Chromosome fragility and the abnormal replication of the FMR1 locus in Fragile X syndrome

2014-01-12 12:35:03

Human Molecular Genetics; 2014 January 12; doi:10.1093/hmg/ddu006

Dmitry Yudkin, Bruce Hayward, Mirit I. Aladjem, Daman Kumari, Karen Usdin


Fragile X Syndrome (FXS) is a learning disability seen in individuals who have >200 CGG-CCG-repeats in the 5' untranslated region of the X-linked FMR1 gene. Such alleles are associated with a fragile site, FRAXA, a gap or constriction in the chromosome that is coincident with the repeat and is induced by folate-stress or thymidylate synthase inhibitors like fluorodeoxyuridine (FdU). The molecular basis of the chromosome fragility is unknown. Previous work has suggested that the stable intrastrand structures formed by the repeat may be responsible, perhaps via their ability to block DNA synthesis. We have examined the replication dynamics of normal and FXS cells with and without FdU. We show here that an intrinsic problem with DNA replication exists in the FMR1 gene of individuals with FXS even in the absence of FdU. Our data suggests a model for chromosome fragility in FXS in which the repeat impairs replication from an origin of replication (ORI) immediately adjacent to the repeat. The fact that the replication problem occurs even in the absence of FdU suggests that this phenomenon may have in vivo consequences, including perhaps accounting for the loss of the X chromosome containing the fragile site that causes Turner syndrome (45, X0) in female carriers of such alleles. Our data on FRAXA may also be germane for the other FdU-inducible fragile sites in humans, that we show here share many common features with FRAXA.


Fragile sites are a common feature of mammalian genomes. These sites are apparent as gaps, constrictions or breaks in the chromosome that are visible in
metaphase spreads prepared from cells grown in the presence of agents like fluorodeoxyuridine (FdU), distamycin, bromodeoxyuridine or aphidicolin (APH). Fragile sites are also frequent sites of chromosome breakage and translocation in vivo, suggesting that chromosome fragility is not an in vitro artifact, but a feature of normally growing cells. While different agents induce fragility at different chromosomal loci, all of these agents interfere with DNA replication in some way, suggesting that the different fragile sites represent sequences that are either difficult to replicate or replicate late in S phase. Cells treated with such agents could then enter mitosis before replication of the fragile site region has been completed. This would result in premature chromatin condensation and the microscopic appearance of the fragile site.

Many of the common fragile sites such as those induced by APH, a DNA polymerase α inhibitor, are spread over megabases of DNA with a heterogeneous composition. While these regions are often enriched for A+Trich sequences and sequences with high flexibility and low stability, as yet no specific sequence responsible for slowing DNA replication has been identified in these regions. Camptothecin (CPT), a topoisomerase I (Topo I) poison, reduces APH-induced chromosome fragility. This has led to the suggestion that APH acts by promoting the uncoupling of the DNA polymerase from the helicase/topoisomerase complex thereby increasing the opportunity for secondary structures to form between the polymerase and the helicase on the exposed template. These structures can then cause replication fork stalling, which in turn results in fragile site production. It has also been suggested that common fragile sites arise from replication stalling at an ORI in an ORI-poor region under conditions of replication stress. Fragile sites are generated because no additional ORIs are available that can be activated to complete replication of the region. On the other hand, there is also data to suggest that structural impediments to DNA synthesis may not be necessary necessary for fragility and that regions with few ORIs may be fragile simply due to their relative lack of origins or because ORIs are utilized that are less efficient than ORIs elsewhere in the genome.

In addition to the common fragile sites, there are a number of rare fragile sites that are only seen in a subset of the human population. A subset of these fragile sites are induced by folate-stress or FdU. Unlike the common APH-inducible fragile sites, FdU-inducible fragile sites all map to much smaller genomic regions that contain long CGG/CCG-repeat tracts that are often methylated. Shorter repeats at the same location are not fragile, suggesting that fragility in these cases cannot simply be the result of their chromosomal context. The best known of this class of fragile site is FRAXA, the fragile site associated with Fragile X syndrome. FXS is an intellectual disability resulting from the inheritance of >200 CGG/CCG-repeats in the 5’ UTR of the X-linked gene FMR1. Such alleles are referred to as full mutations (FMs) and are usually associated with aberrant DNA hypermethylation of the promoter and transcriptional silencing of the gene. Female carriers of such alleles are at high risk of Turner syndrome in which the X chromosome carrying the FM has been lost. Loss of the FM chromosome may be an in vivo consequence of the expression of FRAXA, resulting perhaps from fusion of two broken sister chromatids at the fragile site to generate a dicentric chromatid, followed by its loss from one cell during cytokinesis. Chromosome fragility is seen at much lower frequencies in carriers of FMR1
alleles that are either normal (<55 repeats) or that have 55-200 repeats (Premutation, PM alleles). In fact, prior to the development of a molecular assay
for FXS, chromosome fragility at this locus was a key diagnostic feature of FXS and it is for this site, that the disorder is named.

FdU acts by inhibiting thymidylate synthase (TS). This leads to nucleotide pool imbalances and the slowing of replication. FdU can also become incorporated into DNA leading to the production of FdU:A and, occasionally, FdU:G mispairs. However, the molecular basis of the FdU-inducible fragile sites is unknown. Long CGG/CCG-repeats are known to form a variety of noncanonical structures including hairpins, i-tetraplexes, G-tetraplexes/ quadruplexes and Z-DNA some of which have been shown to be very effective at blocking DNA synthesis in vitro. These sequences are also associated with blocks to DNA synthesis in vivo and methylation stabilizes some of these structures. Chromosome fragility may result if conditions arise that favor the formation of secondary structures by the repeat that disrupt replication through the repeat-containing region. FdU, since it primarily affects pyrimidine pools, may cause replication of the purine-rich strand to proceed more slowly than the pyrimidine-rich strand, possibly resulting in uncoupling of leading and lagging strand synthesis as has been proposed for the formation of some common fragile sites. This uncoupling could create conditions that favor the formation of secondary structures by the repeat that then blocks the replication fork. The effect of FdU may be exacerbated by the known tendency of transcriptionally inactive alleles to show delayed replication and specifically for FM alleles to replicate very late in the cell cycle.

To better understand the basis of chromosome fragility in FXS, we have examined the expression of the FRAXA fragile site in the presence of both FdU
and CPT. We have also ascertained the replication profiles of normal and FXS syndrome alleles in the presence and absence of FdU. Our data shows that there
is an intrinsic problem with replication of FM alleles that is seen even under normal growth conditions and that this problem is not specifically exacerbated by
FdU treatment. While CPT does reduce chromosome fragility, the replication problem is likely to be unrelated to uncoupling of the polymerase from the
helicase/topoisomerase complexes as some have suggested for APH-inducible sites. Rather, we suggest that FRAXA fragility arises because FdU slows
replication globally thus making this region that contains an impediment to DNA synthesis and already replicates late, finish replication even later.

To Access Article, Click Here