Autoubiquitylation of the V(D)J recombinase protein RAG1

Abstract

V(D)J recombination, the rearrangement of gene segments to assemble Ig and T cell receptor coding regions, is vital to B and T lymphocyte development. Here, we demonstrate that the V(D)J recombinase protein RAG1 undergoes ubiquitylation in cells. In vitro, the RING finger domain of RAG1 acts as a ubiquitin ligase that mediates its own ubiquitylation at a highly conserved K residue in the RAG1 amino-terminal region. Ubiquitylation is best supported by a specific ubiquitin-conjugating enzyme, UbcH3/CDC34, and requires an intact RAG1 RING finger motif. Disruption of the RING finger and certain RAG1 N-terminal truncations are associated with immunodeficiency in human patients, suggesting that RAG1's ubiquitin ligase is required for its biological role in lymphocyte development.

Fig. 1.

Full-length RAG1 is ubiquitylated in intact cells. Full-length RAG1 (R1; lanes 2–5 and 7–10) and RAG2 (R2; lanes 4, 5, 9, and 10) were expressed in HeLa cells, and pull-down assays were performed as described in Methods. Products were analyzed by Western blot probed with anti-FLAG antibody (a) and antiubiquitin protein conjugate (b).

Fig. 2.

RAG1-B/RING exhibits E3 activity. (a) In vitro ubiquitylation assays were performed as described in Methods with E1 enzyme, radiolabeled PK-ubiquitin (Ubi*), E2 enzymes (0.3 μM), and RAG1-B/RING as indicated. (b) In vitro ubiquitylation assays were performed as described with the complete (Comp) reaction including Ubi*, RAG1-B/RING (R1), E1, UbcH3 (E2). Components were omitted as indicated (lanes 3–5). In lanes 6 and 7, WT RAG1-B/RING was replaced with the C325G mutant form or with RAG1-B/RING from which Zn had been removed with diethylenetriaminepenta-acetic acid (DTPA). RAG1-B/RING-dependent products are indicated by arrows. (c) In vitro ubiquitylation assays including apyrase (0.1 mg/ml) were staged, with components added at the times (in min) indicated. Buffer components and Ubi* were present from time 0. Reactions were incubated for an additional 2 h after the addition of R1. RAG1-B/RING-dependent products are indicated by arrows.

Fig. 3.

RAG1-B/RING undergoes autoubiquitylation. (a) In vitro ubiquitylation assays were performed as described in Methods with the complete reaction including E1, E2, RAG1-B/RING (R1), and unlabeled PK-ubiquitin (PK-Ubi). Components were omitted as indicated (lanes 3–6). RAG1-B/RING was detected by using anti-Xpress epitope antibody conjugated to horseradish peroxidase; lanes 7 and 8 are a longer exposure of lanes 1 and 2. (b) In vitro ubiquitylation assays were performed as described in Methods for the incubation times indicated. RAG1-B/RING was detected as in a except that unconjugated anti-Xpress antibody was detected with secondary antibody conjugated to Alexa 488 fluorescent probe. (c) In vitro ubiquitylation assays were performed by using PK-Ubi, untagged, reductively methylated ubiquitin (CH₃-Ubi), ubiquitin K48R, or WT ubiquitin.

Fig. 4.

Conserved basic domains are required for RAG1-B/RING E3 activity. In vitro ubiquitylation assays were performed as described in Methods with radiolabeled PK-ubiquitin (a) or unlabeled PK-ubiquitin (b) and RAG1 B/RING (218–389) or RING (264–389) at the concentration indicated (a) or 8 μM (b). (a) Radiolabeled products were detected by phosphor-storage autoradiography. (b) RAG1 218–389 or 264–389 was detected by using anti-Xpress epitope antibody.

Fig. 5.

RAG1-B/RING autoubiquitylation occurs primarily on K233. (a) Alignments of RAG1 segments from various species were conducted by using the National Center for Biotechnology Information blast resource. (b) In vitro ubiquitylation assays were carried out as described in Methods with unlabeled PK-ubiquitin and B/RING WT, K221M, or K233M. RAG1 B/RING was detected by using anti-Xpress epitope antibody. (c) In vitro ubiquitylation assays were carried out by using radiolabeled PK-ubiquitin and RAG1-B/RING WT or K233M.