[Frontiers in Bioscience 3, d944-960, September 1, 1998]

Table of Conents
 Previous Section   Next Section


Jose Russo, Xiaoqi Yang, Yun-Fu Hu, Betsy A. Bove, Yajue Huang, Ismael D.C.G. Silva, Quivo Tahin, Yuli Wu, Nadia Higgy, Abdel Zekri, and Irma H. Russo

Breast Cancer Research Laboratory, Fox Chase Cancer Center, Philadelphia, Pennsylvania, USA

Received 12/17/97 Accepted 7/21/98


7.1. Genomic changes in cell immortalization

Senescence of human fibroblasts has been associated with genetic determinant(s) on chromosome 4 (73) and the long (q) arm of chromosome 1 (74). Karyotyping analyses of 5 spontaneously-immortalized human breast epithelial cell lines have identified several common chromosomal alterations including loss of chromosome 20p and gain of chromosome 1q (75). The most common genomic change in SV40-immortalized human cells is the loss of chromosome 6 (76). Recently, alterations at several other chromosomal loci (e.g. 20q13.2, 6q26-27) have been implicated in immortalization of various epithelial cells with viral oncogenes (77, 78). However, the nature and the function of genes located in these chromosomal loci remains to be defined. In our laboratory, we have determined that the immortalization of the MCF-10F is associated with the mutation at exon 7 of the TP53 locus (47) (Figure 9), supporting the notion that inactivation or loss of function-mutation of the p53 gene is critical in the early stages of breast cancer progression (79, 80). In addition to further evaluation on the role of p53 in immortalization of HBEC, our laboratory is currently pursuing studies to determine further genomic changes such as microsatellite instability (MSI) and its underlying mechanisms that may play a role in the immortalization of HBEC.

Figure 9. Figure 9 A-E. Direct DNA sequencing of the PCR amplified products generated from exon 7 of the p53 gene of A: placenta; B: MCF-10M cells; C: MCF-10F cells; D: D3-1 cells; and E: MCF-7 cells. An insertional mutation of a T at codon 254 was observed in MCF-10F (C) and its derivative D3-1 (D) cells, causing a frame shift following codon 254. Right-hand panel: PCR-SSCP analysis of exons 4-9 of the p53 gene. Exons 4, 5, 6 and 8-9 showed identical appearance in placenta and MCF-10M. A conformational shift in exon 7 was observed in three MCF-10F cell lines tested and the same shift was maintained in MCF-10F-derived transformed cell lines (BP1, BP1E, D3 and D3-1) (Reproduced with permission from Ref. 47).

7.2. Genomic changes in cell transformation

Genomic alterations have been recognized as a hallmark of cancer progression (81-85). The unstable changes of microsatellite, or very short simple repetitive sequences, designated as (CA)n, that are distributed throughout the genome (86, 87) represent such alterations. Its association with human malignancies has been extended from colorectal cancer (88-90) to neoplasms of the neck and head (91), lung (92), skeletal muscle (93), lympho-hematologic system (94), skin such as melanoma (95), prostate (96), gastrointestinal system (97), urinary bladder (98), liver (99), neurologic system (100, 101), cervix (102), endometrium (103), as well as breast (104-112). These findings indicate that MSI is associated with the general process of carcinogenesis.

MSI has been associated with the progression of breast cancer. However, its exact timing is controversial and its specific functional roles are not clear. It may be present as an early (105-108, 112), or a late event (113), or both (114), or not correlated (115, 116) during the breast carcinogenesis, dependent upon the markers (locus specific for each chromosome) used, and samples tested. This assumption is supported by the study of Aldaz et al (114) showing that some chromosomal loci might be involved early, while others late during the progression of human breast cancers. Therefore, the elucidation of this genomic phenomenon can be further clarified by analysis with a comprehensive array of markers in an in vitro system such as ours that is free of affecting factors from variations among individuals and yet consists of various stages of cell transformation. More importantly, such studies may lead to the determination of underlying mechanisms such as defects of DNA mismatch repair genes that have been documented in colorectal, endometrial, ovarian, and prostate cancers (reviews in 117, 118), which have emerged as another type of factors as important as the tumor suppressor genes and oncogenes in breast carcinogenesis. Specifically, several questions exist: 1. Does MSI occur in preference to a particular chromosome(s) at a specific locus? 2. Is this instability really correlated with the phenotypic progression of human breast cancer and thus does occur in a sequential order? 3. What are the specific underlying mechanisms, such as mismatch repair gene defects, or DNA replication errors, or others?

We have pursued the first question by analyzing a total of 466 microsatellite loci on all the chromosomes in transformed HBECs representing the early and intermediate stages of cell transformation (119). These markers were selected to represent 38-96% the banded regions (according to the Human Genome Maps V) (120), taking into consideration of locations where tumor suppressor genes, oncogenes, DNA repair genes, and other cancer or cell growth regulation-associated genes are documented or postulated to be situated. Interestingly, we were able to detect MSI in only a very small number of loci; 0.64% (3/466), or 0.43% (2/466) of the markers analyzed were found in the BP-, or DMBA-transformed HBECs, respectively. These changes were exclusively found in the chromosomal regions of 11q25 at locus D11S912 and 13q12-13 at loci D13S260 and D13S289 in the BP-cells, or the 13q12-13 region at loci D13S260 and 16q12.1 at D13S260 and 16q12.1 at D16S285 in the DMBA-cells (119, 121) (Figure 10). Furthermore, the occurrence of MSI among these loci in the BP-cells seems to reflect a sequential order; i.e., 11q25 (D11S912) in the BP1 cells, followed by 13q12-13 (D13S260) and then another locus of 13q12-13 (D13S289) in the BP1E cells. However, this tendency is not seen in the DMBA cells. Our data have provided direct evidence that MSI is associated with the early and intermediate transformation of HBEC, during which only a very small proportion of loci are affected, and that the involvement of these loci on chromosomes 11 and 13 may be correlated with the progression of HBEC transformation in vitro. This finding supports that carcinogen-transformed HBECs, and presumably sporadic breast carcinomas, are also characterized with a mutator phenotype (122), that appears early as a driving force in tumor progression (review in 84, 123).

Figure 10. Microsatellite instability in the transformed HBECs (1: MCF-10F; 2: BP1; 3: BP1E) and human breast lesions (N: Normal; D: DHP; C: CIS; I: INV).

The carcinogenic mechanisms of benzo(a)pyrene are postulated to be related to its ability to produce G->T mutations, which affect nucleotides in GC-rich stretches of DNA (124), as showed by mutations of p53 at codons 154, 173, 248, 266, 173 and 277 (125, 126). We hypothesize that BP may cause additional mutations in TP53 and other target genes in the transformed cells, which hence contribute to the genomic instability detected in these cells. This is supported by the evidence that germline mutations of p53 are associated with the genomic instability with Li-Fraumeni syndrome (review in 127). Other mechanisms that may contribute to genomic instability include defects in DNA mismatch repair genes (reviews in 117, 118), DNA replication error and DNA repair defects (reviews in 84, 128, 129).