Analysis of radiation induced DNA damage by LC-MS/MS in isolated and cellular DNA

Abstract: It is well established that ionizing radiation induces a variety of damage in DNA by direct effects that are mediated by one-electron oxidation and indirect effects that are mediated by the reaction of water radiolysis products, e.g., hydroxyl radicals (•OH). In cellular DNA, direct and indirect effects appear to have about an equal effect toward DNA damage. We have shown that ϒ-(gamma) ray irradiation of aqueous solutions of DNA, during which •OH is the major damaging ROS can lead to the formation several lesions. On the other hand, the methylation and oxidative demethylation of cytosine in CpG dinucleotides plays a critical role in the gene regulation. The C5 position of cytosine in CG dinucleotides is frequently methylated by DNA methyl transferees (DNMTs) and constitutes 4-5% of the total cytosine. Here, my PhD research work focuses on the analysis of oxidative base modifications of model compounds of methylated and non methylated oligonucleotides, isolated DNA (calf-thymus DNA) and F98 cultured cell by gamma radiation. In addition, we identified a series of modifications of the 2-deoxyribose moiety of DNA arising from the exposure of isolated and cellular DNA to ionizing radiation. We also studied one electron oxidation of cellular DNA in cultured human HeLa cells initiated by intense nanosecond 266 nm laser pulse irradiation, which produces cross-links between guanine and thymine bases (G*-T*). To achieve these goals, we developed several methods based on mass spectrometry to analyze base modifications in isolated DNA and cellular DNA.

Impacts moléculaires d’un excès d’acides gras sur l’androgenèse des cellules surrénaliennes

Le syndrome des ovaires polykystiques (SOPK) touche entre 5 à 10 % des femmes en âge de procréer et est associé à de nombreuses complications. Ce désordre endocrinien est caractérisé par des niveaux circulants élevés d’androgènes, dont la production est principalement modulée par la P450c17 et son cofacteur, soit la P450oxydoréductase (POR). Plusieurs études démontrent que l’hyperandrogénie présente chez les femmes SOPK pourrait être causée par la formation de phénomènes toxiques survenant à la suite de l’exposition des tissus non adipeux à un excès d’acides gras non estérifiés (AGNE), appelé lipotoxicité. Ainsi, l’objectif de cette étude est de déterminer les mécanismes cellulaires sous-jacents à l’hyperandrogénie induite par la surexposition des cellules productrices d’androgènes aux AGNE. Pour y arriver, les cellules surrénaliennes bovines (CSB) ont été exposées en présence de forskoline (Fsk; 1X/2 jours; 10 µM; activateur des adénylates cyclases) et d’oléate (acide gras monoinsaturé, 2X/jour; 200 µM) durant 48 heures. Par la suite, le milieu de culture a été prélevé afin de doser le DHEA (principal androgène surrénalien) par ELISA. De plus, les protéines ont été récoltées afin de déterminer l’expression protéique de la P450c17 et de POR par Western blot. Finalement, pour déterminer les activités 17αhydroxylase et 17,20-lyase de la P450c17, la concentration de plusieurs stéroïdes a été déterminé par LC-MS/MS et le ratio produit/substrat a été effectué. Les résultats sont présentés en moyenne ± SEM. Ainsi, sous stimulation à la Fsk, la présence de 200 µM d’oléate (vs absence d’oléate) augmente la production de DHEA de 114% par les CSB (n=17; 214 ± 20% vs 100 ± 0%; p<0,0001). De plus, l’ajout d’oléate n’affecte pas l’expression de la P450c17 (n=8; 98 ± 6% vs 100 ± 0%; p=0,74) et de POR (n=7; 119 ± 13% vs 100 ± 0%; p= 0,22). Finalement, la présence d’oléate augmente l’activité 17αhydroxylase de la P450c17 de 124% (n=7; 224 ± 19% vs 100 ± 0%; p= 0,02) et tends à augmenter l’activité 17,20-lyase de la P450c17 de 81% (n=7; 181 ± 28% vs 100 ± 0%; p= 0,08). Ainsi, l’augmentation de la production des androgènes induite par les AGNE pourrait être due principalement à leurs effets sur l’augmentation des deux activités enzymatiques de la P450c17, sans effet significatifs sur l’expression de la P450c17 ni de son cofacteur POR. Les mécanismes sous-jacents à l’augmentation des activités de la P450c17 demeurent à élucider.

DNA damage induced by low energy electrons (LEEs)

Abstract : The major objective of our study is to investigate DNA damage induced by soft X-rays (1.5 keV) and low-energy electrons (˂ 30 eV) using a novel irradiation system created by Prof. Sanche’s group. Thin films of double-stranded DNA are deposited on either glass and tantalum substrates and irradiated under standard temperature and pressure surrounded by a N[subscript 2] environment. Base release (cytosine, thymine, adenine and guanine) and base modifications (8-oxo-7,8-dihydro -2’-deoxyguanosine, 5-hydroxymethyl-2’-deoxyuridine, 5-formyl-2’-deoxyuridine, 5,6-dihydrothymidine and 5,6-dihydro-2’-deoxy uridine) are analyzed and quantified by LC-MS/MS. Our results reveal larger damage yields in the sample deposited on tantalum than those on glass. This can be explained by an enhancement of damage due to low-energy electrons, which are emitted from the metal substrate. From a comparison of the yield of products, base release is the major type of damage especially for purine bases, which are 3-fold greater than base modifications. A proposed pathway leading to base release involves the formation of a transient negative ion (TNI) followed by dissociative electron attachment (DEA) at the N-g lycosidic bond. On the other hand, base modification products consist of two major types of chemical modifications, which include thymine methyl oxidation products that likely arises from DEA from the methyl group of thymine, and 5,6-dihydropyrimidine that can involve the initial addition of electrons, H atoms, or hydride ions to the 5,6-pyrimidine double bond.

Alterations of the epigenome in brain cancer: loss of 5-hydroxymethylcytosine

Abstract : 5-Methylcytosine is an epigenetic mark, which can be oxidized to 5-hydroxymethylcytosine (5hmC) in DNA by ten-eleven translocation (TET) oxygenases. It is an initial step in the demethylation of 5mC. Levels of 5hmC is relatively high in the brain compared to other organs, but these levels are known to be significantly reduced during the development of a brain tumor, especially in glioblastoma multiforme (GBM). However, no known mechanisms may fully explain this abnormality. The objectives of my project were to (1) understand the implications of the demethylation pathway mediated by TET, and (2) gain a deeper insight in the epigenetic make-up of brain tumors. (1) U87 cells were incubated with 5mC, 5hmC, 5-formylcytosine (5fC) or co-incubated of 5hmC with 3,4,5,6-tetrahydro-2’-deoxyuridine (dTHU) over a timeline of 0, 24, 48 and 96 hours. (2) 130 brain tumors (GBM= 79; grade II/III= 51) were obtained directly from surgery and immediately suspended in DNA extraction buffer. Both cell samples and tumor tissues underwent DNA extraction and DNA digestion protocols. The percent per cytosine (%/C) was obtained by quantification of 5mC, 5hmC, 5fC, 5-hydroxymethyluracil (5hmU) and 5formyluracil (5fU) using LC-MS/MS. (1) Cellular incubations showed that it is possible to increase levels of 5hmC in DNA, but also a slight increase in 5mC levels throughout the experiment. 5HmC levels dramatically increased by 1.9-fold after 96h. On the other hand, no increase was observed in 5fC levels. Both 5hmC and 5fC incubations were accompanied by high increases in 5hmU and 5fU levels respectively. The addition of dTHU to the 5hmC incubation decreased 5hmU incorporation by 65%. (2) The average levels of 5mC, 5hmC and 5fC, in brain tumors, were 4.0, 0.15 and 0.021 %/C respectively. 5HmU and 5fU levels were present at comparable levels of 5hmC and 5fC. Levels of 5hmC, 5hmU and 5fU were significantly lower in the DNA of GBM specimens. There was a strong correlation between 5mC with 5hmC and 5fC in GBM, but this was absent in low grade tumors. The presence of 5hmU and 5fU in brain tumor and the increase in their levels during cell incubations indicate a deamination activity in these cancerous cells, which may impinge on the cellular levels of 5hmC, in particular. Furthermore, upon the incubations with 5hmC, downstream levels of 5fC did not increase suggesting a TET malfunction. TET activity is maintained in GBMs, but impaired in low grade tumors due to isocitrate dehydrogenase-1 (IDH1) mutations. Therefore, in brain tumors, a strong deamination activity and TET impairment may lead to epigenetic reduction of 5hmC.