Neutrophil collagenase (MMP-8) is expressed during early development in neural crest cells as well as in adult melanoma cells.

Matrix metalloproteinase-8 (MMP-8) is a neutral metalloproteinase of the fibrillar collagenase family that also includes MMP-1 and MMP-13. In contrast to the other collagenases, MMP-8 has a very limited tissue distribution, thought to be restricted to neutrophils and chondrocytes. In a previous study, we observed MMP-8 expression in human melanoma cells. This observation led us to assess in more detail the expression of MMP-8 in normal and malignant melanocytic cells. We found that MMP-8 was expressed by 11 out of 12 human melanoma cell lines tested and all 10 primary melanomas we examined, but was not expressed by four primary neonatal melanocyte strains. Since melanocytes arise from highly motile neural crest cells, we examined the hypothesis that MMP-8 might be expressed by neural crest cells. RT-PCR analysis of post-implantation mouse embryos indicated the presence of MMP-8 transcripts at E9.5. In situ hybridization and immunohistochemistry of mouse embryos between E9.5-E14.5 demonstrated MMP-8 expression in the surface ectoderm, neural crest cells and chondrocytes. MMP-8 was also detected in neural crest cell migration located in the circumference of the neural tube, branchial arches and the notochord. In addition, MMP-8 expression was observed between the somites, in circumscriptive areas of the developing brain, heart, and eye, and in the interdigital zones of the limbs. In summary, we found MMP-8 to be the first fibrillar collagenase expressed during development. In contrast to its restricted tissue expression post-partum, MMP-8 was present in multiple embryonic tissues, including neural crest cells. The production of MMP-8 by migrating neural crest cells may contribute to their ability to degrade fibrillar collagen matrices while in transit.

Chromosomal sublocalization of the transcribed human telomere repeat binding factor 2 gene and comparative mapping in the mouse.

Telomere repeat binding factor 2 (TERF2) is one of two recently cloned mammalian telomere binding protein genes. TERF2 binds as a dimer with high affinity to the double-stranded TTAGGG telomeric repeat through an evolutionarily conserved myb-type DNA binding domain. TERF2 prevents telomere end-to-end fusion and may be important in maintaining genomic stability. We localized the transcribed TERF2 gene to human chromosome 16q22.1, tightly linked to the EST HUM000S343. The mouse Terf2 gene is situated by itself in a newly defined "bin" on chromosome 8 one crossover distal to Psm10 and Sntb2. Human TERF2 and mouse Terf2 are therefore part of a large evolutionarily conserved linkage group comprised of at least 25 known paralogous genes between human chromosome 16q and mouse chromosome 8.

Organization and expression of human telomere repeat binding factor genes.

The ends of mammalian chromosomes terminate in structures called telomeres. Recently a human telomere repeat binding factor (TRF1) that binds the vertebrate TTAGGG telomeric repeat in situ was isolated by Chong et al. (1). TRF1 regulates telomere length (2), which is often altered in cancer cells. To understand their genetic organization, TRF1 genes were localized to human chromosomes 13cen, 21cen, and Xq13 by analysis of human monochromosomal hybrids, and by fluorescent in situ hybridization. We also confirmed the recent localization of a human TRF1 gene to chromosome 8, and provide evidence that this locus is alternatively spliced. In contrast to the TRF1 genes on chromosomes 8 and X, the chromosomes 13 and 21 TRF1 genes contained a 60 bp deletion in the coding region. The results suggest that two distinct forms of TRF1 are expressed and that the TRF1 gene family includes at least three pseudogenes whose dispersal in the human genome may have occurred via cDNA intermediates.

Characterization of the mouse haptoglobin gene.

Plasmids containing mouse cDNA encoding haptoglobin, a major plasma protein that binds free hemoglobin, have been isolated and characterized. The amino acid sequence predicted by the mouse haptoglobin cDNA was 80% identical to human haptoglobin and 90% identical to rat haptoglobin sequence. The mouse haptoglobin probe was used to demonstrate a single haptoglobin gene in the genome of C57BL6 mice mapped to chromosome 8. Sequence analysis of the mouse Hp gene promoter revealed two unique features: the presence of a second TATA box with a 48-bp trinucleotide repeat immediately upstream. The enhancer element and the sequences shown to be required for cytokine and hormonal regulation of the rat Hp gene are highly conserved in mouse. Interestingly, the single nucleotide variation G to A, which completely inactivates the IL-6 responsive element A in the rat Hp gene, is identical in mouse. This suggests that the presence of an inactive IL-6-responsive element A in Hp genes is common in rodents.

Assignment of the mouse tartrate-resistant acid phosphatase gene (Acp5) to chromosome 9.

Tartrate-resistant acid phosphatase is a marker enzyme for osteoclasts, the multinucleated cell responsible for bone resorption. Interspecific somatic whole cell hybrids and karyotypically simple microcell hybrids were used to map the gene encoding tartrate-resistant acid phosphatase (Acp5) to mouse Chromosome 9. Acp5 is therefore a member of a syntenic family of genes that map to human chromosome 19p13.1-p13.3 and mouse Chromosome 9.

Activation of the colony-stimulating factor 1 receptor leads to the rapid tyrosine phosphorylation of GTPase-activating protein and activation of cellular p21ras.

We have previously reported that platelet-derived growth factor (PDGF) induced tyrosine phosphorylation of GTPase-activating protein (GAP) in intact quiescent fibroblasts under conditions in which insulin and basic fibroblast growth factor (bFGF) were ineffective (Molloy et al., 1988). In the present study, we have provided evidence that colony-stimulating factor 1 (CSF-1) is capable of inducing tyrosine phosphorylation of GAP and its associated cellular proteins, p62 and p190, in NIH3T3 cells overexpressing the human CSF-1 receptor (CSF-1R). However, the extent of GAP tyrosine phosphorylation induced by CSF-1 was approximately 10% of that induced by PDGF-BB in the NIH3T3 fibroblasts. Despite this significant difference, both PDGF-BB and CSF-1 increased the activation of p21ras, the extent of which correlated well with the mitogenic response induced by each growth factor in these cells. Taken together, our findings provide evidence for a possible role of tyrosine phosphorylation of GAP and GAP-associated phosphoproteins in regulating transduction of CSF-1-induced mitogenic signals through p21ras activation.

Assignment of tyrosine-specific T-cell phosphatase to conserved syntenic groups on human chromosome 18 and mouse chromosome 18.

Phosphorylation of proteins on tyrosine is crucially involved in signal transduction and mitogenesis and is regulated by both kinases and phosphatases. Recently, a number of soluble and transmembrane receptor-linked protein tyrosine phosphatases (PTPase) have been characterized. Among these is a 48.4-kDa PTPase encoded by a cDNA isolated from a T-lymphocyte library by low-stringency screening with probes derived from placental PTPase 1B. A human T-cell PTPase (PTPT) cDNA and somatic cell hybrids were used to assign a PTPT gene to conserved syntentic groups on human chromosome 18 and on mouse chromosome 18. Two unlinked sequences, one on human chromosome 1, were also detected.

Activation of a phosphatidylcholine-specific phospholipase C by colony stimulating factor 1 receptor requires tyrosine phosphorylation and a guanine nucleotide-binding protein.

Receptor tyrosine kinases couple to multiple intracellular effector molecules that are crucial for normal cell growth and transformation. Stimulation of membrane phospholipid hydrolysis by receptor tyrosine kinases is one such pathway for generating intracellular second messengers that may be important for mitogenesis. Certain receptor tyrosine kinases tyrosine phosphorylate a phosphoinositide-specific phospholipase C that hydrolyses the membrane phospholipid phosphatidylinositol 4,5-bisphosphate. In contrast, the glycoprotein receptor for colony stimulating factor 1, a transmembrane tyrosine kinase, does not utilize this pathway, but rather stimulates the hydrolysis of phosphatidylcholine. Here we show that eluates of antiphosphotyrosine affinity purified lysates of colony-stimulating factor 1-stimulated cells contain elevated levels of phosphatidylcholine-specific phospholipase C activity. The affinity-purified activity is sensitive to tyrosine-specific T-cell phosphatase, and is detected in the membrane fraction of stimulated cells. Recovery of phospholipase C activity in the antiphosphotyrosine protein fraction is reduced by pertussis toxin pretreatment of cells. The phosphatidylcholine phospholipase C activity in isolated membranes of colony-stimulating factor 1-treated cells was also reduced by pertussis toxin treatment and stimulated by guanosine 5'-3-O-(thio)triphosphate. These results indicate that colony stimulating factor 1 receptor-mediated stimulation of phosphatidylcholine-specific phospholipase C requires tyrosine phosphorylation, and might be affected by a G-protein coupled pathway.

Decline of signal transduction by phospholipase C gamma 1 in IMR 90 human diploid fibroblasts at high population doubling levels.

During cellular senescence in vitro, the cells do not respond mitogenically to serum growth factors at high population doubling levels. Phospholipase C activity in low PDL IMR 90 cells showed a 4.7-fold stimulation in response to 10% serum compared to 3.3-fold in high PDL cells when measured in whole cell extracts. Immunoaffinity purified tyrosine phosphorylated protein fraction showed a greater increase (5.2-fold) in phospholipase C activity in low PDL than high PDL cells (2.1-fold) in response to serum. Serum stimulated PLC gamma 1 activity was diminished in high PDL cells. Immunokinase assay of PLC gamma 1 immunoprecipitates from serum stimulated IMR 90 fibroblasts suggested that diminished enzymatic activity in high PDL cells is not due to less receptor coupled tyrosine phosphorylated PLC gamma 1 enzyme. Serum stimulated [3H]thymidine incorporation into DNA declined in parallel with the activity of PLC gamma 1, suggesting that its activation might play significant roles in this in vitro model for cellular senescence.

Chromosomal assignment in mouse of matrix Gla protein and bone Gla protein genes.

Matrix Gla protein (MGLAP) and bone Gla protein (BGLAP) are calcium-binding, vitamin K-dependent proteins produced by cells of the osteoblastic lineage. Sequence homology suggests that the genes for these proteins evolved from a common ancestor. Somatic whole cell hybrids and karyotypically simple microcell hybrids were used to map Mglap to mouse Chromosome 6 and Bglap to mouse Chromosome 3. Human MGLAP has previously been mapped to chromosome 12p, a region with homology to mouse Chromosome 6, and human BGLAP has been mapped to chromosome 1q, a region with homology to mouse Chromosome 3. It appears that BGLAP is the third calcium-binding protein that maps to human chromosome 1q and mouse Chromosome 3.

Mutation in the TP53 gene in colorectal carcinoma detected by polymerase chain reaction.

The human TP53 gene is a possible tumor suppressor since TP53 gene mutations are observed in greater than 70% of sporadic colorectal carcinoma DNAs. In genomic DNAs from seven colon cancer cell samples, a 405 base pair DNA fragment containing exon 5, intron 5, and exon 6 of the TP53 gene was amplified by polymerase chain reaction and analyzed for mutations. One sample [human colon cancer (HCC) 278] was found to have a TP53 mutation altering the amino acid glutamine 167 in exon 5. A deletion of 2 bases changed glutamine 167 (CAG) to alanine (GCA) and the resulting frame-shift produced an in-frame stop codon at amino acid 179. While the normal TP53 gene gives rise to a 53 kD protein, the estimated size of this mutant TP53 protein if expressed would be approximately 20 kD.

Loss of both CSF1R (FMS) alleles in patients with myelodysplasia and a chromosome 5 deletion.

A high proportion of patients with myelodysplasia show characteristic karyotypic abnormalities in bone marrow cells. The most distinctive of the myelodysplastic syndromes is the 5q- syndrome characterized by refractory anemia, poorly lobulated megakaryocytes, and an interstitial deletion of the long arm of chromosome 5 (5q deletion) as the sole karyotypic abnormality. Recently, several genes encoding hemopoietic growth factors and receptors, comprising the interleukins 3, 4, and 5, macrophage colony-stimulating factor, granulocyte/macrophage-colony-stimulating factor, and the receptor for macrophage-colony-stimulating factor [the CSF1R (formerly FMS) gene product], have been localized to the long arm of chromosome 5, and there has been much speculation that deletion of one or more of these genes may be critical to the pathogenesis of the associated myeloid disorders. One candidate gene is CSF1R, which is required for normal proliferation and differentiation of hemopoietic cells of the myeloid lineage. We have carried out a molecular examination of the CSF1R, both on the 5q- chromosome and on the apparently normal homologous chromosome 5, in 10 patients with myelodysplasia and a 5q deletion. We have found, using restriction fragment length polymorphism analysis and gene dosage experiments, that all 10 patients showed deletion of CSF1R; 6 of 10 were hemizygous and 4 of 10 homozygous for CSF1R loss. The homozygous CSF1R loss has been confirmed in 2 patients by an in situ hybridization technique comparing the signal in affected cells to that in control sex-mismatched cells on the same slides. In those patients considered to have homozygous CSF1R loss by DNA experiments the gene was deleted from the 5q chromosome in all cells and from the apparently normal chromosome 5 in a subset of cells. This loss of one CSF1R allele, together with loss in some cells of the remaining allele on the homologous chromosome 5, in patients with myelodysplasia indicates that this is a region of critical gene loss on 5q. The loss of the hemopoietic growth factor receptor gene CSF1R may be important in the pathogenesis of human myeloid leukemia.

Use of the single strand conformation polymorphism technique and PCR to detect p53 gene mutations in small cell lung cancer.

Recent studies have suggested that the p53 oncoprotein might function normally as a tumor suppressor. Mutations in highly conserved regions of the p53 gene have been observed in numerous types of tumors and tumor cell lines. To detect in a more sensitive manner p53 gene mutations in small cell lung cancer (SCLC) we utilized the single strand conformation polymorphism (SSCP) technique of Orita et al., (1989). Using PCR primers for the most highly conserved regions of the p53 gene, including exons 4-9, we have identified p53 mutations in 5 of 9 small cell lung cancer (SCLC) tumor DNA samples and in 1 SCLC cell line. None of the mutations seen in tumor DNA samples were present in normal DNA from the same patients, indicating that mutation of the p53 gene in these tumors was a somatic event. Of the six mutations observed, two were found in exon 7, three were found in the region encompassing exons 8 and 9, and one was found in the region encompassing exons 5 and 6. Nucleotide sequencing of one of the exon 7 mutations and one of the exon 8-9 mutations indicated that each was a C to T transition. In SCLC-6 the mutation resulted in substitution of serine for proline at amino acid 278 and in SCLC-4 substitution of tryptophan for arginine at amino acid 248, both nonconservative amino acid substitutions. Both of these changes are in regions of the p53 gene where mutations have been observed in other tumors. Two additional mutations were observed in SCLC cell lines using conventional PCR techniques. One of these is a mutation which results in altered splicing of the p53 pre-mRNA.

A mutational analysis of phosphatidylinositol-3-kinase activation by human colony-stimulating factor-1 receptor.

Colony-stimulating factor-1 (CSF1) is a cell lineage-specific hemopoietin required for the growth, differentiation, and survival of macrophages and their precursors. The human CSF1 receptor (CSF1R) is a 150-kDa transmembrane glycoprotein whose cytoplasmic tyrosine kinase domain is split by a kinase insert (KI) region of approximately 70 amino acids. We tested the ability of CSF1R KI domain deletion mutants to stimulate phosphatidylinositol-3-kinase (PtdIns-3-kinase), an enzyme whose activity is augmented by tyrosine kinase oncogenes and receptor tyrosine kinases, and to support mitogenesis in transfected cells. Receptor immunoprecipitates from CSF1-stimulated cells contained greater than 5-fold more PtdIns-3-kinase activity compared to nonstimulated cells. High performance liquid chromatography analysis of the PtdIns-3-kinase product scraped from thin layer chromatography plates indicated that PtdIns-3-P was produced. CSF1R KI domain deletion mutants retained tyrosine kinase activity in vitro. Receptor immunoprecipitates of two partially overlapping 28 and 30 amino acid KI deletion mutants of CSF1R retained some PtdIns-3-kinase activity, in contrast to immunoprecipitates of CSF1R lacking 67 amino acids of the KI domain. Each deletion mutant stimulated CSF1-dependent DNA synthesis in transfected cells at much reduced levels compared to wild-type receptor expressing cells. These data suggest a role for the CSF1R KI domain in PtdIns-3-kinase association and for CSF1-induced thymidine incorporation into DNA.

The kinase insert domain of colony stimulating factor-1 receptor is dispensable for CSF-1 induced phosphatidylcholine hydrolysis.

Mouse NIH 3T3 fibroblasts transfected with human colony stimulating factor-1 receptor produced diacylglycerol in response to CSF1 and this correlated with elevated phosphatidylcholine hydrolyzing activity measured in an in vitro assay. Treatment of cells with the isoflavone derivative genistein attenuated PC hydrolysis in vitro suggesting a role for CSF1R tyrosine kinase activity. A CSF1R mutant lacking 67 amino acids of the kinase insert domain, which may affect the association of receptor with certain substrates, stimulated PC hydrolysis in response to CSF1. Coupling to PC hydrolysis is likely a general property of CSF1R and the kinase insert domain is dispensable for this activity.

PCR primers for human chromosomes: reagents for the rapid analysis of somatic cell hybrids.

Rapid analysis of somatic cell hybrids can be facilitated by using the polymerase chain reaction (PCR) to assay for genes assigned to specific human chromosomes. We describe PCR primer pairs for genes on the short and long arms of the 22 autosomes and the X chromosome. Some of the primers were designed from the 3' untranslated region of cDNA sequences, whereas others were derived from genomic sequence. Each primer set was tested for its specificity and mapped to a chromosome by screening a somatic cell hybrid panel. Two of the primer pairs (APOC2 and G6PD) detect CA dinucleotide repeat polymorphisms.

Human colony stimulating factor-1 receptor activates the C-raf-1 proto-oncogene kinase.

The proto-oncogene c-raf-1 encodes a 74 kD serine/threonine kinase. Recently, it has been shown that Raf kinase activity is stimulated by platelet derived growth factor (PDGF) treatment of receptor bearing cells, and that p74 is a direct substrate for PDGF receptor. CSF-1 treatment of BeWo cells, a human choriocarcinoma cell line, and mouse NIH 3T3 cells expressing a transfected human CSF-1 receptor cDNA, was associated with a 3-4 fold increase in phosphorylation of a 74 kD protein immunoprecipitated with affinity purified Raf-1 antibody. The kinase activity of p74 was increased 2-3 fold against two exogenous substrates following CSF-1 treatment of the transfected cells. These observations suggest that Raf-1 protein is a downstream second messenger molecule in CSF-1 mediated signal transduction.