E3 ubiquitin ligase activity of the trifunctional ARD1 (ADP-ribosylation factor domain protein 1)

Abstract

Protein ubiquitinylation plays a key role in many important cellular processes. Ubiquitinylation requires the E1 ubiquitin-activating enzyme, an E2 ubiquitin-conjugating enzyme, and, frequently, a substrate-specific E3 ubiquitin-protein ligase. In one class of E3 ubiquitin ligases, the catalytic domain contains a zinc-binding RING finger motif. ARD1 (ADP-ribosylation factor domain protein 1), with a RING finger domain in the N-terminal region, two predicted B-Boxes, and a coiled-coil protein interaction motif immediately preceding an ADP-ribosylation factor domain at the C terminus, belongs to the TRIM (Tripartite motif) or RBCC (RING, B-Box, coiled-coil) family. The region containing the B-Boxes and the coiled-coil motif acts as a GTPase-activating protein for the ADP-ribosylation factor domain of ARD1. We report here that full-length ARD1 or the RING finger domain (residues 1-110) produced polyubiquitinylated proteins in vitro in the presence of mammalian E1, an E2 enzyme (UbcH6 or UbcH5a, -5b, or -5c), ATP, and ubiquitin. Deletion of the RING region or point mutations within the RING sequence abolished ARD1 E3 ligase activity. All data are consistent with a potential function for ARD1 as an E3 ubiquitin ligase in cells.

Fig. 1.

ARD1 and its mutant forms: structure and SDS/PAGE of recombinant GST-proteins. (Upper) Predicted domain structures of ARD1, a TRIM/RBCC family member, and related GST-tagged recombinant proteins (6, 9). (Lower) Samples (0.5 μg) of purified recombinant GST proteins were subjected to SDS/PAGE in 4–20% Tris-glycine gels and detected by silver staining. The doublet at 26 kDa probably represents GST generated via proteolysis of bacterially synthesized GST-fusion proteins and was present in all GST-ARD1 preparations.

Fig. 6.

Ubiquitinylation of ARD1, free GST, and UbcH6 (E2) in vitro. Standard assay mixture (150 μl) containing 75 nM GST-ARD1 (2.2 pmol/30 μl) (A) or 300 nM His-tagged ARD1 (9 pmol/30 μl) (B and C) was incubated at 30°C with 26 nM E1 (0.8 pmol/30 μl), 0.53 μM UbcH6 (E2, 16 pmol/30 μl), and 3.9 μM recombinant human ubiquitin (117 pmol/30 μl) or mutant human ubiquitin in which arginine replaced all lysines (asterisks) (see methods regarding mutant ubiquitin concentration). At the indicated times, 30-μl samples were removed and added to 10 μl of 4× Laemmli sample buffer. (A) Blot was reacted with anti-GST antibodies before stripping and reaction with anti-ubiquitin antibodies. Free GST was present in all GST-ARD1 preparations, because it was bound along with GST-ARD1 to glutathione-Sepharose during purification. (B and C) Reactions contained His-tagged ARD1 and either wild-type or lysine-free ubiquitin (B, asterisk), or only lysine-free ubiquitin (C). Blots were reacted with anti-ubiquitin, anti-ARD1, or anti-UbcH6 antibodies as indicated. Arrow and arrowheads indicate, respectively, the unmodified and modified forms of the UbcH6 (E2). Data were replicated twice.

Fig. 2.

GST-ARD1-catalyzed formation of ubiquitinylated proteins in vitro required E1 and E2. Samples (2.2 pmol) of native or heat-inactivated (asterisk) GST-ARD1 (1–574), GST-cytohesin-1 (C-1), or GST were incubated with 0.8 pmol of E1 and/or 16 pmol of E2 (UbcH6) in standard ubiquitinylation assays (total volume of 30 μl) for 60 min at 30°C before separation of proteins by SDS/PAGE. Blots were reacted with anti-ubiquitin monoclonal antibody. Dots indicate positions of mono- or diubiquitin; the arrowhead indicates ubiquitinylated E2. Data were replicated at least three times.

Fig. 3.

Specificity of E2 requirement for in vitro ubiquitinylation with recombinant GST-ARD1. Standard ubiquitinylation assays contained 0.8 pmol of E1, 2.2 pmol of GST-ARD1 (Upper) or GST (Lower), and 0.4 μg of the indicated human E2 ubiquitin-conjugating enzyme. Ubiquitinylated proteins were detected with anti-ubiquitin antibodies; arrowhead indicates ubiquitinylated UbcH6 (E2). The experiment was repeated twice with similar results. Activities of UbcH1 and UbcH3/CDC34 were demonstrated by the production of E2-ubiquitin conjugates that were stable under reducing conditions (data not shown).

Fig. 4.

Ubiquitinylation catalyzed by E1, E2, and GST-ARD1 (E3) as a function of time and enzyme concentrations. (Left) Standard assay mixture (270 μl) containing 75 nM ARD1 (2.2 pmol/30 μl) was incubated at 30°C with 26 nM E1 (0.8 pmol/30 μl) and 0.53 μM UbcH6 (E2, 16 pmol/30 μl). At the indicated times, 30-μl samples were removed and added to 10 μl of 4× Laemmli sample buffer. (Center) Standard assays (30 μl) containing 0, 0.1, 0.2, 0.4, 0.8, or 1.6 pmol of GST-ARD1 or 2.2 pmol of GST were incubated for 60 min. (Right) Assays (30 μl) containing 2.2 pmol of GST-ARD1 with the indicated amounts (pmol) of E1 and E2 were incubated for 60 min. Data in Left and Center were replicated at least twice.

Fig. 5.

Requirement of intact ARD1 RING finger domain for E3 ubiquitin ligase activity. (Top Left) Ubiquitinylation activity of GST-ARD1 or the indicated mutant proteins (5 pmol each). (Right) Standard assays (30 μl) contained, as indicated, 0.8 pmol of E1 and 16 pmol of UbcH6 (E2), with 2.2 pmol of the indicated recombinant GST-ARD1 or GST protein. (Middle Left) Standard assays contained, as indicated, His-tagged ARD1 protein (9 pmol) or GST-ARD1 (2.2 pmol) with E1 and/or E2; E1/E2 lane contained no ARD1. In Bottom Left, the same blot reacted with anti-ARD1 antibodies is showing unmodified GST- and His-ARD1. Arrowheads indicate ubiquitinylated E2. Data were replicated at least twice.