br Materials and methods br Results Artemis
Materials and methods
Results Artemis has been reported to possess 5′–3′ exonuclease activity in vitro on ssDNA, as well as DNA-PK-dependent endonuclease activity on single-strand overhang and hairpin DNA structures . However, enzymes within the metallo-β-lactamase family typically contain only one active site that has been shown to be the functional catalytic site for all substrates . Possessing two different nuclease activities that are located within two different active sites would make Artemis unique in the metallo-β-lactamase family. We sought to determine biochemically if in fact the reported 5′–3′ exonuclease activity of Artemis is an intrinsic activity of the Artemis polypeptide. To accomplish this, following cloning and overexpression of Artemis, we undertook the process of fractionating the [His]6-Artemis fusion protein via column chromatography and monitoring exonuclease activity. A three-step protein purification procedure was developed including anionic exchange, nickel-affinity and hydroxyapatite column chromatography (Fig. 1A). Following expression of the human recombinant [His]6-Artemis protein in insect az628 using a baculovirus expression system, a cell-free extract was prepared and Artemis expression analyzed. Analysis of Artemis nuclease enzymatic activity is not possible in crude extracts due to the abundance of endogenous nucleases (data not shown). To detect Artemis expression, we analyzed cell-free extracts prepared from infected cells by SDS-PAGE, western blot and phosphorylation by DNA-PK. Coomassie Blue staining of an SDS denaturing gel revealed expression of the [His]6-Artemis protein (Fig. 1B) that was confirmed by western blot analysis. This revealed a dominant band at the expected molecular mass for the recombinant His-tagged fusion protein using both a monoclonal antibody against the N-terminal [His]6 fusion tag (Fig. 1C) and a polyclonal antibody against the full length Artemis polypeptide (data not shown). Artemis has been shown to possess eleven serine/threonine residues that are phosphorylated by DNA-PK in vitro. To confirm that the baculovirus expressed [His]6-Artemis can be phosphorylated by DNA-PK, we performed an in vitro DNA-PK phosphorylation reaction. The cell-free extract was incubated with purified DNA-PK, DNA and [γ-32P]ATP and the reaction products were separated by SDS-PAGE and radioactive incorporation of the phosphate into protein was detected by PhosphorImager analysis. Again, a prominent band appeared at the same size as that seen in the western blot of Artemis (Fig. 1D), indicative of DNA-PK-dependent phosphorylation of [His]6-Artemis in the cell-free extract. Following preparation of the whole cell extract from insect cells infected with recombinant [His]6-Artemis baculovirus, the concentration of potassium chloride was increased to 0.5M and mixed with phosphocellulose chromatography media. The high-salt concentration allowed for the majority of the protein contained in the extract (including [His]6-Artemis) to flow through the matrix. The flow-through protein was collected and loaded directly onto a 2mL nickel–NTA agarose column. The column was washed extensively with high ionic strength buffer followed by low ionic strength buffer, both containing 5mM imidazole. The bound protein was eluted with 200mM imidazole and collected in 1mL fractions. Analyses of these pools revealed that the majority of [His]6-Artemis bound to the Ni–NTA agarose matrix and was eluted with 200mM imidazole. The load, flow-through and eluate from the nickel column were analyzed by Coomassie Blue staining of SDS-PAGE (Fig. 2A). The presence of [His]6-Artemis in these fractions was assessed by western blotting (Fig. 2B), and phosphorylation by DNA-PK (Fig. 2C). In each case the majority of the Artemis applied to the column (lane 1) was retained in the imidazole elution (lane 3). We found that retention of Artemis on the Ni-matrix was very sensitive to chromatographic conditions and the presence of reducing agents, which should be avoided to maximize yield (data not shown). While these results indicate a degree of purity of [His]6-Artemis can be achieved by nickel-affinity chromatography, the Coomassie Blue stained gel and phosphorylation assay reveal significant impurities in the eluate, indicating the need for further fractionation to achieve a higher degree of purity.