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.

The FRA3B at 3p14.2 is the most highly expressed of the common fragile sites observed when DNA replication is perturbed by aphidicolin or folate stress. The molecular basis for chromosome fragility at FRA3B is unknown. In contrast to the rare fragile sites, including FRAXA, no repeat motifs, such as trinucleotide repeats, have been identified within FRA3B. Several lines of evidence suggest that fragile sites are regions of DNA whose replication is unusually sensitive to interference. We have used fluorescence in situ hybridization to determine the relative timing of replication of FRA3B sequences. Our studies revealed that FRA3B sequences are late replicating. Exposure to aphidicolin, an inhibitor of both DNA polymerase alpha and delta, results in a reproducible delay in the timing of replication, and some cells enter G2without having completed replication of FRA3B sequences. Our results support a model in which common fragile sites are sequences that initiate replication late in S phase or are slow to replicate, and the chromosomal breaks and gaps observed in metaphase cells are due to unreplicated DNA.

Direct cloning of DNA sequences from the common fragile site region at chromosome band 3p14.2.

Despite several lines of evidence suggesting that common chromosomal fragile sites are biologically important as hot spots for recombination, their structure remains unknown. We showed previously that the plasmid pSV2neo preferentially integrates into bands containing fragile sites in cells transfected under conditions of fragile site induction. Here we report the isolation and characterization of the DNA sequences from two such independent integrations into 3p14.2, a common fragile site (FRA3B). These FRA3B region sequences were shown to lie within a 1330-kb YAC, 850A6, approximately 350 kb telomeric of the breakpoint of t(3;8), a constitutional rearrangement. The two integration sites are 10 kb apart, but each integration is associated with a deletion. We have constructed a partial genomic contig of the integration sites and deleted regions spanning approximately 85 kb. Analysis of the DNA sequences immediately surrounding the plasmid integrations revealed no known coding sequences or repeat structures resembling the (CGG)n motif characteristic of the rare fragile sites. In addition, by Southern blotting analysis, none of the phage clones isolated from the FRA3B region were found to contain CGG repeats. Fluorescence in situ hybridization analysis of genomic clones from this contig to metaphase cells induced to express breaks demonstrated hybridization adjoining the chromosome breaks, and occasionally the hybridization signal spanned the break. The results imply that breakage occurs at variable positions within a large region (at least on the order of 85 kb). Together, these data suggest that the structure of FRA3B differs from that of rare fragile sites.

Localization of the Chinese hamster MHC locus to chromosome 1q17-->q18 by fluorescence in situ hybridization.

Comparative gene mapping analysis in mammals suggests that the major histocompatibility complex (MHC) genes map syntenic with genes, such as glyoxylase 1 (GLO1). In man, the MHC locus and other genes of this syntenic group map to chromosome band 6p21.3, and in mouse, these genes map to chromosome 17. In the hamster, however, only the GLO1 gene has been localized; GLO1 maps to chromosome 1, suggesting that the genes within the MHC locus also map to this chromosome. We have localized the hamster MHC class I genes to chromosome band 1q17-->q18 by fluorescence in situ hybridization (FISH). These results suggest that GLO1 and other syntenic genes also lie within this chromosome region.

Increased genetic instability of the common fragile site at 3p14 after integration of exogenous DNA.

We determined previously that the selectable marker pSV2neo is preferentially inserted into chromosomal fragile sites in human x hamster hybrid cells in the presence of an agent (aphidicolin) that induces fragile-site expression. In contrast, cells transfected without fragile-site induction showed an essentially random integration pattern. To determine whether the integration of marker DNA at fragile sites affects the frequency of fragile-site expression, the parental hybrid and three transfectants (two with pSV2neo integrated at the fragile site at 3p14.2 [FRA3B] and specific hamster fragile sites [chromosome 1, bands q26-31, or mar2, bands q11-13] and one with pSV2neo integrated at sites that are not fragile sites) were treated with aphidicolin. After 24 h the two cell lines with plasmid integration at FRA3B showed structural rearrangements at that site; these rearrangements accounted for 43%-67% of the total deletions and translocations observed. Structural rearrangements were not observed in the parental cell line. After 5 d aphidicolin treatment, the observed excess in frequency of structural rearrangements at FRA3B in the cell lines with pSV2neo integration at 3p14 over that in the two lines without FRA3B integration was less dramatic, but nonetheless significant. Fluorescent in situ hybridization (FISH) analysis of these cells, using a biotin-labeled pSV2neo probe, showed results consistent with the gross rearrangements detected cytogenetically in the lines with FRA3B integration; however, the pSV2neo sequences were frequently deleted concomitantly with translocations.(ABSTRACT TRUNCATED AT 250 WORDS)

Preferential integration of marker DNA into the chromosomal fragile site at 3p14: an approach to cloning fragile sites.

Fragile sites are specific regions of chromosomes that are prone to breakage. In cells cultured under conditions that induce fragile site expression, high levels of inter- and intrachromosomal recombination have been observed involving chromosomal bands containing fragile sites. To determine whether expression of specific fragile sites would facilitate preferential integration of exogenous DNA at these recombination hot spots, the vector pSV2Neo was transfected into a Chinese hamster-human somatic cell hybrid containing a derivative chromosome 3 as its only human component. Chromosome 3 contains a common fragile site at band 3p14.2 (FRA3B) that is induced by aphidicolin. Both cells induced to express FRA3B and the uninduced control cells were transfected with the pSV2Neo selectable plasmid. In situ hybridization of a biotin-labeled pSV2Neo probe to metaphase chromosomes revealed one to three integration sites in each stably transfected clone. Four of 13 clones transfected under conditions of FRA3B induction showed integration of pSV2Neo at 3p14; these clones also showed specific integration into hamster chromosome 1 and a rearranged chromosome characteristic of CHO cells (mar2). The 7 control clones, however, showed an apparently random pattern of pSV2Neo integration. Significant hybridization of pSV2Neo to both FRA3B and Chinese hamster chromosomes 1 and mar2 was seen in 100 cells from pooled colonies transfected after treatment with aphidicolin. These results suggest that preferential integration of marker DNA into human and Chinese hamster fragile sites occurs with exposure to aphidicolin. The nature of the DNA sequences at fragile sites is unknown and, despite a number of approaches, these sequences have not yet been isolated; our procedure may represent an approach to the cloning of fragile sites.

Familial Lymphoproliferative Disorders with Chromosomal Fragile Site Analysis.

We have identified a family in which three members developed B-cell lymphoproliferative disorders within a nine month period. The 33 year old proband and his mother have hairy cell leukemia, and his 37 year old brother developed a large cell lymphoma. Chromosomal fragile site analysis of peripheral blood lymphocytes of the three patients as well as two healthy family members was performed. The mean number of rare fragile sites present per cell analyzed was not significantly different from that observed in a group of healthy adults used as controls. However, the level of expression of the common fragile sites detected in each of the study patients was significantly elevated compared to the control population. Although the relationship between the level of expression of common fragile sites and the subsequent development of a malignant process is unknown, many of the agents that induce these sites are known mutagens, and the level of their expression may reflect a genetic susceptibility to mutagenic damage. Thus, it is possible that common exposure to an environmental mutagen may have contributed to the temporal clustering of malignancy in this family.

Rearrangements of chromosome 3 involving bands 3q21 and 3q26 are associated with normal or elevated platelet counts in acute nonlymphocytic leukemia.

Fourteen patients with acute nonlymphocytic leukemia (ANLL) or dysmyelopoietic syndromes were found to have abnormalities involving the long arm of chromosome 3. In eight patients, the structural rearrangements involved both bands 3q21 and 3q26 and included t(3;3) (four patients), inv(3) (three patients), and ins(5;3) (one patient). Before treatment, seven of these eight patients had platelet counts above 100,000 per microliter, five had normal or elevated platelet counts, and four had significantly elevated platelet counts (600,000 to 1,731,000 per microliter). In each of the eight cases, normal or elevated platelet counts were associated with marked abnormalities of megakaryocytopoiesis, including increased numbers of megakaryocytes and numerous micromegakaryocytes. Classification within the French-American-British system was difficult in most of these cases; however, the leukemia in five of the eight patients with abnormalities of chromosome 3 that involved both bands 3q21 and 3q26 was classified as M4. The remaining six of the 14 patients had translocations between chromosome 3 and another chromosome. None involved both bands 3q21 and 3q26, and a break in either q21 or q26 was noted in only two patients. One of the six, who had ANLL (M4) with a normal platelet count, had a 3;5 translocation which involved band 3q25. These data suggest that in patients with ANLL, abnormalities of chromosome 3 which simultaneously involve bands 3q21 and 3q26 are associated with unusually high platelet counts.