Nature Protocols 2, 2734 - 2746 (2007)
Published online: 25 October 2007 | doi:10.1038/nprot.2007.408

Subject Categories: Cell and developmental biology | Cell and tissue culture | Genetic modification | Model organisms | Nucleic acid based molecular biology

Genetic knockouts and knockins in human somatic cells

Carlo Rago1,2, Bert Vogelstein1,2 & Fred Bunz3

Gene targeting by homologous recombination with exogenous DNA constructs is the most powerful technique available for analysis of mammalian gene function. Over the past several years, the methods used to generate knockout and knockin mice have been modified for use in cultured human cells. The most significant innovation has been the adaptation of recombinant adeno-associated viruses (rAAVs) for such targeting. The stages of rAAV-mediated gene targeting include (i) the design and construction of a DNA targeting vector, (ii) the production of an infectious rAAV stock, (iii) the generation of cell clones that harbor rAAV transgenes, (iv) screening for homologous recombinants and (v) the iterative targeting of multiple alleles. The protocol described herein allows the generation of a cell line with a single altered allele in 3 months. A second allele of the same gene can be targeted in an additional 3 months.



The generation of null alleles and more subtle genetic alterations by homologous recombination is a powerful technique that allows the definitive assessment of gene function. Gene targeting technology has yielded important insights into the genetic basis of many human diseases, including cancer. The experimental approaches routinely applied to model organisms such as mice have been applied to the study of human genes1. Human somatic cell knockouts and knockins provide genetic model systems applicable to the study of virtually any cell-based phenomenon. Furthermore, they have conceptual advantages over analogous studies in model organisms when the biochemical and physiologic pathways involved vary among cell types or organisms2.

The most common application of human gene targeting approaches has been the sequential inactivation of all alleles to generate a knockout cell line. Gene targeting can also be used to introduce sequences into an endogenous gene, thereby creating a knockin allele. Knockin alleles can include alterations as small as a single base substitution and as large as a reporter gene.

Several methods have been employed to inactivate genes in cultured cells. Among the most accessible of these is the use of RNA interference (RNAi) to knock down gene expression. Compared with gene targeting by homologous recombination, RNAi-mediated gene knockdown is rapid, relatively inexpensive and thereby adaptable to high-throughput approaches3. However, the results of RNAi can vary between experiments and between laboratories. In addition, RNAi can have off-target effects that can be difficult to predict and interpret4, 5, 6. For these reasons, RNAi is not generally as definitive nor as reproducible as the analysis of a genetically altered cell line. Furthermore, RNAi can only lower, not eliminate, the expression of target genes, and cannot be used to compare genetic variants. Other methods of manipulating gene function such as the overexpression of dominant negative mutations have been used with considerable success, but clearly do not precisely recapitulate naturally occurring genetic alterations. Knockout and knockin cell lines can therefore address many types of experimental questions that are difficult, if not impossible, to address by other means.

Cancer cells and genetic instability

As is the case with any experimental system, human somatic cell knockouts/knockins have both unique attributes and clear limitations. Gene targeting in human cells, as in mice, is unambiguous and the reagents (i.e., cell lines) generated offer a binary means of comparatively evaluating the effects of a defined genetic alteration. The main experimental limitation of human gene targeting lies in the cells that are most often the focus of study: cultured human cancer cells. These widely used reagents contain numerous genetic alterations acquired and selected during tumorigenesis7, 8. The cancer genome is clearly distinct from the 'normal' genome present in most individuals9, 10. Experimental alterations created against a background of incompletely known cancer gene mutations and passenger mutations must be interpreted with caution. Furthermore, cancer cells are inherently genetically unstable and therefore acquire alterations during extended in vitro propagation11, 12. Genetic instability causes subclones of established cancer cell lines to acquire additional, unintended alterations that can theoretically confound phenotypic analysis. In practice, spontaneous alterations occur randomly and at a frequency that is in most cases acceptably low. For example, cells with mismatch repair deficiency have an overall mutation rate that is three orders of magnitude higher than mismatch-repair-proficient cells, but predominantly at repetitive sequence elements and at an overall rate that may be as low as 10−8 per base per generation13. Through careful experimental design and appropriate controls, an investigator can minimize confounding factors related to genetic instability. Direct comparison of parental cell lines with multiple, independently derived knockout/knockin and nontargeted sibling clones allows a locus of interest to be evaluated with a high degree of confidence.

Suitable cell lines

The cells to be used in gene targeting experiments must be chosen with care. Study of a pathway of interest, for example, requires the use of a cell line or cell lines in which that pathway has been characterized and shown to provide an interpretable signal in a cell-based assay. Because the generation of homozygous mutations requires the iterative targeting of multiple alleles, it is most straightforward to target genes in cells that are diploid or near-diploid. Most diploid cancer cell lines with a stable chromosome complement are mismatch repair deficient12. The cell lines that have been utilized most often for the generation of knockouts and knockins are the mismatch-repair-deficient colorectal cancer cell lines HCT116 and DLD1 (and a closely related cell line, derived from the same patient, called HCT15). Genes have been targeted in numerous other cell lines including those that are aneuploid14, those derived from other cancers15, 16 and those from normal tissues17, 18, 19, 20. It is probable that some cell lines will not be amenable to efficient gene targeting, so previously validated cell lines should be used whenever possible. The protocol described below is specifically designed for adherent cells that grow robustly in monolayers.

Human gene targeting

Several methods devised to alter genes have been employed with varying success and efficiency. The most straightforward are techniques that are typically used in mice and exploit homologous recombination between exogenous DNA constructs and endogenous genes21. In recent years, several related techniques have emerged that involve the design of endonucleases that recognize rare target sequences via customized zinc-finger motifs22. Although these alternative methods appear promising, they require specialized expertise, and as a result their use has not yet become widespread.

The method used for generating human somatic cell knockouts and knockins is similar in principle to that employed in the generation of knockout mice. In both methods, there are four basic steps: (i) the design and construction of a targeting vector that contains regions of homology to the locus of interest flanking a selectable marker gene, (ii) the efficient delivery of this targeting vector to target cells, (iii) the growth under selection of cell clones that have stably integrated the targeting vector and (iv) the identification and expansion of clonal homologous recombinants that contain the desired genetic alteration.

Human cells present unique challenges that have necessitated significant modification of standard mouse protocols. Exogenous DNA integrates into the genomic DNA of human cells at a lower rate than mouse cells21, which makes the generation of large numbers of transgenic clones relatively difficult. Furthermore, the human genome is approximately threefold larger than the mouse genome, and there is some evidence suggesting that many cultured human cells exhibit an intrinsically lower rate of homologous recombination21. Finally, human cells contain multiple alleles of most genes, as do mice. In mice, heterozygosity can be reduced to homozygosity by back crossing. Obviously, this strategy is not applicable to cultured human cells, and so multiple alleles must be targeted, one after the other, to achieve homozygosity.

rAAV-mediated gene targeting

The limiting stages in human gene targeting are vector delivery and integration. The frequency of homologous recombination determines the proportion of transgenic clones with the desired alteration, and is also a limiting factor. The rates of both vector delivery and site-specific integration have been significantly improved by the development of rAAV as a gene-targeting tool. Russell and co-workers16 demonstrated that rAAV containing synthetic DNA sequences can integrate into homologous chromosomal loci with high efficiency. It has been shown that rAAV-based vectors are 25-fold more efficient than comparable plasmid vectors23, 24, 25. Because cell-surface AAV receptors are widespread among human tissues, this virus can be used productively to infect many human cell types with high efficiency.

A second technical advance is the promoter trap vector (also called the promoterless vector). The typical targeting vector used to disrupt genes in mouse embryonic stem cells contains a selectable marker gene driven by a strong promoter. The random integration of a promoter-containing construct allows stable marker gene expression and therefore the growth of a transgenic colony. Such vectors have been inefficient when used in human cells. Sedivy and Dutriaux21 demonstrated that configuring a targeting construct so that marker gene expression is dependent on the endogenous gene promoter results in a significant enrichment for homologous recombinants within the pool of transgenic clones. This approach is known as a promoter trap. The inclusion of a splice acceptor26, 27 and an internal ribosomal entry sequence28 upstream of a selectable marker creates the synthetic exon promoter trap (SEPT), a versatile targeting vector cassette that facilitates the construction of promoter trap vectors (Fig. 1a)24. LoxP sites within the cassette that flank the SEPT element facilitate its removal by the transient expression of the cre recombinase29, 30. It is important to note, however, that the promoter trap approach requires an active, endogenous promoter, and is therefore restricted to targeting genes that are expressed under normal growth conditions31, 32.

Figure 1: Overview of the generation of knockout and knockin alleles by homologous recombination.
Figure 1 : Overview of the generation of knockout and knockin alleles by homologous recombination.

(a) Illustration of the functional elements of a gene-targeting vector. A generic construct contains two HAs that flank a central element featuring a selectable marker gene. A versatile cassette, known as the SEPT element, contains a splice acceptor (SA) followed by an internal ribosomal entry sequence (IRES), the coding sequence of neomycin transferase (neo) and a polyadenylation signal sequence (pA). This element is flanked by recognition sites (LoxP) for the cre recombinase in the plasmid pSEPT. Arranged in a head-to-tail configuration, the LoxP sites facilitate the removal of all marker elements. (b) The locations of the homology regions define the function of a targeting vector. In knockout vectors (left), the regions of homology typically flank a critical exon of Your Favorite Gene, thereby creating a deletion upon homologous recombination. To generate a knockin allele, targeting vectors (right) often contain an engineered alteration (indicated with an asterisk) in one of the HAs. In such cases, regions of homology are designed so as to insert the selectable marker into an intron, whereas the engineered alteration is incorporated into a neighboring exon. In both knockout and knockin experiments, transient expression of cre recombinase restores sensitivity of the cells to the selectable drug and facilitates the targeting of multiple alleles.

Full size image (43 KB)

The development of rAAV targeting vectors and, to a lesser extent, the inclusion of promoter trap elements in targeting vectors have significantly streamlined the generation of human somatic cell knockouts and knockins. The combined use of these strategies has optimized the generation of transgenic clones and minimized the scope of screening. The reported frequency of homologous recombinant clones in transgenic clone populations has consistently been greater than 1% and as high as 70% (refs. 16,19,24,33).

rAAV technology has been successfully used to study the role of many genes in cancer biology20, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47. These studies have provided unique insights into tumor suppression, apoptosis, tumorigenicity, metastasis, cell cycle regulation, signal transduction, DNA-damage responses, chromosomal dynamics, chemical genetics and cis-acting elements. Many of the knockout and knockin cell lines developed in these studies also provide excellent reagents for drug discovery and target validation48, 49. In the protocol below, we describe the generation of such cell lines using the pSEPT rAAV shuttle vector, but the principles described are applicable to the generation of cell lines using pNeDaKO19 or any other rAAV shuttle vectors.

Experimental design

The first stage of any gene targeting effort is the careful design of a targeting construct. A targeting construct is composed of both custom-designed and generic elements (Figs. 1a and 2). The customized elements are the regions of homologous DNA, known as homology arms (HAs), which flank the selectable marker cassette. The HAs define the targeted region into which the selectable marker cassette will be inserted (Fig. 1b). The cassette containing the selectable marker can be inserted into any region of a gene.

Figure 2: Assembly of an rAAV targeting construct.
Figure 2 : Assembly of an rAAV targeting construct.

(a) The SEPT element and LoxP sites are flanked by several unique 6-bp restriction sites to facilitate cloning. (b) The final construct is assembled by a four-part ligation that includes the two HAs (HA-1 and HA-2), the SEPT/LoxP cassette and a pAAV shuttle vector that contains AAV inverted terminal repeats (ITRs). The SEPT element is designed to ligate to each HA through base pairing of unique restriction sites (RE-1 and RE-2).

Full size image (25 KB)

For the creation of null alleles, the HAs often contain noncontiguous sequences, and the genomic DNA representing the region between the HAs will be completely deleted in a successful targeting event. In general, large deletions that eliminate multiple exons close to the 5′ coding region are most likely to be inactivating. In the case of a gene with defined domains, a deletion of a highly conserved or catalytic domain is often desirable. It is generally useful to design a deletion in such a way that skipping of the disrupted exon would result in a shift of the correct open reading frame and the generation of a stop codon. In the case of a knockin vector, deletions are usually designed to fall within an intron, so as to minimize disruption of gene structure.

HAs can be amplified from a human genomic DNA library in bacterial artificial chromosome clones, but it is often simpler to amplify HAs from genomic DNA of the cell line to be targeted (Fig. 3). A deletion of approximately 1 kb should be more than sufficient to inactivate any gene of interest. Although the maximum deletion size that can be generated by rAAV-mediated gene targeting is unknown, there is anecdotal evidence from our labs that smaller deletions are most efficiently obtained.

Figure 3: Synthesis of HAs.
Figure 3 : Synthesis of HAs.

The design of HAs will determine the region of deletion or the point of insertion and is therefore of critical importance. The first stage is the generation of a homology template. Primers 1 and 2 are designed to span the entire proposed region of homology. If genomic PCR generates a robust product, primers 1 and 2 can also be used as screening primers (Fig. 4). The homology template is cloned and sequenced. For synthesis of HAs, primers 3–6 are designed to amplify regions within the homology template. Before vector assembly (Fig. 2), embedded restriction sites are digested with restriction endonucleases NotI, RE-1 and RE-2.

Full size image (16 KB)

There are several considerations that specifically pertain to the generation of knockin alleles. Small genetic alterations can be engineered into the HAs (Fig. 1b). These alterations are typically introduced by site-directed mutagenesis50. Because the intended alteration can become unlinked from the selectable marker gene during integration, it is advisable to design the vector so that the knocked in element is as close as possible to the selectable marker cassette. Larger alterations, such as insertion of a reporter gene, can be placed directly adjacent to the selectable marker cassette, outside the paired LoxP elements.

The total size of the targeting construct is limited by the packaging capacity of rAAV. Packaging is most efficient with recombinant genomes between 4.1 and 4.9 kb51. HAs should be as large as possible to maximize targeting efficiency52. The SEPT element used as a standard selectable marker element is approximately 2 kb in size24. This leaves a total of 2.6 kb for HAs, translating to 1.3 kb per arm. We routinely use HAs of 0.9–1.3 kb19.

The generation of homologous recombinant clones with rAAV does not require the production of a purified, high-titer viral stock. When incubated with a low-titer lysate, the majority of rAAV-infected cells do not integrate viral DNA. Because the targeting construct contains a selectable marker, even rare clones with integrated transgenes can be readily expanded. Crude viral lysates have proved to be more than adequate for the generation of hundreds of transgenic clones for subsequent screening19.

Recombinant clones are identified by the analysis of genomic DNA. There are several options for isolating genomic DNA from drug-resistant clones. Commercially available manifolds that employ glass milk technology53, 54 such as those available from Promega and Qiagen are reliable. Crude lysates, such as those generated by Pierce Lyse-N-Go (Pierce, cat. no. 78882), can provide DNA of sufficient quality for most screening purposes at a significantly lower cost. Below, we describe a method using the Qiagen 96 DNA Blood Kit.

Another critical step in the isolation of targeted cell lines involves the screen for proper integration of the construct. Several PCR-based strategies have been successfully employed to identify the cell clones that harbor recombinant alleles. The simplest of these involves the amplification of a diagnostic DNA wherein one primer anneals within the selectable marker and the second primer anneals outside the homology region (Fig. 4a). Such an amplicon spans one of the two HAs and is of predetermined size. The relatively compact HAs typically incorporated into rAAV targeting vectors facilitate this screening approach. It must be emphasized that this approach does not allow a positive control reaction, rendering the screening process essentially 'blind' until a correct recombinant is identified. A useful modification of the standard PCR-integration screen incorporates an extra 20 bp at the 5′ end of the primer used to amplify one of the HAs. The extra 20 bp are identical to those in another position within the targeted locus (Fig. 4b)55, 56, 57, 58. The wild-type allele generates a PCR signal that is distinguishable from the positive (targeted) signal by its different size and thereby produces a positive control for each PCR in the screen. High-throughput Southern blot approaches also provide positive controls for each clone59, although these involve considerably more effort than PCR-based approaches.

Figure 4: Screen for homologous integration.
Figure 4 : Screen for homologous integration.

(a) The simplest and most widely used screening approach employs one primer that anneals within the SEPT element (primer A) and a second primer (primer B) that is outside the homology region (shown in red). If the primers are well chosen, only a recombinant allele will generate a signal, as assessed by agarose gel electrophoresis. Primers that span the upstream HA (primers C and D) can be used to confirm positive clones. (b) In a modified screen devised by Chan et al.57, a short sequence (yellow) derived from the locus of interest is included in a primer used to synthesize an HA. PCR amplification using a primer that anneals to this endogenous sequence (primer E) is paired with the outside primer (primer B). Both wild-type and recombinant alleles will yield a signal; however, the signal derived from the recombinant allele is smaller and therefore more robust. It is advisable to test primers E and B for generation of an expected amplicon from normal genomic DNA before vector construction. (c) PCR can also be used to identify clones that have successfully excised the SEPT element following Ad-cre infection. In this case, a primer that anneals at the site flanking the proposed deletion (primer F) is paired with the outside primer (primer B). Both wild-type and excised alleles will generate a product, but the product amplified from the excised allele will be shorter by an amount that corresponds to the size of the engineered deletion. Alleles that retain the SEPT element after Ad-cre infection will generate a product that is 2 kb larger (the size of the SEPT element) than that generated from the excised allele. This larger product is not robustly amplified and therefore a SEPT-retaining allele is typically not visualized.

Full size image (52 KB)

After identification of a correctly targeted allele, part of the targeting construct is excised using cre. Cre can be introduced into the clone either by transfection of a mammalian expression vector or via infection with an adenovirus expression vector (Vector Biolabs). Following infection, clones are isolated by limiting dilution or colony picking and tested by PCR for the expected products of excision (Fig. 4c). With adenoviruses, more than a third of the resultant clones are generally found to have properly excised the sequences between the two LoxP sites.

A successful first round of gene targeting results in the generation of cell lines that are heterozygous with respect to the desired genetic alteration and sensitive to the drug initially used for selection (geneticin in the case of pSEPT-based approaches). For many purposes, a cell line that harbors a single targeted allele will be the end point. In cases when a homozygous cell line is desired, multiple alleles, and indeed multiple loci, can be targeted by iterative rounds of rAAV infection, drug selection and PCR screening.




  • AAV-293 cells (Stratagene, cat. no. 240073)
  • Somatic cells, for example, colorectal cancer cell lines HCT116 and DLD1 (ATCC); immortalized human fibroblast cell line Tc7 (Coriell Institute for Medical Research); and the immortalized retinal pigment epithelial cell line RPE (Clontech)
  • Platinum Taq DNA polymerase (Invitrogen, cat. no. 10966-018)
  • Platinum Taq HiFi polymerase (Invitrogen, cat. no. 11304-011)
  • 1 kb ladder (Invitrogen, cat. no. 15615-016)
  • pSEPT AAV shuttle vectors (available from authors upon request)
  • Kits for DNA extraction and purification (Qiagen Blood Mini Kit, Qiagen, cat. no. 51104; Qiagen 96-well blood kit, Qiagen, cat. no. 51161 or 51162)
  • TOPO TA cloning kit (Invitrogen, cat. no. K4520-01)
  • Quickchange II site-directed mutagenesis kit (Stratagene, cat. no. 200523)
  • Rapid DNA ligase (Roche, cat. no. 11635379001)
  • AAV Helper-free kit (Stratagene, cat. no. 240071)
  • Geneticin selective antibiotic, liquid (Invtitrogen, cat. no. 10131-027)
  • Luria Bertani (LB) broth (Invitrogen, cat. no. 12780-052)
  • LB-ampicillin plates (100 μg ml−1 ampicillin)
  • LB agar (Invitrogen, cat. no. 22700-025)
  • Ampicillin (Shelton Scientific, cat. no. IB02040)
  • DH10B electrocompetent Escherichia coli cells (Invitrogen, cat. no. 18290015)
  • Trypsin-EDTA, 0.05% (wt/vol) (Invitrogen, cat. no. 25300-054)
  • McCoy's 5A with L-glutamine (Invitrogen, cat. no. 16600-082)
  • Fetal bovine serum (Hyclone, cat. no. SH30070.01)
  • Penicillin–streptomycin (Invitrogen, cat. no. 15140-122)
  • Ad-cre virus stock (Vector Biolabs, cat. no. 1045)
  • Dry ice
  • Methanol (VWR, cat. no. BDH1135-4LG)
  • Phenol–chloroform solution (VWR, cat. no. IB05174)
  • Calf intestinal alkaline phosphatase, 10,000 U ml−1 (NEB, cat. no. M0290S or M0290L)
  • User-specific oligonucleotide primers for HA template amplification (see REAGENT SETUP)
  • User-specific oligonucleotide primers for HA amplification (see REAGENT SETUP)
  • User-specific oligonucleotide primers for PCR screening (see REAGENT SETUP)
  • 1 × Hanks' buffered salt solution (HBSS) without calcium chloride, magnesium chloride and magnesium sulfate (Invitrogen, cat. no. 14170)
  • 20 × TE buffer (Invitrogen, cat. no. T11493)


  • Multichannel pipette (Rainin, cat. no. L12-200)
  • Electronic pipette (Rainin, cat. no. E3-1000)
  • Cell scraper (Corning, cat. no. 3010)

Reagent setup

  • AAV Helper-free kit The kit includes the following components, some of which can be purchased separately: pAAV-MCS, pAAV-RC, pHelper, AAV-293 packaging cell line, HT1080 control cell line.
  • Primer design All gene-specific primers are designed to optimize the efficiency of PCR by including standard parameters such as Tm, primer length, GC content, secondary structure and 3′ end stability. We find that the free, web-based program called NetPrimer (Premier Biosoft) is well suited for this purpose.
  • Primers for synthesis of HAs Two rounds of PCR are employed to create the HAs of the targeting construct. The generation of the homology template requires gene-specific PCR primers that span the genomic regions of homology. PCR primers internal to these primers are used to generate the HAs. These internal primers include embedded restriction sites at the 5′ termini that are compatible with vector assembly (see Fig. 2). For SEPT, the left HA requires a NotI site at the 5′ end and a SEPT-compatible restriction site at the 3′ end. The right HA requires a SEPT-compatible restriction site at the 5′ end and a NotI site at the 3′ end (see Fig. 2). If desired, a unique primer annealing site derived from the endogenous locus can be incorporated into one of the primers (see Fig. 4b). Each primer should include an extra 3 nt at its 5′ end to facilitate efficient endonuclease cleavage. Two different restriction sites must be used. The site used in the primer of each arm must not occur within the HA itself.
  • Primers for DNA sequencing The recommended sequencing of synthesized HAs and LoxP sites requires the design of primers positioned to allow full sequence coverage.


  1. Preparation of targeting vectorTiming: Standard cloning steps require approximately 14 dPrepare template genomic DNA from a target cell line using any method that produces highly purified DNA. We generally use the Qiagen Blood Kit according to the manufacturer's protocol. Genomic DNA extraction kits from other manufacturers may also be suitable.
  2. Design primers for amplification of the genomic region containing the desired HAs (see Fig. 3). This region can often be amplified in a single PCR product, but if it is too large, the region containing each HA can be separately amplified.
  3. Synthesis of homology template DNA: Amplify homology region(s) from genomic DNA of one of the lines to be targeted or from a BAC or other genomic clone. Many reaction conditions can be employed, but the following ones have proved successful in our hands. Set up the following reaction:
    ComponentAmount (per reaction) (μl)Final
    10 × PCR buffer2.51 ×
    50 mM MgSO40.51 mM
    10 mM dNTPs0.5200 μM
    50 μM forward primer0.51 μM
    50 μM reverse primer0.51 μM
    5 U μl−1 platinum Taq HiFi DNA polymerase0.50.1 U μl−1
    50–200 ng μl−1 genomic DNA12–4 ng μl−1
  4. Perform step-down PCR as follows:
    Cycle numberDenatureAnnealExtend
    194 °C for 2 min  
    2–594 °C for 10 s64 °C for 30 s68 °C for 2 min
    6–994 °C for 10 s61 °C for 30 s68 °C for 2 min
    10–1394 °C for 10 s58 °C for 30 s68 °C for 2 min
    14–4394 °C for 10 s56.5 °C for 30 s68 °C for 2 min
  5. Use a 15 μl aliquot of each PCR for electrophoresis through a 0.8% agarose gel in 1 × TAE buffer. Successful amplification of the genomic homology region should result in a single band of DNA of predicted size.Troubleshooting
  6. Excise and purify the homology region(s) using a Qiagen Gel Extraction Kit. Elute in 50 μl of Qiagen Elution Buffer, as per the manufacturer's instructions.
  7. Clone the desired PCR product into the pCR-TOPO vector using the TOPO TA cloning kit (Invitrogen), as per the manufacturer's instructions.
  8. Both PCR fidelity and the extent of polymorphism within the regions of homology should be assessed by sequencing the insert in several independent isolates.
  9. If the goal of the project is to knock in a small genomic alteration, incorporate base substitutions and small indels into the cloned homology region by site-directed mutagenesis using the Stratagene Quickchange kit.
  10. Synthesis of HAs: PCR-amplify each HA from the genomic template prepared in Steps 1–9 using a high-fidelity DNA polymerase (Invitrogen). Set up the following reaction:
    ComponentAmount (per reaction) (μl)Final
    10 × PCR buffer2.51 ×
    50 mM MgSO40.51 mM
    10 mM dNTPs0.5200 μM
    50 μM forward primer0.51 μM
    50 μM reverse primer0.51 μM
    5 U μl−1 platinum Taq HiFi DNA polymerase0.50.1 U μl−1
    100–500 ng μl−1 homology template DNA14–20 ng μl−1
  11. Perform step-down PCR as follows:
    Cycle numberDenatureAnnealExtend
    194 °C for 2 min  
    2–594 °C for 10 s64 °C for 30 s68 °C for 1 min
    6–994 °C for 10 s61 °C for 30 s68 °C for 1 min
    10–1394 °C for 10 s58 °C for 30 s68 °C for 1 min
    14–2994 °C for 10 s56.5 °C for 30 s68 °C for 1 min

    Critical step A low cycle number minimizes the number of artifactual mutations introduced.
  12. Use half of each PCR for electrophoresis through a 0.8% agarose gel in 1 × TAE buffer. Successful amplification of each HA should result in a single band of DNA of predicted size.
  13. Excise and purify the HAs using a Qiagen Gel Extraction Kit. Elute in 50 μl of Qiagen Elution Buffer.
  14. Digest DNA with the appropriate pair of restriction enzymes for each HA.
  15. Preparation of AAV shuttle vector and SEPT cassette: Prepare the two generic elements of the targeting vector construct, that is, the AAV shuttle vector and the SEPT cassette, as follows. First digest 2 μg of the AAV shuttle plasmid pAAV-MCS (Stratagene) with 40 U of NotI for 2 h at 37 °C in 200 μl total reaction volume.
  16. Add 2 μl of calf intestinal alkaline phosphatase (10,000 U ml−1) and incubate at 37 °C for an additional 15 min.
  17. Excise the SEPT element from plasmid pSEPT by cleaving the chosen unique restriction sites (RE-1 and RE-2; see Fig. 2). Use 2 μg of plasmid DNA and 40 U of each restriction enzyme for 2 h at 37 °C in 200 μl total reaction volume.
  18. Purify and concentrate with a Qiagen PCR Purification Kit, or extract with phenol and chloroform, ethanol-precipitate and dissolve in 20 μl water.
  19. Gel-purify the two restriction fragments with a 0.8% agarose gel and a Qiagen Gel Extraction Kit, as in Step 1. The pAAV-MCS backbone is 3.0 kb in size; the SEPT element is 2.0 kb in size. Elute each DNA in 50 μl of Qiagen Elution Buffer.
  20. Assembly of the final targeting construct: Assemble the custom rAAV targeting vector, composed of the two amplified HAs and the SEPT element inserted into the NotI site of pAAV-MCS (see Fig. 2). This is accomplished by four-part ligation using the Rapid DNA Ligation Kit (Roche). Set up the ligation reaction as follows:
    ComponentAmount (per reaction)Final
    pAAV-MCS NotI fragment (10 ng μl−1)1 μl0.5 ng μl−1
    SEPT element (10 ng μl−1)2 μl0.5 ng μl−1
    HA-1 (10 ng μl−1)2 μl0.5 ng μl−1
    HA-2 (10 ng μl−1)2 μl0.5 ng μl−1
    5 × DNA dilution buffer2 μl0.5 ×
    2 × T4 DNA ligase buffer10 μl1 ×
    T4 DNA ligase (1 U μl−1)1 μl0.05 U μl−1
  21. Incubate the ligation reaction at room temperature (25 °C) for 1 h.
  22. Bring the volume to 100 μl in 1 × TE. Extract with 100 μl of phenol–chloroform, precipitate with ethanol and dissolve in 40 μl of water.
  23. Electroporate 20 μl of DNA into 10 μl of DH10B electrocompetent E. coli cells.
  24. Add 700 μl of LB broth to the electroporation cuvette, transfer the contents to a 1.5 ml tube and incubate for 15 min at 37 °C.
  25. Spread the bacteria onto LB agar plates containing ampicillin and incubate at 37 °C overnight.
  26. Expand individual colonies in 5 ml of LB broth containing ampicillin (100 μg ml−1) and incubate for 18 h at 37 °C, rotating at 300 r.p.m.
  27. Prepare plasmid DNA using the Qiagen miniprep kit. Save a small aliquot of each bacterial culture for Step 30.
  28. Screen for desired recombinants by restriction digestion.
  29. Check each recombinant clone that appears to be correct upon restriction digestion by sequencing the HAs and LoxP sites using sequencing primers (see REGENT SETUP).
  30. Inoculate an LB broth culture with the saved material from Step 27 and prepare a plasmid DNA stock from the overnight culture using the Qiagen Plasmid Maxiprep kit.
  31. Production of infectious rAAV stockTiming: 3 dThree plasmids are required for generation of infectious rAAV: the customized targeting vector assembled in the preceding steps (pAAV-YFG) and the plasmids that contain the trans elements required for packaging (pAAV-RC and pHelper). Combine 2.5 μg of each plasmid at the bottom of a 50 ml conical tube. Add Opti-MEM reduced-serum media (Invitrogen) to a total volume of 750 μl. In a separate tube, dilute 54 μl of lipofectamine in Opti-MEM to a total volume of 750 μl. Add the DNA solution dropwise to the lipofectamine solution and incubate at room temperature for 15 min.
    Critical step Plasmids are not linearized before transfection into packaging cells.
  32. Grow AAV-293 cells in a 75 cm2 flask in DMEM supplemented with 10% fetal bovine serum (Hyclone) and penicillin–streptomycin (Invitrogen).
  33. Wash a 70–80% confluent 75 cm2 flask of AAV-293 cells with HBSS and add 7.5 ml of Opti-MEM.
  34. Add 1.5 ml of the DNA-lipofectamine mixture dropwise to the flask of cells and incubate at 37 °C for 3–4 h.
  35. Replace Opti-MEM with AAV-293 growth media (see Step 32) and allow the cells to grow for 48 h.
  36. Aspirate the media from the flask and use a rubber policeman to scrape the AAV-293 cells into 1 ml of PBS.
  37. Transfer the cell suspension to a 2 ml microfuge tube (use multiple tubes if the volume exceeds 1.5 ml) and use three cycles of freeze–thaw to harvest the virus from the cells. Each cycle consists of a 10 min freeze in a dry ice–methanol bath and a 10 min thaw at 37 °C, vortexing after each thaw.
  38. Clarify the lysate by centrifugation at 12,000–15,000g in a microfuge for 10 min at 4 °C.
  39. Aliquot the supernatant containing rAAV and store at −80 °C.
    Critical step The presence of PCR inhibitors in the crude lysate precludes the determination of viral titer by standard PCR-based methods. A functional estimate of rAAV titer is obtained in Steps 41–45.Pause Point The supernatant can be stored at −80 °C for 1 year.
  40. Infection and plating of target cellsTiming: 14–21 dWash a 60–80% confluent 75 cm2 flask of desired target cells with HBSS.
  41. Add 350 μl of the cleared virus-containing lysate directly to the cell monolayer, then add 4 ml of growth media and incubate for 2–3 h at 37 °C.
  42. Add 8 ml of growth media and allow the cells to grow for 48 h.
  43. Harvest the cells by trypsinization and distribute to ten or twenty 96-well plates in media containing geneticin (Invitrogen). The concentration of geneticin at which selection occurs is generally 200–1,000 μg ml−1 but is cell-line specific and must be determined empirically before this stage.
  44. Incubate plates in tissue culture incubator until large colonies appear. For the cells described in the Reagents section, this requires 14–17 d.
  45. The total number of colonies obtained will vary between cell lines and will be linearly dependent on the titer of the viral stock. If the colony density exceeds 15 drug-resistant, transgenic clones per plate, then there is a high probability that individual wells might contain two or more clones. If this is the case, repeat this step, seeding fewer cells to more plates.Troubleshooting
  46. Selection and consolidation of transgenic cell clonesTiming: 2 dIdentify wells that have single colonies and mark the lid above each positive well. Colonies should be transferred when the majority of colonies reach a size that occupies 30–60% of the well bottom.
    Critical step It is important to 'score' the clones early, when they occupy 10–25% of the surface area of each well. This ensures that each marked colony is derived from a single clone.
  47. Decant the media by inverting the plate, shaking into a sink and drying the residual liquid by depressing the inverted plate onto a clean paper towel.
  48. In the biosafety cabinet, place the lid on the bottom of the 96-well plate such that positive wells are aligned with their mark and can be clearly identified by viewing from the top.
  49. Add 50 μl of trypsin to each well that contains a single clone. This is most conveniently accomplished by using an electronic repeat pipettor with a P1000 tip. Incubate at 37 °C for approximately 10 min. Confirm colony detachment by examining several clones under an inverted microscope at low power. Some somatic cell lines may require more than 10 min for detachment.
  50. Add 180 μl of growth media and resuspend each colony by scraping the bottom of the well and pipetting up and down. Transfer 180 μl of the cell suspension to a fresh 96-well plate. Place the filled plate in the incubator and allow the majority of clones to grow to 80–100% confluence before preparing genomic DNA.
  51. Genomic DNA isolationTiming: 3 hDecant the media from the 96-well plate containing the clones as in Step 47.
  52. Add 50 μl of trypsin to each well using a multichannel pipette and incubate for 10 min at 37 °C.
  53. Add 50 μl of growth media using a 12-channel pipette and resuspend the cells by scraping the bottom of the wells and mixing several times.
  54. Transfer half (~50 μl) of the trypsinized suspension to a fresh 96-well plate containing 150 μl of growth media in each well. Incubate at 37 °C.
  55. Transfer the remaining volume (~50 μl) of the suspension to the bottom of a Qiagen round-well block containing 20 μl of Qiagen Protease stock solution and 130 μl of 1 × PBS. Follow the manufacturer's protocol for the remaining DNA purification steps and elute in 100 μl of elution buffer supplied by manufacturer.
  56. PCR screen for homologous recombinant clonesTiming: 5 hThe method used to detect homologous recombinants should have been decided at the stage of vector design (Fig. 4a,b). Perform a PCR to efficiently generate an amplicon that is 0.9–2 kb in size. We recommend the following PCR screening reactions:
    ComponentAmount (per reaction)Final
    dH2O6.08 μl 
    10 × PCR buffer1 μl1 ×
    DMSO0.6 μl6%
    50 mM MgCl20.4 μl2 mM
    10 mM dNTPs0.2 μl200 μM
    50 μM forward primer0.06 μl0.3 μM
    50 μM reverse primer0.06 μl0.3 μM
    5 U μl−1 platinum Taq DNA polymerase0.1 μl0.05 U μl−1
    50–200 ng μl−1 genomic template DNA1.5 μl7.5–30 ng μl−1
  57. Add 8.5 μl of the PCR mix to each well, then add 1.5 μl of each template DNA, then overlay with mineral oil (if needed).
  58. Perform PCR as follows:
    Cycle numberDenatureAnnealExtend
    194 °C for 2 min  
    2–594 °C for 10 s64 °C for 30 s68 °C for 2 min
    6–994 °C for 10 s61 °C for 30 s68 °C for 2 min
    10–1394 °C for 10 s58 °C for 30 s68 °C for 2 min
    14–4394 °C for 10 s56.5 °C for 30 s68 °C for 2 min
  59. Analyze 5 μl of each reaction on a 0.8% agarose gel. A robust amplified band of the expected size identifies positive clones.Troubleshooting
  60. Excision of neo cassetteTiming: ~ 16–18 dExpand positive clones from Step 59.
    Critical step Clones at this stage are valuable. Aliquots should be stored in liquid nitrogen.
  61. Seed approximately 5 × 105 cells to a 25 cm2 cell culture flask (for each clone to be treated). Incubate for 24 h, or until cells are firmly attached.
  62. Add 1 μl of Ad-cre virus stock (107 plaque forming units) to a tube containing 4 ml of cell culture medium. Mix by inversion.
  63. Aspirate the medium from the flask. Replace with Ad-cre solution. Incubate for 24 h. The final multiplicity of infection is approximately 102.
  64. Rinse cells with 1 × HBSS and detach with 1 ml of trypsin-EDTA. Plate cells at limiting dilution in 96-well plates. To do this, transfer 10 μl of the cell suspension in trypsin-EDTA to 1 ml of cell culture medium. Add 1 μl of the diluted cell suspension to 25 ml of medium and seed a 96-well plate (250 μl per well). Repeat plating 5, 25 and 100 μl of the diluted cell suspension into a series of 96-well plates.
  65. Examine plates 7 d after plating. Some of the wells should exhibit small clusters of cells at this stage.
  66. Individual colonies can be expanded when they occupy more than 50% of the well bottom (~ 14 d).
  67. Clones that have undergone excision of the SEPT element will be sensitive to the same concentration of geneticin that was initially used for selection (Step 43). Geneticin-sensitive clones can be identified either by examining geneticin sensitivity directly (phenotypic, option A) or by examining the locus by PCR (genetic, option B).
    1. Phenotypic screen for SEPT excision
      1. Choose 20–40 clones to test for excision. Seed each clone to be tested into a single well in each of two separate 24-well plates. Add geneticin at the same concentration used for selection to the wells in one of the plates (see Step 43).
      2. After approximately 10 d of growth, examine cells microscopically. Clones that are geneticin-sensitive should exhibit numerous dead cells. Geneticin-resistant clones should appear indistinguishable from the replica wells that do not contain geneticin.
      3. Expand geneticin-sensitive clones. After cells are progressively expanded to larger culture dishes, seed 105 cells to a 25 cm2 flask in medium that contains geneticin. Observe the flask after 10 d of growth. The appearance of viable cells indicates the presence of a geneticin-resistant subpopulation, that is, inefficient excision.
    2. PCR screen for SEPT excision
      1. Prepare genomic DNA from 20 to 40 single clones obtained in Step 60. Use the method for purifying DNA described in Step 1.
      2. Design a screening primer that flanks the introduced deletion (Fig. 4c). Alleles that result from cre-mediated excision of SEPT will generate a shorter PCR product than wild-type alleles or SEPT-integrated alleles.
      3. Perform PCR using the flanking primer and the outside primer (Fig. 4c), as described in Steps 51–59.
      4. Expand clones that contain the desired deletion. Confirm clonal homogeneity using a phenotypic screen, as described in Step A(iii). The phenotypic screen is more sensitive than the PCR screen, because one cell with an unexcised SEPT element among 100,000 cells can be identified with the phenotypic screen.Troubleshooting
  68. Targeting multiple allelesTiming: 2 months per alleleThe successful generation of a SEPT-excised clone represents the starting point for the iterative targeting of multiple alleles. If additional alleles are to be targeted, repeat Steps 40–59 using SEPT-excised clone as the target cell line.
    Critical step Because the removal of the SEPT element eliminates geneticin resistance, the original rAAV stock can be used to target the remaining wild-type allele (in diploid cells) or alleles (in hyperdiploid cells). The original PCR screen can also be employed.Troubleshooting


Design of targeting vector and PCR screen for homologous integration (REAGENT SETUP): ~ 7 d
Steps 1–30, preparation of targeting components and assembly of targeting vector: ~ 14 d
Steps 31–39, production of infectious rAAV stock: 3 d
Steps 40–45, generation of transgenic cell clones: 14–21 d
Steps 46–50, consolidation of transgenic clones for screening: 2 d
Steps 51–59, screen of transgenic clones for homologous recombinants: 1 d
Step 60, expansion of PCR-positive clones: 14 d
Steps 60–67, excision of SEPT element: 14 d
Step 68, targeting of additional alleles: 2 months per allele



Troubleshooting advice can be found in Table 1.


Anticipated results

Proper targeting should be confirmed using one of the following methods. The method chosen will depend on the desired genetic alteration:

  • Confirmatory PCR. The homologous integration of a targeting construct can be assessed by PCR amplification across either HA. A single PCR screen is used to identify positives. A second PCR assay can confirm targeting by assessing the position of the other HA (Fig. 4a).
  • Western blot. In knockout experiments, nullizygosity will result in the complete loss of protein expression. Western blotting is therefore a useful confirmatory tool if and only if a suitable antibody whose specificity is beyond repute is available. We have found that many antibodies cross-react with proteins of molecular weight identical to that of their purported antigens, even in cells that have been targeted to nullizygosity. Conversely, once a knockout line has been made and confirmed, it is an excellent reagent to confirm the specificity of antibodies.
  • RT-PCR. Knockout cell lines should exhibit loss of transcripts containing the region of deletion. This can be assessed by RT-PCR, using primers that span the targeted region. Note that alternative transcription start sites as well as antisense transcription is common, so that the mere presence of a transcript from a gene does not necessarily indicate failure of homologous recombination or disruption of the targeted exon.
  • Genomic Southern blot. This is one of the most conclusive means of confirming the presence of a targeted genomic alteration. Only alterations that result in changes in the lengths of genomic restriction fragments can be detected, so the restriction endonucleases used must be chosen carefully.
  • DNA sequencing. In knockin experiments designed to introduce a small genetic alteration, DNA sequencing of genomic DNA and cDNA must be employed to confirm the presence of the desired alteration. DNA sequencing of the PCR product obtained during screening (Steps 56–59) will be diagnostic if the screening primers flank the desired alteration.

Successful gene targeting results in the creation of a single, precisely defined genetic alteration in a cell line. The targeting of an allele results in the alteration of that allele in the target cell and all of its progeny. Altered alleles are as stable as endogenous alleles, even in the absence of ongoing selection. Cell populations that contain altered alleles tend to be homogenous. Knockout cells will exhibit complete and permanent loss of specific RNA and protein expression. The molecular and physiologic phenotypes of both knockout and knockin clones are dependent on the cell line, the gene and the alteration introduced and is often unambiguous. There is clonal variation, however; so the phenotypes of greatest certainty are those that are observed in all independent knockout/knockin clones tested of a given line and, optimally, in all clones of at least two different cell lines. It is also important to understand that the disruption of many genes can result in reductions in cell growth. This is true for many oncogenes and housekeeping genes, but generally not for tumor suppressor genes. Even if the reduction in growth is of small magnitude, the phenotype of the cells may change over time owing to selective overgrowth of cells with compensatory genetic or epigenetic changes. Thus, the phenotype may change with time even though the genotype is completely stable. Thus, it might be important to freeze multiple aliquots of clones after generation of the clones and to use low-passage number clones whenever possible. We generally thaw new aliquots of cells after using cells for a month or two.

The observed targeting frequency in known cell lines using rAAV methods ranges from 0.2% to 70% per round of targeting16, 19, 24, 33. The reasons for the broad range are not obvious, but it is not unreasonable to anticipate a frequency of approximately 5–10% for a previously untested locus. Ad-cre-mediated excision of the SEPT element is generally ~40% efficient.

When sequentially targeting multiple alleles, it is important to realize that homologous integration of the targeting vector can occur in the remaining wild-type alleles as well as in previously targeted alleles. Integration into a previously targeted and cre-excised allele is referred to as retargeting. Assuming that a targeting vector integrates either of two alleles at roughly equal frequency, which is most often the case, one half of the PCR-positive clones obtained during the second screen will harbor the desired alteration of the second allele. Thus, retargeting is not in and of itself a problem. However, the screens employed during the second round of targeting may need to be somewhat larger in scale. Additional methods must be employed to distinguish retargeted clones from homozygous clones.

Genes that are essential for viability are by definition impossible to target to nullizygosity. All PCR-positive clones derived at the second stage of targeting will represent retargeting. In such cases, the use of knockin methods to generate more subtle alterations can be highly informative.



  1. Bunz, F. Human cell knockouts. Curr. Opin. Oncol. 14, 73–78 (2002). | Article | PubMed | ISI |
  2. Cassman, M. Computational biology. Counting on the neuron. Science 300, 756–757 (2003). | Article | PubMed | ISI | ChemPort |
  3. Hannon, G.J. RNA interference. Nature 418, 244–251 (2002). | Article | PubMed | ISI | ChemPort |
  4. Sledz, C.A., Holko, M., de Veer, M.J., Silverman, R.H. & Williams, B.R. Activation of the interferon system by short-interfering RNAs. Nat. Cell Biol. 5, 834–839 (2003). | Article | PubMed | ISI | ChemPort |
  5. Jackson, A.L. et al. Expression profiling reveals off-target gene regulation by RNAi. Nat. Biotechnol. 21, 635–637 (2003). | Article | PubMed | ISI | ChemPort |
  6. Bridge, A.J., Pebernard, S., Ducraux, A., Nicoulaz, A.L. & Iggo, R. Induction of an interferon response by RNAi vectors in mammalian cells. Nat. Genet. 34, 263–264 (2003). | Article | PubMed | ISI | ChemPort |
  7. Vogelstein, B. et al. Genetic alterations during colorectal-tumor development. N. Engl. J. Med. 319, 525–532 (1988). | PubMed | ISI | ChemPort |
  8. Vogelstein, B. & Kinzler, K.W. The multistep nature of cancer. Trends Genet. 9, 138–141 (1993). | Article | PubMed | ISI | ChemPort |
  9. Wang, T.L. et al. Prevalence of somatic alterations in the colorectal cancer cell genome. Proc. Natl. Acad. Sci. USA 99, 3076–3080 (2002). | Article | PubMed | ChemPort |
  10. Sjoblom, T. et al. The consensus coding sequences of human breast and colorectal cancers. Science 314, 268–274 (2006). | Article | PubMed | ISI | ChemPort |
  11. Lengauer, C., Kinzler, K.W. & Vogelstein, B. Genetic instability in colorectal cancers. Nature 386, 623–627 (1997). | Article | PubMed | ISI | ChemPort |
  12. Lengauer, C., Kinzler, K.W. & Vogelstein, B. Genetic instabilities in human cancers. Nature 396, 643–649 (1998). | Article | PubMed | ISI | ChemPort |
  13. Parsons, R. et al. Hypermutability and mismatch repair deficiency in RER+ tumor cells. Cell 75, 1227–1236 (1993). | Article | PubMed | ISI | ChemPort |
  14. Montrose-Rafizadeh, C. et al. Gene targeting of a CFTR allele in HT29 human epithelial cells. J. Cell Physiol. 170, 299–308 (1997). | Article | PubMed | ChemPort |
  15. Itzhaki, J.E., Gilbert, C.S. & Porter, A.C. Construction by gene targeting in human cells of a 'conditional' CDC2 mutant that rereplicates its DNA. Nat. Genet. 15, 258–265 (1997). | Article | PubMed | ISI | ChemPort |
  16. Hirata, R., Chamberlain, J., Dong, R. & Russell, D.W. Targeted transgene insertion into human chromosomes by adeno-associated virus vectors. Nat. Biotechnol. 20, 735–738 (2002). | Article | PubMed | ISI | ChemPort |
  17. Karakas, B. et al. Interleukin-1 alpha mediates the growth proliferative effects of transforming growth factor-beta in p21 null MCF-10A human mammary epithelial cells. Oncogene 25, 5561–5569 (2006). | Article | PubMed | ISI | ChemPort |
  18. Chamberlain, J.R. et al. Gene targeting in stem cells from individuals with osteogenesis imperfecta. Science 303, 1198–1201 (2004). | Article | PubMed | ISI | ChemPort |
  19. Kohli, M., Rago, C., Lengauer, C., Kinzler, K.W. & Vogelstein, B. Facile methods for generating human somatic cell gene knockouts using recombinant adeno-associated viruses. Nucleic Acids Res. 32, e3 (2004). | Article | PubMed |
  20. Cummins, J.M. et al. Tumour suppression: disruption of HAUSP gene stabilizes p53. Nature 428 1 p following 486 (2004). | Article | PubMed |
  21. Sedivy, J.M. & Dutriaux, A. Gene targeting and somatic cell genetics—a rebirth or a coming of age? Trends Genet 15, 88–90 (1999). | Article | PubMed | ISI | ChemPort |
  22. Porteus, M.H. & Baltimore, D. Chimeric nucleases stimulate gene targeting in human cells. Science 300, 763 (2003). | Article | PubMed | ISI |
  23. Yanez, R.J. & Porter, A.C. Therapeutic gene targeting. Gene Ther. 5, 149–159 (1998). | Article | PubMed | ISI | ChemPort |
  24. Topaloglu, O., Hurley, P.J., Yildirim, O., Civin, C.I. & Bunz, F. Improved methods for the generation of human gene knockout and knockin cell lines. Nucleic Acids Res. 33, e158 (2005). | Article | PubMed |
  25. Russell, D.W. & Hirata, R.K. Human gene targeting by viral vectors. Nat. Genet. 18, 325–330 (1998). | Article | PubMed | ISI | ChemPort |
  26. Choi, T., Huang, M., Gorman, C. & Jaenisch, R. A generic intron increases gene expression in transgenic mice. Mol. Cell. Biol. 11, 3070–3074 (1991). | PubMed | ISI | ChemPort |
  27. Huang, M.T. & Gorman, C.M. Intervening sequences increase efficiency of RNA 3' processing and accumulation of cytoplasmic RNA. Nucleic Acids Res. 18, 937–947 (1990). | Article | PubMed | ISI | ChemPort |
  28. Hellen, C.U. & Sarnow, P. Internal ribosome entry sites in eukaryotic mRNA molecules. Genes Dev. 15, 1593–1612 (2001). | Article | PubMed | ISI | ChemPort |
  29. Park, B.H., Vogelstein, B. & Kinzler, K.W. Genetic disruption of PPARdelta decreases the tumorigenicity of human colon cancer cells. Proc. Natl. Acad. Sci. USA 98, 2598–2603 (2001). | Article | PubMed | ChemPort |
  30. Jallepalli, P.V. et al. Securin is required for chromosomal stability in human cells. Cell 105, 445–457 (2001). | Article | PubMed | ISI | ChemPort |
  31. Hanson, K.D. & Sedivy, J.M. Analysis of biological selections for high-efficiency gene targeting. Mol. Cell. Biol. 15, 45–51 (1995). | PubMed | ISI | ChemPort |
  32. Friedel, R.H. et al. Gene targeting using a promoterless gene trap vector ('targeted trapping') is an efficient method to mutate a large fraction of genes. Proc. Natl. Acad. Sci. USA 102, 13188–13193 (2005). | Article | PubMed | ChemPort |
  33. Porteus, M.H., Cathomen, T., Weitzman, M.D. & Baltimore, D. Efficient gene targeting mediated by adeno-associated virus and DNA double-strand breaks. Mol. Cell. Biol. 23, 3558–3565 (2003). | Article | PubMed | ISI | ChemPort |
  34. Wilsker, D. & Bunz, F. Loss of ataxia telangiectasia mutated- and Rad3-related function potentiates the effects of chemotherapeutic drugs on cancer cell survival. Mol. Cancer Ther. 6, 1406–1413 (2007). | Article | PubMed | ISI | ChemPort |
  35. Wang, P., Yu, J. & Zhang, L. The nuclear function of p53 is required for PUMA-mediated apoptosis induced by DNA damage. Proc. Natl. Acad. Sci. USA 104, 4054–4059 (2007). | Article | PubMed | ChemPort |
  36. Larochelle, S. et al. Requirements for Cdk7 in the assembly of Cdk1/cyclin B and activation of Cdk2 revealed by chemical genetics in human cells. Mol. Cell 25, 839–850 (2007). | Article | PubMed | ISI | ChemPort |
  37. Kim, J.S., Lee, C., Bonifant, C.L., Ressom, H. & Waldman, T. Activation of p53-dependent growth suppression in human cells by mutations in PTEN or PIK3CA. Mol. Cell. Biol. 27, 662–677 (2007). | Article | PubMed | ISI | ChemPort |
  38. Hurley, P.J., Wilsker, D. & Bunz, F. Human cancer cells require ATR for cell cycle progression following exposure to ionizing radiation. Oncogene 26, 2535–2542 (2007). | Article | PubMed | ISI | ChemPort |
  39. Burkard, M.E. et al. Chemical genetics reveals the requirement for Polo-like kinase 1 activity in positioning RhoA and triggering cytokinesis in human cells. Proc. Natl. Acad. Sci. USA 104, 4383–4388 (2007). | Article | PubMed | ChemPort |
  40. Gallmeier, E. et al. Targeted disruption of FANCC and FANCG in human cancer provides a preclinical model for specific therapeutic options. Gastroenterology 130, 2145–2154 (2006). | Article | PubMed | ISI | ChemPort |
  41. Dang, L.H. et al. CDX2 has tumorigenic potential in the human colon cancer cell lines LOVO and SW48. Oncogene 25, 2264–2272 (2006). | Article | PubMed | ISI | ChemPort |
  42. Dang, D.T. et al. Hypoxia-inducible factor-1alpha promotes nonhypoxia-mediated proliferation in colon cancer cells and xenografts. Cancer Res. 66, 1684–1936 (2006). | Article | PubMed | ISI | ChemPort |
  43. Cunningham, S.C. et al. Targeted deletion of MKK4 in cancer cells: a detrimental phenotype manifests as decreased experimental metastasis and suggests a counterweight to the evolution of tumor-suppressor loss. Cancer Res. 66, 5560–5564 (2006). | Article | PubMed | ISI | ChemPort |
  44. Samuels, Y. et al. Mutant PIK3CA promotes cell growth and invasion of human cancer cells. Cancer Cell 7, 561–573 (2005). | Article | PubMed | ISI | ChemPort |
  45. Papi, M., Berdougo, E., Randall, C.L., Ganguly, S. & Jallepalli, P.V. Multiple roles for separase auto-cleavage during the G2/M transition. Nat. Cell Biol. 7, 1029–1035 (2005). | Article | PubMed | ISI |
  46. Dang, D.T. et al. Glutathione I-transferase pi1 promotes tumorigenicity in HCT116 human colon cancer cells. Cancer Res. 65, 9485–9494 (2005). | Article | PubMed | ISI | ChemPort |
  47. Cummins, J.M. et al. X-linked inhibitor of apoptosis protein (XIAP) is a nonredundant modulator of tumor necrosis factor-related apoptosis-inducing ligand (TRAIL)-mediated apoptosis in human cancer cells. Cancer Res. 64, 3006–3008 (2004). | Article | PubMed | ISI | ChemPort |
  48. Torrance, C.J., Agrawal, V., Vogelstein, B. & Kinzler, K.W. Use of isogenic human cancer cells for high-throughput screening and drug discovery. Nat. Biotechnol. 19, 940–945 (2001). | Article | PubMed | ISI | ChemPort |
  49. Arena, S., Pisacane, A., Mazzone, M., Comoglio, P.M. & Bardelli, A. Genetic targeting of the kinase activity of the Met receptor in cancer cells. Proc. Natl. Acad. Sci. USA 104, 11412–11417 (2007). | Article | PubMed | ChemPort |
  50. Ho, S.N., Hunt, H.D., Horton, R.M., Pullen, J.K. & Pease, L.R. Site-directed mutagenesis by overlap extension using the polymerase chain reaction. Gene 77, 51–59 (1989). | Article | PubMed | ISI | ChemPort |
  51. Dong, J.Y., Fan, P.D. & Frizzell, R.A. Quantitative analysis of the packaging capacity of recombinant adeno-associated virus. Hum. Gene Ther. 7, 2101–2112 (1996). | PubMed | ISI | ChemPort |
  52. Hirata, R.K. & Russell, D.W. Design and packaging of adeno-associated virus gene targeting vectors. J. Virol. 74, 4612–4620 (2000). | Article | PubMed | ISI | ChemPort |
  53. Vogelstein, B. & Gillespie, D. Preparative and analytical purification of DNA from agarose. Proc. Natl. Acad. Sci. USA 76, 615–619 (1979). | Article | PubMed | ChemPort |
  54. Vogelstein, B. Rapid purification of DNA from agarose gels by centrifugation through a disposable plastic column. Anal. Biochem. 160, 115–118 (1987). | Article | PubMed | ISI | ChemPort |
  55. Chan, T.A., Wang, Z., Dang, L.H., Vogelstein, B. & Kinzler, K.W. Targeted inactivation of CTNNB1 reveals unexpected effects of beta-catenin mutation. Proc. Natl. Acad. Sci. USA 99, 8265–8270 (2002). | Article | PubMed | ChemPort |
  56. Chan, T.A., Hwang, P.M., Hermeking, H., Kinzler, K.W. & Vogelstein, B. Cooperative effects of genes controlling the G(2)/M checkpoint. Genes Dev. 14, 1584–1588 (2000). | PubMed | ISI | ChemPort |
  57. Chan, T.A., Hermeking, H., Lengauer, C., Kinzler, K.W. & Vogelstein, B. 14-3-3Sigma is required to prevent mitotic catastrophe after DNA damage. Nature 401, 616–620 (1999). | Article | PubMed | ISI | ChemPort |
  58. Hwang, P.M. et al. Ferredoxin reductase affects p53-dependent, 5-fluorouracil-induced apoptosis in colorectal cancer cells. Nat. Med. 7, 1111–1117 (2001). | Article | PubMed | ISI | ChemPort |
  59. Ramirez-Solis, R., Davis, A.C. & Bradley, A. Gene targeting in embryonic stem cells. Methods Enzymol. 225, 855–878 (1993). | PubMed | ISI | ChemPort |
  1. Sidney Kimmel Comprehensive Cancer Center, Cellular and Molecular Medicine Program, Ludwig Center for Cancer Genetics and Therapeutics, Johns Hopkins University School of Medicine, 1550 Orleans Street CRB2-453, Baltimore, Maryland 21231, USA.
  2. Sidney Kimmel Comprehensive Cancer Center, Cellular and Molecular Medicine Program, Howard Hughes Medical Institute, Johns Hopkins University School of Medicine, 1550 Orleans Street CRB2-453, Baltimore, Maryland 21231, USA.
  3. Sidney Kimmel Comprehensive Cancer Center, Cellular and Molecular Medicine Program, Department of Radiation Oncology and Molecular Radiation Sciences, Johns Hopkins University School of Medicine, 1550 Orleans Street CRB2-453, Baltimore, Maryland 21231, USA.

Correspondence to: Fred Bunz3 e-mail:


These links to content published by NPG are automatically generated.


Mouse models, immunology, multiple sclerosis and myelination

Nature Genetics News and Views (01 Jan 1996)