Yasuyuki Takagi

Localized Expression of Aromatase in Human Vascular Tissues

The atheroprotective effects of estrogen are well established and the presence of an estrogen receptor in vascular tissues has recently been reported. Therefore, we investigated the localization of the estrogen-producing enzyme aromatase in vascular tissues to assess the possible contribution of endocrine, paracrine, and autocrine modes of action. Aromatase was found in human vascular smooth muscle cells (SMCs) but not in endothelial cells on in situ hybridization. These observations were further supported by quantitative analysis of aromatase mRNA and the activity in 15 human vascular specimens. Only trace levels of expression were detected in the 3 infants examined, whereas 0.0088 to 0.0806 amol/ microg RNA of aromatase mRNA and 12.9 to 122.3 fmol. h-1. mg-1 protein of the activity were detected in 12 of the adult individuals. The switching of tissue-specific exon 1 of the human aromatase gene was also observed in some cases. Aromatase was found to be expressed only in cultured SMCs and not in cultured endothelial cells of human aorta and pulmonary artery and to be regulated through dexamethasone and the signaling pathways of protein kinase A and C. Study results revealed the localized expression of aromatase in vascular SMCs, which indicated a possible direct action of locally produced estrogen in an autocrine or paracrine manner, with possible cross talk between smooth muscle and endothelial cells.

Structural characterization of the human estrogen synthetase (aromatase) gene

The estrogen synthetase (aromatase, cytochrome P-450AROM) gene has been isolated from human genomic libraries and characterized. The restriction map of 43 positive clones obtained indicated that this enzyme is present as a single copy gene. The aromatase gene is unexpectedly large compared with other forms of the cytochrome P-450 superfamily, spanning at least 70 kilobases. The gene consists of 10 exons and its 5-untranslated region is divided into 2 exons by an intron of more than 35 kilobases long. This organization of the first exon in the aromatase gene is unique in the cytochrome P-450 superfamily. All the exon-intron junctional sequences conform to the canonical GT/AG rule. The sequences of a TATA box and a CAAT box are present 27 and 83 base pairs upstream from the transcriptional initiation site. Within 3 kilobases upstream from the initiation site, there are no typical consensus sequences of responsive elements for glucocorticoid and c-AMP, which regulate aromatase expression.

The alternative exons 1 of the mouse aromatase cytochrome P-450 gene

Aromatase cDNA clones were isolated from cDNA libraries of mouse hypothalamus, amygdala and ovary. Analysis of the nucleotide sequences of the 5 regions of the obtained cDNAs suggested that the mouse aromatase gene is tissue-specifically regulated by alternative exons 1. There were obvious differences between the 5 regions of the brain and ovary aromatase cDNAs, but no difference was found between the sequences of the hypothalamus and amygdala ones. We further isolated a mouse genomic DNA clone containing brain- and ovary-specific exons 1. The brain specific exons 1 and their promoters were highly homologous in the human and mouse aromatase genes. In contrast there were several differences in the sequences among the promoter regions of the ovary-specific exons 1 of the mouse, human and rat aromatase genes, significant homology between their sequences was also observed. The present results demonstrate that expression of the mouse aromatase gene is also tissue-specifically regulated through the use of alternative exons 1 and promoters, as reported for man.

Molecular and epidemiological analyses of abnormal expression of aromatase in breast cancer.

One-third of human breast cancers exhibit estrogen-dependent proliferation. It appears that estrogen functions as a mitogenic factor in these carcinomas. As aromatase is the rate-limiting enzyme in estrogen biosynthesis. It could play an important role in the pathogenesis of estrogen-dependent breast cancer. The aromatase gene consists of at least six exons 1, each containing a promoter, and the tissue-specific expression is regulated by alternative use of these multiple promoters. The expression of aromatase in the breast and abdominal adipose tissues is regulated by a promoter flanking exon 1b. Molecular and epidemiological analyses of tissue-specific utilization of multiple exons 1 and promoters revealed a switching from use of the adipose-specific exon 1b to exon 1c in adipose tissues adjacent to the carcinomas in most breast cancer patients. Exon 1c has been shown to be specific for the ovary. Aromatase mRNA in adipose tissues distal to the tumour of the same patients was normally transcribed from exon 1b as was breast tissue in healthy controls. It is noteworthy that a switching from exon 1b to exon 1c was often observed in breast cancer patients having metastatic lymph nodes. These data suggest that switching from an adipose-specific exon 1b to exon 1c could cause a deviation from strict regulation of tissue-specific expression of the adipose aromatase leading to over-expression of the adipose aromatase. Consequently overproduction of local estrogen may promote carcinogenesis or proliferation of breast cancer cells.

Molecular cloning of a cDNA encoding human histidase

We isolated overlapping cDNA clones encoding human histidase (histidine ammonia-lyase) from a human lambda gt10 library. The cDNA predicted a 657 amino acid protein of 72,651 Da. The human histadase amino acid sequence was 93% conserved with both rat and mouse histidase sequences, including four N-glycosylation consensus sites.

Neurotransmitter-mediated regulation of brain aromatase: protein kinase C- and G-dependent induction.

Abstract: Aromatase in the diencephalic neurons, the level of which increases transiently during the prenatal to neonatal period, has been suggested to be involved in control of sexual behavior and differentiation of the CNS. Effects of neurotransmitters on levels of aromatase mRNA in cultured neurons were investigated to determine factors regulating the developmental increase that occurs in level of fetal brain aromatase. The expression of aromatase in diencephalic neurons of fetal mice at embryonic day 13, cultured in vitro, was significantly affected by α1‐adrenergic receptor ligands. Aromatase mRNA levels were higher in neurons treated with the α1‐agonist phenylephrine than in control neurons, whereas prazosin, an α1‐antagonist, suppressed this increase, and ligands for α2‐ or β‐adrenergic receptors did not exert any influence. The profile of α1‐adrenergic receptor subtypes during actual development in vivo suggested that the α1B subtype is in fact responsible for the signal transduction. Substance P, cholecystokinin, neurotensin, and brain natriuretic peptide also increased the level of expression along with phorbol 12‐myristate 13‐acetate and dibutyryl‐cyclic GMP, whereas forskolin and dibutyryl‐cyclic AMP caused a decrease. These data indicate that stimulation via α1 (possibly α1B)‐adrenergic receptors, as well as receptors of specific neuropeptides, controls the expression of aromatase in embryonic day 13 diencephalic neurons through activation of protein kinase C or G. β‐Adrenergic receptors would not appear to participate in the regulation, judging from their developmental profile, although cyclic AMP might be a suppressive second messenger.

Identification of cis-acting elements in the proximal promoter region for brain-specific exon 1 of the mouse aromatase gene

Among multiple exons 1 of the mouse aromatase gene, brain-specific exon 1 is only utilized in the hypothalamus and amygdala regions. In this study, identification of the promoter region necessary for basal transcription of the aromatase gene in the brain was undertaken. Deletions of various lengths were introduced into the overall promoter region, which was fused to the chloramphenicol acetyltransferase gene. The resulting reporters were transfected into cultured neurons from the diencephala of fetal mouse brains on embryonic day 13 and then their CAT mRNA levels were determined. The reporter plasmid containing the promoter region 202 bp upstream from the transcriptional initiation site gave the greatest expression. Then binding of trans-acting factors in a nuclear extract of the diencephala to the -202 bp promoter region was investigated by DNase I footprint analysis, multiple protected areas, referred to as Arom-Aalpha, Abeta, Agamma, B and C, being found. Gel shift assays, performed with oligonucleotides corresponding to the protected areas, showed that nuclear DNA binding factors form specific complexes exhibiting different mobilities. Substitution in the Arom-Aalpha or -B sequence in the promoter region in the CAT reporters decreased the CAT mRNA expression levels to about one-fifth the wild type one. These results suggest that multiple nuclear factors bound to the core promoter region participate in the expression of the aromatase gene in mouse brain neurons.

Autonomous expression of aromatase during development of mouse brain is modulated by neurotransmitters

A transient increase in aromatase activity is known to occur in the hypothalamus of rodents in pre- and postnatal periods. The mechanisms regulating such a developmental increase of brain aromatase was studied in fetal mouse diencephalic cells, by measuring aromatase mRNA levels by a quantitative reverse transcription-polymerase chain reaction (RT-PCR) method. When slices of diencephalon were cultured on embryonic day (E) 12, E13 and E15, the level of aromatase mRNA continued to increase for the first 2 to 3 days. A time-dependent increase of mRNA was also shown for 3 days in E13 neuronal cells dissociated with papain and cultured in chemically defined medium. However, no significant increase was observed in E10 or E11 brain cells cultured by either method. Aromatase mRNA was detected in neither cerebral cortex neurons nor astrocytes. An alpha1-selective adrenergic agonist, phenylephrine, increased aromatase in the E13 diencephalic neurons in culture, whereas prazosin, an alpha1-antagonist, suppressed the mRNA level. Ligands for alpha2- or beta-adrenergic receptors did not alter the mRNA level. Substance P, cholecystokinin, neurotensin, and brain natriuretic peptide as well as phorbol 12-myristate 13-acetate and dibutyryl-cyclic GMP all increased the mRNA level. We concluded that: (a) the developmental increase of aromatase mRNA in diencephalic neurons is an autonomous event and is perhaps genetically regulated after E12; (b) aromatase mRNA is expressed in a cell type- and region-specific manner; and (c) protein kinases C and G activated via receptors of the specific neurotransmitters may be involved in modulation of the developmental expression of aromatase mRNA.

In vitro Increase of Aromatase mRNA in Diencephalic Neurons

This study was designed to (1) examine the ability of fetal diencephalic neurons cultured in vitro to express aromatase mRNA and (2) evaluate the involvement of several environmental factors which may regulate the development and differentiation of the neurons in the central nervous system. Brain cells from fetal mice at various developmental stages were cultured as tissue slices and as primary monolayer cells, and the expression levels of aromatase mRNA in the cultured cells were measured by a quantitative reverse transcription-polymerase chain reaction method using an internal standard. On cultured slices of the diencephalic region from fetal mice on embryonic day 12 (E12), E13, or E15 on collagen-coated membranes, the expression level of the mRNA continued to increase for the initial 2-3 days as that in vivo. Time-dependent increase of aromatase mRNA was also observed for 3 days in E13 neuronal cells dissociated with papain and cultured on poly L-Lys-coated dishes in serum-free medium. However, no significant time-dependent increase of the aromatase mRNA level was observed in E10 or E11 brain cells cultured by either method. These findings suggest that the developmental increase of aromatase mRNA in diencephalic neurons is an endogenous characteristic and probably genetically determined after E12.

Identification of an amino acid residue involved in the substrate-binding site of rat liver uricase by site-directed mutagenesis

Computer analysis has shown that a conserved amino acid sequence (Leu 160 to Lys 164) of rat liver uricase is also present in other enzymes with purine substrates. The significances of the amino acids in this sequence were studied by site-directed mutagenesis. Replacement of Lys 164 by Glu or Ile resulted in loss of uricase activity and decrease in binding of the competitive inhibitor xanthine. The far ultraviolet circular dichroic spectra of the mutant uricases were identical to that of the wild type protein, indicating that the replacement of Lys 164 by other amino acids did not result in serious modification of the conformation of uricase. These findings suggest that this amino acid is involved in the substrate-binding site of the enzyme.