Abstract

The organization and replication of DNA render fragile sites (FSs) prone to breakage, recombination as well as well as preferential targets for mutagenes-carcinogens and integration of oncogenic viruses. For many years, attempts to link FSs and cancer generated mostly circumstantial evidence. The discoveries that chromosome translocations, amplification of proto-oncogenes, deletion of tumor suppressor genes, and integration of oncogenic viruses all result from the specific breakage of genomic DNA at FSs, however, have provided compelling support for such a link, further suggesting a causative role for FSs in cancer.

INTRODUCTION

Over the last several years, researchers opened up the fragile sites (FSs), regions of the genome that are prone to breakage and recombination, dissected their the molecular structure, discovered new cancer-related genes alterations and integrated oncogenic viruses at FSs, thus establishing an association with cancer development. In this article, I will highlight the recent progress in our knowledge of FSs and of their relevance to neoplasia, including their role as preferential sites for the integration of oncogenic DNA viruses, a field to which I have now contributed for a number of years. I recall the submission of my first paper describing the localization, with the use of isotopic in situ hybridization, of the integration sites of human papillomavirus (HPV)–18 to FSs in HeLa cervical carcinoma cells. The reviewer reached an unfavorable decision primarily because, at that time, the integration of oncogenic viruses into tumor cells was thought to be random. The paper was eventually published [1]. A few weeks before its publication, however, a study based on analysis of somatic cell hybrids revealed HPV integration into the genome at sites located near oncogenes in cervical carcinomas [2]. The integration of HPV at FSs and loci of cancer-related genes in cervical cancer has subsequently been amply documented.

Two of several recent excellent review articles [3–6] referred to FSs as “the weakest links” [5] and “still breaking” [4]. FSs are indeed weak and are vulnerable targets for various oncogenic agents, and their damage may potentially result in deleterious consequences for genomic integrity and function. Although FSs are “still breaking,” recent investigations have provided insight into their structure and their involvement in important processes associated with cancer development. We now know, for example, that the amplification of proto-oncogenes and the deletion of tumor suppressor genes frequently occur at FSs. Thus, a link between FSs and cancer development can no longer be denied.

Characteristics of fragile sites

FSs are often sites for the exchange of genetic material between sister chromatids, chromosomal translocations, deletions, gene amplification, and the integration of oncogenic viruses. FSs are visible by simple staining of metaphase chromosomes and comprise large regions (~50 kb to 1 Mb) of chromosomal DNA that are prone to breakage. They appear as gaps or breaks in chromosomes when cultured lymphocytes are exposed to antifolates, distamycin A, bromodeoxyuridine, methotrexate, 5-azacytidine, or aphidicolin (apc), the latter of which is an inhibitor of DNA polymerase a (Figure 1a) [7,8]. Whereas most, if not all, humans harbor common apc-sensitive FSs, ~5% of individuals express rare FSs. At least 104 FSs, 24 rare and 80 common, encompassing ~100 Mb of DNA overall are currently listed in the human genome database. FSs are scattered throughout the genome, but they tend to cluster at G-light chromosome bands [9]. G-negative bands contain GC-rich Alu repeats and are constitutionally more relaxed and unfolded during transcription than are G-positive bands[10]. FSs center around CpG islands and correspond both to sites of nuclease sensitivity and to binding sites for proteins that contain a zinc finger domain, indicating that they reflect an aberrant state of genetic activity at regions associated with transcriptional regulation [11]. FSs tend to replicate late during the cell cycle, thus providing longer-lasting targets for breakage and recombination [12,13]. Characterization of the replication of the fragile site FRA3B indicated that incompletely copied DNA in G2 manifests as gaps and breaks during metaphase [14]. The application of fluorescence in situ hybridization (FISH) to interphase nuclei has revealed an uncoordinated pattern of replication of FRA7H under normal growth conditions or during apc exposure, compared with the replication pattern of two non-FS regions under identical conditions. Apc increases the existing difference in replication time between FRA7H and other genomic regions, underscoring the impact of delayed replication on fragility at apc-sensitive FSs [15]. Chromatin organization and DNA conformation are also important determinants of breakage and recombination at FSs. It has been suggested that recombination between sequences on nonhomologous chromosomes might be promoted at FSs by the unusual left-handed helical structure of Z-DNA [16,17].

All rare folate-sensitive FSs that have been cloned (FRAXA, FRAXF, FRA16A, and FRA11B) exhibit unstable expansion of CCG trinucleotide repeats and their expression results from an excessive copy number of such repeats [18–22]. In contrast, common apc-sensitive FSs that have been completely or partially sequenced (FRA3B, FRA7G, FRA7H, and FRA16D) do not contain unstable repeat sequences, showing that FSs are distinguishable not only by their frequency of expression within the population and their sensitivity to various agents but also by their molecular structure and the mechanisms that underlie their expression [4,23-25]. The cause of the instability of apc-sensitive FSs remains unclear. Both FS expression and sister chromatid exchange (SCE) are inducible by manipulation of culture conditions or by exposure of cells to mutagens or carcinogens [26,27]. Several studies have shown nonrandom induction of SCE at expressed and nonexpressed FSs on chromosome 16, suggesting that a common DNA alteration underlies both SCE and FS expression [28–30].

Fragile sites and cancer development

Cancer arises from a single precursor cell as a result of the accumulation of multiple genetic and epigenetic alterations caused by chemical and physical carcinogenes, oncogenic viruses, errors in replication, or the effects of aging. Genetic alterations and changes in DNA methylation may affect a variety of genes. The accumulation of DNA damage that manifests as point mutations as well as chromosomal rearrangements, amplifications, and deletions results in the acquisition by incipient clonal cancer cells of the potential for unlimited, self-sufficient growth and of resistance to normal homeostatic regulatory mechanisms [31,32]. Chromosome translocations, inversions, deletions, and amplifications are critical in the pathogenesis of human cancer. The molecular and cytogenetic characterization of the breakpoints of cancer-specific chromosome translocations at FSs, the identification of tumor suppressor genes that span FSs and their deletion in cancer cells, and the demonstration both that chromosomal breakage at FSs underlies oncogene amplification and that oncogenic viruses integrate into the genome at FSs constitute the turning points in the linkage of FSs to cancer development.

Cancer-specific chromosome translocations at fragile sites

An association between FSs and cancer emerged as a result of the discovery of chromosome banding in the 1970s, when more and more leukemias, lymphomas, and solid tumors with specific structural chromosome alterations and a new type of apc-sensitive FSs were identified [33,34]. Among the first cancers associated with specific chromosome changes were chronic granulocytic leukemia, Burkitt’s lymphoma (BL), and lung cancer. The localization both of the breakpoints of these cancer-specific chromosomal translocations and of cellular oncogenes to FSs led to the hypothesis that FSs confer a predisposition and contribute to cancer development [35]. Early maps revealed a striking correlation between the location of FSs at the breakpoints of specific chromosomal translocations and deletions in cancer cells and the loci of oncogenes [35,36]. Furthermore, certain individuals with malignancies exhibiting specific chromosomal alterations also were shown to express a rare FS at the breakpoints of the genomic rearrangements [37,38]. Skeptics, however, considered such correlations coincidental.
Fragile site 8C
Many leukemias, lymphomas, and sarcomas exhibit specific reciprocal chromosomal translocations that either result in the activation of proto-oncogenes or generate new oncogenic chimeric genes. Given that many of the products of both oncogenes and gene fusions function as transcription factors, the disruption of transcriptional control likely plays an important role in the development of certain types of neoplasia [39]. BL is characterized by translocation of the c-MYC locus at FRA8C to any of the three immunoglobulin gene loci on chromosomes 2, 14, and 22 [40]. Although all such translocations result in deregulation of MYC expression, the positions of the breakpoints vary among individuals. The breakpoints in the most common translocation, t(8;14), cluster within or near the MYC locus at FRA8C. In certain cases of BL, secondary rearrangements result in transposition of MYC sequences to additional genomic locations [41]. In addition to chromosomal translocations that involve FRA8C, both regional DNA amplification and HPV integration frequently occur at this site. The mechanisms underlying the susceptibility of FRA8C to recombination or DNA amplification and its accessibility to viral integration are unknown. The cloning of FRA8C should provide more insight. Translocations in B and T cell lymphomas occur as a result of recombinase-mediated breakage of DNA strands. Analysis of the breakpoints in several B and T cell malignancies has identified heptamer or nonamer sequences within a transcriptionally active DNA region [42]. Recombination can occur only if such sequences become accessible to the recombinase through an open chromatid conformation. Further detailed sequence and nuclease sensitivity analyses demonstrated that B and T cell translocation breakpoints are located near DNase I–hypersensitive sites and regions that contain alternating purine and pyrimidine tracts [16]. Similar sequences are thought to be components of FSs [36,42,43]. Such sequences are able to alter chromatin structure and to facilitate the access of recombinase by converting the DNA helix from a right-handed B conformation to a left-handed Z conformation (16). Certain chromosome translocations occur at a high frequency because of the physical proximity of the affected sites within the nucleus of a particular cell type [44]. Interactions between FSs after their induction is thus thought to facilitate the translocation process [6,45].
Fragile site 3B An apc-inducible FS located at chromosome 3p14.2, designated FRA3B, is the most highly expressed FS in humans and is associated with deletions in a variety of histologically distinct cancers [5]. Interest in FRA3B increased dramatically after the isolation of the fragile histidine triad gene (FHIT), which spans this FS and is abnormally expressed in various common types of cancer (5). FRA3B is also one of the most frequent sites for balanced or unbalanced chromosome translocations that affect FHIT in a variety of cancers. The breakpoint in the constitutional t(3;8) associated with hereditary renal cell carcinoma (RCC) interrupts FHIT and results in its fusion with the patched-related gene TRC8 [46]. One of four balanced translocations, t(3;20), identified in breast tumor cell lines results in homozygous deletion of exon 5 of FHIT and loss of expression of the encoded protein, and a t(3;12) associated with pleomorphic adenoma of the parotid gland results in the fusion of FHIT and HMGA2 the latter of which encodes high mobility group protein A2 [47,48]. Similarly, in esophageal adenocarcinoma, the breakpoints of translocations involving chromosome 3, t(3;16) and t(3;4), occur within FHIT, at or near the center of the FS. Amplification of cDNA ends with FHIT-specific primers allowed the identification of a noncoding chimeric transcript produced as a result of the t(3;16) [49]. Unbalanced translocations that typicaly cause loss of genetic material and FHIT alterations have also been demonstrated in human hepatocelullar carcinoma (HCC), a neoplasm that is closely associated with exposure to mutagens-carcinogens or oncogenic viruses. Such translocations lead to a reduction in the abundance or loss of FHIT transcripts, intragenic deletions, and the loss of FHIT protein in a significant number of HCCs. Exposure to aflatoxin B1 or alcohol as well as integration of hepatitis B virus (HBV) at FRA3B (FHIT) may generate a background genetic instability early during hepatocarcinogenesis that has important pathological consequences [50,51]. Collectively, these observations show that balanced and unbalanced translocations that affect FRA3B give rise to aberrant FHIT transcripts in various solid tumors.
Fragile site 16D
FRA16D, another highly expressed FS, encompasses the gene WWOX, also known as FOR (52,53). A recurrent translocation, t(14;16), that nonrandomly involves FRA16D has been identified in multiple myeloma (MM) cell lines. In four out of five such translocations, the breakpoints map within WWOX, scattered over an ~500-kb region centromeric to the MAF proto-oncogene, and within the immunoglobulin H locus on chromosome 14q32. Similar to the balanced translocations that affect proto-oncogene and immunoglobulin gene loci in B cell lymphomas, the t(14;16) translocations associated with MM result in deregulation of MAF expression, which might contribute to carcinogenesis [54].
Fragile site 2G
A constitutive translocation involving FRA2G was recently characterized in members of a family in which multifocal clear RCC segregates with the balanced t(2;3). Both yeast (YAC) and bacterial (BAC) artificial chromosome–based contigs encompassing the 2q and 3q breakpoints were constructed, and those clones spanning the breakpoints were partially sequenced. All known regional markers, genes, and expressed sequence tags (ESTs) were mapped relative to the breakpoint sequences. FISH localized the 3q breakpoint to 3q13, possibly near the border with 3q21, and the 2q breakpoint closely telomeric to FRA2G [55]. The genomic map of the 2q breakpoint revealed a nearby EST, and a full-length cDNA of the corresponding novel gene, designated DIRC1, that is disrupted by the translocation was isolated (56). This gene was shown to be expressed at a low level in various tissues. The absence of mutations and rare polymorphisms of DIRC1 in tumors and tumor cell lines, however, has impeded characterization of its role in RCC [56].
Jumping translocations
Jumping translocations (JTs) and segmental jumping translocations (SJTs) constitute a distinct class of unbalanced translocations that involve the fusion of a donor chromosome arm or chromosome segment with multiple recipient chromosomes. The breakpoints of 188 JTs and SJTs in 10 cell lines derived from carcinomas of the bladder, prostate, breast, cervix, and pancreas have been localized and shown to correspond to FSs and integration sites of oncogenic viruses. The acquisition of extra copies of the donated chromosome segments, many of which contain oncogenes, resulted in tumor-specific genomic imbalance and promoted clonal progression [57].

Cancer-related gene alterations at fragile sites
Various genes associated with cell growth, senescence, or apoptosis or with the maintenance of genomic integrity are implicated in cancer development. Tumor suppressor genes are negative regulators of cell proliferation and, with the exception of leukemias and lymphomas, are frequently mutated, deleted, or hypermethylated in human cancers [58,59]. In contrast, overexpression or amplification of proto-oncogenes promotes cell proliferation. Molecular and cytogenetic evidence has demonstrated that, in addition to chromosome translocations, deletion of tumor suppressor genes and oncogene amplification are frequently the consequence of DNA strand breakage at FSs.

Proto-oncogene amplification Like other structural alterations, DNA amplification is associated with genomic instability and contributes to carcinogenesis. Small pairs of extrachromosomal acentric chromatin bodies, referred to as double minute chromosomes (DMs), and abnormally banded or homogeneously stained regions (HSRs) of the genome are associated with the acquisition of drug resistance and tumor progression and manifest a high level of DNA amplification [31,60–63]. The loss of control of gene copy number results in the generation of subpopulations of tumor cells with an increased growth potential and invasiveness that are refractory to chemotherapeutic agents [64,65]. These genetic alterations frequently involve oncogenes and are considered important in cancer development [31]. Although DMs and HSRs were first described over 30 years ago, despite intensive efforts the mechanisms responsible for their formation remain unclear. A breakage-fusion-bridge (BFB) model proposed in the 1950s has proved relevant to the genesis of gene amplification through FSs [66,67]. Evidence for intrachromosomal amplification of drug-selected genes mediated by multiple breakage at FSs was obtained with cultured hamster cells. The positions of FSs relative to the boundaries of early amplicons as well as the organization of the amplicons were also defined [67]. An increase in DNA copy number at region 11q13 occurs frequently in several cancers, and the CCND1 (cyclin D1), FGF3 (INT2), and EMS1 genes at 11q13 are amplified or overexpressed in various types of malignancy [68,69]. Amplification of the RIN gene at this site, possibly as a result of BFB cycles involving FRA11B, was also demonstrated in an oral squamous cell carcinoma [70]. The amplification of a bona fide proto-oncogene, MET, through DNA strand breakage at a FS was for the first time demonstrated in human gastric carcinoma [71]. Among other evidence consistent with repeated breakage at FRA7G in this cancer, large marker chromosomes containing equally spaced and inverted MET amplicons, similar to those obseved in rodent cells with amplification of drug-selected genes, were detected [71]. Such markers, which typically exhibit an abnormal banding pattern, are often apparent in tumor cells and may contain two or more amplified genes, frequently oncogenes. Figure 1c shows a large abnormal chromosome identified in a breast carcinoma cell line that is the result of rearrangement of chromosomes 3, 8, 13, and 17 and contains multiple copies of ERBB2 and MYC oncogenes within clusters on both chromosome arms [72]. Whereas MYC is located within FRA8C, ERBB2 is located distant from FRA17B. Although the signals of the two genes appeared closely spaced in chromosome preparations, FISH with DNA fibers revealed that the two genes are not contiguous and that the clusters most likely contain additional amplified genes [72]. The genesis of such complex abnormalities containing two or more amplified genes cannot be fully explained by BFB cycles and likely involves additional breakage and recombination events at both FSs and non-FS regions. Multiple copies of DMs derived from chromosome 19 were identified in a mouse HCC. A probe generated by PCR from the microdissected DMs was shown by FISH to hybridize to chromosome 19 at two sites separated by several medium-sized G-bands. This organization of extrachromosomal amplified sequences derived from separate loci of the same chromosome is also indicative of a complex mechanism of DNA amplification, possibly involving BFB cycles of two or more genes [73]. It appears, however, that mouse chromosome 19 does not contain a FS [74].

Inactivation of tumor suppressor genes
FRA3B is currently the most well characterized FS. FISH with cosmids corresponding to specific regions of FHIT revealed that most of the apc-induced gaps at FRA3B are located within FHIT, with the largest number occurring in intron 5 of the gene within a region <30 kb telomeric to exon 5. Gaps also occur in intron 4, where HPV-16 sequences have been detected, as well as in intron 3, where the t(3;8) breakpoint associated with familial RCC is located.These observations suggest that cancer-specific deletions of FHIT, which frequently involve introns 4 and 5, result from breaks at the FS [75]. More recently, two-color FISH with pairs of BAC probes confirmed that most of the apc-induced breaks cluster within a 300-kb region between exon 4 and intron 5. However, analysis of hybrid clones containing various deletions of FRA3B sequences suggested that there is no single sequence responsible for breakage or recombination within FRA3B, although the existence of multiple “hot spots” cannot be ruled out [76]. Gene inactivation is considered a primary hallmark of tumor suppressor genes. Although inactivation of FHIT has been detected in most types of cancer, until recently the evidence that FHIT is a tumor suppressor gene appeared conflicting. The antitumorigenic activity of FHIT was demonstrated convincingly with the use of an adenoviral vector to obtain uniform populations of cells expressing the gene without the drug selection required with other vectors. Restoration of FHIT expression by gene transfer with this adenoviral vector in FHIT-negative lung and cervical carcinoma cell lines resulted in apoptosis in vitro as well as complete or partial suppression of tumor formation in vivo [77]. Breakage at FRA3B and FHIT alterations in a multitude of diverse types of cancer, established the first link between FSs and tumor suppressor genes in neoplasia. The mouse ortholog of FHIT has also been isolated and found to contain FRA14A2 [78]. The generation of mice heterozygous for knockout of this gene provided further support for the role of FHIT in neoplastic development. The transgenic animals thus exhibit an increased susceptibility to carcinogenesis and develop tumors that mimic those associated with Muir-Torre syndrome [79]. Introduction of FHIT into the FHIT-deficient mice prevented tumor formation, providing experimental evidence for the recently proposed concept of cancer prevention [80]. The association of BRCA1 and BRCA2 with hereditary breast cancer constituted a milestone in the genetics of cancer. Even though the detection of mutation of these genes provides only limited information on cancer risk [81], substantial insight into the mechanism of carcinogenesis has been obtained from characterization of the encoded proteins, which are implicated in multiple nuclear functions including transcription and maintenance of genomic stability through involvement in DNA repair and recombination. Comparative genomic hybridization (CGH) studies have revealed a twofold increase in the number of genetic imbalances in tumors from carriers of BRCA1 or BRCA2 mutations compared with unselected controls [reviwed in 82]. Furthermore, consistent with the notion that mutation of BRCA2 and loss of its function promote genomic instabiltity, hyperinstability at chromosome 9p and the FRA3B locus has been demonstrated in tumors from individuals with familial cancer linked to this gene [83,84]. The frequency of loss of heterozygosity (LOH) at the FHIT locus, resulting in loss of the FHIT protein, was shown to be significantly greater in breast tumors from women with inherited deleterious mutations of BRCA2 than in sporadic breast tumors, again suggesting that loss of BRCA2 function affects stability of the FHIT (FRA3B) locus [85]. The frequency of apc-induced gaps or breaks in human tumor cell lines or lymphoblastoid cell lines that lack DNA mismatch repair as well as in repair-deficient mouse cell lines was shown to be increased compared with that in control cells, indicating that proteins that contribute to mismatch repair or to the repair of double-strand breaks are important for maintenance of the integrity of FSs. FISH with YAC probes revealed that gaps and breaks occurred more frequently at FRA3B and FRA16D than at FRA7G in both repair-deficient and control human cells (Figure 1 and b) [85]. Studies of FRA16D have also provided evidence for the association of FSs with cancer. The WWOX gene localized at this FS spans a region of >1 Mb and is a candidate tumor suppressor. This gene is aberrantly expressed as a result of homozygous deletions and inhibits both growth in vitro and tumorigenicity in vivo of breast cancer cells [53]. The mouse ortholog of WWOX has also been isolated and colocalizes with FRA8E1. The sequence of mouse FRA8E1 is highly conserved relative to that of human FRA16D over a region of at least 100 kb [86]. Sequence conservation is also apparent for human FRA3B and the mouse ortholog FRA14C [87]. Not all FSs are conserved, however. Only a small proportion of FSs thus appear to be conserved between mouse and rat, for example, and those that are conserved exhibit different degrees of expression [74]. These observations suggest that conserved FSs harbor genes that are vital for development and that damage to these sites is likely to result in various disorders, including cancer. A variety of human tumors exhibits LOH and deletion at FRA16D and, in certain instances, both WWOX and FHIT are deleted. Screening of a series of primary tumors and tumor cell lines for deletion of, or the production of aberrant transcripts from, FSs revealed codeletion of FRA16D and three other FSs (FRAXB, FRA7G, and FRA7H) in an esophageal tumor [88]. Common features of the most often deleted FSs in cancer cells include a high frequency of expression, conservation during evolution, expansion over a broad region and encompassing large genes as well as preferential targeting by mutagens-carcinogens and oncogeneic viruses.

Fragile sites and oncogenic agents
Targeting of fragile sites by mutagens-carcinogens

Genetic alterations in certain cancers are likely the result of exposure to mutagens-carcinogens or oncogenic viruses that target FSs. Evidence for the targeting of FSs by mutagens-carcinogens has been provided by high-resolution banding analysis of chromosomes from cultured human lymphocytes [89]. Breakage of DNA strands induced by mutagens-carcinogens that act by distinct molecular mechanisms was localized and the positions of the breaks were shown to coincide with those of cancer-related translocation breakpoints, proto-oncogenes, and FSs. Only a fraction of the known FSs was targeted by the chemicals, however [89]. Among the agents that interacted with FSs were two chemicals that induce the formation of hypersensitive sites in the regulatory regions of active genes. The relations among chromatin hypersensitivity, chemical induction of FSs, and the accessibility of chromosome domains to transcriptional factors were investigated further by exposure of cultured human lymphocytes to DNase I and restriction endonucleases. More than 60% of the DNase I–induced break sites corresponded to FSs [90]. A similar distribution of DNA strand breaks was observed after exposure of the cells to restriction enzymes or benzo[a]pyrene diol epoxide, which interact with GC-rich and guanine-rich promoters, respectively [90]. Experiments with cultured human lymphocytes exposed to apc and subsequently treated with mutagens for several hours demonstrated that the induced deletions and interchromosomal recombination occurred at FSs. A substantial number of quadriradial and triradial chromatid and chromosome exchanges exhibited breakage primarily at apc-inducible FSs [90]. It has been suggested that nonhomologous interchanges may precede the formation of reciprocal translocations, duplications, or deletions. Moreover, structural rearrangements induced by the action of mutagens-carcinogens at FSs are indistinguishable from those apparent in cancer cells [89]. Small cell lung carcinoma (SCLC) and HCC are common forms of cancer worldwide and are associated with exposure to mutagens-carcinogens such as tobacco smoke, alcohol, and aflatoxin B1. Recurrent deletion of chromosome 3p was first described in SCLC more than 20 years ago, and LOH at 3p was subsequently shown to occur in most, if not all, SCLC tumors [91]. The damage to chromosome 3p affects the FHIT locus and appears to be an early event in the lungs of smokers. The frequency of LOH at the FHIT locus was shown to be markedly higher in lung adenocarcinomas of smokers than in those of nonsmokers as a result of the interaction of tobacco carcinogens and FRA3B [92]. Indeed, the frequency of FS expression, including that of FRA3B, is significantly higher in the lungs of active smokers than in those of nonsmokers or of SCLC patients who have stopped smoking. These observations suggest that persistent tobacco exposure increases FS expression, but that the induced fragility is reversible. Exposure to tobacco carcinogens thus increases the potential for damage at FSs [93]. The expression of FSs varies in the human population and is probably triggered by a combination of genetic factors and exposure to environmental mutagens-carcinogens [5]. It is also associated with a wide spectrum of cancers that are either related or unrelated to exposure to environmental mutagens-carcinogens or oncogenic viruses. Evidence from a variety of cancers has convinced many researchers that FS expression is a sensitive predictor of cancer susceptibility. Others, however, believe that the variation in FS expression apparent within individuals undermines such risk prediction; both environmental factors as well as ingested substances such as caffeine and alcohol are thought to contribute to fluctuations in FS expression in individuals [6]. A cytogenetic-based predictive assay for cancer is unlikely to materialize, however, primarily because of the inherent difficulties of the procedures. Elucidation of the molecular basis of FSs might reveal a relation between the molecular determinants of FS expression and sensitivity to mutagens-carcinogens, and therefore allow development of a predictive assay for cancer risk [6]. The cataloging of the expression profiles of genes representative of FSs might constitute a feasible approach to the development of such an assay. Transcriptional profiling analysis has revealed that a substantial proportion of genes that are down-regulated in ovarian tumors maps to FSs. Nine of these genes were found to be localized within seven previously uncloned FSs, and down-regulation of 7 of 10 genes was confirmed in cell lines as well as in primary ovarian tumors [94].
Preferential integration of oncogenic viruses at fragile sites
Human papilloma virus
DNA- or RNA-containing oncogenic viruses have been implicated in certain types of human cancer either as causative agents or cofactors. Both tumors and cells transformed by oncogenic viruses usually contain viral nucleic acid sequences integrated into their genome and express viral antigens. The specificity of viral integration is an important determinant of the pathological significance of this event [95,96]. The integration sites of various DNA and RNA viruses have been localized with the use of in situ hybridization. The application of isotopic in situ hybridization to HeLa cells provided the first cytogenetic localization of HPV integration at a FS [1]. The most frequent site of HPV integration is FRA8C, which corresponds to the MYC locus. HPV integration at 8q24 is invariably accompanied by amplification of the viral sequences and often by overexpression of MYC [97]. Overexpression of MYC is considered a prognostic indicator of the invasiveness and metastatic potential of cervical cancer [98]. The sites of HPV integration also appear to contribute to chromosomal rearrangements in cervical cancer. An analysis of 1912 chromosomal breakpoints in 148 individuals with this malignancy revealed that 133 breakpoints were located at HPV integration sites [99]. Furthermore, exposure of peripheral lymphocytes derived from HPV-infected subjects to apc revealed a significant difference in the number of induced chromosomal alterations compared with that observed in control cells obtained from women without cervical lesions [100]. Moreover, the chromosomal fragility induced by apc treatment was significantly higher in peripheral lymphoctes isolated from women infected with HPV-16 or -18, which are associated with premalignant and invasive cancer, than in those from women harboring HPV-6 or -11, which are associated with benign genital warts [100]. Integration of the viral genome into cellular DNA affects the initial stages of cell transformation, maintenance of the transformed phenotype, and tumor progression [101]. Transfection of normal genital epithelial cells with HPV DNA resulted in their development into permanent cell lines only after the integration of transcriptionally active viral sequences into the cellular genome [102,103]. Studies with undifferentiated human epithelial cells have suggested that alterations of cellular genes triggered by virus integration may override the influence of negative interfering factors produced by the cell to control viral gene expression [104]. The localization of integrated viral sequences to aberrant chromosomes in foreskin epithelia immortalized by HPV-16 demonstrated that viral integration resulted in stable structural alterations. The sites of HPV integration were frequently located near or at FSs. The selection of FSs as sites of HPV integration was shown to be independent of cell origin and to parallel HPV integration in cervical carcinomas [103]. In situ hybridization analysis of integration sites relative to G-banding and DNA replication patterns in human keratinocyte cell lines transfected with recombinant HPV-16 showed that viral DNA sequences were located at FSs, chromosomal G-negative bands exhibiting late DNA replication, suggesting that the structural and functional features of these sites render them accessible to viral integration [105]. FISH and FISH-based procedures have been applied extensively to map viral integration sites in tumor cells. The integration of a single copy of HPV-18 in an established cervical carcinoma cell line (C4-1) was mapped to region 8q22, which corresponds to the location of a FS and the loci of two oncogenes, MYBL1 and AMLIT1 . This site was further examined for nuclease sensitivity with the use of a probe for cellular flanking DNA. A DNase I–hypersensitive site was detected within 3 kb of the viral DNA. Furthermore, the integration site was found to be undermethylated in cervical carcinoma cells and fully methylated in tumor cells of other origins, suggesting that conformational and functional characteristics of chromatin at FSs are important for viral integration [106]. HPV types other than 16 and 18 also integrate at FSs. HPV-68 has thus been shown to integrate at chromosome 18q21, the location of FRA18B. The gene APM1, which encodes a protein with a BTB/POZ domain and four zinc fingers, was cloned from the integration site and shown to inhibit the growth of two cervical carcinoma cell lines in vitro [107]. HPV-33 was found to integrate at FRAE5 on chromosome 5p14 in a cell line derived from a vaginal dysplasia [108]. The karyotype of HeLa cells has also been examined with a combination of spectral karyotyping, CGH, and FISH with probes specific for HPV-18 and MYC previously characterized by G-banding and chromosome painting. HeLa markers identified by G-banding and HPV-18 sites of integration detected by isotopic hybridization were confirmed and new cryptic rearrangements undetectable by other techniques were fully resolved (1,109). Three rearrangements of chromosomes harboring integrated HPV at the sites of breakage and rejoining were identified, indicating that viral integration triggered genomic instability. HPV-18 and MYC sequences were detected at sites of rejoining on derivatives of chromosomes 5 and 21 [109]. These observations differ from results of isotopic in situ hybridization suggesting that HPV-18 integration at multiple sites occurs independently [97]. Two new procedures for the accurate mapping of viral integration sites have recently been developed. One procedure based on binary ratio labeling FISH, known as Cobra FISH, allows the simultaneous visualization of viral integration sites and multicolor karyotyping. Four previously unknown sites of HPV-16 and -18 integration were identified with this approach at FSs on normal and abnormal chromosomes 8 at 8q24 and normal copies of chromosomes 1 and 17 at regions 1q42 and 17q21-23, respectively [110]. The second procedure combines molecular biology and FISH. BAC clones are selected on the basis of cellular sequences flanking viral integration sites isolated by RS-PCR are used as probes for conventional FISH with metaphase chromosomes. This approach has identified three new sites of HPV-16 integration at FRA6C, FRA17B, and FRA1F. Integration of HPV occurred within the coding region of USP23 at chromosome 6p22.2 and within an intron of a gene encoding a putative RNA-binding protein at 17q23.1, suggesting that these genes may be mutated in cervical and other cancers [111]. With a similar approach, HPV-16 was found to have integrated at chromosome 14q32.3 in the genome of a rapidly progressive, lethal cervical cancer from a 39-year-old woman. This chromosomal region is implicated in many translocations associated with T and B cell lymphomas and also harbors the putative proto-oncogene TCLIA. Virus integration resulted in disruption of TNFAIP2, a cytokine- and retinoic acid–inducible gene. The integrated HPV-16 DNA contained an intact upstream regulatory region as well as E6 and E7 open reading frames. Both 5¢ and 3¢ virus-cell junction regions contained direct repeat and palindromic sequences [112]. Virus integration at chromosome 14q32 has also been detected in an HPV-16–immortalized keratinocyte line . Viral integration in this region may be a contributing factor to the growth advantage of HPV-containing cells (103,113). These various observations suggest that maintenance of E6 and E7 expression, loss of the E2 gene, and disruption of TNFAIP2 by viral integration contribute to the rapid progression of cervical cancer [112]. Overall, studies of the localization of HPV integration sites in transformed cells and cervical tumor cells have demonstrated that FSs are frequently targeted by the virus. Furthermore, certain regions, such as 8q24 and 12q14-15, have been associated with clusters of integration. A comprehensive list of sites of HPV integration in cervical cancer was recently presented [112].
Hepatitis B virus
HPV is not the only virus with a propensity for integration at FSs or a high specificity of integration at other sites. HBV, a DNA virus that is closely associated with the development of HCC, exhibits marked similarities with retroviruses in terms of its integration and requirement for reverse transcriptase and an RNA intermediate during its replication. Woodchuck hepatitis virus, the most oncogenic of the hepandaviruses, integrates near the MYC locus and induces gene activation. Woodchuck hepatitis virus–mediated rearrangements and activation of MYC are similar to those resulting from chromosome translocations in B cell lymphoma [114,115]. A similar scenario also may apply in HCC that develops in individuals chronically infected with HBV. However, given the inherent ability of HBV to cause secondary chromosomal rearrangements, the localization of HBV genomic sequences in HCC may not represent the initial site of integration [116]. A survey of HBV sequences in HCC nevertheless suggested that chromosomes 11 and 17 are frequent targets for integration [117]. HBV integration on chromosome 11 has been shown to result in LOH of one or more markers on the short arm of the chromosome in several HCCs. Deletion of a DNA segment without its replication has been proposed as the most likely mechanism for a LOH that may involve tumor suppressor genes [118]. Like HPV and retroviruses, HBV integrates near oncogenes and thereby induces alterations in their structure or function. For example, the proto-oncogenes INT2 and HST1 are often amplified in HCC [117]. HBV integration into the gene for cyclin A, which plays an important role in the control of cell division, may also contribute to hepatocarcinogenesis [119]. The specific integration of HBV into the telomerase reverse transcriptase (hTERT) gene was recently demonstrated in HCC. HBV integration is an early event in hepatocarcinogenesis and genes affected by virus integration are important determinants of cell immortalization. As a major component of telomerase, hTERT is responsible for maintaining telomere length during cell immortalization. Activation of the hTERT gene in cis as a result of HBV integration into its promoter region was first demonstrated in an HCC cell line, suggesting a causative role for telomerase in hepatocarcinogenesis [120]. Subsequently, three additional cases of HBV integration into the hTERT gene in HCC were identified [121,122]. Amplification of chromosomal region 11q13 is a frequent alteration in HCC, although it is more common in recurrent HCC than in the initial tumors (123,124). In addition, amplification at 11q13 corresponding with FRA11H, appears to be related to virus infection and geographical location. In one study, for example, 11q13 amplification was preferentially detected in HBV-positive HCCs, whereas another study of Japanese individuals revealed a higher incidence of 11q13 gain in hepatitis C virus–related HCCs [125,126]. HBV DNA integration near the HSTF1 (FGF4) gene at 11q13 was also shown to trigger coamplification of both viral and cellular DNA [127].
Epstein-Barr virus
Epstein-Barr virus (EBV) is a lymphotrophic DNA virus that is associated with the development of BL and nasopharingeal carcinoma. EBV also induces immortalization of human B lymphocytes in vitro. Studies with newly established cell lines have shown that EBV frequently persists by integrating into the host DNA [128]. Episomal and integrated EBV are distinguishable by FISH, given that only integrated sequences yield fluorescence signals on both sister chromatids at identical positions. FISH revealed the localization of EBV in two BL lines: The Namalva line, which was originally thought to contain only one copy of EBV, actually harbors two closely spaced copies integrated at chromosome 1p35 [129], whereas a BL cell line derived from a North American patient contains a single copy of EBV integrated at chromosome 2p13. Virus insertion into the chromosomal DNA of the latter cell line resulted in a stable modification site apparent as a distinctive achromatic region adjacent to band 2p13. Consistent with the pattern of integration of other DNA viruses, the EBV integration site at 2p13 overlaps with the locations of FRA2E, the REL proto-oncogene, and the transforming growth factor–a gene [130]. The induction of achromatic lesions may reflect EBV-induced genomic damage that, together with frequent MYC activation due to specific chromosome translocations in BL, might provide a selective growth advantage.
Adeno-associated virus
The introduction of genes into cells to correct hereditary defects or to suppress cancer is an emerging field of research. Retroviruses or genetically engineered vectors containing subgenomic viral sequences are being studied for their ability to mediate such gene delivery and possible adverse effects. The adeno-associated virus (AAV) genome preferentially integrates into a specific region of chromosome 19. Somatic cell hybrid mapping revealed that 15 out of 22 clones of human cells latently infected with AAV contained viral sequences on chromosome 19, whereas subsequent FISH analysis identified the integration site on chromosome 19 to 19q13-qter overalping with FRA19A [131,132]. AAV integration at this same site on chromosome 19 was demonstrated independently with the use of somatic cell hybrids and FISH analysis in several latently infected cell lines. Furthermore, sequence analysis of several independent integration sites revealed breakpoints positioned within a 100-bp region of cellulat DNA [133]. Southern blot and FISH analysis of AAV integration in cervical tumor cells and HeLa cell clones demonstrated that, unlike the wild-type virus, recombinant viruses did not integrate at chromosome 19. Integration of the recombinant viruses did not appear to be random, however, given that three of the five unique clones studied exhibited similar integration sites.AAV integration was detected at sites close to known proto-oncogenes, including MYC, and FSs, showing that infection or transfection with recombinant AAV is a highly efficient means for gene delivery and integration. However, the properties of parental AAV are not retained by recombinant viruses and in certan cases is integrated at loci of cancer promoter genes [134]. The early cumulative cytogenetic data obtained with both DNA and RNA tumor viruses thus indicated that preferential integration at sites of fragility and recombination might be a general phenomenon and may represent a critical event in viral carcinogenesis [135]. More recent cytogenetic and molecular studies have provided considerably more evidence for these conclusions.

Conclusions

Discoveries linking FSs and the development of cancer mark a new era in cancer research and should stimulate interest in these important regions of the human genome. Although not all FSs may be equally important in cancer development, the cloning of additional FSs associated with recurrent genomic alterations will likely lead to the identification of new oncogenes, tumor suppressor genes, and chimeric genes with oncogenic potential. It also may provide key insight into the mechanisms of FS instability and the impact of such instability on gene function as well as into the effects of oncogenic agents on FS expression and gene function. Lastly, the cataloging of expression profiles of genes located at FSs might provide the basis of a system for the detection, prediction, and determination of the prognosis of cancer as well as for identifying new targets for cancer therapy.

References
[1] N.C. Popescu, J.A. DiPaolo, S.C. Amsbaugh, Integration sites of human papillomavirus 18 DNA sequences on HeLa cell chromosomes. Cytogenet. Cell Genet 44 (1987) 58-62.
[2] M. Durst, C.M. Croce, L. Gissmann, E. Schwarz, K. Huebner, Papillomavirus sequences integrate near cellular oncogenes in some cervical carcinomas. Proc Natl Acad Sci U S A 84 (1987) 1070-1074.
[3] D.I. Smith, H. Huang, L. Wang, Common fragile sites and cancer. Int J Oncol 12 (1998) 187-196.
[4] G.R. Sutherland, E. Baker, R.I. Richards, Fragile sites still breaking. Trends Genet 14 (1998) 501-506.
[5] K. Huebner, C.M. Croce, FRA3B and other common fragile sites: the weakest links. Nat Rev Cancer 1 (2001) 214-221.
[6] R.I. Richards, Fragile and unstable chromosomes in cancer: causes and consequences. Trends Genet 17 (2001) 339-345.
[7] T.W. Glover, C. Berger, J. Coyle, B. Echo, DNA polymerase alpha inhibition by aphidicolin induces gaps and breaks at common fragile sites in human chromosomes. Hum Genet 67 (1984) 136-142.
[8] G.R. Sutherland, Chromosomal fragile sites. Genet Anal Tech Appl 8 (1991) 161-166.
[9] F. Hecht, Fragile sites, cancer chromosome breakpoints, and oncogenes all cluster in light G bands. Cancer Genet Cytogenet 31 (1988) 17-24.
[10] L. Manuelidis, A view of interphase chromosomes. Science 250 (1990) 1533-1540.
[11] I. Simonic, G.S. Gericke, The enigma of common fragile sites. Hum Genet 97 (1996) 524-531.
[12] C. Laird, E. Jaffee, G. Karpen, M. Lamb R. Nelson, Fragile sites in chromosomes as regions of late replicating DNA. Trends Genet., 3 (1987) 274-281.
[13] B. Porfirio, B. Tedeschi, P. Vernole, D. Caporossi, B. Nicoletti, The distribution of MspI-induced breaks in human lymphocyte chromosomes and its relationship to common fragile sites. Mutat Res 213 (1989) 117-124.
[14] M.M. Le Beau, F.V. Rassool, M.E. Neilly, R. Espinosa, 3rd, T.W. Glover, D.I. Smith, T.W. McKeithan, Replication of a common fragile site, FRA3B, occurs late in S phase and is delayed further upon induction: implications for the mechanism of fragile site induction. Hum Mol Genet 7 (1998) 755-761.
[15] A. Hellman, A. Rahat, S.W. Scherer, A. Darvasi, L.C. Tsui, B. Kerem, Replication delay along FRA7H, a common fragile site on human chromosome 7, leads to chromosomal instability. Mol Cell Biol 20 (2000) 4420-4427.
[16] T. Boehm, L. Mengle-Gaw, U.R. Kees, N. Spurr, I. Lavenir, A. Forster, T.H. Rabbitts, Alternating purine-pyrimidine tracts may promote chromosomal translocations seen in a variety of human lymphoid tumours. Embo J 8 (1989) 2621-2631.
[17] N. Arnheim, The possible role of Z DNA in chromosomal translocations. Cancer Cells 2 (1990) 121-122.
[18] A.J. Verkerk, M. Pieretti, J.S. Sutcliffe, Y.H. Fu, D.P. Kuhl, A. Pizzuti, O. Reiner, S. Richards, M.F. Victoria, F.P. Zhang, B.E. Eussen, G.J.B. van Ommen, L.A.J. Blonden, G.J. Riggins, J.L. Chastain, C.B. Kunst, H. Galjaard, C.T. Caskey, D.L. Nelson, B.A. Oostra S.T. Warren, Identification of a gene (FMR-1) containing a CGG repeat coincident with a breakpoint cluster region exhibiting length variation in fragile X syndrome. Cell 65 (1991) 905-914.
[19] S.J. Knight, A.V. Flannery, M.C. Hirst, L. Campbell, Z. Christodoulou, S.R. Phelps, J. Pointon, H.R. Middleton-Price, A. Barnicoat, M.E. Pembrey, J. Holland, B.A. Oostra, M. Bobrow, K.E. Davis, Trinucleotide repeat amplification and hypermethylation of a CpG island in FRAXE mental retardation. Cell 74 (1993) 127-134.
[20] J.E. Parrish, B.A. Oostra, A.J. Verkerk, C.S. Richards, J. Reynolds, A.S. Spikes, L.G. Shaffer, D.L. Nelson, Isolation of a GCC repeat showing expansion in FRAXF, a fragile site distal to FRAXA and FRAXE. Nat Genet 8 (1994) 229-235.
[21] J.K. Nancarrow, E. Kremer, K. Holman, H. Eyre, N.A. Doggett, D. Le Paslier, D.F. Callen, G.R. Sutherland, R.I. Richards, Implications of FRA16A structure for the mechanism of chromosomal fragile site genesis. Science 264 (1994) 1938-1941.
[22] C. Jones, P. Slijepcevic, S. Marsh, E. Baker, W.Y. Langdon, R.I. Richards, A. Tunnacliffe, Physical linkage of the fragile site FRA11B and a Jacobsen syndrome chromosome deletion breakpoint in 11q23.3. Hum Mol Genet 3 (1994) 2123-2130.
[23] F.V. Rassool, T.W. McKeithan, M.E. Neilly, E. van Melle, R. Espinosa, 3rd, M.M. Le Beau, Preferential integration of marker DNA into the chromosomal fragile site at 3p14: an approach to cloning fragile sites. Proc Natl Acad Sci U S A 88 (1991) 6657-6661.
[24] H. Inoue, H. Ishii, H. Alder, E. Snyder, T. Druck, K. Huebner, C.M. Croce, Sequence of the FRA3B common fragile region: implications for the mechanism of FHIT deletion. Proc Natl Acad Sci U S A 94 (1997) 14584-14589.
[25] C.M. Wilke, B.K. Hall, A. Hoge, W. Paradee, D.I. Smith, FRA3B extends over a broad region and contains a spontaneous HPV16 integration site: direct evidence for the coincidence of viral integration sites and fragile sites. Hum. Mol. Genet. 5 (1996) 187-195.
[26] N.C. Popescu, D. Turnbull, J. A. DiPaolo, Sister chromatid exchange and chromosome aberration analysis with the use of several carcinogens and noncarcinogens. J. Natl. Cancer Inst 59 (1977) 289-293.
[27] N.C. Popescu, Chromosome fragility and instability in human cancer. Crit. Rev. Oncog. 5 (1994) 121-140.
[28] B. Hirsch, Sister chromatid exchanges are preferentially induced at expressed and nonexpressed common fragile sites. Hum Genet 87 (1991) 302-306.
[29] T. Lukusa, E. Meulepas, J.P. Fryns, H. Van den Berghe, J.J. Cassiman, "Spontaneous" FRA16B is a hot spot for sister chromatid exchanges. Hum Genet 87 (1991) 583-586.
[30] H. Tsuji, A. Hitomi, E. Takahashi, M. Murata, T. Ikeuchi, K. Yamamoto, S. Tsuji, T. Hori, Induction of distamycin A-inducible rare fragile sites and increased sister chromatid exchanges at the fragile site. Hum Genet 87 (1991) 254-260.
[31] J.M. Bishop, The molecular genetics of cancer. Science 235 (1987) 305-311.
[32] D. Hanahan, R.A. Weinberg, The hallmarks of cancer. Cell 100 (2000) 57-70.
[33] A.A. Sandberg, The chromosomes in human cancer and leukemia. New York:Elsevier Publishing Co., Inc., Second Edition (1990
[34] F. Mitelman, F. Mertens, A breakpoint map of recurrent chromosomal rearrangements in human neoplasia. Nat. Genet. 15 (1997) 417-474.
[35] J.J. Yunis, A.L. Soreng, Constitutive fragile sites and cancer. Science 226 (1984) 1199-1204.
[36] J.D. Rowley, Human oncogene locations and chromosome aberrations. Nature 301 (1983) 290-291.
[37] M.M. Le Beau, Chromosomal fragile sites and cancer-specific breakpoints--a moderating viewpoint. Cancer Genet Cytogenet 31 (1988) 55-61.
[38] F. Shabtai, A. Sternberg, D. Klar, I. Halbrecht, R. Reiss, Familial fragile 8q22 involved as a cancer breakpoint in cells of a large bowel tumor. Cancer Genet Cytogenet 31 (1988) 113-118.
[39] T.H. Rabbits, Chromosomal translocations in human cancers. Nature 372 (1994) 143-149.
[40] C.M. Croce, P.C. Nowell, Molecular basis of human B cell neoplasia. Blood 65 (1985) 1-7.
[41] D.B. Zimonjic, C. Keck-Waggoner, N.C. Popescu, Novel genomic imbalances and chromosome translocations involving c-myc gene in Burkitt's lymphoma. Leukemia 15 (2001) 1582-1588.
[42] T.K. Blackwell, F.W. Alt, Molecular characterization of the lymphoid V(D)J recombination activity. J Biol Chem 264 (1989) 10327-10330.
[43] R.L. Nussbaum, S.D. Airhart, D.H. Ledbetter, Recombination and amplification of pyrimidine-rich sequences may be responsible for initiation and progression of the Xq27 fragile site: an hypothesis. Am J Med Genet 23 (1986) 715-721.
[44] M.N. Nikiforova, J.R. Stringer, R. Blough, M. Medvedovic, J.A. Fagin, Y.E. Nikiforov, Proximity of chromosomal loci that participate in radiation-induced rearrangements in human cells. Science 290 (2000) 138-141.
[45] G.R. Sutherland, E. Baker, The clinical significance of fragile sites on human chromosomes. Clin Genet 58 (2000) 157-161.
[46] R.M. Gemmill, J.D. West, F. Boldog, N. Tanaka, L.J. Robinson, D.I. Smith, F. Li, H.A. Drabkin, The hereditary renal cell carcinoma 3;8 translocation fuses FHIT to a patched-related gene, TRC8. Proc Natl Acad Sci U S A 95 (1998) 9572-9577.
[47] C. Popovici, C. Basset, F. Bertucci, B. Orsetti, J. Adélaide, M.J. Mozziconacci, N. Conte, A. Murati, C. Ginestier, E. Charafe-Jauffret, S. P. Ethier, M. Lafage-Pochitaloff, C. Theillet, D. Birnbaum, M. Chaffanet, Reciprocal translocations in breast tumor cell lines: Cloning of a t(3;20) that targets the FHIT gene. Genes Chromosomes Cancer 35 (2002) 204-218.
[48] J.M. Geurts, E.F. Schoenmakers, E. Roijer, G. Stenman, W.J. Van de Ven, Expression of reciprocal hybrid transcripts of HMGIC and FHIT in a pleomorphic adenoma of the parotid gland. Cancer Res 57 (1997) 13-17.
[49] J.M. Fang, M.F. Arlt, A.C. Burgess, S.L. Dagenais, D.G. Beer, T.W. Glover, Translocation breakpoints in FHIT and FRA3B in both homologs of chromosome 3 in an esophageal adenocarcinoma. Genes Chromosomes Cancer 30 (2001) 292-298.
[50] C.L. Keck, D.B. Zimonjic, B.Z. Yuan, S.S. Thorgeirsson, N.C. Popescu, Nonrandom breakpoints of unbalanced chromosome translocations in human hepatocellular carcinoma cell lines. Cancer Genet Cytogenet 111 (1999) 37-44.
[51] B.Z. Yuan, C. Keck-Waggoner, D.B. Zimonjic, S.S. Thorgeirsson, N.C. Popescu, Alterations of FHIT gene in human hepatocellular carcinoma. Cancer Res 60 (2000) 1049-1053.
[52] M. Mangelsdorf, K. Ried, E. Woollatt, S. Dayan, H. Eyre, M. Finnis, L. Hobson, J. Nancarrow, D. Venter, E. Baker, R.I. Richards, Chromosomal fragile site FRA16D and DNA instability in cancer. Cancer Res 60 (2000) 1683-1689.
[53] A.K. Bednarek, C.L. Keck-Waggoner, R.L. Daniel, K.J. Laflin, P.L. Bergsagel, K. Kiguchi, A.J. Brenner, C.M. Aldaz, WWOX, the FRA16D gene, behaves as a suppressor of tumor growth. Cancer Res 61 (2001) 8068-8073.
[54] M. Chesi, P.L. Bergsagel, O.O. Shonukan, M.L. Martelli, L.A. Brents, T. Chen, E. Schrock, T. Ried, W.M. Kuehl, Frequent dysregulation of the c-maf proto-oncogene at 16q23 by translocation to an Ig locus in multiple myeloma. Blood 91 (1998) 4457-4463.
[55] J. Podolsky, T. Byrsky, S. Zajackzek, T. Druck, D.B. Zimonjic, N.C. Popescu, K.G. Lubinski, A. Borowka, J. Gronwald, I.J. Lubinsky, Characterization of a familial RCC-associated T(2;3) (q33;q21) chromosome translocation. J Hum Genet 48 (2001) 685-693.
[56] T. Druck, J. Podolski, T. Byrski, L. Wyrwicz, S. Zajaczek, G. Kata, A. Borowka, J. Lubinski, K. Huebner, The DIRC1 gene at chromosome 2q33 spans a familial RCC-associated t(2;3)(q33;q21) chromosome translocation. J Hum Genet 46 (2001) 583-589.
[57] H.M. Padilla-Nash, K. Heselmeyer-Haddad, D. Wangsa, H. Zhang, B.M. Ghadimi, M. Macville, M. Augustus, E. Schröck, E. Hilgenfeld, T. Ried, Jumping translocations are common in solid tumor cell lines and result in recurrent fusions of whole chromosome arms. Genes Chromosomes Cancer 30 (2001) 349-363.
[58] A.G. Knudson, Antioncogenes and human cancer. Proc Natl Acad Sci U S A 90 (1993) 10914-10921.
[59] P.A. Jones, S.B. Baylin, The fundamental role of epigenetic events in cancer. Nat Rev Genet 3 (2002) 415-428.
[60] J.L. Biedler, B.A. Spengler, Metaphase chromosome anomaly: association with drug resistance and cell- specific products. Science 191 (1976) 185-187.
[61] K. Alitalo, M. Schwab, Oncogene amplification in tumor cells. Adv. Cancer Res 47 (1986) 235-281.
[62] R.T. Schimke, Gene amplification in cultured cells. J Biol Chem 263 (1988) 5989-5992.
[63] B. Windle, B.W. Draper, Y.X. Yin, S. O'Gorman, G.M. Wahl, A central role for chromosome breakage in gene amplification, deletion formation, and amplicon integration. Genes Dev 5 (1991) 160-174.
[64] R.J. Kaufman, R.T. Schimke, Amplification and loss of dihydrofolate reductase genes in a Chinese hamster ovary cell line. Mol Cell Biol 1 (1981) 1069-1076.
[65] M.L. Pardue, Dynamic instability of chromosomes and genomes. Cell 66 (1991) 427-431.
[66] B. McClintock, Chromosome organization and genetic expression. Cold Sring Harb.Symp. Quant.Biol.16 (1951) 13-47.
[67] A. Coquelle, E. Pipiras, F. Toledo, G. Buttin, M. Debatisse, Expression of fragile sites triggers intrachromosomal mammalian gene amplification and sets boundaries to early amplicons. Cell 89 (1997) 215-225.
[68] H. Zitzelsberger, L. Lehmann, M. Werner, M. Bauchinger, Comparative genomic hybridisation for the analysis of chromosomal imbalances in solid tumours and haematological malignancies. Histochem. Cell Biol 108 (1997) 403-417.
[69] E. Schuuring, The involvement of the chromosome 11q13 region in human malignancies: cyclin D1 and EMS1 are two new candidate oncogenes--a review. Gene 159 (1995) 83-96.
[70] M.I. Shuster, L. Han, M.M. Le Beau, E. Davis, M. Sawicki, C.M. Lese, N.H. Park, J. Colicelli, S.M. Gollin, A consistent pattern of RIN1 rearrangements in oral squamous cell carcinoma cell lines supports a breakage-fusion-bridge cycle model for 11q13 amplification. Genes, Chromosomes and Cancer 28 (2000) 153-163.
[71] A. Hellman, E. Zlotorynski, S.W. Scherer, J. Cheung, J.B. Vincent, D.I. Smith, L. Trakhtenbrot, B. Kerem, A role for common fragile site induction in amplification of human oncogenes. Cancer Cell 1 (2002) 89-97.
[72] D.B. Zimonjic, C.L. Keck-Waggoner, B.Z. Yuan, M.H. Kraus, N.C. Popescu, Profile of genetic alterations and tumorigenicity of human breast cancer cells. Int J Oncol 16 (2000) 221-230.
[73] D.B. Zimonjic, H. Zhang, Z. Shan, V.M. Factor, J. Trent, S.S. Thorgeirsson, N.C. Popescu, DNA amplification associated with double minutes originating from chromosome 19 in mouse hepatocellular carcinoma. Cytogenet. Cell Genet (2001)93: 114-116.
[74] F.F. Elder, T.J. Robinson, Rodent common fragile sites: are they conserved? Evidence from mouse and rat. Chromosoma 97 (1989) 459-464.
[75] D.B. Zimonjic, T. Druck, M. Ohta, K. Kastury, C.M. Croce, N.C. Popescu, K. Huebner, Positions of chromosome 3p14.2 fragile sites (FRA3B) within the FHIT gene. Cancer Res 57 (1997) 1166-1170.
[76] S. Corbin, M.E. Neilly, R. Espinosa, 3rd, E.M. Davis, T.W. McKeithan, M.M. Le Beau, Identification of unstable sequences within the common fragile site at 3p14.2: implications for the mechanism of deletions within fragile histidine triad gene/common fragile site at 3p14.2 in tumors. Cancer Res 62 (2002) 3477-3484.
[77] L. Roz, M. Gramegna, H. Ishii, C.M. Croce, G. Sozzi, Restoration of fragile histidine triad (FHIT) expression induces apoptosis and suppresses tumorigenicity in lung and cervical cancer cell lines. Proc Natl Acad Sci U S A 99 (2002) 3615-3620.
[78] T.W. Glover, A.W. Hoge, D.E. Miller, J.E. Ascara-Wilke, A.N. Adam, S.L. Dagenais, C.M. Wilke, H.A. Dierick, D.G. Beer, The murine Fhit gene is highly similar to its human orthologue and maps to a common fragile site region. Cancer Res 58 (1998) 3409-3414.
[79] L.Y. Fong, V. Fidanza, N. Zanesi, L. Lock, L. Siracusa, R. Mancini, Z. Siprashvili, M. Ottey, S.E. Martin, T. Druck, P.A. McCue, C.M. Croce, K. Huebner, Muir-Torre-like syndrome in FHIT deficient mice. Proc. Natl. Acad. Sci. USA 97 (2000) 4742-4747.
[80] C.M. Croce, How can we prevent cancer? Proc Natl Acad Sci U S A 98 (2001) 10986-10988.
[81] M. Schwab, A. Claas, L. Savelyeva, BRCA2: a genetic risk factor for breast cancer. Cancer Lett 175 (2002) 1-8.
[82] N. C. Popescu, and D. B Zimonjic, Chromosome and gene alterations in breast cancer as markers for diagnosis and prognosis as well as targets for therapy. (2002) In press.
[83] L. Savelyeva, A. Claas, I. Matzner, P. Schlag, W. Hofmann, S. Scherneck, B. Weber, M. Schwab, Constitutional genomic instability with inversions, duplications, and amplifications in 9p23-24 in BRCA2 mutation carriers. Cancer Res 61 (2001) 5179-5185.
[84] J.T. Bergthorsson, J. Johannsdottir, A. Jonasdottir, G. Eiriksdottir, V. Egilsson, S. Ingvarsson, R.B. Barkardottir, Chromosome imbalance at the 3p14 region in human breast tumours: high frequency in patients with inherited predisposition due to BRCA2. Eur. J. Cancer 34 (1998) 142-147.
[85] B.C. Turner, M. Ottey, D.B. Zimonjic, M. Potoczek, W.W. Hauck, E. Pequignot, C.L. Keck-Waggoner, C. Sevignani, C.M. Aldaz, P.A. McCue, J. Palazzo, K. Huebner, N.C. Popescu, The fragile histidine triad/common chromosome fragile site 3B locus and repair-deficient cancers. Cancer Res 62 (2002) 4054-4060.
[86] K.A. Krummel, S.R. Denison, E. Calhoun, L.A. Phillips, D.I. Smith, The common fragile site FRA16D and its associated gene WWOX are highly conserved in the mouse at Fra8E1. Genes Chromosomes Cancer 34 (2002) 154-167.
[87] T. Shiraishi, T. Druck, K. Mimori, J. Flomenberg, L. Berk, H. Alder, W. Miller, K. Huebner, C.M. Croce, Sequence conservation at human and mouse orthologous common fragile regions, FRA3B/FHIT and Fra14A2/Fhit. Proc Natl Acad Sci U S A 98 (2001) 5722-5727.
[88] M.F. Arlt, D.E. Miller, D.G. Beer, T.W. Glover, Molecular characterization of FRAXB and comparative common fragile site instability in cancer cells. Genes Chromosomes Cancer 33 (2002) 82-92.
[89] J.J. Yunis, A.L. Soreng, A.E. Bowe, Fragile sites are targets of diverse mutagens and carcinogens. Oncogene 1 (1987) 59-69.
[90] J.J. Yunis, W.R. Hoffman, Nuclear enzymes, fragile sites, and cancer. J Gerontol 44 (1989) 37-44.
[91] J. Whang-Peng, C.S. Kao-Shan, E.C. Lee, P.A. Bunn, D.N. Carney, A.F. Gazdar, J.D. Minna, Specific chromosome defect associated with human small-cell lung cancer; deletion 3p(14-23). Science 215 (1982) 181-182.
[92] G. Sozzi, L. Sard, L. De Gregorio, A. Marchetti, K. Musso, F. Buttitta, S. Tornielli, S. Pellegrini, M.L. Veronese, G. Manenti, M. Incarbone, A. Chella, C.A. Angeletti, U. Pastorino, K. Huebner, G. Bevilaqua, S. Pilotti, C.M. Croce, Association between cigarette smoking and FHIT gene alterations in lung cancer. Cancer Res 57 (1997) 2121-2123.
[93] C.K. Stein, T.W. Glover, J.L. Palmer, B.S. Glisson, Direct correlation between FRA3B expression and cigarette smoking. Genes Chromosomes Cancer 34 (2002) 333-340.
[94] S.R. Denison, N.A. Becker, M.J. Ferber, L.A. Phillips, K.R. Kalli, J. Lee, J. Lillie, D.I. Smith, V. Shridhar, Transcriptional profiling reveals that several common fragile-site genes are downregulated in ovarian cancer. Genes Chromosomes Cancer 34 (2002) 406-415.
[95] R.A. Weinberg, Integrated genomes of animal viruses. Annu. Rev. Biochem. 49 (1980) 197-226.
[96] C.M. Croce, Integration of oncogenic viruses in mammalian cells. Int Rev Cytol 71 (1981) 1-17.
[97] P. Lazo, J.A. DiPaolo, N.C. Popescu, Amplification of the integrated viral transforming genes of human papillomavirus 18 and 5' flanking cellular sequence located near the myc proto-oncogene in HeLa cells. Cancer Res 49 (1989) 4305-4310.
[98] G. Riou, M.G. Le, M. Favre, D. Jeannel, J. Bourhis, G. Orth, Human papillomavirus-negative status and c-myc gene overexpression: independent prognostic indicators of distant metastasis for early-stage invasive cervical cancers. J Natl Cancer Inst 84 (1992) 1525-1526.
[99] M. De Braekeleer, C. Sreekantaiah, O. Haas, Herpes simplex virus and human papillomavirus sites correlate with chromosomal breakpoints in human cervical carcinoma. Cancer Genet Cytogenet 59 (1992) 135-137.
[100] C. Paz-y-Mino, L. Ocampo, R. Narvaez, L. Narvaez, Chromosome fragility in lymphocytes of women with cervical uterine lesions produced by human papillomavirus. Cancer Genet Cytogenet 59 (1992) 173-176.
[101] N.C. Popescu, J.A. DiPaolo, Preferential sites for viral integration on mammalian genome. Cancer Genet Cytogenet 42 (1989) 157-171.
[102] L. Pirisi, K.E. Creek, J. Doniger, J.A. DiPaolo, Continuous cell lines with altered growth and differentiation properties originate after transfection of human keratinocytes with human papillomavirus type 16 DNA. Carcinogenesis 9 (1988) 1573-1579.
[103] N.C. Popescu, J.A. Dipaolo,Integration of human papillomavirus 16 DNA and genomic rearrangements in immortalized human keratinocyte lines. Cancer Res 50 (1990) 1316-1323.
[104] M. von Knebel Doeberitz, T. Bauknecht, D. Bartsch, H. zur Hausen, Influence of chromosomal integration on glucocorticoid-regulated transcription of growth-stimulating papillomavirus genes E6 and E7 in cervical carcinoma cells. Proc Natl Acad Sci U S A 88 (1991) 1411-1415.
[105] D.B. Zimonjic, N.D. Popescu, J.A. DiPaolo, Chromosomal organization of viral integration sites in human papillomavirus-immortalized human keratinocyte cell lines. Cancer Genet Cytogenet 72 (1994) 39-43.
[106] M.I. Gallego, D.B. Zimonjic, N.C. Popescu, J.A. DiPaolo, P.A. Lazo, Integration site of human papillomavirus type-18 DNA in chromosome band 8q22.1 of C4-I cervical carcinoma: DNase I hypersensitivity and methylation of cellular flanking sequences. Genes Chromosomes Cancer 9 (1994) 28-32.
[107] S. Reuter, M. Bartelmann, M. Vogt, C. Geisen, I. Napierski, T. Kahn, H. Delius, P. Lichter, S. Weitz, B. Korn, E. Schwarz, APM-1, a novel human gene, identified by aberrant co-transcription with papillomavirus oncogenes in a cervical carcinoma cell line, encodes a BTB/POZ-zinc finger protein with growth inhibitory activity. Embo J 17 (1998) 215-222.
[108] P. Peitsaro, S. Hietanen, B. Johansson, T. Lakkala, S. Syrjanen, Single copy heterozygote integration of HPV 33 in chromosomal band 5p14 is found in an epithelial cell clone with selective growth advantage. Carcinogenesis 23 (2002) 1057-1064.
[109] M. Macville, E. Schrock, H. Padilla-Nash, C. Keck, B.M. Ghadimi, D. Zimonjic, N. Popescu, T. Ried, Comprehensive and definitive molecular cytogenetic characterization of HeLa cells by spectral karyotyping. Cancer Res 59 (1999) 141-150.
[110] A.A. Brink, J.C. Wiegant, K. Szuhai, H.J. Tanke, G.G. Kenter, G.J. Fleuren, E. Schuuring, A.K. Raap, Simultaneous mapping of human papillomavirus integration sites and molecular karyotyping in short-term cultures of cervical carcinomas by using 49-color combined binary ratio labeling fluorescence in situ hybridization. Cancer Genet Cytogenet 134 (2002) 145-150.
[111] E.C. Thorland, S.L. Myers, D.H. Persing, G. Sarkar, R.M. McGovern, B.S. Gostout, D.I. Smith, Human papillomavirus type 16 integrations in cervical tumors frequently occur in common fragile sites. Cancer Res 60 (2000) 5916-5921.
[112] M.H. Einstein, Y. Cruz, M.K. El-Awady, N.C. Popescu, J.A. DiPaolo, M. van Ranst, A.S. Kadish, S. Romney, C.D. Runowicz, R.D. Burk, Utilization of the human genome sequence localizes human papillomavirus type 16 DNA integrated into the TNFAIP2 gene in a fatal cervical cancer from a 39-year-old woman. Clin Cancer Res 8 (2002) 549-554.
[113] S. Jeon, B.L. Allen-Hoffmann, P.F. Lambert, Integration of human papillomavirus type 16 into the human genome correlates with a selective growth advantage of cells. J Virol 69 (1995) 2989-2997.
[114] T. Moroy, A. Marchio, J. Etiemble, C. Trepo, P. Tiollais, M.A. Buendia, Rearrangement and enhanced expression of c-myc in hepatocellular carcinoma of hepatitis virus infected woodchucks. Nature 324 (1986) 276-279.
[115] Y. Mizuno, S. Murakami, F. Matsushita, M. Unoura, K. Kobayashi, S. Migita, N. Hattori, S. Ohno, Chromosomal assignment of woodchuck hepatitis virus (WHV) DNA integration sites in a woodchuck hepatocellular carcinoma-derived cell line (WH257GE10). Int J Cancer 43 (1989) 652-657.
[116] N.C. Popescu, D.B. Zimonjic, Chromosome-mediated alterations of the MYC gene in human cancer. J Cell Mol Med 6 (2002) 151-159.
[117] T. Tokino, K. Matsubara, Chromosomal sites for hepatitis B virus integration in human hepatocellular carcinoma. J Virol 65 (1991) 6761-6764.
[118] H.P. Wang, C.E. Rogler, Deletions in human chromosome arms 11p and 13q in primary hepatocellular carcinomas. Cytogenet Cell Genet 48 (1988) 72-78.
[119] J. Wang, X. Chenivesse, B. Henglein, C. Brechot, Hepatitis B virus integration in a cyclin A gene in a hepatocellular carcinoma. Nature 343 (1990) 555-557.
[120] I. Horikawa, J.C. Barrett, cis-Activation of the human telomerase gene (hTERT) by the hepatitis B virus genome. J Natl Cancer Inst 93 (2001) 1171-1173.
[121] D. Gozuacik, Y. Murakami, K. Saigo, M. Chami, C. Mugnier, D. Lagorce, T. Okanoue, T. Urashima, C. Brechot, P. Paterlini-Brechot, Identification of human cancer-related genes by naturally occurring Hepatitis B Virus DNA tagging. Oncogene 20 (2001) 6233-6240.
[122] C. Yu, D. Montoya, D.I. Smith, L.R.Roberts. Preferential integration of hepatitis B virus (HBV) into human telomerase reverse transcriptase (hTERT) gene in hepatocellular carcinoma. Proc .Am Ass. Cancer Res. (2002)150.
[123] D.B. Zimonjic, C.L. Keck, S.S. Thorgeirsson, N.C. Popescu, Novel recurrent genetic imbalances in human hepatocellular carcinoma cell lines identified by comparative genomic hybridization. Hepatology 29 (1999) 1208-1214.
[124] Y.J. Chen, S.H. Yeh, J.T. Chen, C.C. Wu, M.T. Hsu, S.F. Tsai, P.J. Chen, C.H. Lin, Chromosomal changes and clonality relationship between primary and recurrent hepatocellular carcinoma. Gastroenterology 119 (2000) 431-440.
[125] N. Kusano, K. Shiraishi, K. Kubo, A. Oga, K. Okita, Genetic aberrations detected by comparative genomic hybridization in hepatocellular carcinomas: their relationship to clinicopathological features. Hepatology 29 (1999) 1858-1862.
[126] N. Wong, P. Lai, E. Pang, L.F. Fung, Z. Sheng, V. Wong, W. Wang, Y. Hayashi, E. Perlman, J.W. Lau, Genomic aberrations in human hepatocellular carcinomas of differing etiologies. Clin. Cancer Res 6 (2000) 4000-4009.
[127] I. Hatada, T. Tokino, T. Ochiya, K. Matsubara, Co-amplification of integrated hepatitis B virus DNA and transforming gene hst-1 in a hepatocellular carcinoma. Oncogene 3 (1988) 537-540.
[128] E.A. Hurley, S. Agger, J.A. McNeil, J.B. Lawrence, A. Calendar, G. Lenoir, D.A. Thorley-Lawson, When Epstein-Barr virus persistently infects B-cell lines, it frequently integrates. J Virol 65 (1991) 1245-1254.
[129] J.B. Lawrence, C.A. Villnave, R.H. Singer, Sensitive, high-resolution chromatin and chromosome mapping in situ: presence and orientation of two closely integrated copies of EBV in a lymphoma line. Cell 52 (1988) 51-61.
[130] N.C. Popescu, M.C. Chen, S. Simpson, S. Solinas, J.A. DiPaolo, A Burkitt lymphoma cell line with integrated Epstein-Barr virus at a stable chromosome modification site. Virology 195 (1993) 248-251.
[131] R.M. Kotin, J.C. Menninger, D.C. Ward, K.I. Berns, Mapping and direct visualization of a region-specific viral DNA integration site on chromosome 19q13-qter. Genomics 10 (1991) 831-834.
[132] R.M. Kotin, M. Siniscalco, R.J. Samulski, X.D. Zhu, L. Hunter, C.A. Laughlin, S. McLaughlin, N. Muzyczka, M. Rocchi, K.I. Berns, Site-specific integration by adeno-associated virus. Proc Natl Acad Sci U S A 87 (1990) 2211-2215.
[133] R.J. Samulski, X. Zhu, X. Xiao, J.D. Brook, D.E. Housman, N. Epstein, L.A. Hunter, Targeted integration of adeno-associated virus (AAV) into human chromosome 19. Embo J 10 (1991) 3941-3950.
[134] E.D. Rivadeneira, N.C. Popescu, D.B. Zimonjic, G.S. Cheng, P.J. Nelson, M.D. Ross, J.A. DiPaolo, Sites of recombinant adeno-associated virus integration. Int. J. Oncol. 12 (1998) 805-810.
[135] N.C. Popescu, D. Zimonjic, J.A. DiPaolo, Viral integration, fragile sites and proto- oncogenes in human neoplasia. Hum. Genet. 84 (1989) 383-386.

Figure 1.Molecular cytogenetic characterization of fragile sites and oncogene amplification in cancer cells. (a) Aphidicolin induction of gaps and breaks in a human lymphoblast cell line deficient in mismatch repair, as revealed by G-banding and indicated by arrows. (b) The same metaphase as that in (a) after FISH with YAC probes for FRA3B (3p14.2) (green), FRA7G (7q32) (red), and FRA16D (16q23) (yellow), revealing lesions at these FSs. (c) Characterization of a large abnormal chromosome from a breast cancer cell harboring clusters of multiple copies of ERBB2 and MYC oncogenes. From left to right: FISH detection of multiple signals of ERBB2 and MYC, SKY hybridization of the abnormal chromosome, the G-banding pattern showing abnormally banded regions, and the classified SKY image depicting the derivation of the abnormality. Arrows indicate that the sites of oncogene clusters correspond to abnormally banded regions containing material from chromosome 17.( Reproduced with the permission of Cancer Research and Journal of Celullar and Molecular Medicine)

Back . . .