Molecular characterization of FRAXB and comparative common fragile site instability in cancer cells

FRA1E common fragile site breaks map within a 370kilobase pair region and disrupt the dihydropyrimidine dehydrogenase gene (DPYD)

Common fragile sites represent components of normal chromosome structure that are particularly prone to breakage under replication stress. Although the cytogenetic locations of 88 common fragile sites are listed in the Genome database, the DNA at only 14 of them has been defined and characterized at the molecular level. Here, we identify the precise genomic position of the common fragile site FRA1E, mapped to the chromosomal band 1p21.2, and characterize the genetic complexity of the fragile DNA sequence. We show that FRA1E extends over 370kb within the dihydropyrimidine dehydrogenase (DPYD) gene, which genomically spans approximately 840kb. The 185kb region of the highest fragility, which accounts for 86% of all observed breaks at FRA1E, encompasses the central part of DPYD including exons 13-16. DPYD encodes dihydropyrimidine dehydrogenase (DPD), which is the first and rate-limiting enzyme in a three-step metabolic pathway involved in degradation of the pyrimidine bases uracil and thymine. Deficiency in human DPD is associated with autosomal recessive disease, thymine-uraciluria, and with severe 5-fluorouracil toxicity in cancer patients. To which extent the disruption of the DPYD gene by the fragile site break is only transient, followed by DNA repair to restore the original structure, or occasionally may result in genomic damage associated with human disease remains to be determined.

The molecular basis of common and rare fragile sites

Fragile sites are specific loci that form gaps and constrictions on chromosomes exposed to partial replication stress. Fragile sites are classified as rare or common, depending on their induction and frequency within the population. These loci are known to be involved in chromosomal rearrangements in tumors and are associated with human diseases. Therefore, the understanding of the molecular basis of fragile sites is of high significance. Here we discuss the works performed in recent years that investigated the characteristics of fragile sites which underlie their inherent instability.

Low-frequency common fragile sites: Link to neuropsychiatric disorders?

Common fragile sites are unstable chromosomal regions that predispose chromosomes to breakage and rearrangements. Recombinogenic DNA sequences encompassing these sites may contribute to both germinal and somatic genomic mutations, and the genomic instability at these regions might cause severe inherited disorders or predispose to cancer. In this review, we discuss the characterization of common fragile site FRA13A within the neurobeachin gene, which is involved in development and function of the central nervous system. We raise the possibility of an implication of common fragile sites in neuropsychiatric disorders and overview previous and recent reports concerning individual variability of expression of common fragile sites in human populations.

Is mammalian chromosomal evolution driven by regions of genome fragility?

BACKGROUND

A fundamental question in comparative genomics concerns the identification of mechanisms that underpin chromosomal change. In an attempt to shed light on the dynamics of mammalian genome evolution, we analyzed the distribution of syntenic blocks, evolutionary breakpoint regions, and evolutionary breakpoints taken from public databases available for seven eutherian species (mouse, rat, cattle, dog, pig, cat, and horse) and the chicken, and examined these for correspondence with human fragile sites and tandem repeats.

RESULTS

Our results confirm previous investigations that showed the presence of chromosomal regions in the human genome that have been repeatedly used as illustrated by a high breakpoint accumulation in certain chromosomes and chromosomal bands. We show, however, that there is a striking correspondence between fragile site location, the positions of evolutionary breakpoints, and the distribution of tandem repeats throughout the human genome, which similarly reflect a non-uniform pattern of occurrence.

CONCLUSION

These observations provide further evidence that certain chromosomal regions in the human genome have been repeatedly used in the evolutionary process. As a consequence, the genome is a composite of fragile regions prone to reorganization that have been conserved in different lineages, and genomic tracts that do not exhibit the same levels of evolutionary plasticity.

Common fragile sites

Common fragile sites are regions showing site-specific gaps and breaks on metaphase chromosomes after partial inhibition of DNA synthesis. Common fragile sites are normally stable in somatic cells. However, following treatment of cultured cells with replication inhibitors, fragile sites display gaps, breaks, rearrangements and other features of unstable DNA. Studies showing that fragile sites and associated genes are frequently deleted or rearranged in many cancer cells have clearly demonstrated their importance in genome instability in cancer. Until recently, little was known about the molecular nature and mechanisms involved in fragile site instability. From studies conducted in many laboratories, it is now known that fragile sites extend over large regions, are associated with genes, exhibit delayed replication, and contain regions of high DNA flexibility. Recent findings from our laboratory showing that the key cell cycle checkpoint genes are important for genome stability at fragile sties have shed new light on these mechanisms and on the significance of these sites in cancer and normal chromosome structure. Since their discovery over two decades ago, much has been learned regarding their significance in chromosome structure and instability in cancer, but a number of key questions remain, including why these sites are 'fragile' and the impact of this instability on associated genes in cancer cells. These and other questions have been addressed by participants of this meeting, which highlighted instability at common fragile sites. This brief review is intended to provide background on common fragile sites that has led up to many of the studies presented in the accompanying reports in this volume and not to summarize the findings presented therein. Some aspects of this review were taken from Glover et al. (T.W. Glover, M.F. Arlt, A.M. Casper, S.G. Durkin, Mechanisms of common fragile site instability, Hum. Molec. Genet. 14 (in press). [1]).

Expression of common chromosomal fragile site genes, WWOX/FRA16D and FHIT/FRA3B is downregulated by exposure to environmental carcinogens, UV, and BPDE but not by IR

Common chromosomal fragile sites are unstable genomic loci susceptible to breakage, rearrangement, and are highly recombinogenic. Frequent alterations at these loci in tumor cells led to the hypothesis that they may contribute to cancer development. The two most common chromosomal fragile sites FRA16D and FRA3B which harbor WWOX and FHIT genes, respectively, are frequently altered in human cancers. Here we report that environmental carcinogens, ultraviolet (UV) light, and Benzo[a]pyrene diol epoxide (BPDE), significantly downregulate expression of both genes. On the other hand, we observe that ionizing radiation (IR) does not affect expression of these genes, suggesting that the effect of repression exerted by UV and BPDE is not just a consequence of DNA damage but may be a result of different signaling pathways triggered by specific DNA lesions. Such downregulation correlates with an induction of an S-phase delay in the cell cycle. Treatment of UV-irradiated cells with caffeine abrogates the S-phase delay while concomitantly overcoming the repression phenomenon. This suggests the involvement of unique cell cycle checkpoint mechanisms in the observed repression. Therefore, it is hypothesized that protracted downregulation of the putative tumor suppressor genes WWOX and FHIT by environmental carcinogens may constitute an additional mechanism of relevance in the initiation of tumorigenesis.

Premature condensation induces breaks at the interface of early and late replicating chromosome bands bearing common fragile sites

Various studies suggest a tight relationship between chromosome rearrangements driving tumor progression and breaks at loci called common fragile sites. Most of these sites are induced after perturbation of the replication dynamics, notably by aphidicolin treatment. We have mapped the majority of these sites to the interface of R and G bands, which calls into question the previous assignment of aphidicolin-sensitive sites to R bands. This observation suggests that most of them correspond to loci that ensure the transition between early and late replicating domains. We show that calyculin A, which triggers chromosome condensation at any phase of the cell cycle but does not markedly impair replication, induces damage in the chromosomes of human lymphocytes treated in G(2) but not in G(1) phase. We demonstrate that these lesions colocalize with those induced by aphidicolin treatment. Hence, common fragile site stability is compromised, whether aphidicolin delays replication or calyculin A advances condensation. We also show that, in cells that go through an unperturbed S phase, completion of their replication and/or replication-associated chromatin reorganization occur all along the G(2) phase, which may explain their inability to condense properly after calyculin A treatment during this phase of the cell cycle.

Homologous recombination and nonhomologous end-joining repair pathways regulate fragile site stability

Common fragile sites are specific loci that form gaps and constrictions on metaphase chromosomes exposed to replication stress, which slows DNA replication. These sites have a role in chromosomal rearrangements in tumors; however, the molecular mechanism of their expression is unclear. Here we show that replication stress leads to focus formation of Rad51 and phosphorylated DNA-PKcs, key components of the homologous recombination (HR) and nonhomologous end-joining (NHEJ), double-strand break (DSB) repair pathways, respectively. Down-regulation of Rad51, DNA-PKcs, or Ligase IV, an additional component of the NHEJ repair pathway, leads to a significant increase in fragile site expression under replication stress. Replication stress also results in focus formation of the DSB markers, MDC1 and gammaH2AX. These foci colocalized with those of Rad51 and phospho-DNA-PKcs. Furthermore, gammaH2AX and phospho-DNA-PKcs foci were localized at expressed fragile sites on metaphase chromosomes. These findings suggest that DSBs are formed at common fragile sites as a result of replication perturbation. The repair of these breaks by both HR and NHEJ pathways is essential for chromosomal stability at these sites.

The neurobeachin gene spans the common fragile site FRA13A

Common fragile sites are normal constituents of chromosomal structure prone to chromosomal breakage. In humans, the cytogenetic locations of more than 80 common fragile sites are known. The DNA at 11 of them has been defined and characterized at the molecular level. According to the Genome Database, the common fragile site FRA13A maps to chromosome band 13q13.2. Here, we identify the precise genomic position of FRA13A, and characterize the genetic complexity of the fragile DNA sequence. We show that FRA13A breaks are limited to a 650 kb region within the neurobeachin (NBEA) gene, which genomically spans approximately 730 kb. NBEA encodes a neuron-specific multidomain protein implicated in membrane trafficking that is predominantly expressed in the brain and during development.

Common chromosomal fragile sites and cancer: focus on FRA16D

A growing body of experimental evidence supports the view that certain human chromosomal fragile sites have roles to play in cancer. The principle lines of evidence are at the level of mutation mechanism and gene function. Most research in this area has previously focussed on the FRA3B common fragile site and the FHIT gene that spans this site. Here we review recent progress in characterising the second most readily observed common fragile site, FRA16D, and the WWOX gene that spans it. Comparative analyses of FRA3B/FHIT and FRA16D/WWOX reveal some striking similarities suggesting that these sites and their associated genes may play a part in a normal protective response of cells to environmental stress.

Mechanisms of common fragile site instability

The study of common fragile sites has its roots in the early cytogenetic investigations of the fragile X syndrome. Long considered an interesting component of chromosome structure, common fragile sites have taken on novel significance as regions of the genome that are particularly sensitive to certain forms of replication stress, which are frequently rearranged in cancer cells. In recent years, much has been learned about the genomic structure at fragile sites and the cellular checkpoint functions that monitor their stability. Recent findings suggest that common fragile sites may serve as markers of chromosome damage caused by replication stress during early stages of tumorigenesis. Thus, the study of common fragile sites can provide insight not only into the nature of fragile sites, but also into the broader consequences of replication stress on DNA damage and cancer. However, despite recent advances, many questions remain regarding the normal functional significance of these conserved regions and the basis of their fragility.

Common fragile sites, extremely large genes, neural development and cancer

Common fragile sites (CFSs) are large regions of profound genomic instability found in all individuals. They are biologically significant due to their role in a number of genomic alterations that are frequently found in many different types of cancer. The first CFS to be cloned and characterized was FRA3B, the most active CFS in the human genome. Instability within this region extends for over 4.0 Mbs and contained within the center of this CFS is the FHIT gene spanning 1.5 Mbs of genomic sequence. There are frequent deletions and other alterations within this gene in multiple tumor types and the protein encoded by this gene has been demonstrated to function as a tumor suppressor in vitro and in vivo. In spite of this, FHIT is not a traditional mutational target in cancer and many tumors have large intronic deletions without any exonic alterations. There are several other very large genes found within CFS regions including Parkin (1.37 Mbs in FRA6E), GRID2 (1.47 Mbs within 4q22.3), and WWOX (1.11 Mbs within FRA16D). These genes also appear to function as tumor suppressors but are not traditional mutational targets in cancer. Each of these genes is highly conserved and the regions spanning them are CFSs in mice. We have now examined lists of the largest human genes and found forty that span over one megabase. Many of these are derived from chromosomal bands containing CFSs. BACs within these genes are being utilized as FISH probes to determine if these are also CFS genes. Thus far we have identified the following as CFS genes: CNTNAP2 (2.3 Mbs in FRA7I), DMD (2.09 Mbs in FRAXC), LRP1B (1.9 Mbs in FRA2F), CTNNA3 (1.78 Mbs in FRA10D), DAB1 (1.55 Mbs in FRA1B), and IL1RAPL1 (1.36 Mbs in FRAXC). Although, these genes are also not traditional mutational targets in cancer they do exhibit loss of expression in multiple tumor types suggesting that they may also function as tumor suppressors. Many of the large CFS genes are involved in neurological development. Parkin is mutated in autosomal recessive juvenile Parkinsonism and deletions in mice are associated with the mouse mutant Quaking (viable). Spontaneous mouse mutants in GRID2 and DAB1 are associated with Lurcher and Reelin, respectively. In humans, alterations in IL1RAPL1 cause X-linked mental retardation and loss of WWOX is associated with Tau phosphorylation. We propose that the instability-induced alterations in these genes contribute to cancer development in a two-step process. Initial alterations will primarily occur within intronic regions, as these genes are greater than 99% intronic. These are not benign. Instead, they alter the repertoire of transcripts produced from these genes. As cancer progresses deletions will begin to encompass exons resulting in gene inactivation. These two types of alterations occurring in multiple large CFS genes may contribute significantly to the heterogeneity observed in cancer. There are also important potential linkages between normal neurological development and the development of cancer mediated by alterations in these genes.

Biallelic deletion and loss of expression analysis of genes at FRA2G common fragile site in tumor-derived cell lines

Common fragile sites (CFS) are regions of chromosome instability that show gaps or breaks when cells are exposed to particular culture condition. Much evidence suggests that CFSs are causally related to cancer as breakpoints in recurrent chromosome mutations and as sites of viral integration. We investigated the FRA2G CFS (2q31) for biallelic deletions and loss of expression in a panel of 19 tumor-derived cell lines. We found that Burkitt lymphoma-derived cell line DAUDI has a biallelic deletion of eight of the nine analyzed genes. Moreover, we observed loss of expression (LOE) of the DHRS9 gene (alias RDHL), one of the deleted genes in the DAUDI cell line, in MOLT-14 and Raji cell lines derived from Burkitt lymphoma and from T-cell acute lymphoblastic leukemia, respectively. DHRS9 is involved in development and differentiation pathways.

Homozygous deletions of CDKN2A caused by alternative mechanisms in various human cancer cell lines

The CDKN2A tumor-suppressor locus on chromosome band 9p21, which encodes p16(INK4A), a negative regulator of cyclin-dependent kinases, and p14(ARF1), an activator of TP53, is inactivated in many human cancers by point mutation, promoter hypermethylation, and, often, deletion. Homozygous deletions are unusually prevalent at this locus in very different human cancers. In the present study, we compared deletions in squamous cell carcinoma of the head and neck (SCCHN) cell lines to those in T-cell acute lymphatic leukemia (T-ALL), glioma, and bladder carcinoma (TCC) cell lines. Of 14 SCCHN lines, 10 showed homozygous deletions of CDKN2A, one displayed promoter hypermethylation with gene silencing, and one had a frameshift deletion in exon 2. Many deletion ends were in or proximal to the repetitive sequence clusters flanking the locus. Breakpoint junctions displayed variable microhomologies or insertions characteristic of DNA repair by nonhomologous end-joining. In general, deletions were much smaller in SCCHN than in TCC and glioma. In T-ALL, breakpoints were near consensus sites for recombination mediated by RAG (recombination activating genes) enzymes, and the structure of the junctions was consistent with this mechanism. We suggest that different mechanisms of CDKN2A deletion prevail in different human cancers. Aberrant RAG-mediated recombination may be responsible in T-ALL, and exuberant DNA repair by nonhomologous end-joining is the likely prevailing mechanism in SCCHN, but a distinct mechanism in TCC and glioma remains to be elucidated.

Increased frequency of multiradial chromosome structures in mouse embryonic fibroblasts lacking functional Werner syndrome protein and poly(ADP-ribose) polymerase-1

To determine whether the mouse Werner syndrome homologue (Wrn) and the poly (ADP-ribose) polymerase-1 (PARP-1) enzymes act in concert to prevent specific chromosomal rearrangements, mice with a mutation in the helicase domain of the Wrn gene (Wrn(Deltahel/Deltahel) mice) were crossed to PARP-1 null mice. Spectral karyotyping of the mouse metaphases was used in correlation with conventional G-banded karyotype analysis to precisely define the chromosomal aberrations in cells. Although there was no recurrent clonal chromosome aberration, PARP-1 null/Wrn(Deltahel/Deltahel) fibroblasts were distinguished by an increased frequency of chromatid breaks. Interestingly, multiradial structures were the only type of DNA rearrangement that was significantly higher in such PARP-1 null/Wrn(Deltahel/Deltahel) cells. These results indicate that Wrn and PARP-1 enzymes may be part of a protein complex involved in the processing of DNA breaks that can ultimately lead to multiradial structures when both enzymes are nonfunctional. Finally, regions of chromosomes known to be fragile sites in the mouse genome are not more prone to DNA rearrangements in the absence of both PARP-1 and functional Wrn proteins. Moreover, the low number of recurrent rearranged chromosome at any given site suggest a random mutagenesis process in PARP-1 null/Wrn(Deltahel/Deltahel) fibroblasts.

Alterations of common chromosome fragile sites in hematopoietic malignancies

Conditions of replication stress affect expression of all common fragile regions, including FRA3B (chromosome 3p14.2), FRA16D (16q23), FRA6E (6q26), FRA7G (7q31.2), and FRAXB (Xp22.3), and a number of cancer cell lines exhibit homozygous deletions in 2 or more common fragile regions. In 1996 the fragile histidine triad (FHIT) gene was isolated from the region encompassing the most active fragile FRA3B locus, and recently the WW domain-containing oxidoreductase gene (WWOX) was identified at FRA16D. These 2 fragile genes are altered or deleted in various epithelial tumors and exhibit tumor suppressor function. Aberration or absence of WWOX expression recently was detected in primary hematopoietic malignancies. The aberration resulted not only from genomic deletions but also possibly from epigenetic modifications associated with expression of fragility. Thus chromosomal aberrations at common fragile sites, in addition to the well-defined hallmark leukemia chromosome translocations, are involved in clinicopathological outcomes of hematopoietic malignancies.

BRCA1 is required for common-fragile-site stability via its G2/M checkpoint function

Common fragile sites are loci that form chromosome gaps or breaks when DNA synthesis is partially inhibited. Fragile sites are prone to deletions, translocations, and other rearrangements that can cause the inactivation of associated tumor suppressor genes in cancer cells. It was previously shown that ATR is critical to fragile-site stability and that ATR-deficient cells have greatly elevated fragile-site expression (A. M. Casper, P. Nghiem, M. F. Arlt, and T. W. Glover, Cell 111:779-789, 2002). Here we demonstrate that mouse and human cells deficient for BRCA1, due to mutation or knockdown by RNA interference, also have elevated fragile-site expression. We further show that BRCA1 functions in the induction of the G(2)/M checkpoint after aphidicolin-induced replication stalling and that this checkpoint function is involved in fragile-site stability. These data indicate that BRCA1 is important in fragile-site stability and that fragile sites are recognized by the G(2)/M checkpoint pathway, in which BRCA1 plays a key role. Furthermore, they suggest that mutations in BRCA1 or interacting proteins could lead to rearrangements at fragile sites in cancer cells.

Effect of exogenous E2F-1 on the expression of common chromosome fragile site genes, FHIT and WWOX

The expression of two tumor suppressor genes, fragile histidine triad (FHIT) and WW domain containing oxidoreductase (WWOX), encompassing common chromosome fragile regions, FRA3B at 3p14.2 and FRA16D at 16q23, is altered in many epithelial tumors. Since DNA sequence search shows that the FHIT gene has the E2F-1 recognition site in 5'] region, which regulates cell cycle, we tested the effect of E2F-1 overexpression in tumor cells. Ectopic E2F-1 expression led to an increase of Fhit and Wwox expression in allele remaining tumor cells and resulted in induction of apoptosis. Reporter assay showed that the E2F-1 site in FHIT 5' region was involved in the down-stream transcription after exogenous E2F-1 introduction. Chromatin immunoprecipitation detected exogenous E2F-1 binding to the recognition site in FHIT 5' region. The data suggest that E2F-1 overexpression plays a role in suppression of tumor, at least in part trough transcriptional regulation of FHIT and relevant activation of WWOX.

How common are common fragile sites in humans: interindividual variation in the distribution of aphidicolin-induced fragile sites

To obtain an estimate of the variation in common fragile sites (CFSs) among individuals, aphidicolin (APC)-induced chromosomal breakage data were analyzed for 20 karyotypically normal adult humans. As it is specifically designed to meet the analytical requirements for considering fragile sites as presence/absence characters in single individuals, the FSM methodology (Böhm et al., 1995) was used to statistically distinguish fragile from nonfragile sites. These analyses indicated that the APC-induced fragile sites are not ubiquitous but vary extensively among individuals; the per-individual number of fragile sites ranged from as few as seven to as many as 20. Of the 45 different sites identified as fragile, 19 (42%) occurred in more than half of the individuals, but only two sites (3p14 and 16q23) were fragile in all of the individuals; 12 (27% of the total) were fragile in single individuals only. Although these analyses provide statistical confirmation (and initial estimates of population variation) for 43 of the 88 APC-inducible fragile sites currently recognized as occurring among humans, they are consistent with the hypothesis that many of the currently recognized human CFSs have been erroneously identified. These results indicate the need for per-individual statistical identification of CFSs for larger samples of individuals and that studies of particular fragile sites should be conducted on individuals documented to be fragile at the loci under consideration.

The role of late/slow replication of the FRA16D in common fragile site induction

The FRA16D, at 16q23, spans the WWOX gene and is one of the most highly expressed common fragile sites observed when DNA replication is perturbed by aphidicolin. Several lines of evidence suggest that fragile sites are regions of DNA that are unusually sensitive to interference during replication. We have determined that the FRA16D alleles replicate in a synchronous fashion and that replication of these sequences occurs primarily in late S phase extending into G2 phase. Exposure to aphidicolin, an inhibitor of DNA polymerase alpha, results in a modest increase in cells with replication of FRA16D sequences in early S phase. This may represent initiation of replication in early S phase coupled with slow replication progression, or, alternatively, these cells may have passed through mitosis, entered the G1-S phase of the next cell cycle, and initiated replication/repair. Our results support a model in which common fragile sites are sequences that may initiate replication in early-mid S phase but are slow to complete replication, and the chromosomal breaks and gaps observed in metaphase cells result from unreplicated DNA.

Alterations in the common fragile site gene Parkin in ovarian and other cancers

The cloning and characterization of the common fragile site (CFS) FRA6E (6q26) identified Parkin, the gene involved in the pathogenesis of many cases of juvenile, early-onset and, rarely, late-onset Parkinson's disease, as the third large gene to be localized within a large CFS. Initial analyses of Parkin indicated that in addition to playing a role in Parkinson's disease, it might also be involved in the development and/or progression of ovarian cancer. These analyses also indicated striking similarities among the large CFS-locus genes: fragile histidine triad gene (FHIT; 3p14.2), WW domain-containing oxidoreductase gene (WWOX; 16q23), and Parkin (6q26). Analyses of FHIT and WWOX in a variety of different cancer types have identified the presence of alternative transcripts with whole exon deletions. Interestingly, various whole exon duplications and deletions have been identified for Parkin in juvenile and early-onset Parkinson's patients. Therefore, we performed mutational/exon rearrangement analysis of Parkin in ovarian cancer cell lines and primary tumors. Four (66.7%) cell lines and four (18.2%) primary tumors were identified as being heterozygous for the duplication or deletion of a Parkin exon. Additionally, three of 23 (13.0%) nonovarian tumor-derived cell lines were also identified as having a duplication or deletion of one or more Parkin exons. Analysis of Parkin protein expression with antibodies revealed that most of the ovarian cancer cell lines and primary tumors had diminished or absent Parkin expression. While functional analyses have not yet been performed for Parkin, these data suggest that like FHIT and WWOX, Parkin may represent a tumor suppressor gene.

Molecular basis for expression of common and rare fragile sites

Fragile sites are specific loci that form gaps, constrictions, and breaks on chromosomes exposed to partial replication stress and are rearranged in tumors. Fragile sites are classified as rare or common, depending on their induction and frequency within the population. The molecular basis of rare fragile sites is associated with expanded repeats capable of adopting unusual non-B DNA structures that can perturb DNA replication. The molecular basis of common fragile sites was unknown. Fragile sites from R-bands are enriched in flexible sequences relative to nonfragile regions from the same chromosomal bands. Here we cloned FRA7E, a common fragile site mapped to a G-band, and revealed a significant difference between its flexibility and that of nonfragile regions mapped to G-bands, similar to the pattern found in R-bands. Thus, in the entire genome, flexible sequences might play a role in the mechanism of fragility. The flexible sequences are composed of interrupted runs of AT-dinucleotides, which have the potential to form secondary structures and hence can affect replication. These sequences show similarity to the AT-rich minisatellite repeats that underlie the fragility of the rare fragile sites FRA16B and FRA10B. We further demonstrate that the normal alleles of FRA16B and FRA10B span the same genomic regions as the common fragile sites FRA16C and FRA10E. Our results suggest that a shared molecular basis, conferred by sequences with a potential to form secondary structures that can perturb replication, may underlie the fragility of rare fragile sites harboring AT-rich minisatellite repeats and aphidicolin-induced common fragile sites.

Novel perspective: focusing on the X chromosome in reproductive cancers

In an XX female, one of the two X chromosomes has been inactivated during early embryonic life to achieve a compensation of X-linked gene products between males and females, leaving only one allele of X-linked genes functional. There are some X-linked genes escaping the X-inactivation, i.e., being expressed from both alleles. Escape from X-inactivation varies at different levels; some genes have both alleles active in some women but only one allele active in others, whereas some other genes have both alleles active in neoplastic tissue but only one allele active normally. The X-inactivation may be considered functionally equivalent to a loss of heterozygosity (LOH) for some genes, whereas escape from X-inactivation may be equivalent to functional gene amplification for others. The physiological LOH may make X-linked tumor suppressor genes lose their function more easily, compared with autosomal tumor suppressor genes, thus predisposing women to cancer formation more easily. Moreover, the human X chromosome contains many genes related to cancer or to sex and reproduction. All these properties of the X chromosome suggest that it may play more important roles than any autosomal chromosome in the development and progression of reproductive and urologic cancers.

Pooled analysis of loss of heterozygosity in breast cancer: a genome scan provides comparative evidence for multiple tumor suppressors and identifies novel candidate regions

Somatic loss of heterozygosity (LOH) has been widely reported in breast cancer as a means of identifying putative tumor-suppressor genes. However, individual studies have rarely spanned more than a single chromosome, and the varying criteria used to declare LOH complicate efforts to formally differentiate regions of consistent versus sporadic (random) loss. We report here the compilation of an extensive database from 151 published LOH studies of breast cancer, with summary data from >15,000 tumors and primary allelotypes from >4,300 tumors. Allelic loss was evaluated at 1,168 marker loci, with large variation in the density of informative observations across the genome. Using studies in which primary allelotype information was available, we employed a likelihood-based approach with a formal chromosomal instability and selection model. The approach seeks direct evidence for preferential loss at each locus compared with nearby loci, accounts for heterogeneity across studies, and enables the direct comparison of candidate regions across the genome. Striking preferential loss was observed (in descending order of significance) in specific regions of chromosomes 7q, 16q, 13q, 17p, 8p, 21q, 3p, 18q, 2q, and 19p, as well as other regions, in many cases coinciding with previously identified candidate genes or known fragile sites. Many of these observations were not possible from any single LOH study, and our results suggest that many previously reported LOH results are not systematic or reproducible. Our approach provides a comparative framework for further investigation of regions exhibiting LOH and identifies broad genomic regions for which there exist few data.

Characterization of FRA6E and its potential role in autosomal recessive juvenile parkinsonism and ovarian cancer

Characterization of FRA6E (6q26), the third most frequently observed common fragile site (CFS) in the human population, determined that aphidicolin-induced instability at FRA6E extends over a very large region (3.6 Mb). Sequence analysis identified eight genes (IGF2R, SLC22A1, SLC22A2, SLC22A3, PLG, LPA, MAP3K4, and PARK2) as mapping within the large FRA6E region. PARK2, the gene associated with autosomal recessive juvenile parkinsonism (ARJP), accounts for more than half of the CFS. Homozygous deletions and large heterozygous deletions have been observed in PARK2 in ARJP patients. RT-PCR analysis of the eight genes localizing to FRA6E indicated that 50% of the genes, including PARK2, were down-regulated in one or more of the primary ovarian tumors analyzed. PARK2 expression was down-regulated in 60.0% of the primary ovarian tumors analyzed. Additionally, we found tumor-specific alternative transcripts of PARK2. Loss of heterozygosity analysis of primary ovarian tumors by use of polymorphic markers in the 6q26 region demonstrated 72% LOH in the center of the PARK2 gene, the highest of any of the markers tested. FRA6E shares many similarities with FRA3B (3p14.2) and FRA16D (16q23.2) in representing a large region of genomic instability and containing an extremely large gene that may play a role in the development of ovarian and many other cancers.

Genetic alterations in cancer as a result of breakage at fragile sites

The organization and replication of DNA render fragile sites (FSs) prone to breakage, recombination as well as becoming preferential targets for mutagens-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.

Common fragile sites are preferential targets for HPV16 integrations in cervical tumors

The development of cervical cancer is highly associated with human papillomavirus (HPV) infection. HPV integration into the genome of infected cervical cells is temporally associated with the acquisition of the malignant phenotype. A relationship between the sites of HPV integration in cervical cancer and the position of the common fragile sites (CFSs) has been observed at both the cytogenetic and molecular levels. To further explore this relationship at the molecular level, we used RS-PCR to rapidly isolate cellular sequences flanking the sites of HPV16 integration in 26 primary cervical tumors. Human bacterial artificial chromosome clones were isolated based on these flanking sequences and used as probes for fluorescence in situ hybridization on aphidicolin-stimulated metaphases. Our data demonstrate that 11/23 HPV16 integrations in cervical tumors occurred within CFSs (P&<0.001). In addition, we show that deletions and complex rearrangements frequently occur in the cellular sequences targeted by the integrations and that integrations cluster in FRA13C (13q22), FRA3B (3p14.2), and FRA17B (17q23). Finally, our data suggest that cellular genes, such as Notch 1, are disrupted by the HPV16 integrations, which may contribute to the malignant phenotype.

Characterization of the human common fragile site FRA2G

Common fragile sites are nonrandom loci that show gaps and breaks when cells are exposed to specific compounds. They are preferentially involved in recombination, chromosomal rearrangements, and foreign DNA integration. These sites have been suggested to play a role in chromosome instability observed in cancer. In this work we used a FISH-based approach to identify a BAC contig that spans the FRA2G fragile site located at the 2q31 region. Our observations indicate that a very fragile region spanning at least 450 kb is present within a large fragile region that extends over 1 Mb. At least seven genes are mapped in the fragile region. One of these seems to be a good candidate as a potential tumor suppressor gene impaired by the recurrent deletions observed at the 2q31 region in some neoplasms. In the fragile region, a considerable number of regions of high flexibility that may be related to the fragility are present.

Characterization of the common fragile site FRA9E and its potential role in ovarian cancer

Common fragile sites (CFSs) are regions of profound genomic instability that have been hypothesized to play a role in cancer. The major aim of this study was to locate a fragile region associated with ovarian cancer. Differential display (DD)-PCR analysis comparing normal ovarian epithelial cultures and ovarian cancer cell lines identified pregnancy-associated plasma protein-A (PAPPA) because of its frequent loss of expression (LOE) in ovarian cancer cell lines. PAPPA is localized to human chromosome 9q32-33.1, a region associated with significant loss of heterozygosity (LOH) in ovarian tumors (>50%) and in close proximity to the FRA9E CFS. FISH analysis determined that PAPPA was contained within the distal end of FRA9E. Characterization of FRA9E determined that aphidicolin-induced instability extended over 9 Mb, identifying FRA9E as the largest CFS characterized to date. Comprehensive LOH analysis revealed several distinct peaks of LOH within FRA9E. Semiquantitative RT-PCR analysis of 16 genes contained within FRA9E indicated that genes showing LOE in ovarian tumors coincided with regions of high LOH. PAPPA displayed the most significant loss (72%). This study provides evidence to suggest that instability within FRA9E may play an important role in the development of ovarian cancer and lends further support for the hypothesis that CFSs may be causally related to cancer.

ATR Regulates Fragile Site Stability

Conditions that partially inhibit DNA replication induce expression of common fragile sites. These sites form gaps and breaks on metaphase chromosomes and are deleted and rearranged in many tumors. Yet, the mechanism of fragile site expression has been elusive. We demonstrate that the replication checkpoint kinase ATR, but not ATM, is critical for maintenance of fragile site stability. ATR deficiency results in fragile site expression with and without addition of replication inhibitors. Thus, we propose that fragile sites are unreplicated chromosomal regions resulting from stalled forks that escape the ATR replication checkpoint. These findings have important implications for understanding both the mechanism of fragile site instability and the consequences of stalled replication in mammalian cells.

Transcriptional profiling reveals that several common fragile-site genes are downregulated in ovarian cancer

Previous transcriptional profiling analysis of 14 primary ovarian tumors identified approximately 12,000 genes as decreased in expression by at least twofold in one or more of the tumors sampled. Among those genes were several known to be mapped to common fragile sites (CFSs), some of which had previously been shown to exhibit a loss of expression in ovarian carcinoma. Therefore, we selected a subset of genes to determine whether they localized within CFSs. Of the 262 genes that were downregulated at least twofold in 13 of 14 tumors, 10 genes were selected based on the following criteria: localization to a CFS band; documented aberrations in at least one malignancy; and feasibility of scoring breakage at the specific CFS. Fluorescence in situ hybridization analysis was performed using bacterial artificial chromosome clones encompassing portions of the genes to determine the position of the genes relative to their corresponding CFSs. Nine genes were determined to localize within seven previously uncloned CFSs. Semiquantitative reverse-transcription/polymerase chain reaction analysis of the cell lines and primary ovarian tumors validated the downregulation of seven of the 10 genes. We identified portions of seven uncloned CFSs and provide data to suggest that several of the genes mapping within CFSs may be inactivated in ovarian cancer.

Direct correlation between FRA3B expression and cigarette smoking

Cytogenetic deletions and/or loss of heterozygosity (LOH) of the short arm of chromosome 3, often with a break at 3p14, are well documented in lung tumors. The coincidence of a chromosomal fragile site, FRA3B, at a common chromosomal breakpoint in lung cancer has suggested that fragility at this site may predispose to breakage that could contribute to multistep carcinogenesis. This idea is supported by the more recent finding that FRA3B maps within the FHIT (fragile histadine triad) gene, and that aberrant transcripts and genomic deletions of FHIT/FRA3B occur in a variety of tumors including lung tumors. To determine whether some individuals have increased fragility of FRA3B that might increase the risk for breakage or deletion in 3p14.2, fragile site expression was examined in smokers, nonsmokers, and small cell lung cancer (SCLC) patients. The data clearly show that active smokers exhibit a significantly higher frequency of fragile site expression, including FRA3B, compared to that of nonsmokers and patients diagnosed with SCLC who have stopped smoking. These results suggest that active tobacco exposure increases chromosome fragile site expression, and that this fragility is transient and reversible. The data support the hypothesis that exposure to tobacco carcinogens increases the potential for chromosome breakage at fragile sites.

FRA3B and other common fragile sites: the weakest links

In 1979, the first chromosome alteration associated with familial cancer was reported. Five years later, a fragile site was observed in the same chromosome region. The product of the fragile histidine triad (FHIT) gene, which encompasses this fragile site, is partially or entirely lost in most human cancers, indicating that it has a tumour-suppressor function. Inactivation of only one FHIT allele compromises this suppressor function, indicating that a 'one-hit' mechanism of tumorigenesis is operative. Are genes disrupted at other fragile sites? And, are these genes also tumour suppressors?