Daphne Pontier

Abstract: X chromosome inactivation (XCI) is a process in mammals that ensures equal transcript levels between males and females by genetic inactivation of one of the two X chromosomes in females. Central to XCI is the long non-coding RNA Xist, which is highly and specifically expressed from the inactive X chromosome. Xist covers the X chromosome in cis and triggers genetic silencing, but its working mechanism remains elusive. Here, we review current knowledge about Xist regulation, structure, function and conservation and speculate on possible mechanisms by which its action is restricted in cis. We also discuss dosage compensation mechanisms other than XCI and how knowledge from invertebrate species may help to provide a better understanding of the mechanisms of mammalian XCI.

Abstract: 1. Clone out 96 single dog-1 animals onto 96 6 cm petridishes, and grow these plates till starvation. 2. Optional: Rinse off half the plate and freeze the animals. Keep these as a backup for this or for future isolations of (other) G4 deletion alleles. 3. Rinse off half of the worms in M9 buffer and transfer these to 96 well PCR plates in duplicate. Remove supernatant. 4. Add 50 μL lysis buffer (50 mM KCl; 2.5% mM MgCl2,; 10 mM Tris-HCl (pH=8); 0.45% Nonidet P40; 0.45% Tween-20; 0.01% gelatin and 200 μg ml-1 proteinase K) to the worm pellet. 5. Lyse worm pellets at 60oC for 60 minutes, and inactivate proteinase K for 30 minutes at 95oC. Some non-dissolved pellet may remain. 6. Perform nested PCRs on 1 μL lysis mix in a total volume of 10 μL. As a control, also perform the PCR on lysed wild type (N2) worms and on a blank (no DNA) sample. PCR conditions are as follows: o PCR reaction: 1x PCR buffer (50 mM KCl, 10 mM Tris-HCl pH8.3 and 1.5 mM MgCl2); 0.5 mM dNTPs, 0.5 μM of each primer and 0.025 U ml-1 Taq polymerase o PCR Program: 1 minute 95oC; 35 cycles of 20 seconds 95oC, 40 seconds 58oC, 1 minute 72oC; 3 minutes 72oC. o Nested PCR: Transfer 0.2 μL mix from the first PCR (primers 1 and 4) into the second PCR (primers 2 and 3) as a template (e.g. by using a 384-well hedge). 7. Load 5 μL of the second PCR on a 1% agarose gel, where duplicates run next to each other. Search for duplicates that have a band of identical size. This is called a �positive population� and contains a candidate germline mutation. 8. Chunk the plate corresponding to the positive population into 12 pieces onto fresh 6 cm plates. Grow till starvation and repeat the procedure from step 3-7. As a positive control, include the lysis DNA from the previous, positive round. 9. If one or more of these plates result in a deletion product of identical size to the original positive sample, then transfer ~10 worms of this plate to 48 new 6 cm plates. Grow till starvation and repeat step 3-7. If several samples are positive for the germline deletion, proceed to step 10. If no samples are positive, return to the original (frozen) plate and start the procedure again. Increasing the number of populations should also increase deletion allele isolation efficiency. 10. Return to the plate out of 48 that contains the positive population, and clone out 48 single worms from this plate on fresh plates. Grow till starvation and repeat step 4-8. Alternatively, grow a few days, then pick the mother into 10 μL lysis buffer and perform PCRs on this sample. 11. A positive population from step 10, initiated from a single animal, may still be heterozygous for the deletion allele. To verify this, perform the nested PCR on ~15 individual animals from the positive population. Only if all individuals are positive for the deletion, the population is likely homozygous (this can be confirmed by designing primers that are within the deleted segment and verify that this DNA is lost from these animals). If not, clone out >6 animals from the positive population and perform the nested PCR on 12 F1 progeny animals for each of these. 12. The remaining 5 μL from the PCR reaction can be used to sequence the deletion. 13. Cross the new strain several times back to N2 to remove dog-1(gk10) and possible secondary mutations.

Abstract: Metazoan genomes contain thousands of sequence tracts that match the guanine-quadruplex (G4) DNA signature G(3)N(x)G(3)N(x)G(3)N(x)G(3), a motif that is intrinsically mutagenic, probably because it can form secondary structures during DNA replication. Here we show how and to what extent this feature can be used to generate deletion alleles of many Caenorhabditis elegans genes.

Abstract: To safeguard genetic integrity, cells have evolved an accurate but not failsafe mechanism of DNA replication. Not all DNA sequences tolerate DNA replication equally well [1]. Also, genomic regions that impose structural barriers to the DNA replication fork are a potential source of genetic instability [2, 3]. Here, we demonstrate that G4 DNA-a sequence motif that folds into quadruplex structures in vitro [4, 5]-is highly mutagenic in vivo and is removed from genomes that lack dog-1, the C. elegans ortholog of mammalian FANCJ [6, 7], which is mutated in Fanconi anemia patients [8-11]. We show that sequences that match the G4 DNA signature G3-5N1-3G3-5N1-3G3-5N1-3G3-5 are deleted in germ and somatic tissues of dog-1 animals. Unbiased aCGH analyses of dog-1 genomes that were allowed to accumulate mutations in >100 replication cycles indicate that deletions are found exclusively at G4 DNA; deletion frequencies can reach 4% per site per animal generation. We found that deletion sizes fall short of Okazaki fragment lengths [12], and no significant microhomology was observed at deletion junctions. The existence of 376,000 potentially mutagenic G4 DNA sites in the human genome could have major implications for the etiology of hereditary FancJ and nonhereditary cancers.

Abstract: To preserve genomic integrity, various mechanisms have evolved to repair DNA double-strand breaks (DSBs) [1]. Depending on cell type or cell cycle phase, DSBs can be repaired error-free, by homologous recombination, or with concomitant loss of sequence information, via nonhomologous end-joining (NHEJ) or single-strand annealing (SSA) [2]. Here, we created a transgenic reporter system in C. elegans to investigate the relative contribution of these pathways in somatic cells during animal development. Although all three canonical pathways contribute to repair in the soma, in their combined absence, animals develop without growth delay and chromosomal breaks are still efficiently repaired. This residual repair, which we call alternative end-joining, dominates DSB repair only in the absence of NHEJ and resembles SSA, but acts independent of the SSA nuclease XPF and repair proteins from other pathways. The dynamic interplay between repair pathways might be developmentally regulated, because it was lost from terminally differentiated cells in adult animals. Our results demonstrate profound versatility in DSB repair pathways for somatic cells of C. elegans, which are thus extremely fit to deal with chromosomal breaks.

Abstract: DNA double-strand breaks (DSBs) can be repaired by three canonical repair pathways. Homologous recombination (HR) uses the sister chromatid or homologous chromosome as a template to repair the DSB in an error-free manner. In non-homologous end-joining (NHEJ), the broken ends are ligated with little or no sequence homology, and this is often accompanied by the loss of a few nucleotides. Single-strand annealing (SSA) uses sequence homology within the same chromosome and leads to deletion of one of the repeats and the intervening sequence. Using the model organism C. elegans, we study DSB repair in the context of a developing animal and in complex genetic backgrounds. We make use of a transgenic approach where the restriction enzyme I-SceI can be expressed in an inducible manner, combined with a reporter transgene that contains the 18-nt recognition site for I-SceI in an out-of-frame LacZ gene. In Chapter 2 of this thesis, we use this assay to reveal the activity of a fourth pathway, which we termed alternative end-joining (alt-EJ). This pathway seems to act as a backup for NHEJ because it predominates repair only in the absence of canonical NHEJ, but its repair products are characterized by frequent use of homology in a way that is similar to SSA. Alt-EJ operates independently of many known repair genes and leads to very efficient DSB repair even in triple mutants that are defective for HR, SSA and NHEJ. Despite its putative function as a backup for classic NHEJ, we show in Chapter 3 that, in contrast to NHEJ, alt-EJ only occurs in replicating cells, leading to DSB persistence in non-replicating NHEJ-deficient somatic cells. Although normally highly toxic, endogenous DSBs are introduced in a regulated manner in meiotic cells in the germline. These DSBs need to be repaired by HR to establish crossover formation between homologous chromosomes which is required for genetic diversity among the offspring and for correct chromosome segregation. In Chapter 4 we show that besides HR, other repair pathways are also active in the germline. Moreover, the response to DSBs is highly dependent on the stage of the cell cycle at the time of DSB induction and differs between different germline zones. Quadruplex of G4 DNA is a stable secondary ssDNA structure that can form in particular G-rich sequences during DNA replication. In mutants for the gene dog-1 (mammalian FANCJ), spontaneous deletions arise at G4 DNA. These deletions always initiate immediately downstream of the G-rich sequence and end at various locations downstream. In Chapter 5, we show that these deletions are likely formed through DSB intermediates, because they resemble DSBs at other locations in many ways. Remarkably, these DSBs are not repaired by one of the canonical repair routes, but may instead be repaired by alt-EJ or another mechanism. In Chapter 6, we show how deletion formation in dog-1 mutant background can be used a tool to isolate deletion alleles of many C. elegans genes.