V(D)J recombination proceeds through a series of protein:DNA complexes mediated in part by the RAG1 and RAG2 proteins. These proteins are responsible for sequence-specific DNA recognition and DNA cleavage, and they appear to perform multiple postcleavage roles in the reaction as well. Here we review the interaction of the RAG proteins with DNA, the chemistry of the cleavage reaction, and the higher order complexes in which these events take place. We also discuss postcleavage functions of the RAG proteins, including recent evidence indicating that they initiate the process of coding end processing by nicking hairpin DNA termini. Finally, we discuss the evolutionary and functional implications of the finding that RAG1 and RAG2 constitute a transposase, and we consider RAG protein biochemistry in the context of several bacterial transposition systems. This suggests a model of the RAG protein active site in which two divalent metal ions serve alternating and opposite roles as activators of attacking hydroxyl groups and stabilizers of oxyanion leaving groups.

A system for tetracycline-regulated inducible gene expression was described recently which relies on constitutive expression of a tetracycline-controlled transactivator (tTA) fusion protein combining the tetracycline repressor and the transcriptional activation domain of VP16 [Gossen, M. & Bujard, H. (1992) Proc. Nati. Acad. Sci. USA 89, [5547][5548][5549][5550][5551]. This system yielded only low levels of transactivator protein, probably because tTA is toxic. To avoid this difficulty, we placed the tTA gene under the control of the inducible promoter to which tTA binds, making expression of tTA itself inducible and autoregulatory. When used to drive expression of the recombination activating genes 1 and 2 (RAG-1 and RAG-2), the autoregulatory system yielded both substantially higher levels of variable (diversity) joining [V(D)J] recombination activity (70-fold on average) and inducible expression in a much larger fraction of transfected cells (autoregulatory, 90%, vs. constitutive, 18%). In addition, this system allowed the creation of transgenic mice in which expression of a luciferase transgene was inducible tens to hundreds of times the basal levels in most tissues examined. Induced levels of expression were highest in thymus and lung and appear to be substantially higher than in previously reported inducible luciferase transgenic mice created with the constitutive system. With the modified system, inducible transactivator mRNA and protein were easily detected in cell lines by RNA and Western blotting, and transactivator mRNA was detected by RNA blotting in some tissues of transgenic mice. This autoregulatory system represents an improved strategy for tetracycline-regulated gene expression both in cultured cells and in transgenic animals.

Several inducible gene expression systems have been developed in vitro in recent years to overcome limitations with traditional transgenic mice. One of these, the tetracycline-regulated system, has been used successfully in vivo. Nevertheless, concerns remain about the ability of this system to direct high levels of transgene expression in vivo and to enable such expression to be turned on and off effectively. We report here the generation, using a modified tetracycline-regulated system under the control of the neuron-specific enolase promoter, of several lines of mice that direct transgene expression to specific brain regions, including the striatum, cerebellum, CA1 region of the hippocampus, or deep layers of cerebral neocortex. Transgene expression in these mice can be turned off completely with low doses of doxycycline (a tetracycline derivative) and driven to very high levels in the absence of doxycycline. We demonstrate this tissue-specific, inducible expression for three transgenes: those that encode luciferase (a reporter protein) or DeltaFosB or the cAMP-response element binding protein (CREB) (two transcription factors). The various lines of transgenic mice demonstrate an inducible system that generates high levels of transgene expression in specific brain regions and represent novel and powerful tools with which to study the functioning of these (or potentially any other) genes in the brain.

The lymphoid cell-specific proteins RAG1 and RAG2 initiate V(D)J recombination by cleaving DNA adjacent to recombination signals, generating blunt signal ends and covalently sealed, hairpin coding ends. A critical next step in the reaction is opening of the hairpins, but the factor(s) responsible has not been identified and had been thought to be a ubiquitous component(s) of the DNA repair machinery. Here we demonstrate that RAG1 and RAG2 possess an intrinsic single-stranded nuclease activity capable of nicking hairpin coding ends at or near the hairpin tip. In Mn 2؉, a synthetic hairpin is nicked 5 nucleotides (nt) 5 of the hairpin tip, with more distant sites of nicking suppressed by HMG2. In Mg 2؉ , hairpins generated by V(D)J cleavage are nicked whereas synthetic hairpins are not. Cleavage-generated hairpins are nicked at the tip and predominantly 1 to 2 nt 5 of the tip. RAG1 and RAG2 may therefore be responsible for initiating the processing of coding ends and for the generation of P nucleotides during V(D)J recombination.V(D)J recombination assembles the variable regions of antigen receptor genes during lymphocyte development by joining together V (variable), J (joining), and in some cases D (diversity), coding gene segments (28). Recombination is specifically directed to coding elements by recombination signal sequences (RSSs) which flank the segments to be joined. These RSSs consist of a conserved heptamer which is contiguous to the coding flank and an AT-rich nonamer. Heptamer and nonamer are separated by a nonconserved spacer of either 12 or 23 bp, yielding the 12-RSS or 23-RSS, respectively. In vivo, recombination primarily occurs between coding elements with RSS spacers of different lengths, thereby preventing the joining of inappropriate elements. This restriction is referred to as the 12/23 rule.Mechanistically, the recombination reaction is envisioned to occur in two stages. In the first stage, initiation of recombination is mediated by the proteins encoded by the recombinationactivating genes, RAG1 and RAG2, which bind directly to RSSs (10,32,47,51). Binding by RAG proteins is followed by RSS synapsis and concerted cleavage at both signals. Cleavage involves hydrolytic nicking at the heptamer-coding flank border, and the the 3Ј-hydroxyl thus generated serves as a nucleophile to attack the phosphodiester bond on the other DNA strand opposite the nick in a direct transesterification reaction (32, 52). The coding ends generated by cleavage are covalently sealed DNA hairpins, while signal ends are blunt and 5Ј phosphorylated. This reaction is stimulated in vitro, especially at the 23-RSS, by addition of DNA-bending proteins HMG1 and HMG2 (44,50). Coordinate cleavage in accordance with the 12/23 rule requires Mg 2ϩ rather than Mn 2ϩ as divalent metal cofactor and is stimulated by HMG1 or HMG2 (12,43,44,50,53). In the second stage of the reaction, coding ends are processed, and coding joints and signal joints form in reactions with similarities to the repair of DNA double-strand breaks by nonhomologo...