Polymorphisms in MDM2 and TP53 Genes and Risk of Developing Therapy-Related Myeloid Neoplasms

One of the most severe complications after successful cancer therapy is the development of therapy-related myeloid neoplasms (t-MN). Constitutional genetic variation is likely to impact on t-MN risk. We aimed to evaluate if polymorphisms in the p53 pathway can be useful for predicting t-MN susceptibility. First, an association study revealed that the Pro variant of the TP53 Arg72Pro polymorphism and the G allele of the MDM2 SNP309 were associated with t-MN risk. The Arg variant of TP53 is more efficient at inducing apoptosis, whereas the Pro variant is a more potent inductor of cell cycle arrest and DNA repair. As regards MDM2 SNP309, the G allele is associated with attenuation of the p53 apoptotic response. Second, to evaluate the biological effect of the TP53 polymorphism, we established Jurkat isogenic cell lines expressing p53Arg or p53Pro. Jurkat p53Arg cells presented higher DNA damage and higher apoptotic potential than p53Pro cells, after treatment with chemotherapy agents. Only p53Pro cells presented t(15;17) translocation and del(5q). We suggest that failure to repair DNA lesions in p53Arg cells would lead them to apoptosis, whereas some p53Pro cells, prone to cell cycle arrest and DNA repair, could undergo misrepair, generating chromosomal abnormalities typical of t-MN.


Cell culture and plasmids
To directly assess the biological effect of the TP53 polymorphism on cells treated with chemotherapy agents, we established isogenic cell lines expressing p53Arg or p53Pro. Jurkat cells, which are TP53 null and which are derived from a human lymphoblastic T-cell leukemia, were used (American Type Culture Collection, ATCC). Unfortunately, MDM2 null cells are not viable and such a model could not be constructed for the MDM2 gene. Jurkat cells were grown in RPMI-1640 medium containing 2 mM L-glutamine (Invitrogen, Carlsbad, CA) supplemented with 15% fetal bovine serum and 1% penicillin-streptomycin.
Plasmids expressing either p53Arg (pcDNA3.1-p53Arg) or p53Pro (pcDNA3.1-p53Pro) were kindly donated by Dr. Lawrence Banks and Dr. Miranda Thomas from the International Centre for Genetic Engineering and Biotechnology, Trieste, Italy 3 . Inserts from these plasmids were subcloned into the pLNCX2 retroviral vector to construct pLNCX2-p53Arg and pLNCX2-p53Pro (HindIII and NotI sites) following standard methods.

Cell transfection and retroviral transduction
To generate cell lines that stably express p53Arg or p53Pro, retroviral production and infection was carried out. First, pLNCX2-p53Arg and pLNCX2-p53Pro constructs were transiently transfected into the Phoenix packaging cell line with jetPEI ® (Polyplus, Illkirch, France) according to the manufacturer's protocol. For retroviral infection, Jurkat cells were incubated in the presence of the retrovirus-containing supernatant and 4 µg/mL polybrene (Sigma-Aldrich, Taufkirchen, Germany) for 24 h. Infection was repeated the next day. Twenty-four hours after the second infection, medium supplemented with G418 (1 mg/mL, Sigma-Aldrich) was added, and cells underwent selection for 3 days to eliminate uninfected cells. Standard Western blot analysis was carried out to confirm p53 expression.

Immunofluorescence staining and analysis of γ-H2AX foci
To assess the kinetics of γ-H2AX foci induction and disappearance following drug treatment, immunostaining of foci and microscopic analysis was performed. Early log phase Jurkat p53Pro and Jurkat p53Arg cells were treated for 2 h with doxorubicin (500 nM) or busulfan (500 µM). After treatment, cell cultures were maintained at 37°C for 2.5, 5, 12, 24 and 48 h.
Then, cells (30,000 cells/mL) were cytospun onto glass slides (Menzel-Glaser, Badalona, Spain) at 500 g for 5 min and immunofluorescence staining was performed as previously described 4 . Automated slide scanning was done with a Zeiss Axio Imager.Z2 epifluorescence microscope (Metasystems, Altlussheim, Germany) and the MetaCyte software module of the Metafer4 Slide Scanning System v3.10.2 (Metasystems). The images were captured using a 63x PlanApo objective and an SpOr filter. All signals were acquired as a z-stack with 10 focal planes and a z-step size of 0.35 µm between planes. A unique classifier was used to count a minimum of 200 cells for each particular experimental condition (a previous experiment with this classifier showed that the number of foci scored in 100 cells is enough to obtain a satisfactory result 4 ). These experiments were performed two independent times.

Chromosome breakage assay
After treatment with doxorubicin (500 nM) or busulfan (500 µM) for 2 h, cells were incubated for 24 h in the presence of colcemid (0.15 µg/mL, Gibco Thermofisher Scientific, Barcelona, Spain). Then, cells were collected by centrifugation and hypotonic shock was induced by a 0.075 M potassium chloride solution. Cell suspension was fixed in a mixture of methanol and glacial acetic acid (v/v 3:1) and cells were dropped onto slides and air-dried before staining with Leishman stain (Leishman eosin methylene blue solution modified, Merck, Madrid, Spain).
One hundred metaphases with 46 chromosomes were analyzed for each cell line and for each treatment in a Zeiss Axio Imager Z2 microscope coupled to a Metafer® Slide Scanning System v3.10.2. Chromosome gaps (chrg) and breaks (chrb), as well as chromatid gaps (chtg) and breaks (chtb) were classified according to the International System for Human Cytogenetic Nomenclature 5 . The experiment was performed two independent times.

SCE assay
Cells were treated with doxorubicin (500 nM) or busulfan (500 µM) for 2 h and then incubated at 37°C for 48 h under the presence of the thymine analogue 5-bromo-2´-deoxyuridine (BrdU, 12 µg/mL, Sigma-Aldrich). Colcemid was added at a final concentration of 0.15 µg/mL 24 h before harvest. Harvesting, staining and microscope observation was performed as previously mentioned. However, to be able to observe SCE and chromosome breaks in the same metaphase, treatment to distinguish dark and pail chromatids was softer than standard methods for SCE analysis. Before Leishman staining, slides were treated with 0.15 mg/mL Hoechst 33258 (Sigma-Aldrich) for 15 min and then exposed to 254-nm Ultraviolet Crosslinker These experiments were performed two independent times.

Cell proliferation
Cell proliferation was measured by the mitotic and proliferation indexes. The mitotic index was calculated as the ratio of the number of mitotic cells in 1000 stimulated nuclei, using the cultures from the chromosome breakage assay and the SCE assay. The proliferation index was determined as (MI+2MII+3MIII)/100 cells, using the cultures from the SCE assay. MI, MII, and MIII are the number of metaphase cells from the first, second and third cell cycle, respectively, according to the pattern of chromatid staining.

Apoptosis assay
To measure apoptosis after drug treatment, the Annexin-V-FLUOS Staining Kit (Roche, Basel, Switzerland) was used following manufacturer's instructions. Cells were treated with 500 nM of doxorubicin or 500 µM of busulfan during 16 h and 24 h in a cellular incubator at 37°C. After labeling with Annexin-V-FLUOS, we determined the proportion of apoptotic cells using an Olympus BX-60 epifluorescence microscope (Olympus, Shinjuku, Tokyo, Japan) equipped with FITC, Cy3 and DAPI filters, and a 50x Plan objective. Three independent experiments were performed analyzing 1000 cells for each particular experimental condition.

Fluorescence in situ hybridization (FISH)
Cell cultures treated with doxorubicin (500 nM) or busulfan (500 µM) for 2 h were grown for 100 days. Then, colcemid (0.15 µg/mL) was added 2 h before harvesting as previously mentioned. FISH assay was performed with the following probes for chromosomal alterations typical of t-MN: MLL dual color break apart rearrangement probe, t(15;17) PML-RARA dual color translocation probe, XL 5q31/5q33/5p15 locus-specific probe and a probe centromerespecific for chromosome 7 (CEP7). All the probes used were from Vysis, Downers Grove, IL, except chromosome 5 probe, which is from MetaSystems Probes, Heidelberg, Germany.

Polymerase chain reaction (PCR)
Detection of FLT3-ITD and NPM1 mutations was performed on cell cultures treated with doxorubicin or busulfan and grown in long-term culture; two independent experiments were performed. DNA was extracted following standard methods. The presence of FLT3-ITD was analyzed by PCR on genomic DNA as previously reported 7 . The type A mutation of the gene NPM1 was studied with the commercial kit ipsogen® NPM1 mutA MutaQuant® (Qiagen) according to the manufacturer's instructions.

Statistical analysis
Hardy-Weinberg equilibrium was explored for the two SNPs using a chi-square test. Test of association between TP53 or MDM2 polymorphisms and risk of developing t-MN was performed using the R 3.3.1 package (http://www.r-project.org/). Specifically, SNPassoc library 8 , which implements logistic regression methods under five different genetic models (codominant, dominant, recessive, overdominant and log-additive), was used. Chi-square or Fisher exact test was used to assess association combining TP53 and MDM2 polymorphisms.
To compare the number of ɣH2AX foci, chromosome/chromatid breaks, SCE or the number of apoptotic cells between Jurkat p53Pro and Jurkat p53Arg cells after drug treatment, first, normality of distribution of these parameters was tested with a Kolmogorov-Smirnov test (with Lilliefors correction) or a Shapiro-Wilk test. In case of compliance with normal distribution, a parametric test (t-student) was used. Otherwise, the nonparametric Mann-Whitney test was applied. All these statistical analyses were carried out using SPSS v22.0 software. To compare the mitotic index between Jurkat p53Pro and Jurkat p53Arg cells after drug treatment, a z-test was applied with a z-score calculator for two population proportions (http://www.socscistatistics.com/tests/ztest).

SUPPLEMENTARY TABLES
Supplementary