Chromosomal localization of seven PAX genes and cloning of a novel family member, PAX-9.

In the human paired box-containing (PAX) gene family, only two members, PAX-3 and PAX-6, which are associated with Waardenburg's syndrome and aniridia, respectively have been mapped to human chromosomes. We have now isolated cosmids for six additional human PAX genes (PAX-1,-2,-5,-7,-8,-9) and a polymerase chain reaction fragment for PAX-4. PAX-9 is a novel family member which is closely related in its paired domain to PAX-1. The chromosomal location of all cloned PAX genes was determined by analysis of somatic cell hybrids and (except PAX-4) by fluorescence in situ hybridization to metaphase chromosomes. PAX-1 and PAX-7 map to chromosomal regions containing previously assigned disease loci.

Independent regulation of the two Pax5 alleles during B-cell development.

The developmental control genes of the Pax family are frequently associated with mouse mutants and human disease syndromes. The function of these transcription factors is sensitive to gene dosage, as mutation of one allele or a modest increase in gene number results in phenotypic abnormalities. Pax5 has an important role in B-cell and midbrain development. By following the expression of individual Pax5 alleles at the single-cell level, we demonstrate here that Pax5 is subject to allele-specific regulation during B-lymphopoiesis. Pax5 is predominantly transcribed from only one allele in early progenitors and mature B cells, whereas it switches to a biallelic transcription mode in immature B cells. The allele-specific regulation of Pax5 is stochastic, reversible, independent of parental origin and correlates with synchronous replication, in contrast with imprinted and other monoallelically expressed genes. As a consequence, B-lymphoid tissues are mosaics with respect to the transcribed Pax5 allele, and thus mutation of one allele in heterozygous mice results in deletion of the cell population expressing the mutant allele due to loss of Pax5 function at the single-cell level. Similar allele-specific regulation may be a common mechanism causing the haploinsufficiency and frequent association of other Pax genes with human disease.

A crucial function of PDGF in TGF-beta-mediated cancer progression of hepatocytes.

Polarized hepatocytes expressing hyperactive Ha-Ras adopt an invasive and metastatic phenotype in cooperation with transforming growth factor (TGF)-beta. This dramatic increase in malignancy is displayed by an epithelial to mesenchymal transition (EMT), which mimics the TGF-beta-mediated progression of human hepatocellular carcinomas. In culture, hepatocellular EMT occurs highly synchronously, facilitating the analysis of molecular events underlying the various stages of this process. Here, we show that in response to TGF-beta, phosphorylated Smads rapidly translocated into the nucleus and activated transcription of target genes such as E-cadherin repressors of the Snail superfamily, causing loss of cell adhesion. Within the TGF-beta superfamily of cytokines, TGF-beta1, -beta2 and -beta3 were specific for the induction of hepatocellular EMT. Expression profiling of EMT kinetics revealed 78 up- and 235 downregulated genes, which preferentially modulate metabolic activities, extracellular matrix composition, transcriptional activities and cell survival. Independent of the genetic background, platelet-derived growth factor (PDGF)-A ligand and both PDGF receptor subunits were highly elevated, together with autocrine secretion of bioactive PDGF. Interference with PDGF signalling by employing hepatocytes expressing the dominant-negative PDGF-alpha receptor revealed decreased TGF-beta-induced migration in vitro and efficient suppression of tumour growth in vivo. In conclusion, these results provide evidence for a crucial role of PDGF in TGF-beta-mediated tumour progression of hepatocytes and suggest PDGF as a target for therapeutic intervention in liver cancer.

Transcripts from amplified sequences of an inherited homogeneously staining region in chromosome 1 of the house mouse (Mus musculus).

Several populations of the house mouse, Mus musculus, are polymorphic for the presence or absence of an inherited homogeneously staining region (HSR) in chromosome 1. The HSR consists of highly amplified DNA sequences, present in low copy numbers in the HSR- genome. A cloned HSR-derived genomic sequence detected transcripts of about 1.3 and 4.5 kb on blots of poly(A)+ RNA from liver of HSR+ mice but not from that of HSR- mice. A cDNA library was established from RNA of HSR+ mice and screened with the HSR-derived genomic clone. Positive clones were isolated and shown to be complementary to the 1.3-kb RNA species and to amplified DNA sequences in the HSR+ genome. The combined sequence of four overlapping cloned cDNAs is 959 nucleotides long and includes an open reading frame encoding a putative protein of 208 amino acids. The pertinent gene is unidentified. No homologous sequence is stored in the EMBL data base. A stretch of 109 nucleotides at the 3' end of the 1.3-kb RNA homology region in the same genomic fragment, as indicated by hybridization data and sequence motifs resembling promoter elements. Thus, our data suggest that at least two genes or gene families are encoded in the HSR.

Human colorectal cancer: high frequency of deletions at chromosome 1p35.

Cytogenetic analyses of human colon cancer cells have revealed a high frequency of chromosome 1p deletions among other chromosomal abnormalities. In order to find out whether these chromosomal alterations are manifestations of loss of genetic material, we surveyed DNA of 62 primary tumors, 7 metastases, and matching peripheral blood cells with a panel of polymorphic DNA probes that detect different loci on chromosome 1p. A portion of the probes was derived from a microclone bank generated by microdissection and microcloning of 1p35----pter DNA. In 42% of the colon carcinomas allelic loss was observed with at least one probe. The deletions were of different sizes but always included a region involving band 1p35, except for two tumors in which allelic loss was detected more proximally. The frequency of 1p deletion in the metastases was higher than in the primary tumors. These data indicate that genetic information related to tumorigenesis is located within or nearby region 1p35 and that deletion of this region occurs later in tumor development. Our results add to the number of genetic changes presumably involved in colon cancerogenesis.

Abnormalities of chromosome 1 and loss of heterozygosity on 1p in primary hepatomas.

Cytogenetic analysis of eight human hepatoma-derived cell lines and one primary hepatocellular carcinoma biopsy revealed multiple chromosome abnormalities; however, only chromosome 1 was consistently affected by rearrangements. Pseudopolysomy 1 as well as chromosome 1 deletions and/or translocations that resulted in loss of the distal 1p region from at least one copy of chromosome 1 were observed in all but one of the cell lines analysed. Molecular analyses of tumor-derived and normal genomic DNA from six cases of hepatocellular carcinoma and from two of hepatoblastoma, using a panel of chromosome 1p-specific DNA probes indicated allelic loss in the distal 1p region in five of the six hepatocellular carcinomas but not in either hepatoblastoma. These results suggest the location of a gene in the distal 1p region whose functional loss may be involved in hepatocellular carcinogenesis.

Dominant and recessive molecular changes in neuroblastomas.

Of all human tumors, neuroblastomas bear the most prominent genetic changes. Amplifications and deletions of chromosomal DNA can be identified by light microscopy on chromosomal spreads of neuroblastoma cells with remarkable frequency and consistency. Consequently, extensive studies have been undertaken to elucidate the molecular basis of these cytogenetic changes. A rich body of information has accumulated on the role played by dominant oncogenes and recessive tumor suppressor genes in the pathogenesis of this disease. Most notably, it was found that amplification of N-myc is responsible for the presence of double minutes and homogeneously staining regions in neuroblastoma chromosomes. It has also been discovered that N-myc amplification is a prognostic sign of malignancy. More recently, recessive genetic alterations in neuroblastoma, such as deletion of putative tumor-suppressing genes have received increasing attention, and considerable efforts are being made to identify such genetic elements. Finally, the susceptibility of neuroblastoma cells to differentiating stimuli has made them a popular in vitro system for neurobiological and pharmacological research. The need for suitable in vivo systems has spurred the development of several animal models employing tumor viruses and transgenic technologies.

Analysis of location and integrity of the human PITSLRE (p58(cdc2L1)) genes in neuroblastoma cell genomes.

The family of PITSLRE kinase genes, located in chromosome 1p36, has recently been associated with neuroblastoma tumorigenesis. In order to evaluate the role of these genes as putative tumor suppressor genes, we have analyzed the integrity of the coding region in primary tumors and its location relative to a neuroblastoma consensus deletion. A subset of aggressive neuroblastoma tumors with allelic loss of different parts of chromosome 1p were investigated. Single-strand conformation polymorphism (SSCP), heteroduplex (HD) and sequencing analysis of tumor DNA did not reveal any significant changes in the coding region. In particular, a primary tumor with an interstitial allelic deletion in 1p36 did not reveal concomitant loss of heterozygosity of the PITSLRE gene region when analyzed with a C/T DNA sequence polymorphism in exon 5 of PITSLRE1. FISH analysis on neuroblastoma cell lines with small interstitial deletions and with a balanced translocation in 1p36 revealed that the PITSLRE gene cluster was localized distal to the neuroblastoma consensus deletion. against an involvement of the PITSLRE genes in neuroblastoma tumorigenesis.

[Reproductive and developmental toxicity studies of lactitol (NS-4) (3)--Teratogenicity study in rabbits by oral administration].

Lactitol, a hepatic encephalopathy drug, was administered by oral gavage to pregnant New Zealand White rabbits during organogenesis from day 6 to day 18 of gestation inclusive, at dosages of 0, 0.25, 0.75 or 4.5 g/kg/day. On day 29 of gestation, females were killed to allow examination of their uterine contents. There was a slight reduction in food intake and faecal output among females receiving 4.5 g/kg. One female receiving 4.5 g/kg aborted following a prolonged period of weight loss. No adverse effects on litter parameters were recorded that could be attributed to treatment. Foetal morphogenesis was unaffected by treatment with lactitol. The results show that no-effect dose levels of lactitol are 0.75 g/kg in mother rabbits for general toxicity and for reproductive functions, and 4.5 g/kg for their fetuses.

Continuous high-repetition-rate operation of collisional soft-x-ray lasers with solid targets.

We have generated a laser average output power of 2 microW at a wavelength of 13.9 nm by operating a tabletop laser-pumped Ni-like Ag laser at a 5 Hz repetition rate, using a solid helicoidal target that is continuously rotated and advanced to renew the target surface between shots. More than 2 x 10(4) soft-x-ray laser shots were obtained by using a single target. Similar results were obtained at 13.2 nm in Ni-like Cd with a Cd-coated target. This scheme will allow uninterrupted operation of laser-pumped tabletop collisional soft-x-ray lasers at a repetition rate of 10 Hz for a period of hours, enabling the generation of continuous high average soft-x-ray powers for applications.

The fine structure of euchromatin and centromeric heterochromatin in Tenebrio molitor chromosomes.

The fine structure of constitutive heterochromatin and euchromatin was compared in electron microscope whole-mount preparations of Tenebrio molitor (Insecta, Coleoptera) spermatocyte nuclei. Tenebrio molitor pachytene chromosomes display extended segments of centromeric heterochromatin and thus are especially suitable for this purpose. When nuclei were incubated in solutions containing different concentrations of NaCl or of MgCl2, two levels of chromatin fine structures were observed in the euchromatic segments: nucleosome fibers (0.1 mM-20 mM NaCl) and supranucleosomal fibers with 28 nm in diameter (40 mM-100 mM NaCl, 0.2 mM-1.0 mM MgCl2). The fine structure in the heterochromatic segments was the same as that in the euchromatic segments in all NaCl concentrations and in MgCl2 concentrations up to 0.4 mM. In higher MgCl2 concentrations the heterochromatin remained more compact than the euchromatin and consisted of 37-nm-thick fibers in 0.6 mM MgCl2 and of 65-nm-thick fibers in 1.0 mM MgCl2. After the 37-nm and the 65-nm fibers had been dispersed in Mg2+-free solutions they could be recondensed by incubation in 0.6 mM and 1.0 mM MgCl2, respectively. It is concluded that a Mg2+-sensitive component of the heterochromatin is responsible for the folding of the nucleosome chain to heterochromatin-specific supranucleosomal structures.

Mg2+-dependent compactness of heterochromatic chromosome segments.

The stability of heterochromatic blocks in pachytene chromosomes of Tenebrio molitor (Insecta, Coleoptera) was analysed at the chromosomal level using a modified Miller spreading technique. Incubation of nuclei in solutions of different ionic strength and composition revealed that the characteristic compactness of heterochromatic segments was preserved in the presence of at least 0.6 mM MgCl2. The compactness of these segments was lost in solutions of different ionic strength and composition revealed that the characteristic compactness of heterochromatic segments was preserved in the presence of at least 0.6 mM MgCl2. The compactness of these segments was lost in solutions containing NaCl (0.1-100 mM), but no MgCl2. They then resembled the euchromatic segments. The decondensed heterochromatic segments could be recompacted by adding MgCl2 to a final concentration of 1.0 mM. The characteristic compactness of heterochromatin of pachytene chromosomes therefore depends on the presence of Mg2+, but is independent of Na+.

Expression of the helix-loop-helix gene Id3 during murine embryonic development.

The family of dominant-negative helix-loop-helix (dnHLH) transcriptional modulators consists of four mammalian genes known to date: Id1, Id2, Id3 (Id3 was formerly designated HLH462 or HEIR1 or HLH1R21), and Id4. We have performed a detailed study of Id3 expression in mouse embryos in order to gain insight into the possible developmental control functions of this gene and to compare its expression to that of other mammalian dnHLH genes. Employing RNA in situ hybridization on sectioned mouse embryos of gestational stages E9.5-17.5 and neonatal head, we generally detected a high abundance of transcripts at early stages that gradually declined at most sites toward E15.5. Expression of Id3 in embryos was detected in brain, spinal cord, olfactory system, branchial arches, limbs, sclerotome, endocardiac cushions, the outer lining of the gut, lung, retina, the collecting system of the kidney, and in tooth anlagen. Although the abundance of mRNA decreased toward later stages in most tissues, it remained high in teeth and kidney. This expression pattern suggests that Id3 functions both in undifferentiated tissues and in organs which are in the process of differentiation. When compared to the expression of other dnHLH genes, it becomes obvious that the pattern of Id3 mainly coincides with that of Id1. This may reflect a partial redundancy in gene function. Furthermore, previous experiments suggested a mutually exclusive expression of the proto-oncogene N-myc an Id3. The results of the present study confirm these data.

Nonradioactive in situ nick translation combined with counterstaining: characterization of C-band and silver positive regions in mouse testicular cells.

The DNase I sensitivity of three different chromatin regions in mouse testicular cells was analysed by in situ nick translation with biotin-dUTP combined with various counterstaining techniques. The regions were: (i) the constitutive centromeric heterochromatin, (ii) an interstitial C-band positive insertion on chromosome 1, Is(HSR1;C5)1Lub, and (iii) the chromatin containing rDNA (designated nucleolar chromatin herein). Incorporated biotin was detected either by the horseradish peroxidase reaction with diaminobenzidine (DAB) or the alkaline phosphatase reaction with fast red. The latter resulted in a water insoluble red precipitate, which was easily removable by any organic solution thus allowing the application of various counterstaining protocols. DNase I sensitivity of the three chromatin regions was screened in different cell types of the mouse testis. The interstitial Is(HSR) region was highly DNase I sensitive when it was recognizable by strong mithramycin fluorescence. The centromeric heterochromatin was DNase I resistant when it was compacted into microscopically visible chromosomal structures (mitosis, pachytene, metaphase I and II). In interphase nuclei from Sertoli cells and spermatogonia it became highly DNase I sensitive. In round spermatids it displayed medium DNase I sensitivity. Nucleolar chromatin was not labelled by in situ nick translation when silver staining demonstrated strong protein production. Sperm cells were highly DNase I sensitive from stages 11 to 15, but resistant as mature spermatozoa.

Evidence for in situ amplification of a germ line homogeneously staining region in the mouse.

A cloned DNA sequence that is specific for a germ line homogeneously staining region (HSR) on chromosome 1 of the mouse was found to be homologous to a single copy sequence in non-HSR mice. By in situ hybridization, the sequence in non-HSR mice was localized to approximately the same site as the insertion site of the HSR on chromosome 1 of HSR mice, indicating in situ amplification of the HSR.

A physical map of the region encompassing the human interleukin-1 alpha, interleukin-1 beta, and interleukin-1 receptor antagonist genes.

We have constructed a restriction map of the human genomic region containing the genes encoding the three members of the interleukin-1 (IL-1) family, IL-1 alpha, IL-1 beta, and IL-1 receptor antagonist (IL-1 ra). For this purpose, pulsed-field gel electrophoresis blots were hybridized with probes derived from the three genes. The genes (IL1A, IL1B, and IL1RN, respectively) were found to map to a common restriction fragment of approximately 430 kb that is flanked by two clusters of sites for methylation-sensitive rare-cutter restriction enzymes (putative CpG islands). A likely third internal CpG island was marked by two rare-cutter sites. CpG and non-CpG-specific enzymes were used to map the three genes. Relative to one terminal CpG island, the three genes were mapped to the following intervals: IL1A was between +0 and +35 kb, IL1B was between +70 and +110 kb, and IL1RN was between +330 and +430 kb.

The gene coding for the B cell surface protein CD19 is localized on human chromosome 16p11.

The CD19 gene codes for one of the earliest markers of the human B cell lineage and is a target for the B lymphoid-specific transcription factor BSAP (Pax-5). The transmembrane protein CD19 has been implicated in controlling proliferation of mature B lymphocytes by modulating signal transduction through the antigen receptor. In this study, we have employed Southern blot and fluorescence in situ hybridization analyses to localize the CD19 gene to human chromosome 16p11.

Chromosome 1 deletions in human neuroblastomas: generation and fine mapping of microclones from the distal 1p region.

Human neuroblastomas show a high incidence of deletions in the distal region of the short arm of chromosome 1. In pursuit of a molecular analysis of these deletions, we have generated a microclone bank from microdissected 1p35-pter chromosomal fragments. To allow a rapid localization of the microclones, we have also generated a panel of (human x mouse) hybrid cell lines through microcell-mediated chromosome transfer. The hybrid cells contained different portions of the human chromosome 1 on a murine background. A total of 20 randomly chosen single or low-copy microclones were localized by Southern analysis on DNA of the hybrid panel: All probes were derived from chromosome I. Sixteen mapped in region 1p36.1-pter, two in 1p22-p36.1, and another two in 1cen-qter. The mapping of ten of these microclones was further refined by in situ hybridization. Cells of the neuroblastoma line GI-ME-N carry two types of chromosome 1, one cytogenetically normal and another with a translocation reported to be in 1p36.2, i.e., a t(1;?) (p36.2;?) marker. Using cell hybridization, we separated the two chromosome 1 types of GI-ME-N into different hybrid cell clones. Southern hybridization of three microclones from distal Ip to DNA of the hybrid cell clones revealed that the breakpoint in the translocated chromosome I was located in 1p36.1.

Analysis of chromosome band 1p36 alterations by chromosomal in situ suppression hybridization with a microclone DNA bank.

Alterations of the distal portion of chromosome Ip are a recurrent abnormality of several types of human cancer. In this study we show that chromosomal in situ suppression hybridization with a regional 1p36 DNA bank generated by microdissection and microcloning can be employed to detect translocations involving 1p36.

Large-scale isolation of human 1p36-specific P1 clones and their use for fluorescence in situ hybridization.

A series of 80 microclone probes derived from the chromosomal region 1p36 was used to isolate corresponding clones from the ICRF human P1 library (see Francis et al., this issue). Hybridization screenings were performed using probe pools on high-density filter grids. A total of 87 P1 clones specific for 1p36 were isolated. This large-scale approach allowed a detailed evaluation of the complexity, quality, and utility of this library. The isolated P1 clones were used both for size determination by pulsed-field gel electrophoresis and as probes for fluorescence in situ hybridization (FISH) analysis. FISH of P1 clones is shown to be both easy and efficient to perform on metaphase chromosomes and interphase nuclei. This observation is expected to reveal new avenues for diagnosis of disease-related chromosomal changes. The use of P1 clones as a tool in clinical and tumor interphase cytogenetics is discussed and compared with FISH data of other long insert clones such as cosmids and YAC clones.