XB-ART-47797PLoS One January 1, 2013; 8 (12): e82097.
Simultaneous in vitro characterisation of DNA deaminase function and associated DNA repair pathways.
During immunoglobulin (Ig) diversification, activation-induced deaminase (AID) initiates somatic hypermutation and class switch recombination by catalysing the conversion of cytosine to uracil. The synergy between AID and DNA repair pathways is fundamental for the introduction of mutations, however the molecular and biochemical mechanisms underlying this process are not fully elucidated. We describe a novel method to efficiently decipher the composition and activity of DNA repair pathways that are activated by AID-induced lesions. The in vitro resolution (IVR) assay combines AID based deamination and DNA repair activities from a cellular milieu in a single assay, thus avoiding synthetically created DNA-lesions or genetic-based readouts. Recombinant GAL4-AID fusion protein is targeted to a plasmid containing GAL4 binding sites, allowing for controlled cytosine deamination within a substrate plasmid. Subsequently, the Xenopus laevis egg extract provides a source of DNA repair proteins and functional repair pathways. Our results demonstrated that DNA repair pathways which are in vitro activated by AID-induced lesions are reminiscent of those found during AID-induced in vivo Ig diversification. The comparative ease of manipulation of this in vitro systems provides a new approach to dissect the complex DNA repair pathways acting on defined physiologically lesions, can be adapted to use with other DNA damaging proteins (e.g. APOBECs), and provide a means to develop and characterise pharmacological agents to inhibit these potentially oncogenic processes.
PubMed ID: 24349193
PMC ID: PMC3857227
Article link: PLoS One
Species referenced: Xenopus
Genes referenced: aicda ccno lgals4.2 pcna ube2k ung
Article Images: [+] show captions
|Figure 2. AID-induced damage repair in the absence of replicative polymerases.(A) Replicative DNA polymerases were inhibited with aphidicolin (aph) during Xenopus laevis egg extracts (FE) incubation. Supercoiled DNA (scDNA) was added to FE (with bio-dC) in the presence or absence of aph, and biotinylated DNA isolated and quantitated by qPCR. The bars represent fold change as the difference between the qPCR Ct value of each sample normalised to the treated (+ aph) sample, which was set to 1. Error bars indicate ± SD (n = 3). Time line of experiment shown above the graph indicates the order of addition of substrates/proteins/nucleotides/extract/etc. or treatments. (B) AID-induced lesions are repaired in Xenopus laevis egg extracts. scDNA plasmids were treated (or not - bar 1) with the indicated proteins and then incubated in FE (or not - bar 2), isolated, and quantified by qPCR. The blue bars represent the ratio (fold change) of the amount of recovered plasmids from reactions carried out in the presence of G-AID (bar 3), untagged AID (AID; bar 4), GAL4 DNA binding domain (G-DBD; bar 5), or mutant GAL4-AID C87R (G-AIDmt; bar 6) versus levels of plasmids recovered from reactions that did not contain G-AID (FE alone, set to 1; bar 1). Open pink bars represent the absolute recovery of each treated sample in relation to its input. Error bars indicate ± SD (n = 3). Statistical analysis (t-test) was performed on differences of indicated fold change (brackets), with p values shown. Time line of the experiment is shown above the graph.|
|Figure 3. DNA topology during the IVR.(A) Various forms of the substrate plasmid were analysed on a 0.8% agarose gel at 5–10 V/cm for 16 h at 4°C. scDNA (sc) migrated faster than linear (L), while nicked open circular or relaxed circular DNA (oc/rc) migrated slowest. (B) Each topological form of DNA substrates was subjected to IVR without AID-induced damage. Experiments were done as in Figure 2, and results expressed as % of input. Error bars indicate ± SD (n = 3). (C) Modulation of DNA topology due to FE. 1 µg of supercoiled DNA (lane 1) was incubated with FE (lane 2) and analysed as in (A).|
|Figure 4. Correlation between lesion size, biotin incorporation, and plasmid recovery.(A) Incorporation of 1 or 2 biotinylated dCTP molecules is sufficient to recover the targeted plasmid by streptavidin purification. Schematic of the experiment is depicted at the top of the figure. Plasmids were untreated (scDNA) or nicked at one position with Nt.BsmAI (nicked DNA), followed by treatment with Klenow for 30 min at 25°C in the presence or absence of dTTP (dT) and varying ratios of biotinylated (bio-dC) and normal dCTP (dC). As indicated in the schematic, the lack of dGTP and dATP only allowed for the incorporation of 1 or 2 bio-dC molecules per plasmid. The bars (fold change) represent the difference between the qPCR Ct value for each sample before and after the streptavidin purification, normalised to the sample (scDNA or nicked DNA) treated with the Klenow and bio-dC (no dT and set to 1). Error bars indicate ± SD (n = 3). Time line of the experiment is shown above the graph. (B) Patch length of incorporated bio-dC does not bias the recovery of plasmids from IVR. The amount of bio-dC incorporation depending on the presence of other dNTPs was monitored by an AID-induced IVR assay. The fold change represents the difference between the qPCR Ct value of each sample normalised to the FE-treated sample (no G-AID) that was set to 1.|
|Figure 5. Quality control of AID used for the IVR assay.(A) G-AID deaminates cytosine to uracil during in vitro oligonucleotide deamination assay . ssDNA oligonucleotide deamination assay was performed using an oligonucleotide SPM163 containing a single cytosine. Two concentrations (0.05 µg and 0.5 µg) of G-AID and G-AIDmt, as well as untagged AID (0.05 µg) and BSA (1 µg) were incubated with oligonucleotide for 30 min at 37°C, followed treatment with UNG and NaOH, and separated on a 17.5% PAGE gel. (B) Time course of G-AID activity during an IVR. G-AID was incubated with substrate for 5 to 90 min (37°C) before addition of the FE (30 min 23°C). Analysis was done as in Figure 2. Error bars indicate ± SD (n = 3). (C) G-AID does not induce topological changes on the supercoiled plasmid over time. Samples were processed as in (B), but prior to FE addition they were treated with SDS and proteinase K for 2 h at 56°C and analysed for changes in DNA topology as in Figure 3A. (D & E) G-AID and G-AIDmt were incubated with scDNA (as in C) and analysed for topological changes (D) or subjected to an IVR reaction (E). Quantitations of the topological forms of the substrate are shown in green in (E), while IVR results are shown in blue as % of input recovery. IVR analysis was performed as in Figure 2. Error bars indicate ± SD (n = 3).|
|Figure 6. Quality control of FE in the IVR assay.(A) DNA topology changes upon FE treatment. scDNA was treated with G-AID, followed by incubation with FE for 0 to 60 min before deproteinisation with SDS and proteinase K at 56°C overnight. Deproteinised DNA was analysed as in Figure 3A. Various topoisomers of the DNA are indicated on the left; oc/rc (open circular/relaxed circular); sc (supercoiled). (B) Time course of DNA repair activity during an IVR. G-AID was incubated with scDNA for 30 min 37°C before addition of the FE and incubation for the indicated time (23°C). Analysis was performed as in Figure 2 with untreated sample (no AID) set to 1. Error bars indicate ± SD (n = 3). (C) Bio-dC incorporation by FE quantitatively correlates with the number of DNA damages (nicks). FE repair activity was monitored by incorporation of bio-dC after incubation with damaged plasmid (nicked with Nt.AlwI or Nt.BsmAI, which cut the plasmid 15 or 1 times, respectively). Analysis was performed as in Figure 2 with untreated sample (no FE - not shown) set to 1. (D) Repair specificity in FE. Repair activity was monitored by the incorporation of bio-dC after incubation of the nicked (Nt.AlwI or Nt.BsmAI) or re-ligated nicked (Lig - T4 ligase) plasmid with FE. The effect of T4 ligase on a G-AID-treated scDNA plasmid sample was also tested. Analysis was performed as in Figure 2 with untreated sample (no ligation) set to 1. Error bars indicate ± SD (n = 3). (E) The incorporation of bio-dC or bio-dA from a nick by FE is equivalent. FE DNA repair activity was monitored by incorporation of bio-dC or bio-dA after incubation of either a supercoiled (scDNA) or nicked (Nt.BsmAI) plasmid. Analysis was performed as in Figure 2 with untreated sample (no FE - not shown) set to 1. Error bars indicate ± SD (n = 3). (F) Time course of FE activity on a nicked damaged plasmid. DNase I (0.001 U) treated scDNA was incubated with FE for 0 to 60 min (23°C) in the presence of bio-dC or bio-dA. Analysis was performed as in Figure 2 with untreated sample (no FE - not shown) set to 1. Error bars indicate ± SD (n = 3).|
|Figure 7. Substrate characterisation for IVR activity.(A) Schematic of plasmid used for IVR. GAL4 binding sites (5 x UAS) in red and fragments (A–I) which are monitored by qPCR are indicated. All fragments are flanked by HinfI sites. (B) Table indicating size, distance from UAS, and average number of WRC per sequence for each fragment. (C–E) Graphic representation of data from the table in (B). (F) AID-induced IVR fragment analysis. The substrate plasmid was cleaved with HinfI after FE repair prior to streptavidin isolation (see timeline). Analysis was performed as in Figure 2 with uncut (full length - FL) plasmid sample set to 1. Error bars indicate ± SD (n = 3). (G) Correlation of fragment length and IVR activity. The fold change values from (F) were plotted against the fragment length and a line of best fit generated. The r2 value indicated a positive correlation of length and IVR recovery.|
|Figure 8. Multiple DNA repair pathways resolve AID-induced lesions.(A) Inhibiting the UNG activity of FE with UGI. Top - Increasing amounts of UGI were added to the FE during oligonucleotide based deglycosylation . After NaOH treatment the fragments were separated using PAGE. Bottom - Quantification of deglycosylation activity of UNG2. Activities of lane 2 and 10 were set to 100% (no inhibition) and 0% (maximum inhibition), respectively. (B) UNG2 activity in FE is near saturation. 5 or 10 units of recombinant UNG2 were added to the FE reaction during the IVR reaction. Analysis was performed as in Figure 2 with untreated (no G-AID - not shown) set to 1. Error bars indicate ± SD (n = 3). (C) DNA repair pathway inhibitors alter IVR of AID lesions. AID-induced damaged plasmids were subject to FE IVR in the presence of bio-dC (bars 1–4) or bio-dA (bars 5–8). BER inhibitor UGI was added to the FE either alone (bars 2 & 6) or in combination with PCNA inhibitor T2AA (bars 4 & 8), while T2AA was also added alone (bars 3 & 7). Analysis was performed as in Figure 2 with untreated FE (bars 1 & 5) set to 100% (blue bar). For absolute comparisons, analysis was also shown as % of input (pink bars). Error bars indicate ± SD (n = 3). Statistical analysis (t-test) was performed on differences of indicated fold change (brackets), with p values shown.|
|Figure 1. AID-induced lesions repair in vitro by the IVR system.Schematic description of the IVR assay. A supercoiled DNA plasmid (pGL4.31) containing 5 x GAL4 binding sites (UAS) is incubated with a recombinant fusion GAL4-AID protein (G-AID, represented by a yellow triangle and a red circle). Incubation at 37°C deaminates dC to create dU lesions in single stranded DNA (green star). The repair phase (yellow box) - relaxation and lesion repair - is carried out by the addition of frog egg extract (FE) in the presence of biotinylated dCTP (bio-dC) or biotinylated dATP (bio-dA) - (blue arrow), along with normal dNTPs. After DNA repair of the dU lesion the biotinylated-tagged DNA is isolated via magnetic streptavidin beads. Eluted products are subject to quantitative real-time PCR (red bar), and compared to input values of the same reaction prior to streptavidin isolation.|
References [+] :
Besmer, The transcription elongation complex directs activation-induced cytidine deaminase-mediated DNA deamination. 2006, Pubmed