Generation of Cre Recombinase-Expressing Transgenic Mice Using Bacterial Artificial Chromosomes

Abstract

Generation of genetically modified mice is one of the primary methods for understanding gene function. In particular, approaches that allow for restricting the effects of a mutation to defined cell-types are fundamental for understanding the roles of genes in specific cells or tissues. The Cre/loxP recombination system is the most robust approach to produce cell-type-specific gene inactivation. When the Cre recombinase is expressed from a transgene containing a tissue-type-specific promoter it will delete genomic segments flanked by loxP sequences in this tissue only. In this regard, the selectivity and reproducibility of Cre expression is absolutely critical for the result. To meet these requirements large constructs based on bacterial artificial chromosomes (BACs) have been successfully used. Here we present a protocol for the generation of constructs in which the Cre recombinase or a tamoxifen-inducible Cre fusion protein, are inserted at the translation start sequence of a BAC-derived gene. We describe all the critical steps, including construct-design, recombineering, and preparation of the transgene-containing genomic fragment for pronuclear injection and identification of “founder” animals among the resulting offspring. In our experience, the use of this protocol typically results in specific and transgene copy number-dependent expression of the Cre recombinase.

1 Introduction

The Cre/loxP recombination system is the most widely used system for cell-type-specific gene inactivation (1–3). The core of this system is the Cre recombinase, a protein from bacteriophage P1, which mediates recombination of specific sequences called loxP sites. When two loxP sites are introduced into the genome in the same orientation and the Cre recombinase is expressed, the Cre will recombine the loxP sites and delete the sequence between the sites. In this way the Cre/loxP system can be used to delete genes or critical parts of genes. The system is also widely used for conditional gene activation. In this setting a cassette that is blocking transcription, a so-called “stop-cassette” that is flanked by loxP sites, is introduced between a promoter and the translational start site of a gene. Thus, the gene is silent under basal conditions but activated upon Cre-mediated removal of the stop-cassette. For a review of advanced applications of the Cre/loxP recombination system see (4, 5). Cre fusion proteins have been developed which are inactive in the basal state since they are fused to a modified ligand-binding domain of the estrogen receptor and thus kept in the cytoplasm (6–9). Upon binding of tamoxifen, a synthetic steroid, the fusion protein is translocated to the nucleus and the Cre is able to perform the recombination of the loxP sites. Furthermore, in order to restrict gene inactivation in a certain cell-type or tissue, the Cre fusion protein can be expressed under a tissue-type-specific gene promoter.

The usefulness of this approach is critically dependent on the quality of the mouse line carrying the Cre transgene. Thus, the recombinase must be expressed in the vast majority of the cells that should be targeted (high efficiency) and no recombination should be observed in other cells, or in the absence of the ligand (low leakiness). Classic approaches, based on plasmid-derived constructs containing short (10–15 kb) genomic fragments, often do not meet these requirements. In addition they commonly suffer from inconsistent expression patterns between different lines harboring the same construct and the expression patterns often change between generations as well as between different mouse backgrounds. A major step forward for the generation of high-quality transgenic mice was the development of methods for using large pieces of genomic DNA propagated by yeast artificial chromosomes (YACs) (10, 11) and bacterial artificial chromosomes (BACs) (12, 13). A second breakthrough along this line was the development of systems for manipulating BAC-based genomic DNA fragments by homologous recombination in bacteria (14). These systems allow the manipulation of large genomic fragments, which due to their considerable size (typically 100–250 kb) in most cases contain all of the regulatory elements necessary for the expected expression pattern of the transgene. Collectively, this typically makes the expression pattern independent of the site of integration and the strength of expression is mainly dependent on the number of transgenes integrated into the genome (“copy-number”). These advantages have made BAC-based transgenesis the state-of-the-art method for the generation of Cre-expressing lines. Here we describe a detailed protocol for the generation of constructs for Cre lines using BAC recombineering. In our experience, the use of this protocol typically results in specific, reproducible and copy number-dependent expression of the Cre.

2 Materials

2.1 Software and Web Resources

• 1. Vector NTI software package is available from Invitrogen and is free for academic use (registration required). pDraw is distributed by Acaclone Software (www.acaclone.com) and is provided as “freeware”.

• 2. The interface for searching BAC clones is provided at www.ncbi.nlm.nih.gov/projects/genome/clone/. A similar functionality is also provided by the Ensembl genome browser: www.ensembl.org

• 3. The website of CHORI, which provides a comprehensive collection of mouse BAC clones for academic use worldwide is found at bacpac.chori.org

2.2 Plasmids and Bacterial Strains

• 1. Bacterial strains EL250 and SW105 harboring inducible recombinases are available from the NCI (see http://recombineering.ncifcrf.gov/).

• 2. The pIndu and pConst plasmids were cloned by Erich Greiner and are available to academic institutions from our laboratory. Maps of the plasmids are shown in Fig. 17.1 .

Fig. 17.1.

Maps of plasmids pConst and pIndu. The Cre in the pConst plasmid is the codon-optimized (“improved”) Cre variant by Shimshek and colleagues (2002) (22). The CreER^T2 on the pIndu plasmid is the fusion of the improved Cre and a modified ligand-binding domain of the human estrogen receptor (23). Abbreviations: CMV, cytomegalovirus; pA, sequence containing the polyadenylation signal; bla, beta-lactamase; frt, binding sites for the Flp-recombinase.

2.3 Equipment

• 1. Shaking incubators for bacterial cultures set to 37°C and 32°C.

• 2. Incubators set to 37°C and 32°C for LB agar plates.

• 3. A water bath set to 42°C for steps requiring the induction of the temperature-sensitive expression of the λ prophage.

• 4. Refrigerated centrifuge(s) and rotors for 1.5-mL tubes and 50-mL tubes.

• 5. A spectrophotometer for measuring DNA concentration and monitoring growth of bacteria (wavelengths 260 nm and 600 nm, respectively).

• 6. A complete setup for running DNA agarose gels with ethidium bromide.

• 7. An electroporation device and 0.1-cm electroporation cuvettes (i.e., BIORAD Gene Pulser and Gene Pulser Cuvettes 0.1 cm #1652089).

• 8. Equipment for pulse-field electrophoresis (i.e., BIORAD CHEF-DR III System).

2.4 Reagents

• 1. LB medium.

• 2. Chloramphenicol 25 mg/mL in ethanol.

• 3. Ampicillin 100 mg/mL in sterile deionized water.

• 4. LB agar plates containing 25 μg/mL chloramphenicol.

• 5. LB agar plates containing 100 μg/mL ampicillin (Note: Do not use ampicillin plates that were stored longer than 2 months).

• 6. LB agar plates containing both 25 μg/mL chloramphenicol and 100 μg/mL ampicillin.

• 7. 10% (v/v) glycerol in deionized water sterilized by autoclaving. Pre-chill the solution to 0°C for several hours before use.

• 8. Liquid nitrogen, a plastic device for removing tubes after freezing and appropriate protective gear (goggles and gloves).

• 9. Buffers P1, P2, and P3 from Qiagen (available with DNA purification kits or separately).

• 10. Qiagen Qiafilter Plasmid Midi Kit for plasmid DNA isolation (or equivalent).

• 11. BD Falcon serological 5-mL pipette #357529 (for the gel filtration column).

• 12. Sepharose CL-4B from Amersham Biosciences.

• 13. 10% (w/v) L-arabinose in water, filter-sterilized.

• 14. Injection buffer: 10 mM Tris-HCl pH 7.5, 0.1 mM EDTA pH 8.0, 100 mM NaCl, autoclaved and filtered.

• 15. TE buffer pH 8.0 (10 mM Tris-HCl pH 8.0, 1 mM EDTA, filter-sterilized or autoclaved.

• 16. Filter paper for removing precipitated proteins during BAC purification.

• 17. Restriction enzymes for analysis of the BACs and preparation of the construct for injection.

3 Methods

The protocol below is based on the methods described by (14–19) and the lab protocols of Tim Wintermantel, Erich Greiner and Stefan Berger.

3.1 Guidelines for Constructing the Plasmid with the Cre

The Cre (or CreER^T2) encoding sequence is introduced into the genomic DNA fragment by recombineering. Therefore, it has to be flanked by “homology arms” to allow for homologous recombination mediated by the λ prophage recombinase system (extensive description of the bacterial strains harboring the temperature-sensitive λ prophage is found in (15, 20)). Correct recombination in the target locus is essential. Introducing any additional nucleotides in the promoter region may affect the specificity or efficiency of transcription. We use the following guidelines for cloning the homology arms.

• 1. Each homology arm should be around 200–500 bp. The “left” homology arm (upstream, 5′) should include the genomic fragment 5′ from the start codon. On the 3′ end it should include an ATG sequence followed by an EcoRV “half-site” (see Fig. 17.2 ). The ATG sequence will correspond to the start codon of the Cre recombinase. The 5′ end point of the genomic sequence included should be a “half-site” for a blunt cutting enzyme (i.e., SnaBI). Additionally, a restriction site (i.e., KpnI) should be introduced on the 5′ end for subsequent cloning into the plasmid (see Note 1).

• 2. The “right” homology arm (downstream, 3′) should contain the genomic sequence starting with the ATG and ending with a “half-site” for a blunt cutting enzyme (i.e., SnaBI) (see Fig. 17.2). The arm needs to be flanked with restriction sites that will be used for cloning (i.e., NheI/XbaI).

• 3. An example of a plasmid construct with homology arms is shown in Fig. 17.2.

Fig. 17.2.

Design example for cloning homology arms into the plasmid carrying the recombinase gene. The left homology arm was amplified by PCR, phosphorylated with polynucleotide kinase and then cloned into the BAC using the KpnI and EcoRV sites. A particular advantage of the EcoRV site is the presence of a guanosine residue at the +4 position, which should enhance translation rate (“strong” Kozak sequence). The right homology arm was amplified by PCR and digested with XbaI and NheI. The resulting sticky ends are compatible with each other, and were cloned into the NheI site in the vector. The advantage of using the XbaI and NheI versus NheI sites alone is the greatly facilitated screening for orientation of the insert after ligation (the NheI site will be reconstituted only on one side of the insert). The pIndu-D1 plasmid shown on the right is the final product of the cloning procedure.

3.2 Selecting the BAC

While screening BAC libraries of the mouse genome has been the classic approach to selecting genomic fragments of interest for the last decade, recently the indexing (sequencing of the BAC “arms”) has been almost completed. The collection of indexed BAC clones is available among others from CHORI (bacpac.chori.org) and several commercial suppliers (see Note 2). Two tools allow for easy finding of a BAC clone containing the promoter region of the gene of interest: the Clone Registry at www.ncbi.nlm.nih.gov/projects/genome/clone/ and the Ensembl genome browser at www.ensembl.org. The BAC display may be enabled while in Contig View by selecting an appropriate option from the DAS resources pull-down menu. An advantage of using both databases for querying BAC clones is that they complement each other. On independent genome assembly projects and thus do not necessarily provide identical results.

When selecting the optimal genomic fragment on a BAC clone, consider having large stretches (>30 kb) upstream and downstream from the gene used to drive the Cre expression. At the same time, avoid including other genes on the BAC clone, as this may result in a phenotype associated with overexpression of the genes included. An example of a BAC clone used for constructing a transgene is shown in Fig. 17.3 . In the example, the gene of interest is Drd1a, the mouse D1 dopamine receptor gene. Upstream is the Sfxn1 gene, but only part of the gene fits on the clone and we assume it will yield no functional product. The Drd1a gene consists of two exons, with the translational start encoded in exon 2. In some cases a BAC meeting the criteria above might not be available and then ET-cloning (14, 16) may be used to remove unwanted portions of the BAC.

Fig. 17.3.

Recombineering of the Cre-encoding sequence into the target genomic sequence. (A) Example of a BAC clone (RP24-179E13) used for construction of the transgene. The boxes represent the Drd1a and Sfxn1 genes. The dashed line around the Sfxn1 gene indicates that the sequence is incomplete: the last three exons are missing. The arrows above the boxes indicate the direction of transcription. The BAC “backbone”“ containing genes and sequences necessary for replication and selection in bacteria lies between the NotI sites indicated. (B) Schematic of recombineering of the CreER^T2 sequence into the coding region of the Drd1a gene. The plasmid fragment containing the recombinase is inserted by homologous recombination catalyzed by the λ prophage recombinases into the site corresponding to the start of translation in exon 2 of the Drd1a gene.

Select one or preferably two BACs that would be suitable for constructing the transgene. It is recommended to prepare exact maps of the BACs using software like Vector NTI or pDraw, as it will be essential to analyze restriction enzyme cleavage patterns at several of the subsequent steps.

3.3 Purification of the BAC for Restriction Analysis and Electroporation

• 1. Use a sterile pipette tip to gently scratch the stab, and then transfer the tip to a sterile 1.5-mL centrifuge tube containing 1 mL of LB medium.

• 2. Perform serial dilutions so you have 10^–3, 10^–5 and 10^–7 dilutions of the BAC clone-harboring bacteria.

• 3. Plate 100 μl of the dilutions on LB agar plates containing 25 μg/mL chloramphenicol. Let the plates dry and then incubate them overnight at 37°C.

• 4. On the next day pick a few single colonies and use them to inoculate 5 mL overnight cultures (37°C with shaking). Use up to 4 mL of the culture for the next step, but keep 1 mL in case a clone will be selected for making a glycerol stock.

• 5. Pour ˜1.4 mL of the BAC clone overnight culture into a centrifuge tube (1.5 mL) and spin it down 5 min at ˜10,000g.

• 6. Discard the supernatant, add more culture and spin down. Repeat these steps until 4 mL of culture had been spun down.

• 7. Carefully remove traces of LB and then add 300 μl of Qiagen buffer P1. Resuspend the bacteria by gentle pipetting or mixing.

• 8. Add 300 μl of Qiagen buffer P2, mix 6–8 times by inverting the tube and incubate for 5 min at room temperature (do not extend this step).

• 9. Add 300 μl of Qiagen buffer P3, mix by inverting and place the tube for 5′ on ice. Spin 10′ at maximum speed (>10,000g) at 4°C.

• 10. Carefully remove 800 μl of the supernatant and transfer it to a new tube. To precipitate the BAC DNA, add 540 μl of isopropanol and mix gently by inverting.

• 11. Spin the tube 10′ at maximum speed (>10,000g) at 4°C. Carefully remove supernatant without disturbing the pellet.

• 12. Add 500 μl 70% (v/v) ethanol to wash the pellet, and spin 5′ at maximum speed (>10,000g) at 4°C. Remove supernatant, spin shortly again at room temperature, and remove remaining traces of supernatant.

• 13. Leave the tubes open for 2–3′ to allow the pellet to dry, do not use a speedvac.

• 14. Dissolve the isolated BAC in 40 μl TE pH 8.0. The BAC is much more difficult to dissolve than plasmids, leave the tube at 4°C for 1–2 h and mix it by tipping occasionally. Do not try to suspend the pellet by pipetting since this will result in shearing of the BAC. Store the BACs at 4°C.

• 15. The purified BACs may be used for a diagnostic restriction cut, in order to confirm that a predicted pattern from the map of the BAC matches the actual result. Select a restriction enzyme (or a combination of enzymes) that will produce several easy-to-separate products in the range of 300–3000 bp apart from larger bands. Larger bands will resolve poorly, but when necessary they may be resolved by pulse-field electrophoresis (see Fig. 17.4 ).

• 16. Cut the BAC with selected restriction enzymes and run the products of the reaction on a 1% (v/w) agarose gel.

• 17. Allow the bromophenol blue band (corresponding to ˜200 bp on 1% agarose gel) to migrate two-thirds of gel length. When the electrophoresis is finished, post-stain the gel with EtBr or any other dye. The post-staining step is often necessary to visualize relatively faint bands.

• 18. Acquire an image of the band patterns on the gel and compare the results against the theoretically predicted pattern. In case there is a perfect match then the BAC clone is suitable for further work.

• 19. Make a glycerol stock (preferably more than one) from the culture with validated BAC. The purified and verified BAC will be used for electroporation into cells allowing for inducible recombination of large DNA fragments.

Fig. 17.4

Examples of BAC clone analysis. (A) Restriction fragment pattern analysis on agarose gel. Enzymes used for cleavage of the BAC clone are indicated above the gel image.(B) Pulse-field gel electrophoresis (PFGE) analysis of restriction fragment patterns. Compared to agarose gels this method allows for resolving the larger fragments. The image shown as the example represents the restriction patterns of the initial BAC clone (Cre–) and the final result of cloning with inserted Cre and removed beta-lactamase (Cre+). The presence of only two bands in the lanes corresponding to the NotI cleavage products (the lower one is very faint) confirms removal of the beta-lactamase. The difference in the two bands after XhoI treatments confirms the CreER^T2 integration. (C) Detection of Cre insertion site by Southern blotting. The presence of a single band of predicted size in the Southern blotting result confirms a correct and unique integration of the CreER^T2.

3.4 Preparation of Competent Cells for Electroporation

• 1. Prepare serial dilutions from SW105 or EL250 bacterial cells (see Note 3) and plate them on LB agar plates (no antibiotic). Incubate for ˜20 h at 32°C.

• 2. Inoculate a 5 mL overnight culture with a single colony picked from one of the plates.

• 3. In the morning, cool 500 mL sterile 10% (v/v) glycerol in H₂O on ice (for at least 3 h before use) (see Note 4). Transfer 2 mL of the overnight cultures to Erlenmeyer flasks containing 50 mL of LB. Grow the cultures 3–5 h, until they reach absorbance at 600 nm of 0.45–0.5. Then transfer them to 50-mL Falcon tubes and spin down at 4000g for 10′ at 0°C in a pre-cooled rotor.

• 4. Carefully discard the supernatant, and place the inverted tube on a stack of paper tissues in a 4°C refrigerator, to dry remaining supernatant. Do not leave the tubes unattended, after 2–3 min the supernatant will be drenched and the pellets will start “sliding” down.

• 5. Take the tubes back on ice, add 50 mL of ice-cold 10% (v/v) glycerol and resuspend the cells (see Note 5). Spin down again at 4000g for 10′ at 0°C and perform the washing with 10% (v/v) glycerol two more times for a total of three washes.

• 6. After the last wash the remaining volume should be around 0.6–0.7 mL. Mix gently and aliquot 50 μl into pre-cooled 1.5-mL centrifuge tube tubes. Flash freeze in liquid nitrogen for future use, or proceed immediately to electroporation.

3.5 Electroporation of the BAC into Competent Bacterial Cells

• 1. Place clean 100 μl electroporation cuvettes on ice 5 min before use. Put the tube with purified BAC and aliquots of competent bacteria on ice as well.

• 2. Prepare the electroporations by pipetting 50 μl of the competent bacteria and 1 μl of the purified BAC into the cuvette. Put the cuvette in the electroporation device and deliver a pulse at 2.3 kV with 25 μF and pulse controller set to 200 Ω (BioRAD Gene Pulser). After the pulse, add 1 mL LB to the cuvette and then transfer the whole volume to a 1.5-mL centrifuge tube. Perform additional two electroporations, using 2 μl and 5 μl of the purified BAC.

• 3. The three tubes containing electroporated bacteria in LB should be incubated for 1 h at 32°C with shaking.

• 4. Plate from each tube 100 μl of the culture from each tube and then spin the tube shortly to pellet the bacteria, remove 800 μl medium, resuspend the bacteria in the remaining medium, and plate them on LB agar with 25 μg/mL chloramphenicol. Incubate the plates for at least 20 h at 32°C.

• 5. Pick colonies, use them to inoculate 5-mL cultures and purify the BACs as described above (Section 3.3). Verify the integrity of the BAC by restriction analysis and prepare glycerol stocks of the BAC clone in the SW105 (or EL250) cells.

3.6 Preparation of Competent Cells for Homologous Recombination

Once the BAC is transformed into the SW105/EL250 cells it is necessary to induce the λ prophage recombinase system that will mediate the homologous recombination between the targeting construct containing the Cre and the site in the genomic sequence containing the start of translation of the gene of interest. The procedure follows the same steps as the standard protocol for preparing electrocompetent cells (Section 3.4) with the following modifications:

• 1. When the cells reach absorbance at 600 nm of ˜0.5, transfer the culture to a shaking water bath set to 42°C for 15′ (we simply place the flask in a water bath and shake it by hand) to induce the bacterial recombinase activity.

• 2. Move the flask to an ice-slurry and shake it gently to cool the medium to 0°C. Leave the culture on ice for ˜20′. Further steps are performed according to the standard protocol (Section 3.5).

3.7 Preparation of a Construct for Recombination with the BAC

• 1. Purify ˜0.5 pmol (typically ˜1.5 μg) of the plasmid harboring the construct prepared for recombination (a standard miniprep is usually enough).

• 2. Set up a reaction to excise the fragment with the blunt-cutting enzyme of choice (i.e., SnaBI). Set the conditions in a way that there is a considerable “over-cut”; for example, extend the time of the reaction to overnight (see Note 6). Isolate the excised fragment by preparative gel electrophoresis.

• 3. Electroporate 0.1 pmol of the fragment into competent cells with induced λ prophage recombinase activity and harboring the target BAC clone. Use identical conditions as described in Section 3.5 (2.3 kV). Add 1 mL of LB medium to the electroporation cuvette, then transfer the culture to a 1.5-mL tube and incubate the bacteria for 1 h at 32°C.

• 4. Plate the bacteria on LB agar containing 25 μg/mL chloramphenicol and 50 μg/μl ampicillin. Incubate the plates for ˜20 h at 32°C.

• 5. Pick several colonies from the plates and prepare small overnight cultures for BAC isolation.

• 6. Perform restriction analysis on isolated BACs from each of the cultures to verify which of the clones appear to have correctly integrated the Cre cassette (see Fig. 17.4 and Note 7). Select a validated BAC clone for further steps.

3.8 Removal of Antibiotic Resistance Gene

The beta-lactamase gene conferring ampicillin resistance is flanked by frt sites and may be removed inducing the Flp-recombinase expression (see Fig. 17.3), which is placed under an arabinose operon-controlled promoter in the genome of SW105 or EL250 cells.

• 1. Prepare an overnight culture (LB + 25 μg/mL chloramphenicol) from a recombined and validated BAC clone containing the Cre cassette.

• 2. In the morning, inoculate 50 mL of LB supplemented with 25 μg/mL chloramphenicol with 1 mL of the overnight culture. Grow the cells at 32°C with shaking until they reach absorbance at 600 nm of ˜0.5. Add 0.5 mL 10% (w/v) L-arabinose.

• 3. Incubate the culture for one more hour and then dilute it by transferring 5 mL of the culture to a new flask containing 50 mL of LB supplemented with 25 μg/mL chloramphenicol. Grow the diluted culture for one more hour at 32°C.

• 4. Prepare serial dilutions from the culture and plate 100 μl from 10^–3, 10^–5 and 10^–7dilutions on LB agar plates containing 25 μg/mL chloramphenicol. Allow the colonies to grow for ˜20 h at 32°C.

• 5. Pick several colonies to prepare small-scale BAC preparations and perform restriction fragment analysis (see Note 8). Select a couple of clones (2 to 4) that display the desired band pattern and perform a medium-scale BAC preparation from each, using for example Qiagen large-construct kits (see Note 9).

Before proceeding to the next step it is necessary to perform sequencing of the candidate BAC, in particular regions containing the ends of the recombination arms introduced from the plasmid as well as the Cre coding sequence. This can be accomplished by using PCR to amplify fragments of the BAC or direct sequencing of the BAC with primers against the regions of interest.

3.9 Preparation of the BAC Fragment for Injection

• 1. Use 20–50 μg of the final BAC for restriction enzyme (i.e., NotI) treatment; the total volume should not exceed 100 μl (see Note 10). After the reaction is finished, denature the restriction enzyme by heating at 70°C for 10 min. Spin down the samples and store them at 4°C for further use.

• 2. Purification of the BAC is performed using a Sepharose CL-4B column. Pour ˜20 mL of Sepharose CL-4B into a 250-mL beaker. Add 150 mL of injection buffer (see Note 11), mix by gentle swirling, and allow the gel to settle completely. Decant the solution, add another 150 mL of injection buffer and perform the procedure three more times for a total of four washes. These steps are necessary for removal of ethanol and sodium azide that are used as gel preservatives.

• 3. Resuspend the Sepharose in ˜100 mL of injection buffer and transfer to a vessel that allows to degas the gel. Connect the Erlenmeyer flask to the pump and evacuate the gel for 15′, swirling occasionally. Use protective eyewear; the pressure may cause the flask to shatter explosively. Transfer the degassed gel to a beaker without introducing air bubbles.

• 4. Prepare the column using a 5-mL Falcon serological pipette. Remove most of the cotton bud and push the remaining bit to the tip of the pipette using compressed air; it will prevent the gel from flowing out. If the piece of cotton is too big it will slow the flow of buffer through the column.

• 5. Attach the column firmly to a holder and check that it is perfectly vertical. Close the bottom of the column (i.e., a pipette tip with a piece of parafilm at the end). Fill the column with injection buffer up to two-thirds of height.

• 6. Take the beaker with the gel and swirl it gently to resuspend the Sepharose. Apply the gel slurry with a Pasteur pipette to the column and fill it up to the top. The gel will start to set on the bottom of the column.

• 7. Once ˜1 cm of gel matrix settles down, open the outflow and place a beaker (50–100 mL is usually convenient) below the column to collect the eluate.

• 8. Keep applying the gel on top of the column (swirl the beaker with the gel occasionally). Pack the column until the gel is 2–3 cm from the top of the column (see Note 12).

• 9. Once the packing is finished connect with a piece of parafilm a 50-mL syringe (without the piston) to the top of the column and fill it with injection buffer without allowing the gel to dry or disturbing its surface.

• 10. Wash the column with at least 30 mL of injection buffer before starting the BAC purification. Once the column is washed, wait until the upper reservoir is empty and then remove it.

• 11. When there is almost no buffer left on top of the gel (be careful, the gel will start shrinking before drying) apply the BAC sample (˜100 μl) without disturbing the gel surface. Let the sample enter the column, and just before the column dries apply 100 μl of the injection buffer.

• 12. Again, allow the solution to enter the column, and repeat this step one more time. This should prevent the sample from diluting during loading and is important for good separation.

• 13. Load the column with injection buffer till the top of the pipette, re-attach the syringe-reservoir and fill it with injection buffer. Start collecting 300-μl fractions.

• 14. Collect ˜35–40 fractions and stop the column. Measure the absorbance at 260 nm of all fractions (if the spectrophotometer allows for measuring in ˜60-μl volume, you may measure the samples directly).

• 15. In most cases a first smaller peak of absorbance will separate from a second larger one, corresponding to the backbone and transgene, respectively. Load the fractions corresponding to the second peak (or every other fraction if absorbance measurements give no clear conclusion) on a 1% (w/v) agarose gel.

• 16. Load 20-μl aliquots of selected samples onto a 1% (w/v) pulse-field agarose gel and separate the fragments. After the electrophoresis stain the gel with ethidium bromide. If based on this analysis the separation was successful, and the transgene does not appear degraded, the selected fraction(s) are suitable for injection. An example of pulse-field result is shown in Fig. 17.5 (see Note 13).

Fig. 17.5.

Analysis of fractions collected from the gel column. Aliquots (20 μl) from the fractions (as indicated above the images) were run on a normal agarose gel (A) and pulse-field gel electrophoresis (B, PFGE). On panel B the upper band corresponds to the genomic fragment with the inserted Cre, and the lower band corresponds to the excised BAC backbone.

3.10 Guidelines for identification and characterization of founder animals

The genotyping of animals for identification of founders can be performed by diagnostic PCR. For the detection of the Cre or CreER^T2 recombinases the following primers may be used (21):

Cre-forward:	ctg cca ggg aca tgg cca gg
Cre-reverse:	gca cag tcg agg ctg atc agc
Product size:	316 bp

and for amplification of the CreER^T2exclusively:

CreERT2-forward:	ggc tgg tgt gtc cat ccc tga a
CreERT2-reverse:	ggt caa atc cac aaa gcc tgg ca
Product size:	406 bp

Reaction conditions are initial denaturation for 3′ at 95°C followed by 30 cycles of 30′′ at 95°C, 30′′ at 60°C and 60′′ at 72°C with additional 10′ at 72°C at the end of the PCR.

4 Notes

• 1. The genomic sequence in the BAC clone is a convenient template for amplifying the “homology-arms”. Use a high-fidelity polymerase (i.e., Pfu) for amplification of the arms.

• 2. BACs ordered from CHORI arrive as agar-stabs and are stable for days at 4°C, nevertheless take steps to preserve the BAC immediately upon arrival. Validate the BAC by analyzing restriction enzyme cleavage patterns and/or sequencing of the “arms”.

• 3. Incubations of SW105 or EL250 cells are done at ≤ 32°C, to prevent the induction of the bacterial recombinase. As a result these bacterial strains tend to grow slowly and it is often necessary to wait for more than 20 h to obtain suitable colonies on agar plates. For the purpose of the procedure described here the two strains are equivalent. A more complete description of the strains is found in (15, 19).

• 4. While preparing electroporation-competent bacteria all steps have to be performed as close to 0°C as possible. When using glass pipettes pre-cool them by pipetting a 0°C cold solution before pipetting bacteria. Do not remove the bacteria from ice. If possible, work in a cold room (and still keep everything on ice). The glycerol used in the procedure allows for storing the bacteria for future use. If the competent cells will be only used freshly, it is possible to omit it.

• 5. After subsequent glycerol (or water) washes the bacteria will be forming an increasingly less solid pellet. While it makes resuspending them easier, discarding the supernatant without disturbing the pellet becomes difficult.

• 6. When recombineering the Cre-containing fragment into the genomic sequence, removing traces of the plasmid is absolutely essential. The intact plasmid is greatly preferred by cells for acquiring resistance to ampicillin. It is not unreasonable to “re-cut” the agarose-purified fragment with the appropriate restriction enzyme and perform another preparative agarose gel (though make sure that the final yield will be sufficient).

• 7. Typical conditions for pulse-field electrophoresis would be 6 V/cm, included angle 120, initial switching time 0.5′′, final switching time 20′′, run time 14 h and cooling set to 14°C.

• 8. It is often convenient to use NotI for verifying the presence of the beta-lactamase gene in the BAC construct, as a NotI site is present in its sequence. It might be necessary to use pulse-field electrophoresis to separate the large fragments generated by NotI treatment. Additionally, Southern blotting may be performed using a probe targeting the Cre-encoding sequence to confirm that the recombinase was integrated in the correct location, though in this case a restriction enzyme(s) producing several fragments from the BAC is recommended.

• 9. The purification of BACs may be performed using commercially available kits (i.e., Qiagen large-construct kits), with the following modifications: (a) Be very careful while pipetting the BAC-containing solutions to prevent shearing. Cut away the ends of pipette tips to widen them a little. Pipette slowly. (b) Do not use syringe-mounted filters, remove the precipitate by passing the solution through a piece of filter-paper mounted on top of the purification column. (c) Do not over-dry the DNA pellet after precipitation. Allow the BAC to resuspend for 1 – 2 hours or overnight at 4°C.

• 10. In most cases NotI will remove the backbone of the BAC without fragmenting the genomic sequence. Usually an overnight restriction reaction at 37°C using 1 U of Not I per 1 μg of DNA will result in efficient cleavage of the BAC, which should be verified on an analytical pulse-field gel electrophoresis (PFGE).

• 11. The injection buffer must be of highest purity; any kind of contamination might result in toxicity to the oocytes. We cannot stress enough how important meticulous care is in its preparation.

• 12. The packing of the Sepharose column has to be continuous: the gel should not be allowed to settle completely until the packing is finished. Do not introduce bubbles into the column, as they would impair the separation. Typical flow rate through the column varies around 5–10 mL/h. You may leave the column overnight for washing, but make sure that the amount of buffer is sufficient and the column will not dry out. If the flow through the column is stopped, the washing should be performed again. Do not store the column at 4°C, as it will cause bubbles to appear.

• 13. The purification of the BAC involves a gel filtration matrix but the BAC fragments probably exceed the size of exclusion of the pores of the matrix and the order of elution of the fragments is opposite to what could be expected. Despite the fact that the separation is obviously not due to size exclusion, the method allows to separate the BAC fragments without the need of preparative pulse-field electrophoresis followed by buffer exchange.