Chromosome 18 suppresses the tumorigenicity of prostate cancer cells

Microcell-mediated chromosome transfer allows for the introduction of normal chromosomes into tumor cells in an effort to identify putative tumor suppressor genes. We have used this approach to introduce an intact copy of chromosome 18 into the prostate cancer cell line DU145, and independently to introduce human chromosomes 8 and 18 into the prostate cancer cell line TSU-PR1. Introduction of an extra copy of human chromosome 8 had no effect on the growth properties in vitro or the tumorigenicity in vivo of TSU-PR1 cells. However, microcell hybrids containing an introduced copy of human chromosome 18 exhibited a longer population doubling time, retarded growth in soft agar, and slowed tumor growth in athymic nude mice. These experiments provide functional evidence for the presence of one or more tumor suppressor genes on human chromosome 18 that are involved in prostate cancer.     

肝癌细胞微细胞介导染色体转移方法学的建立与探讨

目的建立肝癌细胞微细胞介导染色体转移方法，为肝癌转移抑制基因的染色体功能定位建立技术平台。方法人单染色体供体细胞通过微核化、出核、融合步骤将随机标记有耐药neo基因的正常人8号染色体导入到大鼠肝癌高转移细胞系C5F中，对微细胞杂交克隆进行药物筛选和单细胞克隆，并填序列标签位点-PCR和全染色体涂染荧光原位杂交方法验证人染色体转移的结果。结果获得具有G418和HAT双重抗性的微细胞杂交细胞，通过单细胞分离克隆方法获得15个具有双重抗性的微细胞杂交克隆，序列标签位点-PCR结果发现导入染色体的随机片段丢失，全染色体涂染荧光原位杂交结果发现导入的人8号染色体与大鼠染色体发生了稳定的重组。结论成功建立微细胞介导的染色体转移技术，为肝癌转移抑制基因的染色体功能定位奠定了技术基础。     

Evidence that wild-type TP53, and not genes on either chromosome 1 or 11, controls the tumorigenic phenotype of the human fibrosarcoma HT1080

The specific transfer of normal chromosomes via microcell fusion has been instrumental in identifying putative tumor suppressor gene loci in a variety of human cancers. Using this same technique it has been proposed that the tumorigenicity of the human fibrosarcoma cell line HT1080 is controlled by functionally distinct tumor suppressor genes on human chromosomes 1 and 11. To address these results and perhaps further localize the suppressive effect to particular regions on these two chromosomes, we transferred into HTI080 seven different fibroblast-derived human chromosomes containing either intact or discrete portions of chromosome I or II. Interestingly, we found no evidence of genes on these chromosomes that could alter the growth of HT1080 either in vitro or in vivo. Based on these results we were left with the possibility that a gene, or genes, residing on an entirely different chromosome(s) was involved in the tumorigenesis of HTI080. Since TP53 mutation has been documented in a variety of human tumor types, and we found both copies of TP53 to be mutated in HTI080, we were prompted to examine its role by both cDNA transfection and chromosome transfer. Although by cDNA transfection we found that expression of exogenous wild-type TP53 was incompatible with continued proliferation of HTI080 cells in vitro, chromosome 17 transfer studies revealed that a more physiologic expression of exogenous wild-type TP53 could be tolerated in vitro while being completely incompatible with growth in vivo. These studies demonstrate a differential effect of TP53 growth inhibition and clearly show that TP53 tumor suppressing function can be independent from its potent growth suppressing effect in vitro. Genes Chrom Cancer 9:266-281 (1994). © 1994 Wiley-Liss, Inc.     

A functional investigation of tumor suppressor gene activities in a nasopharyngeal carcinoma cell line HONE1 using a monochromosome transfer approach

Monochromosome transfers of selected chromosomes into a nasopharyngeal carcinoma (NPC) cell line were performed to determine if tumor suppressing activity for NPC mapped to chromosomes 9, 11, and 17. Current information from cytogenetic and molecular allelotyping studies indicate that these chromosomes may harbor potential tumor suppressor genes vital to NPC. The present results show the importance of CDKN2A on chromosome 9 in NPC development. There was no functional suppression of tumor development in nude mice with microcell hybrids harboring the newly transferred chromosome 9 containing an interstitial deletion at 9p21, whereas transfection of CDKN2A into the NPC HONE1 cells resulted in obvious growth suppression. Whereas intact chromosome 17 transfers into HONE1 cells showed no functional suppression of tumor formation, chromosome 11 was able to do so. Molecular analysis of chromosome 11 tumor segregants indicated that at least two tumor suppressive regions mapping to 11q13 and 11q22-23 may be critical for the development of NPC.     

Mapping of metastasis suppressor gene(s) for rat prostate cancer on the short arm of human chromosome 8 by irradiated microcell-mediated chromosome transfer

Our previous studies demonstrated that human chromosome 8 contains metastasis suppressor gene(s) for rat prostate cancer. However, it is still unknown which portion of human chromosome 8 is associated with suppression of metastatic ability, because all of the clones in which metastatic ability is suppressed contain at least one copy of intact human chromosome 8. In the present study, we used the irradiated microcell-mediated chromosome transfer technique to enrich for specific chromosomal arm deletions of selected chromosomes. The resultant series of human chromosomes 8 with a variety of chromosomal deletions was introduced into highly metastatic Dunning rat prostate cancer cells. All of the resultant microcell hybrids showed reduced metastatic ability. To obtain a smaller size of human chromosome 8 and to locate further the region of metastasis suppressor gene(s), the most reduced size of human chromosome 8 that was generated with the initial irradiated chromosome transfer was retransferred into the Dunning cancer cells without irradiation. The resultant microcell hybrids were analyzed to determine which portion of human chromosome 8 suppressed the metastatic ability of the recipient cells. This analysis demonstrates that the portion of human chromosome 8 containing metastasis suppressor gene(s) for rat prostate cancer cells lies on human chromosome segment 8p21-p12, where frequent allelic losses have been detected in allelotype analyses of human prostate cancer. This suggests that one of the metastasis suppressor genes for rat prostate cancer on human chromosome 8 may also play an important role in the progression of human prostate cancer. Genes Chromosom Cancer 17:260–268 (1996). © 1996 Wiley-Liss, Inc.     

Monochromosome transfers to syrian hamster BHK cells via microcell fusion provide functional evidence for suppressor genes on human chromosome 9 both for anchorage independence and for tumorigenicity

We previously identified an anchorage independence-suppressor gene, SAII, on rat chromosome (RNO) 5. RNO5 is homologous to human chromosomes (HSA) I and 9. In order to find the human homolog of the SAII gene, we transferred HSAI and HSA9 to an anchorage-independent and tumorigenic Syrian hamster BHK 191-5C cell line by microcell fusion. For HSA9, we used a t(X; 9)-derivative chromosome to force the retention of this chromosome in hybrids by hypoxanthine-aminopterin-thymidine (HAT) selection. To study the possible effect of the X portion of the der(9)t(X; 9), we also transferred a normal X to 191-5C cells. For HSAI, a neo-tagged chromosome was introduced. Following the transfer of der(9)t(X; 9) to 191 -5C cells, the hybrid cells became anchorage dependent and nontumorigenic, and, upon the loss of this chromosome, the cells regained their tumorigenic and anchorage-independent phenotypes. The transfer of HSAX or HSAI, on the other hand, affected neither of these phenotypes. These results provide functional proof of suppressor genes on HSA9 involving both anchorage independence and tumorigenicity. In addition, our data suggest the presence of another gene on HSA9 that causes a negative growth effect and whose phenotypic expression, contrary to the suppressor genes, is dosage dependent. © 1995 Wiley-Liss, Inc.     

Suppression of tumorigenicity in human ovarian carcinoma cell line SKOV-3 by microcell-mediated transfer of chromosome 11

To determine the pathogenic role of chromosomes 11 and 17 in the carcinogenesis of human ovarian cancers, neo(R)-tagged chromosome 11 or 17 was transferred from cell lines A9H11 or A9H17, respectively, into the ovarian carcinoma cell line SKOV-3 using microcell-mediated chromosome transfer. The chromosome transfer was verified by polymerase chain reaction detection of the neo(R) gene, fluorescence in situ hybridization detection of an extra chromosome 11, and microsatellite polymorphism detection of an exogeneous chromosome 11. Five SKOV-3/A9H11 hybrids and five SKOV-3/A9H17 hybrid clones were generated. For the chromosome 11 transfer, complete suppression of tumorigenicity was observed in four clones, (11)9-8 and 11(H)7-2, 11(H)8-3, and 11(H)7-2, 100 days post implantation. For the chromosome 17 transfer, no complete suppression of tumorigenicity was observed. However, an increased latency period ranging from 25 to 49 days in contrast to 7 days for the SKOV-3 parental line, and a significant reduction in tumor size was observed. There was no correlation between the in vitro growth rate and the tumorigenicity or length of latency period. Our results demonstrate functionally that chromosome 11 may carry a tumor suppressor gene(s) while chromosome 17 may carry a tumor growth-inhibitor gene(s) for the ovarian carcinoma cell line, SKOV-3.     

Integration of a dominant selectable marker into human chromosomes and transfer of marked chromosomes to mouse cells by microcell fusion

A method for the production of stable mouse-human cell hybrids containing a single human chromosome is described. As a first step in this method, a cloned selectable marker, the E. coli xanthine-guanine phosphoribosyltransferase (Ecogpt) gene, was transferred to human cells to generate cell lines each carrying Ecogpt integrated into a different site. Human chromosomes marked with Ecogpt were transferred further into mouse cells by microcell fusion. Monochromosomal hybrids, in which the human chromosome is maintained by selection, have been produced for chromosomes 2, 5, 16, and a rearranged chromosome involving a translocation between chromosomes 1 and 2. In addition to these monochromosomal hybrids, we have also obtained monochromosomal hybrids for human chromosomes 6, 12, and 17 by selection for the loss of marked chromosome from the microcell hybrids each containing two human chromosomes. Although the human chromosome present in these hybrids cannot be maintained by selection, 80–90% of cells retained the transferred chromosome on continuous growth for 15 days. Monochromosomal hybrids would provide biological materials to construct genetic maps of human chromosomes. In addition, chromosomes marked with dominant selectable markers can be transferred further to any cell line of interest in inter- or intra-species combination.     

Normal human chromosome 2 induces cellular senescence in the human cervical carcinoma cell line SiHa

For identification of the chromosome carrying cellular senescence-inducing activity, normal human chromosome 2, 3, 6, 7, 9, 11, or 12 tagged with a selectable marker gene (neo) was introduced into the human cervical carcinoma cell line SiHa via microcell-mediated chromosome transfer. Seventy-six percent (158/207) of the G418-resistant clones obtained by the transfer of chromosome 2 showed a remarkable change in morphology (cells were flat), and 93% (147/158) of them ceased to divide (senesced) prior to 6–9 population doublings, whereas most of the clones generated by the transfer of other chromosomes exhibited a morphology similar to that of the parental cells and continued to grow. Chromosome analyses suggested that cells which escaped from senescence contained only a small fragment derived from the transferred chromosome 2, whereas the transferred chromosomes were apparently intact in most of the continuously growing microcell hybrids with introduction of other chromosomes. These results indicate that the normal human chromosome 2 carries a gene or genes that induce cellular senescence in SiHa cells.     

Regional mapping of suppressor loci for anchorage independence and tumorigenicity on human chromosome 9

By microcell-mediated chromosome transfer to the malignant Syrian hamster cell line BHK-191-5C, we previously identified two suppressor functions on human chromosome 9 (HSA9), one for anchorage independence and another for tumorigenicity. However, the precise chromosomal locations of these suppressor functions were not determined. The present study was undertaken to define the regional location of these suppressor loci using a panel of microcell hybrids containing structurally altered HSA9 with different deleted regions in the BHK-191-5C background. DNA derived from the cell hybrids was analyzed by PCR for verification of the presence of HSA9 genetic material by amplifying 62 microsatellite markers and 13 genes, covering the entire length of HSA9. Our deletion mapping data on anchorage independent and tumorigenic hybrids suggest that the suppressor function for anchorage independence is located in the region between 9q32 to 9qter. The suppressor for tumorigenicity may be located in one of three deleted regions on HSA9, the first one between the markers D9S162 and D9S1870, the second one between the markers D9S1868 and TIGRA002I21, and the third one between the markers D9S59 and D9S155.     

Human chromosome 5 carries a putative telomerase repressor gene

Telomerase, the ribonucleoprotein enzyme that maintains the telomere, is active in human germ and stem cells and in a majority of tumor tissues and immortalized cell lines. In contrast, telomerase activity is not detected in most somatic cells, suggesting that normal human cells contain a regulatory factor(s) to repress this activity. To identify which human chromosomes carry a gene or genes that function as telomerase repressors, we investigated telomerase activity in hybrids of the B16-F10 cell line, which contain individual human chromosomes transferred previously by microcell fusion and therefore represent a hybrid panel for the entire genome except for the Y chromosome. Microcell hybrids with an introduced normal human chromosome 5 showed inhibition of telomerase activity, but clones at a late passage exhibited reactivation of telomerase activity. Reactivation of telomerase activity was accompanied by deletion and/or rearrangement of the transferred human chromosome 5. The introduction of other human chromosomes did not significantly affect the telomerase activity of B16-F10 cells. The effect of suppression of telomerase activity in microcell hybrids containing chromosome 5 was accompanied by a reduction in the level of mTERT mRNA, which encodes a component of the telomerase complex. The putative telomerase repressor gene was mapped to human chromosome bands 5p11-p13 by a combination of functional analysis using transfer of subchromosomal transferable fragments of chromosome 5 into B16-F10 cells and deletion mapping of revertant clones with reactivated telomerase activity. Thus, these results suggest that loss of a gene(s) on this chromosome was responsible for telomerase reactivation, indicating that human chromosome 5 contains a gene or genes that can regulate the expression of mTERT in B16-F10 cells.     

Genetic background of different cancer cell lines influences the gene set involved in chromosome 8 mediated breast tumor suppression

Several lines of evidence suggest that chromosome 8 is likely to harbor tumor-suppressor genes involved in breast cancer. We showed previously that microcell-mediated transfer of human chromosome 8 into breast cancer cell line MDA-MB-231 resulted in reversion of these cells to tumorigenicity and was accompanied by changes in the expression of a breast cancer-relevant gene set. In the present study, we demonstrated that transfer of human chromosome 8 into another breast cancer cell line, CAL51, strongly reduced the tumorigenic potential of these cells. Loss of the transferred chromosome 8 resulted in reappearance of the CAL51 phenotype. Microarray analysis identified 78 probe sets differentially expressed in the hybrids compared with in the CAL51 and the rerevertant cells. This signature was also reflected in a panel of breast tumors, lymph nodes, and distant metastases and was correlated with several prognostic markers including tumor size, grading, metastatic behavior, and estrogen receptor status. The expression patterns of seven genes highly expressed in the hybrids but down-regulated in the tumors and metastases (MYH11, CRYAB, C11ORF8, PDGFRL, PLAGL1, SH3BP5, and KIAA1026) were confirmed by RT-PCR and tissue microarray analyses. Unlike with the corresponding nontumorigenic phenotypes demonstrated for the MDA-MB-231- and CAL51-derived microcell hybrids, the respective differentially expressed genes strongly differed. However, the majority of genes in both gene sets could be integrated into a similar spectrum of biological processes and pathways, suggesting that alterations in gene expression are manifested at the level of functions and pathways rather than in individual genes.     

A hamster-human subchromosomal hybrid cell panel for chromosome 2

We have constructed hamster-human hybrid cell lines containing fragments of human chromosome 2 as their only source of human DNA. Microcell-mediated chromosome transfer was used to transfer human chromosome 2 from a monochromosomal mouse-human hybrid line to a radiation-sensitive hamster mutant (XR-V15B) defective in double-strand break rejoining. The human chromosome 2 carried theEcogpt gene and hybrids were selected using this marker. The transferred human chromosome was frequently broken, and the resulting microcell hybrids contained different sized segments of the q arm of chromosome 2. Two microcell hybrids were irradiated and fused to XR-V15B to generate additional hybrids bearing reduced amounts of human DNA. All hybrids were analyzed by PCR using primers specific for 27 human genes located on chromosome 2. From these data we have localized the integratedgpt gene on the human chromosome 2 to the region q36–37 and present a gene order for chromosome 2 markers.     

Metastasis suppressor gene(s) for rat prostate cancer on the long arm of human chromosome 7

Allelotype analyses of human prostate cancer indicate that allelic losses on human chromosome arms 7q, 8p, 10q, 13q, 16q, 17q, and 18q are observed frequently. For the study of the possible biological significance of the frequently observed deletions on chromosome arm 7q in human prostate cancer, human chromosome 7 was introduced into highly metastatic rat prostate cancer cells by use of a microcell‐mediated chromosome transfer technique. The introduction of human chromosome 7 resulted in the suppression of metastatic ability of the microcell hybrids, whereas no suppression of tumorigenicity was observed. To identify the portion of chromosome 7 containing the metastasis‐suppressive function gene, the derivative chromosome 7 that was generated with the initial transfer was retransferred into rat prostate cancer cells. Human chromosome 7‐containing rat prostate cancer cells could be used as the donor cells, because rodent cells produced a sufficient number of microcells with colchicine treatment. Cytogenetic and molecular analyses of these clones demonstrated that loss of segments on 7q was related to the reexpression of the metastatic phenotype. These results show that human 7q contains a metastasis suppressor gene or genes for rat prostate cancer. The findings also suggest that this gene may play an important role in the progression of human prostate cancer. Genes Chromosomes Cancer 24:1–8, 1999. © 1999 Wiley‐Liss, Inc.     

Cellular senescence of a human bladder carcinoma cell line (JTC-32) induced by a normal chromosome 11

Human chromosome 11 is expected to carry tumor suppressor genes for a variety of human cancers, including bladder carcinoma. To examine the functional role of a putative tumor suppressor gene(s) on this chromosome in the development of bladder carcinoma, we performed microcell-mediated transfer of chromosome 11 into the bladder carcinoma cell line, JTC-32. Fifteen of 20 colonies formed by the transfer experiment showed a remarkable change in cell morphology. They flattened and ceased growing, or senesced, prior to 10 population doublings. The presence of transferred chromosome 11-derived fragments in the growth-arrested cells was confirmed by PCR-based polymorphism analyses. The remaining 5 microcell hybrid clones exhibited a parental cell-like morphology, and presumably escaped from senescence, which was accompanied by deletions and/or rearrangements of the transferred chromosome 11. On the other hand, a transferred normal chromosome 7 neither changed the cell morphology nor arrested the cell growth. These results support the hypothesis that chromosome 11 contains a gene or genes which restore the senescence program lost during the immortalization process of JTC-32 cells.     

Chromosome 13 transfer provides evidence for regulation of RBI protein expression

The human retinoblastoma susceptibility gene (RBI) located on chromosome 13 has been shown to function as a growth/tumor suppressor gene in a large number of human cancers. Although constitutive expression has been observed in most cultured cells and normal tissues, overexpression of RBI protein has not been well documented. Perhaps regulating the level of normal RBI protein expression is one of several ways of controlling its function. To test this hypothesis we transferred normal copies of chromosome 13 via microcell fusion into the human fibrosarcoma cell line HT1080. Microcell hybrids were generated that contained one, two, or three extra copies of the transferred fibroblast chromosome 13. Compared to the parental cell line, the hybrids were completely unaltered with respect to several properties in vitro and in vivo, including morphology, growth rate, and tumor formation. Northern blot analysis revealed a stepwise increase in RBI mRNA expression which increased in proportion to the number of alleles present in each cell line. Although RBI protein exhibited correct nuclear localization and was phosphorylated in a normal cell cycle-dependent manner in the hybrids, the increased level of protein expression in each hybrid was nearly identical and did not increase beyond a threshold amount, although mRNA expression continued to increase. These results demonstrate that HT1080 cells can tolerate an increased level of RBI protein, but that expression beyond a certain level may be down-regulated. These transfer studies provide evidence for regulation of RBI protein expression and may suggest an alternative form of monitoring and controlling normal RBI functioning. Genes Chrom Cancer 9:251-260 (1994). © 1994 Wiley-Liss, Inc.     

Human chromosome II inhibits tumorigenicity of a murine squamous cell carcinoma cell line

Loss of heterozygosity (LOH) of mouse chromosome 7 has been consistently demonstrated in chemically induced murine squamous cell carcinomas (SCCs). The region of this chromosome presenting LOH in the mouse tumors is syntenic to human chromosome segments II p I5 and II q. To determine whether the introduction of human chromosome (Hchr) II can suppress the growth of murine SCC, we injected four clones of a chemically induced murine SCC cell line bearing an Hchr II into athymic BALB/c nude mice. All microcell hybrid clones with Hchr II (CH721Hchr II) had latency periods twice as long as those of the parental CH72 cells and control hybrids containing a Hchr 12. Tumor-derived cells from CH721Hchr II hybrids had lost centromeric and telomeric sequences from Hchr II. All repressed cell lines grew significantly m e slowly in vitro than did the controls. These results suggest that Hchr II contains a tumor-suppressor gene capable of inhibiting tumorigenicity in chemically induced SCC, confirming common pathways in the development of human neoplasias and the murine model. © 1995 Wiley-Liss, Inc.     

Caspase cascade of Fas-mediated apoptosis in human normal endometrium and endometrial carcinoma cells

Human endometrial epithelial cells undergo apoptosis immediately before the menstrual period. Apoptotic signalling was analysed using human endometrial tissue and a human endometrial carcinoma cell line (HHUA). Activity levels of caspase-3, -8, and -9 were elevated in human endometrium during the late secretory phase and in HHUA cells incubated with an anti-Fas monoclonal antibody (mAb). Fas-mediated apoptosis of HHUA cells was blocked by prior exposure to inhibitors of caspase-9, -8 and -3. In HHUA cells treated with anti-Fas mAb, a release of cytochrome c was detected in the cytosolic fraction, in addition a full-length Bid was degraded. Full-length FLIP(L) (p55) was degraded during apoptosis, and p29 (regarded as the product of p55 cleavage) appeared instead of FLIP(L). In normal human endometrial tissue, Bid degradation was also observed in a cyclic manner with a peak during the early secretory phase of the menstrual cycle. Furthermore, the release of cytochrome c was seen in the early secretory phase. However, expression of FLIP(S) was only observed during the menstrual cycle in normal endometrial tissue. We concluded that the main apoptotic signalling in both normal human endometrial tissue and HHUA cells exposed to anti-Fas mAb is the mitochondrial pathway via Bid degradation. Although the function of FLIP is still unknown on normal endometrial tissue, it may be regulated by FLIP(L) expression on HHUA cells derived from human endometrial carcinoma.     

Tissue-specific extinguisher loci in the murine genome: A screening study based on a rat/mouse microcell hybrid panel

Extinction of tissue-specific traits in intertypic somatic cell hybrids is a well-known phenomenon. In the past few years, microcell hybrids have been used in attempts to dissect this phenotype genetically, and tissue-specific extinguisher loci have been mapped to two different mouse chromosomes. When transferred from fibroblasts into hepatoma cells by microcell fusion, these loci down-regulate expression of specific liver genes intrans. However, other liver genes that are extinguished in genotypically complete hybrids seem not to be extinguished in monochromosomal hybrids. To assess the generality of monochromosomal extinction phenotypes, we assembled a collection of rat hepatoma/mouse fibroblast microcell hybrids that represent most of the mouse chromosome complement, and we screened them for expression of a large number of liver-specific genes. Phosphoenolpyruvate carboxykinase gene expression was down-regulated in hybrids containing mouse chromosome 7 or mouse chromosome 11, but other extinction phenotypes were not readily apparent. These results indicate that extinction of many liver genes may be a polygenic trait.