Fumiko Saito-Ohara

Alteration in Copy Numbers of Genes as a Mechanism for Acquired Drug Resistance

Chemoresistance is a major obstacle for successful treatment of cancer. To identify regions of the genome associated with acquired resistance to therapeutic drugs, we conducted molecular cytogenetic analyses of 23 cancer-cell lines, each resistant to either camptothecin, cisplatin, etoposide (VP-16), Adriamycin, or 1-β-d-arabinofuranosylcytosine, although the parental tumor lines were not. Subtractive comparative genomic hybridization studies revealed regions of gain or loss in DNA-copy numbers that were characteristic of drug-resistant cell lines; i.e., differences from their drug-sensitive parental cell lines. Thirteen ATP-binding cassette (ABC) transporter genes [ABCA3, ABCB1 (MDR1), ABCB6, ABCB8, ABCB10, ABCB11, ABCC1 (MRP1), ABCC4, ABCC9, ABCD3, ABCD4, ABCE1, and ABCF2] were amplified among 19 of the resistant cell lines examined. Three genes encoding antiapoptotic BCL-2 proteins (BCL2L2, MCL1, and BCL2L10) were also amplified and consequently overexpressed in three of the derivative lines. Down-regulation of BCL2L2 with an antisense oligonucleotide sensitized a VP-16 resistant ovarian-cancer cell line (SKOV3/VP) to VP-16. A decrease in copy numbers of genes encoding deoxycytidine kinase, DNA topoisomerase I, and DNA topoisomerase II α reduced their expression levels in one cytosine arabinoside-resistant line, two of three camptothecin-resistant lines, and two of five VP-16-resistant cell lines, respectively. Our results indicated that changes in DNA-copy numbers of the genes mentioned can activate or down-regulate them in drug-resistant cell lines, and that such genomic alterations might be implicated in acquired chemoresistance.

A Novel Target Gene, SKP2, within the 5p13 Amplicon That Is Frequently Detected in Small Cell Lung Cancers

We investigated DNA copy-number aberrations in 22 cell lines derived from small cell lung cancers (SCLCs) using comparative genomic hybridization. A minimal common region at 5p13, within the 5p11-p13 amplicon that was most frequently involved, harbored the CDH6, PC4, and SKP2 genes. These three genes showed amplification and consequent overexpression in the SCLC cell lines. SKP2 positively regulates progression of cell cycle by targeting several regulators, such as the cell-cycle inhibitor p27(KIP1), for ubiquitin-mediated degradation. SKP2 was amplified in 7 (44%) of 16 primary SCLC tumors, and consequently overexpressed in 10 (83%) of the 12 of those tumors we examined. Expression levels of SKP2 protein were cell cycle-dependent in SCLC cells as well as in normal cells, and were correlated with the DNA copy-number of the gene. There was an inverse correlation between the expression of SKP2 and p27(KIP1) proteins. Down-regulation of SKP2 using an anti-sense oligonucleotide remarkably suppressed the growth of SCLC cells. Our results indicate that SKP2 is likely to be a target of the 5p13 amplification and to play an important role in the growth of SCLC cells.

The Novel Gene Encoding a Putative Transmembrane Protein Is Mutated in Gnathodiaphyseal Dysplasia (GDD)

Gnathodiaphyseal dysplasia (GDD) is a rare skeletal syndrome characterized by bone fragility, sclerosis of tubular bones, and cemento-osseous lesions of the jawbone. By linkage analysis of a large Japanese family with GDD, we previously mapped the GDD locus to chromosome 11p14.3-15.1. In the critical region determined by recombination mapping, we identified a novel gene (GDD1) that encodes a 913-amino-acid protein containing eight putative transmembrane-spanning domains. Two missense mutations (C356R and C356G) of GDD1 were identified in the two families with GDD (the original Japanese family and a new African American family), and both missense mutations occur at the cysteine residue at amino acid 356, which is evolutionarily conserved among human, mouse, zebrafish, fruit fly, and mosquito. Cellular localization to the endoplasmic reticulum suggests a role for GDD1 in the regulation of intracellular calcium homeostasis.

Reduced Levels of ATF-2 Predispose Mice to Mammary Tumors

ABSTRACT Transcription factor ATF-2 is a nuclear target of stress-activated protein kinases, such as p38, which are activated by various extracellular stresses, including UV light. Here, we show that ATF-2 plays a critical role in hypoxia- and high-cell-density-induced apoptosis and the development of mammary tumors. Compared to wild-type cells, Atf-2−/− mouse embryonic fibroblasts (MEFs) were more resistant to hypoxia- and anisomycin-induced apoptosis but remained equally susceptible to other stresses, including UV. Atf-2−/− and Atf-2+/− MEFs could not express a group of genes, such as Gadd45α, whose overexpression can induce apoptosis, in response to hypoxia. Atf-2−/− MEFs also had a higher saturation density than wild-type cells and expressed lower levels of Maspin, the breast cancer tumor suppressor, which is also known to enhance cellular sensitivity to apoptotic stimuli. Atf-2−/− MEFs underwent a lower degree of apoptosis at high cell density than wild-type cells. Atf-2+/− mice were highly prone to mammary tumors that expressed reduced levels of Gadd45α and Maspin. The ATF-2 mRNA levels in human breast cancers were lower than those in normal breast tissue. Thus, ATF-2 acts as a tumor susceptibility gene of mammary tumors, at least partly, by activating a group of target genes, including Maspin and Gadd45α.

MOLECULAR CHARACTERIZATION OF HUMAN AQUAPORIN-7 GENE AND ITS CHROMOSOMAL MAPPING

The cDNA for the seventh mammalian aquaporin (AQP7) was isolated from rat testis, and its expression demonstrated at the tail of late spermatids (Ishibashi et al., J. Biol. Chem. 272 (1997) 20,782-20,786). Here we report the isolation of the mouse and the human AQP7 cDNA and the human AQP7 gene. The human AQP7 gene is identical with human adipose AQP (AQPap or AQP7L). The deduced amino acid sequences of human and mouse AQP7 were 68% and 79% identical to those of rat AQP7, respectively. The mouse AQP7 is 67% identical to the human AQP7. Such a lower conservation of AQP7 among species is unusual in the aquaporin family. The human AQP7 gene is composed of six exons distributing over 6.5 kb. The exon-intron boundaries are identical to those of the human AQP3 gene. The intron sizes are also similar. Moreover, chromosomal localization of AQP7 was assigned to 9p13 by fluorescent in situ hybridization, where AQP3 is also localized, suggesting that 9p13 may be another site of an aquaporin cluster.

Pax-5 Is Essential for κ Sterile Transcription during Igκ Chain Gene Rearrangement

Pax-5 is the key regulator in B cell development. Pax-5-deficient mice show defects in B cell commitment and recombination of IgH chain gene rearrangement from DJ to VDJ. Previously, we found that Pax-5 bound to KI and KII sites, which play a crucial role in κ-chain gene rearrangement. However, the function of Pax-5 in Igκ chain gene rearrangement has not been investigated. To address this issue, we newly established pre-BI cell lines expressing the pre-B cell receptor from Pax-5-deficient mice and used them in an in vitro culture system, in which κ-chain gene rearrangement is induced by removing IL-7. By examining the Pax-5-deficient pre-BI (knockout (KO)) cells, we show in this study that, despite recombination-activating gene 1 and 2 expression, these KO cells did not rearrange the κ-chain gene following the absence of κ sterile transcription. Consistent with these data, fluorescent in situ hybridization analyses revealed that the Jκ locus in KO cells was located at the nuclear periphery as a repressive compartment. Transfection of KO cells with Pax-5 constructs indicated that the transactivation domain of Pax-5 was required for κ sterile transcription and κ-chain gene rearrangement. Moreover, the hormone-inducible system in KO cells demonstrated that Pax-5 directly functioned in κ sterile transcription. These results indicate that Pax-5 is necessary for κ sterile transcription during Igκ chain gene rearrangement.

Isolation, characterization and mapping of the mouse and human RB1CC1 genes

RB1CC1 (RB1-inducible Coiled-Coil 1), a putative transcription factor implicated in the regulation of RB1 (retinoblastoma 1) expression, was recently identified in a screen for genes involved in multi-drug resistance to anticancer agents. Information about the RB1CC1 gene is limited, however, and its biological function is not determined. Here we report the isolation, characterization and mapping of the mouse RB1CC1 gene (Rb1cc1), together with further characterization of the human RB1CC1 gene. Mouse Rb1cc1 encodes 1588 amino acids, sharing 89% identity and key sequence motifs with its human counterpart. Rb1cc1 is expressed abundantly in heart and testis, with lower levels detected in lung and spleen. Immunohistochemical analysis revealed the Rb1cc1 and Rb1 proteins are co-localized in the cell nuclei of NIH3T3-3 cell and various mouse tissues. The human and mouse RB1CC1 genes, both of which contain 24 exons, span 74 kb on chromosome 8q11.2 and 57 kb on chromosome 1A2-4, respectively. Conserved sequence motifs and nuclear localization suggest that the RB1CC1 proteins function as transcription factors.

The human caspase-activated DNase gene (hCAD) : genomic structure, exonic single-nucleotide polymorphisms, and a highly polymorphic dinucleotide repeat at the hCAD locus

AbstractCaspase-activated DNase (CAD) cleaves chromosomal DNA during apoptosis. We determined its genomic structure and identified single-nucleotide polymorphisms (SNPs) within exons 5 and 7, as well as a highly polymorphic dinucleotide repeat of (CT)m(CA)n within the 5′ region of the human CAD gene (hCAD). The genomic structure of hCAD presented here, together with information concerning SNPs within the gene, as well as a highly polymorphic (CT)m(CA)n repeat fragment at the hCAD locus, may assist in the construction of genetic maps for exploring gene(s) that play pivotal roles in carcinogenesis.

ASSIGNMENT OF THE HUMAN GENE FOR KBF2/RBP-Jk TO CHROMOSOME 9p12-13 AND 9q13 BY FLUORESCENCE IN SITU HYBRIDIZATION

SummaryThe transcription factor KBF2 has been characterized as a factor that binds to the NFkB site of mouse major histocompatibility complex (MHC) class I genes and its amino acid sequence has been shown to be identical to those of members of the recombination signalsequence binding protein (RBP-Jk) family. Previous studies by Amakawa et al. (Genomics17, 306–315, 1993) demonstrated that the functional gene is localized at human chromosome 3q25. However, in the present study we showed by in situ hybridization with the functional KBF2/RBPJk cosmid clone that the gene is localized at 9p12-13 and 9q13, namely, at the same loci as pseudogenes that were reported previously (Zhang et al., Jpn J Human Genet 39, 391–401, 1994).

High-resolution chromosome R-banding in lymphoblastoid cell lines by the combined use of cell synchronization and ethidium bromide treatment

SummaryA reliable method for obtaining high-resolution R-banded chromosomes from lymphoblastoid cell lines is described. The cell cultures are subjected to S-phase synchronization in the presence of excess thymidine (300 μg/ml) for 17 to 19 hr, followed by BrdU treatment (30 μg/ml) for 6.5 hr. Prior to harvest, they are exposed to ethidium bromide (7.5 μg/ml) for 1.5 hr and Colcemid (0.02 μg/ml) for 30 min. Using this method, high-resolution R-banded chromosomes at the 550–850 band level were obtained with frequencies at high as 70% of all mitotic cells.