Functional localization of a melanoma tumor suppressor gene to a small (< or = 2 Mb) region on 11q23.

We have previously demonstrated the existence of a melanoma tumor suppressor gene(s) on the long arm of chromosome 11 through suppression of tumorigenicity assays. Although loss of heterozygosity studies also support this finding, only a large critical region (44 cM) has been identified to date on 11q22-25. To further localize a tumor suppressor gene(s) within this region, we have now generated and characterized nine melanoma microcell hybrids, each retaining an introduced fragment of 11q. Of the nine hybrids, four were suppressed for tumor formation in nude mice, while five formed tumors at the same rate as the parental melanoma cell line (UACC 903). Molecular analysis of the hybrids with 118 microsatellite markers narrowed the location of a putative suppressor gene to a small (< or =2 Mb) candidate region on 11q23 between the markers D11S1786 and D11S2077 and within the larger region frequently deleted in melanoma tumors and cell lines. While multiple tumor suppressor genes are likely to reside on 11q22-25, the presence of this region in all four suppressed hybrids supports the simplest model that a single locus is responsible for the suppressed phenotype observed in UACC 903.

In vitro loss of heterozygosity targets the PTEN/MMAC1 gene in melanoma.

Gross genetic lesions of chromosome 10 occur in 30-50% of sporadic human melanomas. To test the functional significance of this observation, we have developed an in vitro loss of heterozygosity approach in which a wild-type chromosome 10 was transferred into melanoma cells, where there was selection for its breakage and regional deletion to relieve its growth suppressive effects. The overlap of these events was at band 10q23, the site of the recently isolated PTEN/MMAC1 tumor suppressor gene, suggesting it as a potential target. Although the gene was expressed in the parental cells, both of its chromosomal alleles contained truncating mutations. In vitro loss of heterozygosity resulted in loss of the chromosomally introduced wild-type PTEN/MMAC1, and ectopic expression of the gene caused cell growth suppression. Thus, this approach identified PTEN/MMAC1 as a target in malignant melanoma and may provide an alternative means to localizing tumor suppressor genes.

Normal human melanocytes that express a bFGF transgene still require exogenous bFGF for growth in vitro.

The expression of basic fibroblast growth factor (bFGF) has been implicated as an important factor in the development of malignant melanoma. The timing of this expression suggests that bFGF plays a role early in melanoma tumor progression. Benign nevi produce bFGF, and cells cultured from these lesions show a loss of dependence on exogenous bFGF for growth. We have examined the effects of constitutive bFGF expression on the in vitro growth requirements of normal human melanocytes. bFGF was overexpressed in normal human epidermal melanocytes through genomic insertion of a human bFGF cDNA in a retroviral vector. These melanocytes produced the 18 kDa bFGF isoform as well as the higher molecular weight isoforms. The bFGF was not released into the culture medium, but it was present in the cell nucleus. The bFGF produced by these cells was mitogenic for 3T3 fibroblasts and therefore possessed functional activity; however, melanocytes producing bFGF had the same appearance and growth patterns as those infected with control virus or uninfected melanocytes. Expression of bFGF did not confer independence from the exogenous mitogen, nor would these cells form colonies in a soft-agar medium. These results indicate that expression of bFGF alone is not enough to cause aberrant growth of normal human melanocytes.

A panel of transferable fragments of human chromosome 11q.

Cytogenetic and molecular studies have implicated one or more tumor suppressor genes on the long arm of human chromosome 11 in the malignant progression of several human solid tumors, including malignant melanoma and carcinomas of the breast, cervix, ovary, and lung. Microcell-mediated chromosome transfer of an intact copy of chromosome 11 into tumor cell lines has provided additional evidence of tumor suppressor gene function in melanoma, breast cancer, and cervical cancer. However, sublocalization of the region(s) conferring the tumor suppressive effect has been difficult. To facilitate mapping of tumor suppressor gene(s) on chromosome 11, we have generated a panel of 25 mouse donor cell lines containing neo-tagged fragments of human chromosome 11q which can be transferred into cell lines to test for tumor suppressor activity. The chromosome fragments in these cell lines have been characterized by fluorescence in situ hybridization with probes to human DNA and to the centromere of chromosome 11, and also by analysis of microsatellite markers spanning chromosome 11. Finally, to demonstrate the usefulness of these cell lines as donors for microcell-mediated chromosome transfer, two fragments were transferred into the human melanoma cell line UACC 903. This panel of selectable subchromosomal fragments, derived from the long arm of human chromosome 11, will be useful for the regional localization of tumor suppressors and other genes by means of functional assays.

A malignant melanoma tumor suppressor on human chromosome 11.

Considerable molecular genetic and cytogenetic evidence indicates that chromosome 11 is a target for chromosome breakage, rearrangement, and loss during the development of human malignant melanomas. Abnormalities of the long arm of chromosome 11 are also evident in a wide variety of other human solid tumors, including carcinomas of the breast, ovary, cervix, and lung. In melanomas, these abnormalities tend to cluster in the lower half of the long arm of chromosome 11, indicating the possible presence of a melanoma tumor suppressor gene in this region. We tested this possibility by using microcell-mediated chromosome transfer to introduce normal copies of human chromosome 11 into two human malignant melanoma cell lines. In one cell line, MelJuSo, the presence of an additional copy of chromosome 11 severely reduced the ability of the cells to grow in culture. In a second cell line, UACC 903, there was a moderate reduction in cell growth in vitro, and the ability of the hybrid cells to form tumors in animals was suppressed. Suppression of tumorigenicity was even more strongly pronounced in a microcell hybrid that received an isochromosome 11q derived from the donor copy of chromosome 11. The formation of tumors was accompanied by a reduction in the copy number of chromosome 11. This provides functional evidence that a melanoma tumor suppressor resides on the long arm of chromosome 11. Thus, a third distinct locus, in addition to those previously defined on chromosomes 6 and 9, appears to play a role in the development of human malignant melanoma.

Cleavage arrest of early frog embryos by the G protein-activated protein kinase PAK I.

PAK I is a member of the PAK (p21-activated protein kinase) family and is activated by Cdc42 (Jakobi, R., Chen, C.-J., Tuazon, P. T., and Traugh, J. A. (1996) J. Biol. Chem. 271, 6206-6211). To examine the effects of PAK I on cleavage arrest, subfemtomole amounts of endogenously active (58 kDa) and inactive (60 kDa) PAK I and a tryptic peptide (37 kDa) containing the active catalytic domain were injected into one blastomere of 2-cell frog embryos. Active PAK I resulted in cleavage arrest in the injected blastomere at mitotic metaphase, whereas the uninjected blastomere progressed through mid- to late cleavage. Injection of other protein kinases at similar concentrations had no effect on cleavage. Endogenous PAK I was highly active in frog oocytes, and antibody to PAK I reacted specifically with protein of 58-60 kDa. PAK I protein was decreased at 60 min post-fertilization, with little or no PAK I protein or activity detectable at 80 min post-fertilization or in 2-cell embryos. At the 4-cell stage PAK I protein increased, but the protein kinase was present primarily as an inactive form. Rac2 and Cdc42, but not Rac 1, were identified in oocytes and throughout early embryo development. Thus, PAK I appears to be a potent cytostatic protein kinase involved in maintaining cells in a non-dividing state. PAK I activity is high in oocytes and appears to be regulated by degradation/synthesis and through autophosphorylation via binding of Cdc42. PAK I may act through regulation of the stress-activated protein kinase signaling pathway and/or by direct regulation of multiple metabolic pathways.

Mechanisms of human melanoma cell growth and tumor suppression by chromosome 6.

Breaks and deletions of chromosome 6 are among the most frequent karyotypic abnormalities that appear in human malignant melanoma cells, and chromosome transfer experiments have provided functional evidence for the presence of a melanoma tumor suppressor locus on chromosome 6[J.M. Trent et al., Science (Washington DC), 247: 568-571, 1990]. We have investigated the genetic mechanism of this suppression. We have found that the suppression of tumorigenicity that follows the introduction of a normal copy of chromosome 6 into the UACC 903 human melanoma cell line is correlated with increased chromosome 6 dosage, rather than with the presence of the transferred normal copy of the chromosome. Transfer of chromosome 6 into another human melanoma cell line, MelJuSo, does not result in suppression of primary tumor formation, although the additional copy of chromosome 6 does reduce the cell growth rate in vitro. Finally, we have identified a substantial portion of the chromosome that is evidently not involved in tumor suppression. We have observed that a copy of chromosome 6 derived from a normal cell but with a deletion involving chromosome bands 6q22-6q24 suppresses primary tumor formation in UACC 903 cells as effectively as the intact chromosome. We have thus provided additional information about the chromosomal location of this tumor suppressor and about its mode of action.

Metastasis suppressed, but tumorigenicity and local invasiveness unaffected, in the human melanoma cell line MelJuSo after introduction of human chromosomes 1 or 6.

Progression of human melanoma toward increasing malignant behavior is associated with several nonrandom chromosomal aberrations, most commonly involving chromosomes 1, 6, 7, 9, and 10. We previously showed that introduction of human chromosome 6 into the highly metastatic human malignant melanoma cell line C8161 completely suppressed metastasis without altering tumorigenicity (Welch DR, Chen P, Miele ME, et al., Oncogene 9:255-262, 1994). Alterations of chromosome 1 are the most frequent chromosome abnormality observed in melanomas, and they frequently arise late in tumor progression. The purpose of the study presented here was to compare the effects of chromosomes 1 and 6 on malignant melanoma metastasis. By using microcell-mediated chromosome transfer, single copies of neo-tagged human chromosomes 1 or 6 were introduced into the human melanoma cell line MelJuSo. The presence of the added chromosome was verified by G banding of karyotypes, fluorescence in situ hybridization, and screening for polymorphic markers on each chromosome. The incidence and number of metastases per lung after intravenous or intradermal injection of parental MelJuSo cells was significantly (P<0.01) greater than those of hybrids containing either chromosome 1 or chromosome 6, although chromosome 1 was a less potent inhibitor of metastasis than chromosome 6. Cultures established from primary tumors and metastases remained neomycin resistant, suggesting that portions of the added chromosomes were retained. These results strengthen the evidence for the presence of a melanoma metastasis suppressor gene on chromosome 6. neo6/MelJuSo hybrids expressed 2.4- to 3.4-fold more of the melanoma differentiation-associated gene mda-6 (previously shown to be identical to WAF1/CIP1/Sdi1/CAP20) than parental metastatic cells. mda-6/WAF1 is among the candidate genes on chromosome 6. These results also demonstrate, for the first time, the existence of metastasis suppressor genes on human chromosome 1, although these genes appear to be less potent than the one encoded on chromosome 6.

Growth control by tumor suppressors in malignant melanoma.

The development of malignant melanoma is accompanied by an accumulation of genetic damage that is evident within tumor cells at both cytogenetic and molecular levels, and mutations at several gene loci are thought to contribute to malignant progression. Some of these loci are known oncogenes and tumor suppressor genes; others remain to be identified, although their chromosomal locations have been determined. Gene mapping studies indicate the presence of melanoma tumor suppressor genes on chromosomes 1, 6, and 9. The presence of a tumor suppressor gene on a particular chromosome can be demonstrated by transfer of an intact, normal copy of the chromosome into tumor cells. We have used this approach to investigate the mechanisms by which chromosome 6 suppresses the growth and tumorigenicity of human malignant melanoma cells.

Functional analysis of the AUG- and CUG-initiated forms of the c-Myc protein.

Activation of the c-myc proto-oncogene by chromosomal translocation or proviral insertion frequently results in the separation of the c-myc coding region from its normal regulatory elements. Such rearrangements are often accompanied by loss or mutation of c-myc exon 1 sequences. These genetic alterations do not affect synthesis of the major c-myc protein, p64, which is initiated from the first AUG codon in exon 2. However they can result in mutation or loss of the CUG codon located in exon 1 that normally serves as an alternative translational initiation codon for synthesis of an N-terminally extended form of c-Myc (p67). It has been hypothesized that p67 is a functionally distinct form of c-Myc whose specific loss during c-myc rearrangements confers a selective growth advantage. Here we describe experiments designed to test the functional properties of the two c-Myc protein forms. We introduced mutations within the translational initiation codons of a normal human c-myc cDNA that alter the pattern of Myc protein synthesis (p64 vs. p67). The functions of each of these proteins were experimentally addressed using co-transformation and transcriptional activation assays. Both the p64 and p67 c-Myc proteins were independently able to collaborate with bcr-abl in the transformation of Rat-1 fibroblasts. In addition, both the exon 1- and exon 2-initiated forms of the c-Myc protein stimulated transcription of a Myc/Max-responsive reporter construct to a similar level. Given the apparent absence of functional differences between p64 and p67, we conclude that the basis for c-Myc oncogenic activation lies primarily in the overall deregulation of its expression and not in alterations in the protein. The existence of the CUG translational initiator may reflect a mechanism for the continued synthesis of c-Myc protein under conditions where AUG initiation is inhibited.

Distinct deletions of chromosome 9p associated with melanoma versus glioma, lung cancer, and leukemia.

Deletions of DNA on chromosome 9p21-22 are frequently observed in cells derived from melanomas, gliomas, non-small cell lung cancers, and acute lymphoblastic leukemia. The minimal deletion shared by the latter three cancers extends from the interferon-alpha locus towards the centromere; its centromeric end is flanked by the gene encoding methylthioadenosine phosphorylase. We have determined that the telomeric end of the minimal homozygous deletion shared by two melanoma cell lines does not include the methylthioadenosine phosphorylase locus. Thus, a distinct region of DNA is lost in melanoma. The physical size of this region remains to be defined precisely, but it may extend over several million base pairs.

Regulation of chimeric phosphoenolpyruvate carboxykinase genes by the trans-dominant locus TSE1.

Extinction of phosphoenolpyruvate carboxykinase (PCK) gene expression in hepatoma x fibroblast hybrids is mediated by a trans-acting genetic locus designated tissue-specific extinguisher 1 (TSE1). To identify PCK gene sequences required for extinction, hepatoma transfectants expressing PCK-thymidine kinase (TK) chimeric genes were fused with TK- fibroblasts and PCK-TK expression in the resulting hybrids was monitored. Expression of a PCK-TK chimera containing PCK sequences between base pairs -548 and +73 was extinguished in four of five hepatoma transfectants tested, although hybrids derived from one transfectant clone failed to extinguish PCK-TK expression. In contrast, crosses between hepatoma transfectants expressing the herpesvirus TK gene from its own promoter and TK- fibroblasts produced TK+ hybrids; extinction of the transfected TK gene was not observed. Thus, rat PCK gene sequences between base pairs -548 and +73 are sufficient for tissue-specific extinction in hybrid cells. Extinction of PCK-TK gene expression in transfectant microcell hybrids mapped specifically to human chromosome 17, the site of human TSE1.

Tyrosine kinase activity and transformation potency of bcr-abl oncogene products.

Oncogenic activation of the proto-oncogene c-abl in human leukemias occurs as a result of the addition of exons from the gene bcr and truncation of the first abl exon. Analysis of tyrosine kinase activity and quantitative measurement of transformation potency in a single-step assay indicate that variation in bcr exon contribution results in a functional difference between p210bcr-abl and p185bcr-abl proteins. Thus, foreign upstream sequences are important in the deregulation of the kinase activity of the abl product, and the extent of deregulation correlates with the pathological effects of the bcr-abl proteins.

The BCR-ABL oncogene transforms Rat-1 cells and cooperates with v-myc.

The tyrosine kinase P210 is the gene product of the rearranged BCR-ABL locus on the Philadelphia chromosome (Ph1), which is found in leukemic cells of patients with chronic myelogenous leukemia. It has a weakly oncogenic effect in immature murine hematopoietic cells and does not transform NIH 3T3 cells. We have found that P210 has a strikingly different effect in Rat-1 cells, another line of established rodent fibroblasts. Stable expression of P210 in Rat-1 cells caused a distinct morphological change and conferred both tumorigenicity and capacity for anchorage-independent growth. The introduction of v-myc into Rat-1 cells expressing P210 led to complete morphological transformation and enhanced tumorigenicity. No such interaction took place in NIH 3T3 cells. Thus, Rat-1 cells can be used to detect cooperation between BCR-ABL and other oncogenes and may prove useful for the identification of secondary oncogenic events in chronic myelogenous leukemia.

Chromosomal localization of genes encoding guanine nucleotide-binding protein subunits in mouse and human.

A variety of genes have been identified that specify the synthesis of the components of guanine nucleotide-binding proteins (G proteins). Eight different guanine nucleotide-binding alpha-subunit proteins, two different beta subunits, and one gamma subunit have been described. Hybridization of cDNA clones with DNA from human-mouse somatic cell hybrids was used to assign many of these genes to human chromosomes. The retinal-specific transducin subunit genes GNAT1 and GNAT2 were on chromosomes 3 and 1; GNAI1, GNAI2, and GNAI3 were assigned to chromosomes 7, 3, and 1, respectively; GNAZ and GNAS were found on chromosomes 22 and 20. The beta subunits were also assigned--GNB1 to chromosome 1 and GNB2 to chromosome 7. Restriction fragment length polymorphisms were used to map the homologues of some of these genes in the mouse. GNAT1 and GNAI2 were found to map adjacent to each other on mouse chromosome 9 and GNAT2 was mapped on chromosome 17. The mouse GNB1 gene was assigned to chromosome 19. These mapping assignments will be useful in defining the extent of the G alpha gene family and may help in attempts to correlate specific genetic diseases with genes corresponding to G proteins.

Chromosomal localization of the mouse homolog of the Huntington's disease linked G8 (D4S10) marker.

The human DNA probe G8 defines D4S10, a polymorphic locus tightly linked to the Huntington's disease gene on human chromosome 4. A subclone of G8, pSC33, showed significant cross-hybridization to discrete restriction fragments in total genomic mouse DNA. The probe detected restriction fragment length polymorphisms (RFLPs) with the enzymes Taq I and Msp I, permitting chromosomal localization of the mouse G8 homolog by linkage analysis using three sets of recombinant inbred mouse strains: BXH, BXD, and AKXD. The mouse locus was mapped to the central region of chromosome 11 at 1 centiMorgan from the SPARC gene, a locus whose human counterpart is on human chromosome 5.

Isolation of microcell hybrid clones containing retroviral vector insertions into specific human chromosomes.

We sought an efficient means to introduce specific human chromosomes into stable interspecific hybrid cells for applications in gene mapping and studies of gene regulation. A defective amphotropic retrovirus was used to insert the gene conferring G418 resistance (neo), a dominant selectable marker, into the chromosomes of diploid human fibroblasts, and the marked chromosomes were transferred to mouse recipient cells by microcell fusion. We recovered five microcell hybrid clones containing one or two intact human chromosomes which were identified by karyotype and marker analysis. Integration of the neo gene into a specific human chromosome in four hybrid clones was confirmed by segregation analysis or by in situ hybridization. We recovered four different human chromosomes into which the G418 resistance gene had integrated: human chromosomes 11, 14, 20, and 21. The high efficiency of retroviral vector transformation makes it possible to insert selectable markers into any mammalian chromosomes of interest.

Parasexual approaches to the study of human genetic disease.

We have used two different strategies to construct hybrid cells in which specific, individual human chromosomes or fragments thereof are maintained by direct selective pressure. Our first approach was to introduce a drug-resistance gene into human chromosomes using a retroviral vector, and to transfer the marked chromosomes via microcells into mouse cells. The second method was to fuse gamma-irradiated human cells with rodent cells to produce hybrids containing fragments of the human X chromosome. Such hybrid cell lines should greatly facilitate both human gene mapping and the isolation of human genes by molecular cloning. The gene-transfer technologies described here can also be used to construct cell lines in which the expression of genes involved in human diseases can be studied in vitro.

Chromosome-mediated transfer of murine alleles for hypoxanthine-guanine phosphoribosyl transferase (HGPRT) and ouabain resistance into human cell lines.

Genetic drug-resistance markers were transferred via purified metaphase chromosomes from mouse L cells into the human fibrosarcoma line HT1080 and HeLa S3 cells. Interspecific chromosome-mediated transfer of hypoxanthine-guanine phosphoribosyl transferase (HGPRT; EC 2.4.2.8) from mouse L cells into HGPRT- HT1080 cells occurred at a frequency of approximately 1 x 10(-7). The presence of the mouse allele for HGPRT in transferent isolates was confirmed by isoelectric focusing. Transfer of ouabain resistance from mouse L cells to HT1080 and HeLa S3 cells occurred at an average frequency of approximately 4 x 10(-7). Expression of the mouse trait in transferent isolates was confirmed by their ability to withstand doses of ouabain which would be lethal to spontaneous ouabain-resistant mutants of the human cells but not to mouse L cells, ouabain-resistant transferents of human cells showed 10(4)-to greater than 10(5)-fold enhanced drug resistance, characteristic of either wild-type or mutant alleles, respectively, from ouabain-resistant donor L cells. Unstable expression of the transferred phenotypes in the absence of selection was seen in some isolates, but expression was lost at slow rates.

Identification of a cAMP regulatory region in the gene for rat cytosolic phosphoenolpyruvate carboxykinase (GTP). Use of chimeric genes transfected into hepatoma cells.

cAMP stimulates the transcription of the gene for the cytosolic form of phosphoenolpyruvate carboxykinase (GTP) (EC 4.1.1.32) (PEPCK) in rat liver. We have investigated the nucleotide sequences required for regulation of PEPCK gene expression by cAMP. A chimeric gene was constructed in which a 620-base pair fragment of the 5'-end of the PEPCK gene (including 547 base pairs of 5'-flanking sequence) was ligated to the herpes simplex virus thymidine kinase (TK) structural gene. The PEPCK promoter fragment was introduced either in the proper orientation for transcription of the TK gene or in the opposite orientation. These fusion genes and the parent vector, pOPF, which contains the intact TK gene, were transfected individually into TK-deficient FTO-2B rat hepatoma cells. FTO-2B cells contain an active endogenous PEPCK gene which is stimulated by cAMP. Cells were selected in HAT medium and grown either as mass cell cultures or as individual clones. Dibutyryl cyclic AMP (Bt2cAMP) plus theophylline (16 h) stimulated TK activity 1.6-6.1-fold in cell lines transfected with the PEPCK-TK fusion gene containing the PEPCK promoter fragment in the correct orientation. However, the intact TK gene was not induced by Bt2cAMP after transfection, nor was there any expression of the PEPCK-TK fusion gene in cells which contained the PEPCK promoter fragment in the wrong transcriptional orientation. Bt2cAMP also increased the levels of TK mRNA in cells transfected with the PEPCK-TK fusion gene, but not in cells transfected with the intact TK gene. The chimeric PEPCK-TK mRNA initiated at the PEPCK start site, as determined by S1 nuclease mapping. There was no relationship between the number of copies of the PEPCK-TK gene integrated in the various cell lines and either the basal level of TK activity or its inducibility of Bt2cAMP.