Figure 1.
ARTD15 is expressed in different human cell lines.
(A) Schematic diagram of ARTD15 domain structure. CD: catalytic domain; TD: transmembrane domain. (B) Levels of ARTD15 transcript determined by qRT-PCR, normalized to GAPDH RNA and then reported as arbitrary units relative to the ARTD15 transcript in HK2 cells (taken as 1). Data shown represent the mean (±SD) of two independent experiments performed in triplicate. (C) Expression of ARTD15. Western blotting (WB) showing endogenous ARTD15 protein and actin levels in 50 µg protein from total cell lysates. The expression levels of the ARTD15, normalized to actin, are shown in the histogram relative to those of HEK293 cells (taken as 100). Data shown represent the mean (±SD) of two independent experiments performed in triplicate. (D) Ponceau S staining and Western blotting (WB) of 100 µg protein from total HEK293 cell lysate are shown. (E) Immunofluorescence staining of endogenous ARTD15 (green) in combination with PDI (red) shows co-localization of ARTD15 protein with the ER. Bar, 20 µm.
Figure 2.
ARTD15 is an ER resident protein.
HeLa cells were transfected with FLAG-ARTD15. (A) Immunofluorescence staining of ARTD15 (green) in combination with the indicated markers (red) shows co-localization of ARTD15 protein with the ER. Bar, 20 µm. (B) Electron microscopy analysis of immuno-gold staining of ARTD15 (black dots). Arrows show ER tubular structures and nuclear envelope; cellular organelles are indicated by abbreviations (n: nucleus; pm: plasma membrane; m: mitochondria; ne: nuclear envelope). Bar, 500 nM. Data shown are representative of three independent experiments.
Figure 3.
ARTD15 is an ER tail-anchored protein.
HeLa cells were transiently transfected with full-length (ARTD15) or deleted (ΔTM-ARTD15) FLAG-ARTD15 and analyzed either by (A) immuno-fluorescence microscopy with an anti-FLAG antibody; Bar, 20 µm or by (B) Western blotting of total lysates (60 µg), cytosol and total membrane (30 µg) proteins using anti-FLAG antibody to visualize ARTD15 and anti-GRK2 and anti-calnexin antibodies as a control of cell fractionation. (C) Protease protection assay performed with HeLa cells transfected with N-termini or C-termini GFP-ARTD15. Western blotting of ARTD15 revealed with an anti-GFP antibody, and of calnexin revealed with antibodies raised against the N-termini or C-termini of calnexin. (D) Schematic representation of ARTD15 (based on our results) and calnexin protein orientation. The N-termini (N) and C-termini (C) of the proteins with respect to the endoplasmic reticulum are indicated. The data shown in A, B, and C are representative of at least three independent experiments.
Figure 4.
ARTD15 is an active ADP-ribosyl transferase.
(A & C) Two µg of recombinant, purified GST-ARTD15 or GST-ARTD15-H152A/Y254A (ARTD15 dm) were [32P]-ADP-ribosylated in vitro in the presence of increasing concentration of NAD (as indicated in A) or with 10 µM NAD (C) and analysed by autoradiography (AR). Ponceau red staining is a control of protein loading. (B) Recombinant, purified GST-ARTD15 (300 ng) was [32P]-ADP-ribosylated and analysed by autoradiography (AR). The graph indicates the picomoles of [32P]-ADP-ribosylated ARTD15. The inset shows the Lineweaver-Burk analysis of the data. The data shown are the means (±SD) of four independent experiments performed in triplicate. (D) Total membrane proteins (50 µg) obtained from HeLa cells transiently transfected with empty vector (control), ARTD15 or ARTC1, as indicated, were [32P]-ADP-ribosylated in vitro in the absence (left) or presence (right) of 1 mM agmatine. The supernatants and, as a further control, [32P]-NAD (NAD) and [32P]-ADP-ribose (NADase) were analysed by TLC and autoradiography. The data shown are representative of at least five independent experiments. (E) Two µg of recombinant, purified GST-ARTD15 (ARTD15) or plasma membranes added with 250 ng of purified ßγ dimer (ß subunit) were ADP-ribosylated with [32P]-NAD+ for 1 h at 37°C, in the presence or absence of MIBG. The loading control is shown (WB). Data shown are representative of at least three independent experiments. (F) ADP-ribosylated ARTD15 was blotted and the filters were treated with the indicated compounds. Data reported in the graph are the mean (±SD) of three independent experiments performed in duplicate (control: untreated sample). The inset shows a representative experiment.
Figure 5.
(A) Lysates (6 mg protein) from HeLa cells transfected with empty vector (control) or FLAG-ARTD15 were immunoprecipitated with a polyclonal anti-FLAG antibody. Proteins were separated by 10% long SDS-PAGE and proteins revealed by silver staining. Differential proteins were excised from the gel and identified by MALDI-ToF mass spectrometry (boxed bands). (B) Cell lysates from HeLa cells (106 cells/assay) transfected with FLAG-ARTD15 were immunoprecipitated with an anti-Kapß1 antibody or with control IgG. The input is shown (1/20 of the total sample). (C) GST or GST-ARTD15 (0.1 µM) were incubated with increasing amounts of His-Kapß1. GST proteins were pulled down with gluthathione resin. Precipitated Kapß1 protein was probed with an anti-His antibody. (D) Immunofluorescence staining of endogenous ARTD15 (green) and endogenous Kapß1 (red) in HeLa cells. Bar, 20 µm.
Table 1.
Protein identification.
Figure 6.
ARTD15 catalyses Kapß1 mono-ADP-ribosylation.
(A) His-Kapß1 was [32P]-ADP-ribosylated in vitro for 6 hours at 37°C in presence of GST-ARTD15, separated by SDS-PAGE and analysed by autoradiography (AR). Recombinant proteins are shown by immunoblotting with anti-GST (ARTD15) and anti-His antibodies. (B) [32P]-ADP-ribosylation products derived from [32P]-ADP-ribosylated GST-ARTD15, His-Kapß1 and from [32P]-ADP-ribosylated agmatine and [32P]-NAD, as controls, were analysed by high-resolution PAGE to visualise ADP-ribose chain length. (C) GST-ARTD15 and His-Kapß1 were [32P]-ADP-ribosylated in vitro for 6 hours at 37°C in presence of 10 µM NAD, in presence of increasing concentration of the PARP1 inhibitor PJ34 (0.1–100 µM), separated by SDS-PAGE and analysed by autoradiography (AR). [32P]-ADP-ribosylated PARP1 was used as control. Immunoblotting with anti-PARP1, anti-GST (ARTD15) and anti-His (Kapß1) showed loaded protein. (D) Total membranes (50 µg) from HeLa cells transfected with empty vector (control) or ARTD15 were [32P]-ADP-ribosylated in vitro in presence of 10 µM NAD. Solubilised proteins were analysed by SDS-PAGE and autoradiography (AR). The single differential target (*) was recognized with an anti-Kapß1 specific antibody (WB). WB also shows over-expressed ARTD15, and the loading control GRASP55. The data shown are representative of five to ten experiments. (E) Quantification of [32P]-ADP-ribosylated Kapß1 in total membranes prepared from both control (scrambled) and ARTD15-silenced Hela cells. The data are means (±SD) of three experiments. The inset shows an example of the inhibition of Kapß1 ADP-ribosylation (AR) and of the ARTD15 knock-down (WB). WB also shows Kapß1, and GRASP55 as a loading control.
Table 2.
List of primers used in this study.