DNA Amplification Provides a Facile Means for Repeat Expansion for SCA10 dATTCT dAGAAT Repeats

With the exception of repeated replication slippage or reiterative DNA synthesis, all the models described so far can account for small changes in repeat lengths, as observed in somatic cells, but not expansions of tenfold or greater in length. DNA amplification provides a simple, reasonable model for the large repeat expansions that occur on intergenerational transmission in some diseases. The amplification of specific DNA regions by repeated replication occurs in several systems, including the amplification of chorion genes during normal Drosophila development, puff II/9A in Sciara, and drug-resistance genes in tumor cells (Schimke 1988; Liang et al. 1993; Spradling 1999; Calvi and Spradling 1999; Tower 2004). For amplification to occur, the normal controls that limit replication to once per cell cycle must be abrogated (Spradling 1999; Calvi and Spradling 1999). Amplification has been proposed to occur by an onion-skin mechanism in which repeated initiation leads to multiple replication forks (Baran et al. 1983; Stark et al. 1989; Schimke 1992; Spradling 1999), followed by recombination, or nonhomologous end joining, to generate linear tandem arrays (Fig. 8, left panel) from the amplified DNA (Syu and Fluck 1997).

Amplification is frequently associated with replication origins, A+T-rich regions, inverted repeats, or polypurine ■ polypyrimidine tracts (Baran et al. 1987; Kirschner 1996; Spradling 1999). Fragile sites have also been implicated as a causative factor in oncogene amplification (Hellman et al. 2002). While commonly thought to arise by strand breakage, reinitiation at an aberrant origin could also generate abnormal DNA ends, leading to recombinational amplification (Syu and Fluck 1997).

Potaman et al. (2003) proposed that unwound DNA structures in long d(ATTCT) ■ d(AGAAT) repeats drive repeat amplification. The formation of an unwound DNA structure from superhelical energy in DNA may bypass the steps of pre-RC assembly that normally require the low cyclin dependent kinase (CDK) activity environment of the G1 phase, and allow poly-merase a/primase to initiate replication in the high CDK environment of the S phase without the association of origin-bound checkpoint proteins. The observation that the binding affinity of the Drosophila replication initiator origin recognition complex (ORC) is 30-fold higher for supercoiled DNA compared with relaxed DNA (Remus et al. 2004) suggests that a topo-logical equivalence between superhelical and unwound states could allow DUEs to act as replication switches. In addition, an increasing body of evidence suggests that transcription is a critical component of a replication origin (Ghosh et al. 2004; Kouzine et al. 2004; Jenke et al. 2004; MacAlpine et al. 2004; Danis et al. 2004; Casper et al. 2005; Nieduszynski et al. 2005). Perhaps transcription supplies the superhelical energy required to unwind

Amplification, end generation Onion skin amplification


y•;■■■: ' ______v..—^-*'''' "¿■'■'■T

10.B |

Fig. 8 Amplification models for repeat expansion. Models for repeat instability based on the utilization of an unstable DNA repeat as an aberrant replication origin initially described for the A+T-rich SCA10 d(ATTCT)n • d(AGAAT)n repeats by Potaman et al. (2003) are shown. Amplification with the generation of DNA ends (left). A+T-rich repeats can unwind and replication may start within the unwound bubble (step 2). The DNA repeats are denoted by the lighter shaded line. Nascent strands are shown in intermediate shading. Following synthesis (step 3), DNA unwinding again occurs within the repeats (step 3) and replication again starts at the unpaired regions (step 4). The second nascent strands are shown as the darker dashed lines. In step 5.A, lagging-strand replication from the first replication event is shown as the lighter dashed lines. When the nascent strand from the second origin firing reaches the 3' end of the first nascent strand, the strand will become displaced. This results in the formation of a branched molecule with free ends (step 6.A). The DNA ends may participate in recombination leading to expansion (Cromie et al. 2001). Onion-skin amplification (right). The DNA molecule shown in step 5.B follows from step 4. Onion-skin replication can occur by repeated initiation within the A+T-rich repeat. An eightfold amplification is shown in step 7.B. When fork movement ceases or slows at the first and second forks, a requirement for amplification, continued replication from the third fork will lead to a displacement of four DNA molecules consisting of pure repeats, (if synthesis is limited to the repeat tracts) (step 8.B). Pairs of these molecules have complementary single-strand ends that can drive hybridization into longer repeat tracts. These can then be joined by homologous recombination into even longer repeat tracts (step 9.B). These molecules can become integrated into the repeat tracts in the original chromosome, leading to massive expansion (step 10.B). The length of the repeat expansion would be dependent on the number of cycles of amplification. This model alone can easily explain very large repeat expansion using a well documented biological phenomenon. Although this model was described for the A+T-rich SCA10 repeat, other DNA repeats may possibly act in a similar fashion the DNA, allowing for the assembly of replication proteins (Gilbert 2004). The fact that unstable disease-associated repeats are associated with tran-scriptionally active genes suggests that transcription may reflect a significant cis-acting factor for repeat expansion driving aberrant replication initiation events. Alternatively, supercoiling-induced structures may be recognized as distortions by proteins involved in DNA repair (e.g., RPA, XPA, XPC, MSH2) (Pearson et al. 1997; Patrick and Turchi 1999; Wakasugi and Sancar 1999; Volker et al. 2001; Panigrahi et al. 2005; Owen et al. 2005), and may generate a 3'- OH primer by strand breakage or enzymatic nicking. Repetitive rounds of slipped mispairing during replication could then lead to repeat amplification and recombination (Cromie et al. 2001). These events may occur even more frequently during early embryogenesis or gametogenesis, where chro-matin structure and replication differ from that in somatic cells, and dynamic epigenetic modifications are occurring (Fuentes-Mascorro et al. 2000; Santos et al. 2005).

Aberrant replication initiation could also be responsible for the instability observed in somatic cells. Unrestrained superhelical tension measured at active genes in living cells is sufficient to support DNA unwinding (Ljung-man and Hanawalt 1992; Kramer and Sinden 1997; Kramer et al. 1999). The easily unwound pentanucleotide repeat sequence d(ATTCT)n ■ d(AGAAT)n is located at the transcribed SCA10 locus, and plasmids containing d(ATTCT)n ■ d(AGAAT)n repeats supported initiation and replication in HeLa cell extracts without the addition of a specific initiation protein (Potaman et al. 2003). In cells, unwinding of d(ATTCT)n ■ d(AGAAT)n repeats may support repetitive initiation of DNA replication and amplification of the repeat tract. If d(ATTCT)n ■ d(AGAAT)n acts as a replication origin, fractious DNA replication and amplification could lead to repeat expansion as shown in Fig. 8.

Was this article helpful?

0 0

Post a comment