects.

A sandwich enzyme-linked immunosorbent assay (ELISA) using rabbit anti-hepatocyte growth factor (HGF) IgG for human HGF, also known as the scatter factor, has previously been developed for determining increases in serum HGF levels in various liver diseases. The sensitivity limit of the ELISA is, however, approximately 0.2 ng/ml sample, and HGF concentrations in about 50% of normal subjects are not accurately measurable by this method, because the mean level of HGF in normal serum is close to the sensitivity limit. In the present study, chicken Fab' from egg yolk anti-HGF immunoglobulin Y and rabbit Fab' from rabbit anti-HGF IgG were conjugated with beta-D-galactosidase. With these conjugates as the second antibodies, we developed two sandwich ELISAs for human HGF and found that the sensitivities were about 20 pg/ml with the former conjugate and 2 pg/ml with the latter. The HGF concentration in sera from 138 normal subjects determined by the ELISA with the rabbit conjugate was 244+/-65 (SD) pg/ml serum, and it correlated very well with the number of leukocytes. Moreover, the ELISA with the rabbit conjugate permitted the determination of HGF levels in urine from normal subjects without first concentrating the sample. The determination of HGF in various biological fluids other than blood and urine by these ELISAs may aid the diagnosis and prognosis of various diseases.

Synthesis and biochemical studies of 16- or 19-substituted androst-4-enes as aromatase inhibitors.

Androst-4-en-17-one derivatives [19-acetoxide 4, 16-bromides 14 and 15, 19,19-difluoride 18, and (19R,S)-19-acetylenic alcohol 25] and androst-4-en-17 beta-ol derivatives 3, 5, 10, 12, and 19 were synthesized and tested for their ability to inhibit aromatase in human placental microsomes. All the 17-oxo steroids, except compound 25 and 17,19-diol 3 of this series, were effective competitive inhibitors with apparent Ki's ranging from 170 to 455 nM. 19,19-Difluoro steroid 18 and 19-acetylenic alcohol 25, a weak competitive inhibitor (Ki = 7.75 microM), caused a time-dependent, pseudo-first-order inactivation of aromatase activity with kinact's of 0.0213 and 0.1053 min-1 for compounds 18 and 25, respectively. NADPH and oxygen were required for the time-dependent inactivation, and the substrate, androst-4-ene-3,17-dione, prevented it, but a nucleophile, L-cysteine, did not in each case. The results strongly suggest that aromatase would attack the 19-carbon of steroids 18 and 25.