|
Figure 1. Expression and purification of recombinant aprataxin in baculovirus expression system. (A) Construct of His-tagged long-form aprataxin (His-LA) expressed using Bac-to-Bac® Baculovirus Expression System. (B) Chromatogram of gel-filtered aprataxin. Following immobilized metal affinity chromatography, the aprataxin-rich fraction was purified by gel filtration column chromatography. A major peak was observed for each of the fractions from 13 to 20 in the chromatogram. (C) The fractionated extracts were separated by SDS-PAGE and stained with Coomassie brilliant blue (CBB). A 39-kDa single band is detected for each of the fractions from 14 to 19. Western blot analysis using the anti-His antibody (D) and anti-aprataxin antibody (E) shows a 39-kDa immunoreactive product in each of the fractions from 13 to 21.
|
|
Figure 2. 3â²-End processing by aprataxin. (A) Aprataxin removes 3â²-phosphate. The 5â²-FITC-labeled 3â²-phosphate (3â² â PO3â â) oligonucleotide was incubated in the absence (lane 2) or presence of aprataxin at different concentrations (25, 50 and 100ânM, lanes 3â5). A band with the same size as that of the 3â²-hydroxyl (3â²-OH) oligonucleotide (lane 1) appears in lanes with aprataxin (lanes 3â5). 5â²-Polynucleotide kinase 3â²-phosphatase (PNKP) was used as the positive control (lane 6). Reaction products were separated by 20% PAGE and visualized using a fluorescence gel scanner. (B) Aprataxin removes DNA 3â²-phosphoglycolate. The 5â²-FITCâlabeled 3â²-phosphoglycolate (3â²-PG) oligonucleotide was incubated in the absence (lane 1) or presence of aprataxin at different concentrations (25, 50 and 100ânM, lanes 2â4). The amount of the 3â²-OH oligonucleotide increases with aprataxin concentration (lanes 2â4). Apurinic/apyrimidinic endonuclease (APE1) was used as the positive control (lane 5). (C) Aprataxin fails to remove 3â²-α, β-unsaturated aldehyde. The 5â²-FITC-labeled 3â²-α, β-unsaturated aldehyde (3â²-UA) oligonucleotide was incubated in the absence (lane 1) or presence of aprataxin at different concentrations (25, 50 and 100ânM, lanes 2â4). The amount of the 3â²-UA oligonucleotide does not decrease with increasing aprataxin concentration (lanes 2â4). The 3â²-OH oligonucleotides were generated in the presence of APE1 (lane 5). The faint smear corresponding to the 3â²-UA oligonucleotide in lanes 1â5 is an artifact generated under the electrophoresis conditions employed. (D) Aprataxin fails to remove 3â²-phosphotyrosine end. The 5â²-FITC-labeled 3â²-phosphotyrosine (3â²-Y) oligonucleotide was incubated in the absence (lane 1) or presence of aprataxin at different concentrations (25, 50 and 100ânM, lanes 2â4). The amount of the 3â²-Y oligonucleotide do not decrease with increasing aprataxin concentration (lanes 2â4). 3â²-PO3 oligonucleotides were generated in the presence of tyrosyl-DNA phosphodiesterase 1 (TDP1) (lane 5).
|
|
Figure 3. Expression of recombinant GST-aprataxin fusion protein. (A) Constructs of GST-aprataxin fusion proteins. Constructs of GST fusion protein containing full-length aprataxin (long-form aprataxin, LA), the C-terminal region of aprataxin (short-form aprataxin, SA), the N-terminal FHA domain of aprataxin (FHA) and full-length aprataxin with P206L or V263G (P206L, V263G). (B) Expression and purification of GST-aprataxin fusion proteins. Recombinant GST fusion proteins containing LA (lanes 1, 6 and 11), SA (lanes 2, 7 and 12), FHA (lanes 3, 8 and 13), P206L (lanes 4, 9 and 14) and V263G (lanes 5, 10 and 15) were expressed in the bacterial expression system. Purified products were analyzed by CBB staining (left panel), and western blotting using the anti-GST antibody (middle panel) or anti-aprataxin antibody (right panel).
|
|
Figure 4. Disease-associated mutant forms of aprataxin lack their 3â²-end processing activity. (A) Mutant forms of aprataxin fail to remove 3â²-phosphate. The 5â²-FITC-labeled 3â²-phosphate (3â² â PO3â â) oligonucleotide was incubated in the presence of 50ânM recombinant GST fusion proteins containing LA (lanes 2â4), SA (lanes 5â7), FHA (lanes 8â10), P206L (lanes 11â13), and V263G (lanes 14â16) at different incubation times (0, 30 and 60âmin). A band of the same size as that corresponding to the 3â²-hydroxyl (3â²-OH) oligonucleotide (lane 1) appears in lanes with LA (lanes 3 and 4). SA showed a weak phosphatase activity (lanes 5â7). Neither FHA, P206L nor V263G showed phosphatase activity (lanes 8â16). (B) Mutant forms of aprataxin fail to remove 3â²-phosphoglycolate. The 5â²-FITC-labeled 3â²-PG oligonucleotide was incubated in the presence of 50ânM recombinant GST fusion proteins containing LA (lanes 3â5), SA (lanes 6â8), FHA (lanes 9â11), P206L (lanes 12â14) and V263G (lanes 15â17) at different incubation times (0, 30 and 60âmin). A band of the same size as that corresponding to the 3â²-OH oligonucleotide (lane 2) appears in lanes with LA (lanes 3 and 4). SA showed a weak 3â²-PG hydrolase activity (lanes 6â8). Neither FHA, P206L nor V263G removed 3â²-PG (lanes 9â17).
|
|
Figure 5. Aprataxin 3â²-end processing activities on ss and ds DNA substrates. (AâC) 3â²-Phosphatase activity of aprataxin on ss and ds DNA substrates. (A) The ss, recessed, one-nucleotide gapped and nicked DNA substrates with 3â²-phosphate ends used are shown schematically. (B) Aprataxin preferentially acts on ss DNA. The substrates were incubated with 20ânM LA for the indicated times (0, 30, 60 and 90âmin) at 37°C. Products were separated by denaturing PAGE and visualized using a Typhoon 9400 scanner (GE Healthcare). (C) Production rates in each reaction were quantified by ImageQuant TL (GE Healthcare). Error bars indicate standard errors for more than three independent experiments. (DâF) 3â²-PG hydrolase activity of aprataxin on ss and ds DNA substrates. (D) The ss, recessed, one-nucleotide gapped and nicked DNA substrates with 3â²-PG ends used are shown schematically. (E) Aprataxin preferentially acts on ss and gapped DNA. The substrates were incubated with 20ânM LA for the indicated times (0, 30, 60 and 90âmin) at 37°C. (F) Production rates in each reaction were quantified as described above.
|
|
Figure 6. Recombinant aprataxin fails to efficiently hydrolase GpppBODIPY or ApppBODIPY. GpppBODIPY (A) or ApppBODIPY (B) was incubated with recombinant His-tagged long-form aprataxin obtained from the baculovirus expression system (His-LA, lanes 4â6), recombinant GST fusion proteins containing LA (lanes 7 and 8), SA (lanes 9 and 10) and FHA (lanes 11 and 12). None of them showed lysine hydrolase activity (lanes 4â12). Fhit at 10 and 100âmU as the positive control showed GMP-lysine hydrolase activity (lanes 2 and 3).
|
|
Figure 7. Removal of adenylate residues from 5â²-ends by aprataxin. (A) Aprataxin removes AMP from 5â²-ends of nicked ds DNA. The 45-mer ds DNA harboring a nick with 5â²-AMP ends was incubated with the indicated amounts of aprataxin for 1âh. A band of the same size as that corresponding to the 5â²-phosphate (5â² â PO3â â) oligonucleotide appears in lanes with aprataxin. PNKP was used as the negative control. (B) Mutant forms of aprataxin fail to remove 5â²-AMP. The 45-mer ds DNA harboring a nick with 5â²-AMP ends was incubated in the presence of 50ânM recombinant GST fusion proteins containing LA (lanes 2â4), SA (lanes 5â7), FHA (lanes 8â10), P206L (lanes 11â13) and V263G (lanes 14â16) at different incubation times (0, 30 and 60âmin). SA showed a lower 5â²-AMP hydrolysis activity (lanes 5â7) than LA. Neither FHA, P206L nor V263G removed 5â²-AMP (lanes 8â16). (C) Production rates in each reaction were quantified. Error bars indicate standard errors for more than three independent experiments.
|
|
Figure 8. Aprataxin repairs SSBs with damaged 3â²-ends. (A) DNA repair assay employing gapped dsDNA with 3â²-phosphate ends as substrate. The 45-mer ds DNA substrate harboring a 1-nt gap with 3â²-phosphate (3â² â PO3â â) ends was incubated in the absence (lane 3) or presence of each of the indicated recombinant human proteins for 90âmin (lanes 4â9). The 45-mer ds DNA substrate, harboring a 1-nt gap with 3â²-hydroxyl (3â²-OH) ends, was incubated in the absence (lane 1) or presence of each indicated recombinant human protein (lane 2). Complete repair is indicated by the generation of the 5â²-FITC-labeled 45-mer oligonucleotide. The amount of 5â²-FITC-labeled 45-mer increased with the concentration of aprataxin (lanes 6â8). PNKP was used as the positive control (lane 9). (B) DNA repair assay employing gapped dsDNA with 3â²-phosphoglycolate ends as substrate. The 45-mer duplex substrate harboring a 1-nt gap with 3â²-phosphoglycolate (3â²-PG) ends was incubated in the absence (lane 3) or presence of each of the indicated recombinant human proteins for 90âmin (lanes 4â9). The 45-mer duplex substrate harboring a 1-nt gap with 3â²-OH ends was incubated in the absence (lane 1) or presence of each indicated recombinant human protein (lane 2). The amount of the FITC-labeled 45-mer oligonucleotide increases with aprataxin concentration (lanes 6â8). APE1 was used as the positive control (lane 9). The structures of the substrates employed in these experiments are shown on the right side of each panel.
|
|
Figure 9. Model of aprataxin-dependent SSBR pathway. Four SSBR pathways defined by the type of enzyme that removes damaged 3â²-ends are shown (a, b, c and d). SSBs can arise directly from sugar damage or TOP1 cleavage or indirectly from base damage. Red circles denote the damaged ends, the specific types of which are dependent on the source of the break. [1,2] PARP detects SSBs, thereby recruiting the XRCC1 and Lig3 complex. XRCC1 then replaces PARP. [3] The processing of damaged 3â²-ends is mediated by either APE1 (a), aprataxin (b), PNKP (c) or TDP1 (d), depending on the type of damaged 3â²-end. These damaged 3â²-ends should be converted to 3â²-OH ends for subsequent repair processes. In the pathway for repairing indirectly induced SSBs, damaged 3â²-α, β unsaturated aldehyde ends are removed by APE1 (a). In the pathway for repairing directly induced SSBs, 3â²-PG ends might be removed by aprataxin (b) and 3â²-phosphate ends by aprataxin or PNKP (b,c). In the pathway for repairing TOP1-mediated SSBs, TOP1 covalent complexes at the 3â²-ends are restored to 3â²-phosphate ends by TDP1 (d). [4] After removing damaged 3â²-ends, Pol β fills the gap (red dot line). [5] Lig3 seals the single-strand nick (red line).
|
