Gene targeting using a promoterless gene trap vector (‘‘targeted trapping’’) is an efficient method to mutate a large fraction of genes Roland H. Friedel*†, Andrew Plump*†‡, Xiaowei Lu*†, Kerri Spilker*†, Christine Jolicoeur*†, Karen Wong*†, Tadmiri R. Venkatesh§, Avraham Yaron*†, Mary Hynes*, Bin Chen*, Ami Okada*, Susan K. McConnell*, Helen Rayburn*†, and Marc Tessier-Lavigne*†¶ *Department of Biological Sciences, †Howard Hughes Medical Institute, Stanford University, Stanford, CA 94305; and §Department of Biology, City College, City University of New York, New York, NY 10031 Contributed by Marc Tessier-Lavigne, July 11, 2005 A powerful tool for postgenomic analysis of mammalian gene expressed at a sufficient level in ES cells to be accessible by this function is gene targeting in mouse ES cells. We report that method. homologous recombination using a promoterless gene trap vector In contrast to random gene trapping, gene targeting provides (‘‘targeting trapping’’) yields targeting frequencies averaging a directed strategy that exploits homologous recombination to above 50%, a significant increase compared with current ap- modify a specific gene locus. Traditional vectors used for ho- proaches. These high frequencies appear to be due to the strin- mologous recombination usually contain an antibiotic resistance gency of selection with promoterless constructs, because most gene driven by a constitutive promoter (such as the Pgk pro- random insertions are silent and eliminated by drug selection. The moter). The advantage of this method is that a particular gene promoterless design requires that the targeted gene be expressed of interest can be targeted regardless of whether it is expressed in ES cells at levels exceeding a certain threshold (which we in ES cells. The disadvantage is that the vector can confer estimate to be Ϸ1% of the transferrin receptor gene expression antibiotic resistance to transfected ES cells, whether it has level, for the secretory trap vector used here). Analysis of 127 genes recombined into the targeted locus or has inserted in a nonspe- that had been trapped by random (nontargeted) gene trapping cific way. As a consequence of this lack of stringent selection, the with the same vector shows that virtually all are expressed in ES rate of correctly targeted ES cell clones (the ‘‘targeting fre- quency’’) in a typical gene-targeting experiment is usually only cells above this threshold, suggesting that targeted and random low to moderate, in the range of 0.5- 5%, making it a more trapping share similar requirements for expression levels. In a labor-intensive procedure than random gene trapping. random sampling of 130 genes encoding secretory proteins, about An alternative strategy for gene targeting is to use promot- half were expressed above threshold, suggesting that about half erless gene-targeting vectors, the kind used in random gene of all secretory genes are accessible by either targeted or random trapping, and to rely on the endogenous promoter of the targeted gene trapping. The simplicity and high efficiency of the method gene to drive expression of drug resistance. Because the target- facilitate systematic targeting of a large fraction of the genome by ing vector does not have a promoter to drive antibiotic resistance, individual investigators and large-scale consortia alike. most nonspecific insertions are eliminated, so that this approach should in principle be able to reach much higher targeting ES cells ͉ gene trapping frequencies than conventional gene targeting. The utility of promoterless gene targeting (an approach we call ‘‘targeted ene trapping by random insertion of a vector into the trapping’’) was recognized in the early years of gene targeting Ggenome of mouse ES cells has allowed the generation of a and was successfully applied to mutate several genes (4–10). For large collection of mutations that are an invaluable resource for the method to work, however, the targeted gene must be the study of mammalian biology (1, 2). In a typical 5Ј gene expressed at a sufficiently high level in ES cells to confer trapping strategy, the trapping vector comprises a 5Ј splice antibiotic resistance. In these early studies, no attempt was made acceptor site and a promoterless antibiotic resistance gene [such to assess the level at which a gene has to be expressed in ES cells as a gene for neomycin phosphotransferase (neo)] fused to a to make it a successful target for a promoterless construct, and marker gene (such as a gene for ␤-gal). When the vector inserts the predicted improvement in targeting frequencies expected in-frame in an intron of a gene expressed in ES cells, it traps the from the approach was not systematically explored (4–10), upstream exons, resulting in a transcript that directs the expres- potentially explaining why this approach has not been widely sion of a fusion protein comprising the sequences encoded in the adopted by other laboratories in the intervening decade. trapped exons and those in the vector. This allows for isolation We therefore set out to define the level of gene expression of productive insertions by selection with the antibiotic and required for promoterless gene targeting (targeted trapping) to be successful and to assess whether this method can be used to subsequent screening for ␤-gal-positive colonies, provided the access a significant fraction of genes in the genome. Our results, trapped gene is expressed at a sufficiently high level in ES cells. focused primarily on using the secretory gene trap vector, The advantage of random gene trapping is the extreme ease with which a large number of mutations can be generated. In addition, specific vector designs can enrich for insertions in specific classes Freely available online through the PNAS open access option. of genes; thus, a ‘‘secretory gene trap vector’’ was designed to Abbreviations: Trfr, transferrin receptor; TM, transmembrane domain; neo, neomycin target genes encoding proteins with a signal sequence (secreted phosphotransferase; GO, Gene Ontology. and membrane anchored proteins) (3). The disadvantage of See Commentary on page 13001. random gene trapping is that the specific genes that are trapped ‡Present address: Merck Research Laboratories, 126 East Lincoln Avenue, Rahway, NJ 07065. cannot be specified in advance, so there is no guarantee that any ¶To whom correspondence should be sent at the present address: Genentech, Inc., 1 DNA particular gene of interest will be mutated. In addition, to date, Way, South San Francisco, CA 94080. E-mail: [email protected]. it has been unclear what fraction of the genes in the genome are © 2005 by The National Academy of Sciences of the USA 13188–13193 ͉ PNAS ͉ September 13, 2005 ͉ vol. 102 ͉ no. 37 www.pnas.org͞cgi͞doi͞10.1073͞pnas.0505474102 establish that this method is likely to be useful for a large fraction Table 1. Targeting frequencies of targeted trapping constructs of genes and define an approximate level of gene expression in Expression level, SEE COMMENTARY ES cells that is required for successful targeting with this Targeted gene Targeting frequency, %* %(%ofTrfr)† strategy. Genes encoding secreted or membrane anchored proteins‡ Methods Plexin-B2 95 (112͞118) 35 (Ϯ3.6) Targeting Constructs. The vectors used for targeted trapping of Coxadr 88 (102͞116) 13 (Ϯ0.5) genes encoding secretory proteins (pTT0TM, pTT1TM, and Kremen-1 83 (100͞120) 5.8 (Ϯ0.6) pTT2TM for the three reading frames, respectively) were Ryk 74 (83͞112) 109 (Ϯ9.3) derived from the gene trap secretory vector pGT0-2TMpfs Plexin-B1 69 (83͞120) 14 (Ϯ1.8) (11) by flanking the vector cassette with AscI sites. A version SynCAM 68 (67͞99) 59 (Ϯ5.8) of the targeted trapping vector without transmembrane do- Robo-1 61 (36͞59) 17 (Ϯ2.5) main (TM) (pTT0, pTT1, and pTT2), which was used for Nectin-3 58 (63͞108) 73 (Ϯ1.8) targeted trapping of genes encoding cytosolic͞nuclear pro- Teneurin-4 58 (84͞144) 21 (Ϯ5.4) teins, was derived from pTT0-2TM by introducing the ScaI͞ Sema4C 55 (40͞73) 6.4 (Ϯ1.9) ClaI fragment of pGT0-2 (12). CDO 55 (55͞100) 5.1 (Ϯ0.8) Targeting constructs were designed by using the University of Sema7A 52 (61͞116) 2.0 (Ϯ0.3) California, Santa Cruz, and Celera mouse genome databases. PTP mu 46 (30͞65) 2.1 (Ϯ0.5) For genes encoding secretory proteins, the insertion of the BOC 45 (65͞144) 2.0 (Ϯ0.1) pTT0-2TM secretory trap cassette was usually positioned into Neto2 17 (15͞88) 4.6 (Ϯ0.6) the middle of the second-to-last intron upstream of the exon TEM7 15 (11͞75) 2.1 (Ϯ0.3) encoding the type I TM or the GPI signal, respectively. For type 8D6 antigen 6.3 (6͞94) 46 (Ϯ7.5) II TM proteins, the insertion was positioned downstream of the PTP ␭ 0.8 (1͞120)§ 3.5 (Ϯ0.6) exon encoding the TM. For cytosolic͞nuclear genes, the inser- Protocadherin-21 0 (0͞120)§ 4.8 (Ϯ1.8) tion of the pTT0-2 cassette was placed in one of the first introns. Sema3B 0 (0͞132)§ 4.1 (Ϯ2.0) Homology arms flanking the insertion site were typically 5 and Sema5A 0 (0͞99)§ 3.9 (Ϯ0.6) 3 kb in length (for the 5Ј and 3Ј arms, respectively) and were Plexin-A4 0 (0͞87)§ 1.0 (Ϯ0.1) generated by PCR (Expand High Fidelity PCR kit, Roche Teneurin-1 Low colonies¶ 0.1 (Ϯ0.04) Applied Science, Indianapolis) from genomic DNA of Netrin-G1 Low colonies¶ 0.02 (Ϯ0.005) E14Tg2A.4 ES cells. Suitable restriction sites were added to the Genes encoding cytosolic or nuclear proteins primers to allow successive cloning of the homology arms and the Etv5 97 (104͞107) 419 (Ϯ157) targeted trap pTT0-2TM͞pTT0-2 AscI-cassette into pBluescript Grg4 66 (79͞120) 37 (Ϯ36) (Stratagene). Exons on the 5Ј homology arm were sequenced to Grg3 65 (68͞105) 17.0 (Ϯ5.7) control for nonsense mutations. A detailed protocol of the MBTL 5.0 (5͞100) 5.0 (Ϯ1.9) cloning procedures is available on request.