Epitope Tagging of Proteins at the Native Chromosomal Loci of Genes in Mice and in Cultured Vertebrate Cells

J. Mol. Biol. (2006) 361, 412–419

doi:10.1016/j.jmb.2006.06.052

Epitope Tagging of Proteins at the Native Chromosomal Loci of Genes in Mice and in Cultured Vertebrate Cells Yen-I G. Chen 1 , Shanna D. Maika 2 and Scott W. Stevens 1,2,3 ⁎ 1

Graduate Program in Microbiology, University of Texas at Austin, 1 University Station #A4800, 2500 Speedway, Austin, TX 78712, USA 2

Institute for Cellular and Molecular Biology, University of Texas at Austin, 1 University Station #A4800, 2500 Speedway, Austin, TX 78712, USA 3

Section of Molecular Genetics and Microbiology, University of Texas at Austin, 1 University Station #A4800, 2500 Speedway, Austin, TX 78712, USA

Adding epitope tags to proteins is an important method for biochemical analyses and is generally accomplished in metazoan cells using ectopically expressed, tagged trans-genes. In Saccharomyces cerevisiae, the addition of epitope tags to proteins is easily achieved at the genomic locus of a gene of interest due to the high efficiency of homologous recombination in that organism. Most metazoan cells do not exhibit this high homologous recombination efficiency, and therefore trans-genes with in-frame epitope tags are used. Although epitope tagged trans-genes have proven useful, replacing the native promoter with a heterologous promoter introduces numerous artifactual possibilities. These include overexpression, which can lead to promiscuous interactions, and the loss of native transcriptional control, which in live animals often leads to developmental defects and embryonic lethality. We describe an efficient method that overcomes the problems encountered using epitope tagged trans-genes by introducing the epitope tag into the native chromosomal gene locus in vertebrate cells, embryonic stem cells and live mice. These tagged proteins are physically associated with the expected relevant particles, and highly sensitive as shown by co-purification of homologues of the yeast pre-mRNA splicing factors Prp38p and Prp39p, not previously shown to be associated with metazoan snRNPs. These techniques will enhance the validity of conclusions made regarding epitope-tagged proteins and improve our understanding of proteomic dynamics in cultured vertebrate cells and live animals. © 2006 Elsevier Ltd. All rights reserved.

*Corresponding author

Keywords: epitope tagging; DT40; spliceosome; snRNPs; mass spectrometry

Introduction Epitope tagging technology has enhanced the repertoire of the molecular biologist by allowing the facile detection, tracking and purification of virtually any polypeptide in many model organisms. These tagging procedures have, in effect, allowed researchers to perform experiments that once required the generation of monoclonal or polyclonal antibodies, which can be expensive and timeconsuming. Additionally, polyclonal antibodies are generally not useful long-term reagents for preparative purification purposes. Abbreviations used: ES, embryonic stem; CLEP, chromosomal locus epitope; TAP, tandem affinity purification. E-mail address of the corresponding author: [email protected]

In bacteria and unicellular eukaryotes, such as Saccharomyces cerevisiae, straightforward procedures for introducing epitope tags directly into the genomic locus of a desired gene have become routine even on a genomic scale.1–4 In metazoan cells, similar epitope tags incorporated into chromosomal loci have not been reported. Ectopically expressed trans-genes containing epitope tags are routinely used, however. Although the use of trans-genes has proven useful, the technology is inherently prone to artefacts. This is particularly important when considering the use of an epitope-tagged transgene in a live animal. Trans-genes in vertebrate and mammalian cells are generally controlled at the transcriptional level by highly active, heterologous promoters such as the human cytomegalovirus (CMV) immediate-early regulatory DNA sequence termed the CMV promoter, or those based on retroviral long terminal repeats (LTRs). Loss of the

0022-2836/$ - see front matter © 2006 Elsevier Ltd. All rights reserved.

Chromosomal Gene Tagging in Vertebrates

native transcriptional control of a gene of interest can lead to the loss of cell-cycle regulation or regulation of other sorts, or overexpression which can drive the formation of inappropriate interactions between the protein of interest and polypeptides or other molecules with which that polypeptide would not normally interact. Trans-genes are intronless constructs derived from a single cDNA, thus the natural alternative splicing pattern that would otherwise result from expression of that gene from its native chromosomal locus is lost. Additionally, viral promoters can be repressed or inactivated over time, making the use of cell lines with epitopetagged trans-genes less useful as long-term biochemical tools. In several examples, ectopic expression of transgenes in mice has led to embryonic lethality or other unwanted phenotypes.5–12 Importantly, functional coverage of potentially non-functional trans-genes by endogenous chromosomal copies of that gene raises concerns about the validity of the conclusions made from trans-gene experimentation. Expression of a tagged protein using the native promoter in the natural genomic locus will result in the proper expression levels and patterns from that gene and likely will maintain the alternative splicing program for the gene of interest as long as the 3′-terminal exon is common to all spliced isoforms for the tagged gene. We describe a method for epitope tagging, which we term CLEP tagging (chromosomal locus epitope tagging), that eliminates the above-mentioned potential artefacts encountered with the use of epitope-tagged trans-genes. By using homologous recombination techniques,13 we have inserted an epitope tag prior to the STOP codon of the SART-1 gene (U5-110K)14 in mouse embryonic stem (ES) cells. These TAP-tagged SART-1 ES cells were then injected into mouse blastocysts. This allowed us to generate three live mice possessing a TAP15 tagged SART-1 gene, which we showed was expressed in all of the organs tested. In similar experimentation, we employed chicken DT40 pre-B cells to TAP- or (His)8-tag the SmD3, SF3b155 or Lsm3 genes in this model vertebrate cell line.16 With the recent and

413 future progress in genome sequencing, these experiments will be easily applicable to other metazoan organisms for which extensive genomic sequencing has been performed, and for which cells possessing efficient homologous recombination are available.

Results Targeting vector construction The general strategy for the CLEP tagging procedure is shown in Figure 1. To create the targeting vector, three DNA regions from the gene of interest were PCR amplified from genomic DNA purified from the cell type of interest and cloned in the pGEM-T-Easy (Promega) vector. The AB fragment (Figure 1) contains the region of 1–4 kb upstream of the stop codon of the gene of interest and possesses a BamHI or BglII site immediately upstream of the STOP codon. We generally attempt to make the 5′ end of the AB fragment start in an intronic region, away from predicted splice site signals. The CD fragment typically is composed of the 1 kb of genomic DNA downstream of the STOP codon of the gene of interest, although this may be made longer if the sites of cleavage and polyadenlyation are determined to be further downstream than 1 kb from the STOP codon. The EF fragment contains the region 2–4 kb further downstream of the CD fragment. All fragments were sequenced to verify maintenance of the coding region and splice site signals in AB and to verify the identity of the CD and EF fragments. The targeting vector backbone used here is pOSDUPdel (a kind gift from W. Kuziel) and the restriction enzyme sites used in these constructions are flexible. The AB and CD fragments were cloned into the first polylinker region containing the following sites, in order: NotI, PmeI, XbaI, HinDIII, KpnI, HpaI, SalI, BamHI and PacI. The EF fragment was cloned into the second polylinker region containing the following sites in order: ClaI,

Figure 1. Strategy for the CLEP tagging technology in mammalian cells and live mice. The schema for the targeting of the CLEP-tagged vectors into an idealized gene: Top panel, idealized gene structure of a vertebrate gene. Exons are represented by filled rectangles, introns by thin lines. The epitope tag is represented by a grey rectangle. Middle panel, targeting vector fragment containing the sequences required for the introduction of the epitope tag, and the regions of homology for the targeting of the gene. Bottom panel, idealized gene sequence with targeting construct inserted. Locations of oligonucleotides used in PCR and RT-PCR (PCR5 and PCR3) are noted by arrows.

414

Chromosomal Gene Tagging in Vertebrates

XhoI, PmlI, BclI, NheI. For simplicity, we try to use NotI and SalI for cloning of the AB fragment and SalI and PacI for the cloning of the CD fragment as it eliminates the BamHI site in the polylinker such that a BamHI fragment containing the epitope tag can be inserted or replaced as needed. Alternatively, the epitope tag-containing BamHI fragment can be added to the AB fragment in the pGEM-TEasy vector. The cloning strategy should take into account the order in which the fragments are inserted into the targeting vector, and ensure that the sites used for cloning are absent from the fragments yet to be inserted. Live mammals carrying CLEP tags After construction and clone validation, the targeting vector is linearized at an appropriate site upstream of the AB fragment (usually NotI) or downstream of the EF fragment (if not using ganciclovir selection). The SART1-TAP targeting vector was linearized with NotI and electroporated into mouse ES cells. There are no reports of spliced isoforms of this gene that do not contain the terminal tagged exon, making it an ideal candidate for the CLEP tagging procedure. After selection with geneticin and ganciclovir, 141 drug-resistant colonies were selected and expanded. The presence of correctly targeted cells was tested by PCR using one oligonucleotide primer designed outside the targeting region (PCR5 in Figure 1) and one oligonucleotide primer designed inside the targeting vector (PCR3 in Figure 1). Seven out of 141 ES cell clones were positive for the SART1-TAP construct (Figure 2(a)) and all seven were positive for the TAP tag on an appropriately sized band

Figure 3. Demonstration of targeting and functional insertion of epitope tags into DT40 genes. (a) RT-PCR analysis of SF3b155-His8 and Lsm3-His8 targeting construct insertion into the respective DT40 genes. Lower band represents the wild-type allele and upper band represents the targeted allele. (b) RT-PCR analysis of SmD3-TAP targeted clones. Wild-type and targeted alleles are as in (a). (c) Western blot analysis of SmD3-TAP clones. SmD3-TAP polypeptide is noted.

(∼130 kDa) by Western blotting (Figure 2(b)). Three chimeric mice were produced from injected blastocycts and tissue samples from brain, heart, kidney and spleen were analyzed for SART1-TAP by Western blotting. While wild-type mice were negative for TAP signal, the chimeric mice were positive in all tissues analyzed (Figure 2(c)). Cultured vertebrate cells carrying CLEP tags

Figure 2. Validation of correctly targeted SART1-TAP in ES cells and live mice. (a) Detection of SART1-TAP targeted mouse ES cell clones by PCR. Wild-type and targeted alleles are noted. (b) Western blot analysis of five of the seven mouse ES cell clones positive for SART1-TAP from (a). The 130 kDa SART1-TAP is noted. (c) Western blot analysis of SART1-TAP from organs of the mice generated from the mouse ES cells. Left panel, wild-type untagged mouse organs; right panel, SART1-TAP mouse organs.

We next explored using this chromosomal epitope tagging procedure in a model vertebrate system, the DT40 chicken pre-B cell line.16 The DT40 cell line has many features that make it ideal for these studies: (1) high rates of homologous recombination, (2) an 8 h doubling time, (3) a completed chicken genome sequence and (4) suspension growth. We chose three target genes in DT40: the SmD3, SF3b155 and Lsm3 genes encode polypeptides participating in premRNA splicing and/or RNA turnover. Similar to the situation described above for SART1, there are no reports of spliced isoforms lacking the terminal tagged exon. The design of the targeting vectors was performed as described above. Lsm3-HIS8 was positively identified in 19/30 geneticin-resistant colonies (Figure 3(a)), SF3b155-HIS8 was identified in 12/30 geneticin-resistant colonies (Figure 3(a))

Chromosomal Gene Tagging in Vertebrates

415 were identified in the salt-stable snRNP fractions demonstrating the sensitivity of the procedure. Previously uncharacterized metazoan snRNP polypeptides co-purify with SmD3 The yeast genes PRP38 and PRP39 encode polypeptides that are associated with the yeast U4/ U6•U520 and U1 snRNPs,21 respectively. These polypeptides have not yet been shown to be stable components of metazoan snRNPs when purified using a monoclonal antibody demonstrating the robustness of the procedure. We present sequence alignments of the vertebrate homologues of the yeast Prp38p and Prp39p in Figures 5 and 6, Table 1. SmD3-associated polypeptides Identified protein

Figure 4. SmD3-TAP is incorporated into pre-mRNA splicing snRNPs. Untagged DT40 and tagged SmD3-TAP cell nuclei were subjected to TAP purification procedures15 and affinity-purified polypeptides resolved by SDS-PAGE. TEV elution and calmodulin elution profiles are shown for each. An asterisk marks the location of the TEV protease.

and SmD3-TAP was recovered in 2/20 geneticinresistant colonies (Figure 3(b) and (c)). We note that the smaller size of the His8 tag is a likely reason they were more efficiently recovered than TAPtagged constructs (10% of positives in the TAPtagged construct versus 63% and 40% for Histagged constructs). Analysis of TAP-tagged SmD3 To demonstrate association of the tagged material with its native cellular machinery, six liters of SmD3TAP DT40 cells were grown to a density of 7.5 × 105 cells/ml. Cells were harvested and nuclei were fractionated and disrupted by sonication. SmD3-TAP associated material was purified by the two-step TAP procedure15 and associated salt-stable snRNP polypeptides were separated by SDS-PAGE (Figure 4) and analyzed by mass spectrometry peptide analysis. 17 The SmD3-associated polypeptides from selected bands are shown in Table 1. The identified polypeptides correlated to most of the known SmD3-associated spliceosomal snRNP proteins18 demonstrating that the SmD3-TAP polypeptide is being incorporated into complexes known to associate with SmD3, and includes factors previously shown to be only loosely associated with snRNPs19 (e.g. PRP43, SR140, SAP45 and SAP30). Indeed even very low-abundance U11/U12 snRNP proteins (U11/U12-65K and U11/U12-48K)

ENSEMBL name

snRNP association

U1-70K U1C U1A SF3b155 SF3b145 SF3b130 SF3b125 SF3a120 SF3a60 DDX15/Prp43 SR140 SPF45 SPF30 U2B'' U2A' SF3b14 U5-220K U5-200K U5-116K U5-110K U5-102K U5-100K U5-40K U5-15.5K U4/U6-90K U4/U6-61K U4/U6-60K U4/U6-20K U4/U6-15K U4/U6•U5-65K SmB/SmB' SmD3 SmD2 SmD1 SmE SmF SmG SART3 Lsm2

ENSG00000104852a ENSGALG00000002773 ENSGALG00000008729 ENSGALG00000008038 ENSG00000087365a ENSGALG00000002531 ENSGALG00000000581 ENSGALG00000007937 ENSGALG00000001540 ENSGALG00000014395 ENSGALG00000002612 ENSGALG00000006332 ENSGALG00000008561 ENSGALG00000008729 ENSGALG00000007170 ENSGALG00000016501 ENSGALG00000002943 ENSGALG00000015465 ENSGALG00000000988 ENSG00000175467a ENSGALG00000006001 ENSG00000174243a ENSGALG00000000615 ENSGALG00000017396 ENSGALG00000000465 ENSG00000105618a ENSGALG00000008857 ENSGALG00000004874 ENSGALG00000011931 ENSG00000168883a ENSGALG00000007250 ENSGALG00000006596 ENSG00000125743a ENSGALG00000011842 ENSGALG00000000137 ENSGALG00000011409 ENSG00000143977a ENSGALG00000004887 ENSG00000111987a

Lsm6

ENSGALG00000009985

Lsm8

ENSGALG00000009110

U11/U12-65K U11/U12-48K Prp38 Prp39

ENSGALG00000005162 ENSGALG00000013005 ENSGALG00000010627 ENSGALG00000012468

U1 U1 U1 U2 U2 U2 U2 U2 U2 U2 U2 U2 U2 U2 U2 U2 U5, U4/U6•U5 U5, U4/U6•U5 U5, U4/U6•U5 U5, U4/U6•U5 U5, U4/U6•U5 U5, U4/U6•U5 U5, U4/U6•U5 U5, U4/U6•U5 U4/U6, U4/U6•U5 U4/U6, U4/U6•U5 U4/U6, U4/U6•U5 U4/U6, U4/U6•U5 U4/U6, U4/U6•U5 U4/U6•U5 U1, U2, U4/U6, U5 U1, U2, U4/U6, U5 U1, U2, U4/U6, U5 U1, U2, U4/U6, U5 U1, U2, U4/U6, U5 U1, U2, U4/U6, U5 U1, U2, U4/U6, U5 U6, U4/U6 U6, U4/U6, U4/U6•U5 U6, U4/U6, U4/U6•U5 U6, U4/U6, U4/U6•U5 U11/U12 U11/U12 Yeast U4/U6•U5b Yeast U1b

a Not annotated for Gallus gallus. Identification was based on database calls to orthologues from other species. b Not previously shown to be snRNP-associated in metazoans.

416 respectively.

Chromosomal Gene Tagging in Vertebrates

polypeptide of interest exists in multiple complexes with mutually exclusive binding partners.

Discussion Materials and Methods We have demonstrated the feasibility and promise of incorporating epitope tags into the chromosomal loci of essential genes in cultured vertebrate cells and into live mammals. We also have shown that the procedure is robust and highly sensitive, as loosely associated factors and extremely low-abundance material such as the U11/U12 snRNP proteins, were purified and identified by mass spectrometry. These procedures will be of great use where circumstances require that the gene of interest be controlled by its native promoter, especially in live animals where expression of proteins at inappropriate levels can lead to developmental defects or unwanted interactions. CLEP-tagged animals should also prove useful for the study of virtually any process and serve as a biochemical tool to study the organization of protein complexes during different stages of development and in different tissues, especially those of disease relevant gene products. With tissue culture cells, once a tagged cell line has been established using CLEP, an unlimited supply of cells can be grown for experimentation for which precious and expensive antibodies were previously required. The use of different epitope tags on multiple genes will allow additional flexibility and experimentation such as co-immunoprecipitation or the separation of subsets of complexes in which a

Targeting vector creation The targeting vectors were constructed in the backbone of a positive-negative selection vector, pOSDUPdel, containing a neomycin phosphotransferase gene (neo) cassette flanked by loxP sites and a thymidine kinase gene conferring ganciclovir resistance. Three DNA regions, named AB, CD, and EF fragments, homologous to the gene of interest were PCR amplified (Expand High Fidelity PCR kit, Roche Applied Science) from the genomic DNA purified from the cell type of interest and cloned in the pGEM-T-Easy (Promega) vector. The AB, CD, and EF fragments for each gene were amplified using oligonucleotides: mSART1-A (5′- GCGGC CGCGAAATAGGAGCAGCTGTGAACATGG-3′), mSART1-B (5′- TCTAGAGGATCCTTT GGTGATGGTGTTCCTGCAGGG-3′), mSART1-C (5′-GTCGACAGCCGCCCTCCTCCCT GGCCCAGATG-3′), mSART1-D (5′-TTAATTAAGATCCACAGGCACAGCCGAGAACAC-3′), mSART1-E (5′ATCGATTGGCTAGTGATGGTTGTGCAGGAGTG-3′), m S A RT 1 - F ( 5 ′ - T G AT C A AT T G A C C C AT T C T G TCAATGGGTGTG-3′); GgLSM3-A (5′-GCGGCCGCGGT GAGGACAGATCCACGATGCACTGG-3′), GgLSM3BH I S ( 5 ′ - G T C G A C G G AT C C T TA G T G G T G AT G A T G AT G AT G AT G AT G G G AT C C G C C A A C C C TCAGTGGGGGAGCTACAAGCAC-3′), GgLSM3-C (5′-GTCGACAGCAACCAAGGATTGAACCTTCTTGGAAG-3′), GgLSM3-D (5′-TTAATTAAGGATCCTAGGCAAGGTATGTCAAGAAATGCCTG-3′), GgLSM3-E (5′-CTCGAGGGATCCGGTTTTAGACACGTTCTGTAC Q

Figure 5. Sequence alignment of polypeptide homologues to yeast Prp38p. Sequences from human (HsPrp38), chicken (GgPrp38) and worm (CePrp38) were aligned with the yeast Prp38p (ScPrp38) polypeptide. Identical amino acid residues are highlighted with white text on a black block. Similar amino acid residues are highlighted with black text on a gray block.

Chromosomal Gene Tagging in Vertebrates

417

Figure 6. Sequence alignment of polypeptide homologues to yeast Prp39p. Sequences from human (HsPrp39), chicken (GgPrp39) and dog (CfPrp39) were aligned with the yeast Prp39p (ScPrp39) polypeptide. Identical amino acid residues are highlighted with white text on a black block. Similar amino acid residues are highlighted with black text on a gray block.

AATCATC-3′), GgLSM3-F (5′-GCTAGCGCAACTTCTGGGCAAAAAGGGAAGGCAAATG-3′), GgSF3b155-A (5′- GCGGCCGCCTTCCAACCCAAACATTCTGTGATTCTC-3′), GgSF3b155-B-HIS (5′-GTCGACGGATCCTTAGTGGTGATGATGATGATGATGATGGGATCCTAAG AT G TA G T C A A G T T C ATA A C G A ATATA G G - 3 ′ ) GgSF3b155-C (5′-GTCGACTCTTCTTGTTCTGTTGTTTGTGTTTAATGC-3′), GgSF3b155-D (5′-TTAATTAAGGATCCAACACTACCAGAATGACGTAACCTTGTG-3′), GgSF3b155-E (5′-CTCGAGGGATCCGGGTAGTGTTTCATGTCCATAAGAAAAC-3′), GgSF3b155-F (5′-GCTAGCGTACTTTCCAAGACAGTTTAGAGTTTGGCAG-3′), Gg SmD3-A (5′-GCGGCCGCAGGCTGTATGCTTGAC-

AGGGCTTTGAG-3′), GgSmD3-B (5′-GTCGACAGATCTTCTTCGCTTCTGGAAGATGTTGCCACGACC-3′), Gg SmD3-C (5′-GTCGACGACCAGTATGCTTTTTTTTATTAGAGG-3′), Gg SmD3-D (5′-TTAATTAAGGATCCTCATTCATACATGTATAGACTGATGAC-3′), Gg SmD3-E (5′-CTCGAGGGATCCTATAAAGACCCCCTGTGTGCAGACATG-3′), Gg SmD3-F (5′-GCTAGCCCTTCTATCCACAGGTGTGATCTTCC-3′). The epitope tag, TAP or His8, is contained on a BamHI restriction fragment and cloned into the BamHI or BglII site incorporated into the AB fragment. After the targeting vectors containing the epitope tagging fragment were created, the plasmid DNA (20–

418 30 μg) was linearized with a unique site (generally NotI or NheI) and electroporated into the cells of interest. ES electroporation and generating transgenic mice To generate transgenic mice using mouse ES cells, SM1 mouse embryonic stem cells (129S6) were electroporated with the linearized targeting construct. ES clones that survived in double selection media containing geneticin and ganciclovir were isolated and expanded into 6-well plates until confluent. PCR analysis of purified genomic DNA (DNeasy tissue kit, Qiagen) was performed to screen for the epitope-tag insertion into the right chromosomal locus. Western blot analysis using peroxidase-anti-peroxidase (PAP) to detect bands of the correct size was performed as described.15 Correctly targeted ES clones were microinjected into E3.5 C57Bl/6 blastocysts and implanted into (CD1) pseudopregnant females. Resulting chimeric males were mated to C57Bl/6 females for germ line transmission of the altered allele. All mice were housed in accordance with protocols approved by the Animal Care and Use Committee of the University of Texas at Austin. DT40 cell culture and DNA transfection DT40 cells were grown in Dulbecco's modified Eagle media supplemented with 5% (v/v) chicken serum, 5% (v/v) fetal bovine serum, and antibiotics (penicillin/streptomycin, 100×, Gibco/Invitrogen). For DNA transfections into DT40 cells, 107 DT40 cells, resuspended in 300 μl of PBS, were transferred into cuvettes and electroporated with 20 μg of linearized targeting constructs (BioRad Genepulser II, 700 V, 25 μF, 0.4 cm cuvette). After electroporation, the cells were recovered in 15 ml non-selective media for 24 h, then harvested and resuspended into 40 ml of selective media containing 1.5 mg/ml of G418 for neomycin selection and 100 μl aliquots were distributed into four 96-well plates. G418-resistant cells formed colonies after 10–14 days in selective media. They were expanded into 48 wells and grown in G418 selection media until confluent. RT-PCR or Western blot analyses were performed to confirm the presence of the epitope-tag insertion into the correct chromosomal locus. Screening for epitope-tagged gene by PCR and RT-PCR analysis Genomic DNA was extracted (DNeasy tissue kit, Qiagen) from geneticin and ganciclovir-resistant ES colonies to screen mouse ES cells positive for mSART-1TAP. PCR reactions were performed using one oligonucleotide designed outside the targeting construct region (PCR5) and the other one inside the targeting vector (PCR3): mSARTPCR5 (5′-CCTC TCACTGCCTCTCCTGCTGCGGGG-3′) and mSARTPCR3 (5′-GAGGAGGCTCAGCGACTTGCCAGATGG-3′). For screening DT40 cells positive for SmD3-TAP, Lsm3His8, SF3b155His8, total RNA was isolated (RNAwiz, Ambion) from DT40 cells for RT-PCR analysis using oligonucleotides: Gg LSM3PCR5 (5′-ACGCATATGATCAGCATTTAAATATGATTC-3′), Gg LSM3PCR3 (5′-AGCCTGAAACCTTCCAAGAAGGTTCAATCC-3′), Gg SF3b155PCR5 (5′- TTGCAGTATTGTTTGCAGGGTTTGTTTCAC3′), Gg SF3b155PCR3 (5′- TTATGTGAAGAACAGCTGT-

Chromosomal Gene Tagging in Vertebrates

GCATTAAACAC-3′), Gg SmD3PCR5 (5′-CGAGGAAAAGCAGCTATTCTCAAAGCTCAG-3′), Gg SmD3PCR3 (5′GACAGAAGTTACCAACATATATGTAGAC-3′). Western blotting analysis Mouse ES cells, DT40 cells and organs from sacrificed mice were processed by homogenization in LDS sample buffer. 105 cells were lysed by addition of 200 μl of SDSPAGE loading buffer and approximately 50 mg of each organ was homogenized in 1 ml of SDS-PAGE loading buffer. Homogenized samples (20 μl) were electrophoresed through SDS-PAGE gels and transferred to nylon membranes for western blotting analysis. PAP antibody (Sigma) was used for detecting TAP-tagged proteins. Extract preparation, TAP purification and mass spectrometry analysis of SmD3 associated proteins Six liters of SmD3-TAP DT40 cells were grown in Dulbecco's modified Eagle media supplemented with 5% chicken serum and 2.5% Fetalplex (Gemini Bio-Products) to a density of 7.5 × 105 cells/ml for TAP purification. Cells were harvested by centrifugation (1000g for 5 min), washed twice with ice-cold PBS, allowed to swell in 10 ml of TM buffer (10 mM Tris-Cl (pH 7.5), 3 mM MgCl2) with 0.2 mM PMSF, 1 μg/ml leupeptin, and 1 μg/ml pepstatin for 10 min one ice, and lysed with 25 strokes of a Dounce homogenizer at 4 °C. The nuclei were pelleted and washed twice with 10 ml of TM buffer containing 0.1% NP40, resuspended in 5 ml of buffer (30 mM Tris-Cl, 300 mM KCl, 5 mM MgCl2, 0.5% Triton-X100), and sonicated at the maximum output, twice for 20 s on ice with 1 min in ice between sonications. The sonicated mixture was centrifuged at 5000 rpm for 10 min and the supernatant, was used for TAP purification. TAP-tagged protein material for SmD3-TAP DT40 cells was affinity purified by the TAP procedure.15 SDS-PAGE separated affinity purified polypeptides were excised from the gel analyzed by mass spectrometry (MS/MS) as described.17

Acknowledgements The authors acknowledge Stevens Laboratory members for thoughtful discussions and assistance; Henry Bose, Will Bargmann and Andrew Liss for DT40 system advice and plasmids; William Kuziel for plasmids, and Phil Tucker and Karen Artzt for helpful conversations. We thank Roger Moore, Helen Gu and Terry Lee for mass spectrometry. This work was supported by grants to S.W.S. from the Welch Foundation (F-1564), the National Science Foundation (MCB-0448556) and the American Cancer Society (RSG-05-137-01-MCB).

References 1. Hearn, M. T. W. & Acosta, D. (2001). Applications of novel affinity cassette methods: use of peptide fusion handles for the purification of recombinant proteins. J. Mol. Recog. 14, 323–369.

Chromosomal Gene Tagging in Vertebrates

2. Ghaemmaghami,S.,Huh,W.,Bower,K.,Howson,R.W., Belle, A., Dephoure, N. et al. (2003). Global analysis of protein expression in yeast. Nature, 425, 737–741. 3. Huh, W. K., Falvo, J. V., Gerke, L. C., Carroll, A. S., Howson, R. W., Weissman, J. S. & O'Shea, E. K. (2003). Global analysis of protein localization in budding yeast. Nature, 425, 686–691. 4. Stevens, S. W. (2000). Analysis of low-abundance RNPs from yeast by affinity chromatography and mass spectrometry microsequencing. Methods Enzymol. 318, 385–398. 5. Biniszkiewicz, D., Gribnau, J., Ramsahoye, B., Gaudet, F., Eggan, K., Humpherys, D. et al. (2002). Dnmt1 overexpression causes genomic hypermethylation, loss of imprinting and embyronic lethality. Mol. Cell. Biol. 22, 2124–2135. 6. Sum, E. Y. M., Segara, D., Duscio, B., Bath, M. L., Field, A. S., Sutherland, R. L. et al. (2005). Overexpression of LMO4 induces mammary hyperplasia, promotes cell invasion, and is a predictor of poor outcome in breast cancer. Proc. Natl Acad. Sci. USA, 102, 7659–7664. 7. Zaina, S., Pettersson, L., Thomsen, A. B., Chai, C. M., Qi, Z. Q., Thyberg, J. & Nilsson, J. (2003). Shortened life span, bradycardia, and hypotension in mice with targeted expression of an Igf2 transgene in smooth muscle cells. Endocrinology, 144, 2695–2703. 8. Mungrue, I. N., Gros, R., You, X. M., Pirani, A., Azad, A., Csont, T. et al. (2002). Cardiomyocyte overexpression of iNOS in mice results in peroxynitrite generation, heart block, and sudden death. J. Clin. Invest. 109, 735–743. 9. Conkright, J. J., Na, C. L. & Weaver, T. E. (2002). Overexpression of surfactant protein-C mature peptide causes neonatal lethality in transgenic mice. Am. J. Resp. Cell Mol. Biol. 26, 85–90. 10. Hsu, W., Shakya, R. & Costantini, F. (2001). Impaired mammary gland and lymphoid development caused by inducible expression of Axin in transgenic mice. J. Cell Biol. 155, 1055–1064. 11. Li, J. & Hoyle, G. W. (2001). Overexpression of PDGF-A in the lung epithelium of transgenic mice

419

12.

13.

14.

15.

16. 17.

18. 19.

20.

21.

produces a lethal phenotype associated with hyperplasia of mesenchymal cells. Dev. Biol. 239, 338–349. Agah, R., Prasad, K. S., Linnemann, R., Firpo, M. T., Quertermous, T. & Dichek, D. A. (2000). Cardiovascular overexpression of transforming growth factorbeta(1) causes abnormal yolk sac vasculogenesis and early embryonic death. Circ. Res. 86, 1024–1030. Doetschman, T., Gregg, R. G., Maeda, N., Hooper, M. L., Melton, D. W., Thompson, S. & Smithies, O. (1987). Targetted correction of a mutant HPRT gene in mouse embryonic stem cells. Nature, 330, 576–578. Makarova, O. V., Makarov, E. M. & Luhrmann, R. (2001). The 65 and 110 kDa SR-related proteins of the U4/U6•U5 tri-snRNP are essential for the assembly of mature spliceosomes. EMBO J. 20, 2553–2563. Puig, O., Caspary, F., Rigaut, G., Rutz, B., Bouveret, E., Bragado-Nilsson, E. et al. (2001). The tandem affinity purification (TAP) method: a general procedure of protein complex purification. Methods, 24, 218–229. Buerstedde, J. M. & Takeda, S. (1991). Increased ratio of targeted to random integration after transfection of chicken B-cell lines. Cell, 67, 179–188. Davis, M. T. & Lee, T. D. (1998). Rapid protein identification using a microscale electrospray LC/MS system on an ion trap mass spectrometer. J. Am. Soc. Mass Spectrom. 9, 194–201. Jurica, M. S. & Moore, M. J. (2003). Pre-mRNA splicing: awash in a sea of proteins. Mol. Cell. 12, 5–14. Will, C. L., Urlaub, H., Achsel, T., Gentzel, M., Wilm, M. & Luhrmann, R. (2002). Characterization of novel SF3b and 17S U2 snRNP proteins, including a human Prp5p homologue and an SF3b DEAD-box protein. EMBO J. 21, 4978–4988. Stevens, S. W. & Abelson, J. (1999). Purification of the yeast U4/U6•U5 small nuclear ribonucleoprotein particle and identification of its proteins. Proc. Natl Acad. Sci. USA, 96, 7226–7231. Gottschalk, A., Tang, J., Puig, O., Salgado, J., Neubauer, G., Colot, H. V. et al. (1998). A comprehensive biochemical and genetic analysis of the yeast U1 snRNP reveals five novel proteins. RNA, 4, 374–393.

Edited by J. Karn (Received 2 May 2006; received in revised form 16 June 2006; accepted 21 June 2006) Available online 7 July 2006