Folate deficiency drives mitotic missegregation of the human FRAXA locus

Significance Dietary folate deficiency is associated with fetal neural tube defects, psychological disorders, and age-associated dementia. However, it remains unclear how folate deficiency could be a causative factor in such a diverse range of disorders. Through analysis of the FRAXA locus, which contains an extensive CGG repeat sequence, we show that folate deprivation triggers extensive mitotic missegregation of the locus. Moreover, the entire chromosome X becomes unstable during a period of long-term folate deprivation. Considering that the human genome contains several loci associated with extensive CGG repeat regions, these findings suggest a mechanism by which folate deficiency contributes to the onset of a wide range of human diseases.

from the RO3306 arrest, incubated for 30 mins or 45 mins in pre-warmed (37°C) cell culture medium, and then seeded onto Poly L-Lysine coated slides (Sigma Aldrich) prior to fixation. For cells cultured in RPMI 1640 without folic acid (No folate), RO3306 was added to the medium after 72h of incubation, and the cells were released after 9 hours as described above. To inhibit RAD51 activity in G2 phase, RAD51 inhibitor RI-1 (60 µM) (ab144558, Abcam) was added during the last 2h of RO3306 synchronization. G2 arrested cells or anaphases/telophases were then harvested for further analysis.
Probes were labeled using the BioNick labeling system (Thermo Fisher Scientific) or DIG-nick translation mix (Sigma Aldrich). FISH was carried out using standard procedures. Briefly, samples were treated with RNase A, followed by dehydration in an ethanol series (70%, 90%, 96%) at RT for 2 mins each. Samples were denatured in 70% formamide at 75°C for 3 mins, and dehydrated again in an ice-cold ethanol series (70%, 90%, 96%) for 2 mins each. FISH-probe hybridization was carried out at 37°C for 16 to 72h. Biotin-conjugated probes were detected using avidin-FITC (ThermoFisher Scientific) and biotin-labeled anti-avidin (Vector Labs). DIG-conjugated FISH probes were detected using anti-digoxegenin-rhodamine (Roche) and Alexa Fluor 568 donkey 4 anti-sheep IgG (Thermo Fisher Scientific). Slides were mounted using Vectashield mounting medium with DAPI. Images were captured using an Olympus BX63 microscope and analyzed using CellSens (Olympus) or Fiji/ImageJ software.
To analyze anaphase cells, following the treatment, cells were seeded onto Poly L-Lysine slides (Sigma-Aldrich) and harvested at low speed (300 rpm) using Cytospin and fixed in 4% PFA fixation buffer for 20 mins at 4°C. Samples were then blocked for 2h in 5% bovine serum albumin (BSA) and incubated with primary antibodies overnight at 4°C, and then with secondary antibodies for 1h at RT. Slides were either further processed for with FISH combined with IF, or mounted using Vectashield with or without DAPI (Vector Laboratories). Images were captured using Olympus microscopes (BX63 or IX83) and then analyzed using CellSens (Olympus) or Fiji/ImageJ software. Primary

FISH combined with IF
First IF was carried out as described above. For antibodies that could tolerate the FISH conditions, samples were re-fixed in 8% PFA for 20 mins on ice prior to FISH analysis.
For FISH, samples were denatured at 88°C for 5 mins and then processed as described above in FISH section. For antibodies that could not tolerate the FISH conditions, images were captured before and after FISH analysis using CellSens stage navigation. For these samples, the re-fixation step was carried out with 4% PFA fixation buffer and the DNA was denatured at 80°C for 5 mins. Images were captured using an Olympus BX63 microscope and analyzed using CellSens (Olympus) or Fiji/ImageJ software.

Flow cytometry
Cells were harvested and fixed in 70% ethanol (added dropwise while vortexing) and stored at -20°C for a minimum of 2h. Cells were centrifuged at 500 g for 5 mins at 4°C and then washed once in 1% BSA in PBS, centrifuged again, and stained with staining solution (2 mg/ml propidium iodide (Thermo Fisher Scientific), 10 mg/ml RNase A (Sigma-Aldrich) in 1 x PBS for 30 mins at 37°C. Fluorescence-activated cell sorting analysis was carried out on a FACSCalibur flow cytometer (BD Biosciences). For each condition, at least 20,000 events were recorded, and data analysis was performed using FlowJo software.

Western blot analysis
Cells were lysed using cell extraction buffer (FNN0011; Life technologies), supplemented with Protease inhibitor cocktail (PIC, Roche). Following protein quantification of the whole cell lysate, aliquots of samples were incubated for 10 mins at 70°C with NUPAGE SDS sample buffer (ThermoFisher Scientific), run on a SDS-PAGE gel and transferred to a Hybond-PVDF membrane (Amersham Pharmacia). The membrane was blocked in PBST with 5% non-fat dry milk (Sigma Aldrich) for 1h at RT and then incubated overnight with a primary antibody re-suspended in PBST (with 5% non-fat dry milk) at 4°C. Following 3 x 15 min washes using PBST, the membrane was incubated for up to 2h at RT with a secondary antibody re-suspended in PBST (with 5% non-fat dry milk) at room temperature, followed by 4 x 15 min washes in PBST.

The FMR1 CGG allele assay in sub-populations of GM06891 cells
To analyze the effect of folate stress on CGG repeat stability in FRAXA pre-mutation allele, GM06891 cells were cultured under the following conditions: untreated (Unt) or exposed to FdU (0.5 µM) for 17 h, or in RPMI 1640 medium without folic acid for 5 days

Statistical analysis
At least three independent experiments were carried out to generate each data set.
Statistical significance in each case was calculated using chi-squared test. In the FMR1 CGG allele assay in GM06981 cells, 23 pooled-clones per condition were analyzed and the Student's t-test was performed to calculate statistical differences.     In S phase, the replication of the FM allele is perturbed, leading to a greater chance of collapse of replication forks within FRAXA. This fork breakage initiates HR repair late in the cell cycle. A proportion of the HR intermediates (Holliday junctions, HJ; or D-loops) fail to be resolved prior to anaphase, leading to the formation of ssDNA bridges bound by RPA. This failure might be driven by the inability of the HR-derived DNA synthesis to progress through the CGG repeats. The FRAXA-containing bridges are ultimately either lost or retained in micronuclei in the next generation of G1 cells. Unresolved bridges might also trigger cytokinesis failure and hence promote the generation of binucleated cells, which tend to generate aneuploid progeny.