PMC

Untreated and H₂O₂-treated HeLa cells were fixed and stained with anti-human NDK1 antibody 24 hours after DNA damage induction. The NDK1 signals were visualized in the right panels. Control nuclear staining was performed with DAPI (left panels).

Untreated and DNA damaging agents-treated HeLa cells were fixed and stained with anti-human NDK1 antibody 20 min after DNA damage induction. The stained signals were visualized by Alexa Fluor 488-conjugated secondary antibody (middle). Control nuclear staining was performed with DAPI (left). Merged images are also shown (right).

A. Ni-NTA column-purified human wild type and mutant NDK1 proteins on an SDS-PAGE gel. B. Exonuclease activity of purified human wild type and mutant NDK1 proteins was tested with a single-stranded 30-mer oligonucleotide and equal amounts (100 ng) of purified proteins. C. Enzymatic assay for nucleoside diphosphate kinase activity. D54A, E5Q and E5A mutants retained relatively high kinase activity.

Oligonucleotide substrates with single stranded, double stranded, or partially single stranded configurations were incubated with no protein (control) or with human NDK1 (200 ng). The cleavage products were separated on polyacrylamide gels. A. Composition of the substrates. The asterisks indicate the positions of the label. B. Cleavage assays. Exonuclease activity of human NDK1 was most pronounced with single-stranded substrates or with substrates containing single-stranded 3' overhangs. Similar data were obtained with human NDK5 and NDK7 (data not shown).

A. The sequence alignment of E. coli NDK protein and the NDK domains found in eight human homologues is shown. Alignment was performed with the Clustal W program. For NDK7, the N-terminal NDK domain was used for alignment and for NDK8, the central domain was used. B. Domain structures of E. coli NDK and the human NDK domain proteins. Additional domains indicated are a Dpy30 motif (probably formed of two alpha-helices) in NDK5, a DM10 motif (unknown function) in NDK7, and a thioredoxin domain in NDK8.

A. The single-stranded 30-mer oligonucleotide (5 pmoles) was labeled at the 5' end. Purified recombinant E. coli NDK and human NDKs (100 ng each) were tested for exonuclease activity. No, no enzyme. B. Ni-NTA column-purified human NDK1 was eluted from a DEAE sepharose column and the fractions were assayed for exonuclease activity (top panel) and protein content by SDS-PAGE (bottom panel).

A. The competition effect of E129A and E129Q mutants on wild type NDK1 exonuclease activity is shown. Various amounts of E129A and E129Q proteins were preincubated with the single-stranded oligonucleotide substrate followed by incubation with 0.5 μg of wild type human NDK1. B. E5A and E5Q mutants were used for the competition assay. C. Increasing amounts of wild type NDK1 were used with the same substrate. D. Quantified data of the competition effects by human NDK1 mutants. The cleaved substrates were quantified by phophoimager analysis and converted to percentage values relative to wild type NDK1.

Overlay of a well-resolved region of the ¹H-¹⁵N HSQC spectra of NDK1, free and bound to DNA. Resonances in black correspond to the spectrum of free NDK1 while those in red correspond to the DNA-bound protein when the protein to DNA ratio is 1:2. The cross peak that shows a specific chemical shift change in this region is indicated by a blue arrow. Two other cross peaks in this region disappeared upon binding to DNA, suggesting that chemical perturbation also occurs in these resonances.

A. The recombinant proteins, obtained after Ni-NTA-agarose and DEAE sepharose chromatography, were resolved on an SDS-polyacrylamide gel and stained with Coomassie Blue. Preparations of NDK2, NDK3, and NDK6 contained a contaminant of 45 kDa, which could not be separated from the NDK proteins. B. Phosphoenzyme essay. Catalytically active NDKs contain a histidine phosphotransferase activity. This could be demonstrated for E. coli NDK, and human NDK1, NDK2, NDK4, and NDK8. C. Enzymatic assay for nucleoside diphosphate kinase activity. A decrease in absorbance at 340 nm is indicative of the presence of enzymatic activity. This could be demonstrated for E. coli NDK, and human NDK1, NDK2, and NDK4.

A. Cytoplasmic and nuclear extracts from DNA damaging agent-treated HeLa cells were separated on SDS-PAGE. Endogenous human NDK1 in both cytoplasmic (C) and nuclear extracts (N) was detected with anti-human NDK1 antibody. HDAC2 was used as a control for the nuclear protein fraction and as a loading standard. β-tubulin was used as a cytoplasmic specific and loading control. Cells were lysed 20 min after DNA damage induction. The “increased ratio” refers to the relative increase in the percentage of nuclear localized protein in damaging agent-treated cells relative to untreated cells, corrected for the values of the loading controls. B. Localization analysis of human NDK1 in untreated and DNA damaging agents-treated ARPE-19 cells.

A. Cleavage of an abasic site-containing oligonucleotide by combined reaction with UDG, human APE1 and human NDK1 proteins. The abasic site was created by eUDG that excised uracil from a double stranded oligonucleotide containing a U/A mispair. The abasic site was then cleaved by human APE1 followed by reaction with human NDK1 protein. Human NDK1 showed specific exonuclease activity emanating from the cleaved abasic site. B. Cleavage of an abasic site-containing oligonucleotide by combined reaction with eNth and human NDK1 proteins. The double stranded oligonucleotide containing a Tg/A mispair was preincubated with eNth followed by the addition of human NDK1 protein. C. Cleavage of an abasic site-containing oligonucleotide by combined reaction with E. coli UDG or human UDG, human APE1 and human NDK1, NDK5 or NDK7 proteins. eUDG (lane 3-4) or hUDG (lane 1-2 and 5-8) were preincubated with double stranded oligonucleotides containing U/A followed by adding hAPE1. Human NDKs were then added to the reaction mixture. D. The exonuclease activity of human NDK1 in the presence or absence of UDG and hAPE1. The reaction conditions were as described in panel A. Preincubated hUDG and hAPE1 were removed by phenol/chloroform extraction and then human NDK1 protein was added (lanes 7-10). The exonuclease activity of human NDK1 was much higher in the presence of UDG and hAPE1 (lanes 3-6).