C 2014 Wiley Periodicals, Inc. V
genesis 52:907–915 (2014)
TECHNOLOGY REPORT
Generation of Point-Mutant FAK Knockin Mice B. Tavora,1 S. Batista,1 A.N. Alexopoulou,1 V. Kostourou,2 I. Fernandez,1 S.D. Robinson,3 D.M. Lees,1 B. Serrels,4 and K. Hodivala-Dilke1* 1
Adhesion and Angiogenesis Laboratory, Centre for Tumour Biology, Barts Cancer Institute, -A CR-UK Centre of Excellence, Queen Mary University of London, London, United Kingdom
2
Vascular Adhesion Lab, BSRC Al. Fleming, Athens, Greece
3
School of Biological Sciences, University of East Anglia, Norwich, United Kingdom
4
Edinburgh Cancer Research UK Centre, Institute of Genetics and Molecular Medicine, University of Edinburgh, Western General Hospital, Edinburgh, United Kingdom
Received 17 December 2013; Revised 18 September 2014; Accepted 18 September 2014
Summary: Focal adhesion kinase is a non-receptor protein tyrosine kinase with signaling functions downstream of integrins and growth factor receptors. In addition to its role in adhesion, migration, and proliferation it also has non-kinase scaffolding functions in the nucleus. Focal adhesion kinase (FAK) activation involves the following: (1) ligand bound growth factors or clustered integrins activate FAK kinase domain; (2) FAK autophosphorylates tyrosine (Y) 397; (3) Src binds pY397 and phosphorylates FAK at various other sites including Y861; (4) downstream signaling of activated FAK elicits changes in cellular behavior. Although many studies have demonstrated roles for the kinase domain, Y397 and Y861 sites, in vitro much less is known about their functions in vivo. Here, we report the generation of a series of FAK-mutant knockin mice where mutant FAK, either kinase dead, non-phosphorylatable mutants Y397F and Y861F, or mutant Y397E—containing a phosphomimetic site that results in a constitutive active Y397, can be expressed in a Cre inducible fashion driven by the ROSA26 promoter. In future studies, intercrossing these mice with FAKflox/flox mice and inducible cre-expressing mice will enable the in vivo study of mutant FAK function in the absence of endogenous FAK in a spatially and temporally regulated fashion within the whole organism. genesis 52:907–915, C 2014 Wiley Periodicals, Inc. 2014. V Key words: focal adhesion kinase; mutants; mice
Focal adhesion kinase (FAK) is one of the central players in integrin and growth factor receptor mediated signaling and has also been reported to have scaffolding
functions. FAK is a 125-kDa protein with a central tyrosine kinase domain flanked by FERM and FAT domains. FAK is maintained in an inactive conformation in which the FERM domain is bound to the kinase domain. This impairs access to the tyrosine 397 residue (Y397) of FAK, thus leading to auto-inhibition. Integrin or growth factor receptor activation initiates a signal that releases the auto-inhibitory interaction between the FERM and kinase domains, leading to kinase activation and subsequent autophosphorylation of Y397 (Lechertier and Hodivala-Dilke, 2012; Luo and Guan, 2010; Mitra and Schlaepfer, 2006; Parsons et al., 2008; Schaller et al., 1995). Phosphorylation of Y397 provides an Sh2binding site allowing Src to interact with FAK. This interaction stimulates Src to transphosphorylate FAK on a number of key tyrosine residues, including FAK– Y861. Phosphorylation at FAK–Y861 allows p130Cas binding, activating the PI3-kinase signaling pathway. Despite this apparently clear pathway of events, some Additional Supporting Information may be found in the online version of this article. * Correspondence to: K. Hodivala-Dilke, Adhesion and Angiogenesis Laboratory, Centre for Tumour Biology, Barts Cancer Institute, -A CR-UK Centre of Excellence, Queen Mary University of London, Charterhouse Square, London, UK. E-mail: [email protected] B.T., S.B., and A.N.A. contributed equally to this work. Contract grant sponsor: Cancer Research UK, Contract grant number: C8218/A12007; Contract grant sponsor: Breast Cancer Campaign, Contract grant number: 2010MayPR39; Contract grant sponsor: The Association of International Cancer Research, Contract grant number: 12–1068 Published online 20 September 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/dvg.22823
908
TAVORA ET AL
results have demonstrated unexpected pathways for FAK function. For example, growth factor stimulation can induce a Src-dependent increase in pY861 levels but not pY397 levels (Abu-Ghazaleh et al., 2001); FAK knockdown or deletion can increase invadopodia formation, while still impairing invasive cell migration (Ilic et al., 1995; Shen et al., 2005); and re-expression of the kinase dead mutant of FAK (K454R) can restore normal invadopodia formation while mutations in FAK–Y397 do not. Thus, understanding how these different residues of FAK control cell behavior both in vitro and in vivo become important. FAK can be transphosphorylated and thus overexpression of mutant FAK in cells or mice where endogenous FAK is still present makes it impossible to demonstrate the effect of the FAK mutation precisely. To overcome this issue, we have developed a series of Cre inducible mutant FAK-knockin mice which when crossed with FAK-floxed mice will generate mice where endogenous FAK is knocked-out and mutant FAK simultaneously knocked in driven by the ROSA 26 promoter. Other groups have produced kinase defective FAK mutant knockin mice by making point mutations in the FAK gene (Lim et al., 2010; Zhao et al., 2010). These mutations might affect not only downstream signaling pathways but also the regulation of FAK gene transcription (i.e., controlled by the endogenous functional promoter). In contrast, our knockin/knockout system is able to dissect the effect of FAK mutations in signaling without influencing FAK transcription as endogenous FAK is genetically ablated. In addition, our system has the advantage over some of being used with inducible Cre lines. RESULTS Using chicken FAK cDNA constructs we have generated knockin mutants of FAK where: the Y397 site cannot be phosphorylated (Y397F) or mimics a constitutively phosphorylated form (Y397E); the Y861 site cannot be phosphorylated (Y861F); or the kinase domain is disabled whereby the ATP binding site is mutated (lysine (AAA) in the position 454 to arginine (AGA). In addition, double mutants of FAK Y397E/kinase dead (KD), which generates a constitutively active Y397 site with an inactive kinase domain, were generated to assess kinase function independently of Y397 phosphorylation and possibly Src binding. The wild type (WT) chicken FAK construct was used as a negative control (Supporting Information Fig. S1a). Many of these mutants have been used in in vitro studies and their functions validated (Abouzahr-Rifai et al., 2008; Lie et al., 2012; Shi et al., 2009; Zhao et al., 1998). Mutant chicken-FAK constructs (preceded by a STOP sequence flanked by loxP sites) were targeted to the ubiquitous Rosa26 locus to generate a Cre-inducible sys-
tem for mutant FAK expression (Supporting Information Fig. S1b). Targeting the Rosa26 locus, with similar constructs, has been used to induce the expression of various cDNAs in various cell types after Cre-mediated recombination (Halder et al., 2007; Mao et al., 2001; Soriano, 1999). For this purpose, the following cloning steps were required: (1) FAK constructs were cloned into pBluescript KS to allow for subsequent cloning steps; (2) these vectors were inserted into pBigT downstream of a STOP sequence flanked by loxP sites; and finally cloned into the pRosa26-1 vector to allow homologous recombination into the Rosa26 locus (Supporting Information Fig. S1b). FAK WT, Y397F, Y397E, KD, Y397E/KD, and Y861F constructs were electroporated into hybrid 129S6.C57BL6J ES cells and neomycin-resistant clones were selected for DNA extraction. Southern blot screening of ES cell clones was performed using EcoRV and AvrII digested DNA with the 50 R26 probe (Probe A), an external probe to the targeting vector that lies 50 of the Rosa26 locus (Soriano, 1999). EcoRV digested DNA probed with Probe A revealed both a WT (11 kb) and a targeted band (4 kb—indicative of homologous recombination in the Rosa 26 locus) whereas AvrII digested DNA probed with probe A shows a WT (5.5 kb) and a targeted band (8.4 kb). These ES cell clones were expanded for validation screens using Probe B, a probe internal to the targeting vector. EcoRV digested DNA screened using internal Probe B revealed a 9 kb band and AvrII digested DNA showed a 4.2 kb band and also confirmed homologous recombination (Supporting Information Fig. S2a,b). Confirmed positive ES clones, containing more than 70% of cells with normal chromosome counts (data not shown), were injected into blastocysts. Following injection, embryos were transferred into pseudopregnant foster mice, which gave birth 18 days later. High percentage chimeras (95–100%) were bred to C57/BL6 mice to generate offspring. Tail snip DNA was genotyped by Southern blot analysis for homologous recombination using Probes A and B. Mice that displayed both WT and targeted band with Probe A and targeted band with internal Probe B were chosen as founders for each FAK-mutant knockin (R26FAKKI) colony (Supporting Information Fig. S2c). Founder R26FAKKI/1 were bred together to generate homozygous R26FAKKI/KI mice that only reveal a 4 kb targeted band by Southern Blot in EcoRV digested DNA from tail snips of these animals (Fig. 1a). Sequencing mutation sites in DNA extracted from each homozygous FAKknockin mutant mouse lines confirmed the persistence of mutations in vivo (Fig. 1b). R26FAKKI/KI mice were then bred with cell-type specific Creert;FAKfl/fl mice (Tavora et al., 2010) to generate mice with tamoxifen-inducible mutant FAK-knockin and endogenous FAK-knockout in selected cell types (Fig. 2a). As an example, tamoxifen-inducible endothelial
POINT-MUTANT FAK KNOCKIN MICE 1/1
FIG. 1. Generation of FAK-mutant knockin mice. (a) Southern blot analysis of EcoRV-digested DNA extracted from tail snips of WT/WT, 397F/397F, 397E/397E, KD/KD, 397E/KD/397E/KD, and 861F/861F homozygous mice. Homologous recombination was confirmed in all mutations with Probe A. (b) Examples of sequencing of mutation sites in DNA extracted from tails of the same progeny. Tyrosine 397 is replaced by a phenylalanine (TTT) or glutamic acid (GAG) generating Y397F and Y397E, respectively. Lysine 454 is replaced by an arginine (AGA) generating a kinase dead (KD K454R). Tyrosine 861 is replaced by a phenylalanine (TTT) generating Y861F.
specific mutant-FAK-knockin and endogenous FAKknockout mice have been generated (Pdgfb-iCreert;FAKfl/fl;R26FAKKI/KI). Pdgfb (platelet-derived growth factor b) is produced predominantly by endothelial cells (Hellstrom et al., 1999). The efficiency of tamoxifeninduced Cre recombinase activity under the control of Pdgfb promoter (Pdgfb-iCreert mice) has been tested using Rosa26-lacZ reporter crosses and cre activity was shown to be endothelial specific (Claxton et al., 2008). Mice were genotyped using PCR and examples of these are given in Figure 2b. Pdgfb-iCreert PCR shows mice positive (Pdgfb-iCreert1) or negative (Pdgfb-iCreert2) for Creert. FAK-floxed PCR shows mice with both (FAKfl/ fl ) or one (FAKfl/1) FAK floxed alleles and WT mice non-
909
floxed (FAK ). Rosa 26 PCR shows mice homozygous (R26FAKKI/KI) and heterozygous (R26FAKKI/1) for Rosa 26 targeting and mice non-targeted in the Rosa 26 locus (R26FAK1/1) (Fig. 2b). We have isolated, cultured, and immortalized endothelial cells from Pdgfb-iCreert1;FAKfl/fl;R26FAKWT/WT and Pdgfb-iCreert2;FAKfl/fl;R26FAKWT/WT mice that were grown for a minimum of 2 weeks in the presence of 4-hydroxytamoxifen (OHT) to induce the recombination of the loxP sites in vitro to generate endothelial cells with the following genotypes, respectively, msFAKKO;chFAKKIWT and msFAKWT. Real-time PCR shows that OHT-treatment induces the deletion of endogenous mouse FAK and the simultaneous expression of the chicken FAK at the mRNA level (Fig. 3a). Immunoprecipitation with an antibody that recognizes both mouse and chicken FAK followed by blotting with an antibody that recognizes the Myc-tag of the chicken protein shows that chicken FAK is only expressed in the msFAKKO;chFAKKIWT cells (Fig. 3b). Transient Cre transfection in vitro may increase chFAK levels (data not shown). Western blot analysis of total FAK protein expression levels, using an antibody that recognizes both mouse and chicken FAK, shows that total FAK levels in msFAKKO;chFAKKIWT endothelial cell lysates are slightly less but comparable to the levels expressed in msFAKWT cells (Fig. 3c,d). In addition, immunofluorescence analysis of FAK and Myc was performed in cultured cells to further examine the expression pattern of knockin Myc-tagged FAK in endothelial cells. Results showed that although FAK was detectable in msFAKWT endothelial cells Myc was not detectable (Fig. 4a). In contrast, FAK and Myc colocalized, especially in focal contact sites, in cultured msFAKKO;chFAKKIWT endothelial cells (Fig. 4b). Furthermore, in vivo validation of Myc-tagged chFAK knockin expression was performed using whole heart lysates enriched in endothelial cells (Chen et al., 2012) from PDGFb-iCreert2;FAKfl/fl;R26FAKWT/WT and PDGFbiCreert1;FAKfl/fl;R26FAKWT/WT mice treated with tamoxifen for 22 days. Immunoprecipitation of FAK followed by Western blotting for Myc only showed a signal in the PDGFb-iCreert1;FAKfl/fl;R26FAKWT/WT whole heart samples. In contrast, the total levels of FAK immunoprecipitated were similar in both genotypes as shown by an IP for FAK followed by Western blotting for FAK (Fig. 4c). These data suggest strongly that the Myc-tagged FAK knockin is being expressed in whole tissue. In summary, we have generated inducible pointmutant FAK knockin–knockout mice that permit a temporal control of the knockin/knockout as well as tissuespecific activation by crossing with different Cre lines. This constitutes a powerful tool that will allow more directed studies on the mechanism of FAK regulation in different cell types in vivo.
910
TAVORA ET AL
FIG. 2. Generation of Cre-specific mouse FAK knockout/mutant-chicken FAK-knockin strategy. (a) Schematic of Cre-specific mouse FAK knockout/mutant-chicken FAK knockin generation. Cell type-specific CreERT;FAKfl/fl mice can be crossed with mutant chicken FAKknockin lines (R26FAKKI/KI) to generate mice where tamoxifen induction leads to the simultaneous excision of endogenous mouse FAK (msFAK) and of the STOP codon that precedes both alleles of mutant chFAK-cMyc resulting in the simultaneous deletion of mouse endogenous FAK and induction of homozygous expression of the mutant chFAK-cMyc in selected cell types. (b) Examples of PCR genotyping performed using DNA extracted from ear snips. Pdgfb-iCreert PCR shows mice positive (Pdgfb-iCreert1) or negative (Pdgfb-iCreert2) for CreER in the Pdgfb locus; FAK-floxed PCR shows mice with both FAK-floxed alleles (FAKfl/fl), WT mice non-floxed (FAK1/1), and mice heterozygous for the floxed allele (FAKfl/1); and Rosa 26 PCR shows mice with both (R26FAKKI/KI) or just one (R26FAKKI/1) of the alleles targeted in Rosa 26 locus with chicken FAK or WT mice non-targeted in the Rosa26 locus (R26FAK1/1).
MATERIALS AND METHODS Targeting Vector Construction pWZL Hygro vector containing FAKWT, Y397F, Y397E, KD and Y397E/KD inserts, all of which carrying a myc tag and Rcas vector containing the Y861F insert were subcloned into pBlueScript KS2 (Clontech). FAK constructs, originally in pWZL Hygro, were digested using the BamHI and SalI restriction enzymes, while the Y861F construct was digested using the BamHI and EcoRV restriction enzymes prior to insertion into pBlueScript KS2. These were then transferred from pBluescript KS2 into pBigT vector (gift from Shanhar
Srinivas) which contains a neomycin selection cassette followed by a triple polyadenylation sequence (transcriptional stop) preceded by a neomycin selection cassette and flanked by loxP sites. The pBigT vector was previously modified by the insertion of the IRES-GFP sequence in the XhoI–NotI sites. The pBigT was digested sequentially with SalI and NheI restriction enzymes, while the FAK constructs in pBlueScript KS2 were sequentially digested with Sal I and SpeI. The generated SpeI and NheI compatible restriction sites allowed the directional cloning of the SalI—SpeI FAK mutant fragments upstream of the IRES-GFP. Finally, PacI and AscI were used to subclone a 7.8 kb FAK-IRES
POINT-MUTANT FAK KNOCKIN MICE
911
FIG. 3. Loss of mouse FAK and expression of mutant chicken FAK in endothelial cells isolated from Pdgfb-iCreert;FAKfl/fl;R26FAKWT/WT mice. Endothelial cells were isolated from the lungs of Pdgfb-iCreert1;FAKfl/fl;R26FAKWT/WT and control Pdgfb-iCreert2;FAKfl/fl;R26FAKWT/WT mice, immortalized and treated with 4-hydroxytamoxifen (OHT) in vitro to generate msFAKKO;chFAKKIWT and msFAKWT, respectively. (a) Real-time PCR using mouse-specific or chicken-specific primers. Bar charts show that mRNA levels of endogenous mouse FAK are decreased and mutant chicken FAK expressed in msFAKKO;chFAKKIWT endothelial cells while msFAKWT cells express only mouse FAK. Data represent mean of two experiments performed in duplicate. (b) RIPA lysates from endothelial cells were used for immunoprecipitation of FAK followed by either Myc or FAK blot. Only msFAKKO;chFAKKIWT cells show Myc expression indicating the presence of chicken FAK in these cells, while the amount of immunoprecipitated FAK is similar between msFAKWT and msFAKKO;chFAKKIWT. Lysates from the immunoprecipitation were used for blots using an anti-HSC70 antibody as loading control. (c) Total endothelial cell lysates were used for Western blotting with an anti-FAK and anti-HSC70 antibodies. (d) Mean densitometric analysis from two endothelial cell preps.
GFP construct, including the lox-neomycin-STOP-lox region, into pRosa26-1 (gift from Shanhar Srinivas) which contains the necessary homologous regions for recombination into the Rosa26 genomic locus (1.2 kb
on the 50 and 4.3 kb on the 30 Rosa 26). All cloning steps were verified by DNA sequencing. Please note, that despite the presence of the IRES-GFP insert, analysis has indicated that if it is working, the levels of GFP
912
TAVORA ET AL
FIG. 4. Reporter expression in vitro and in vivo. Endothelial cells were isolated from the lungs of Pdgfb-iCreert1;FAKfl/fl;R26FAKWT/WT and control Pdgfb-iCreert2;FAKfl/fl;R26FAKWT/WT mice, were immortalized, and were treated with 4-hydroxytamoxifen (OHT) in vitro to generate msFAKKO;chFAKKIWT and msFAKWT cells, respectively. (a) Immunofluorescence staining for FAK (green) and Myc (red) shows no Myc expression in msFAKWT cells. (b) In contrast, immunofluorescence staining for FAK (green) and Myc (red) shows colocalization at focal contacts in msFAKKO;chFAKKIWT and provides a demonstration of the knockin expression in cultured cells. Arrows, regions of colocalization of FAK and c-Myc in vitro. (c) RIPA lysates from hearts were used for immunoprecipitation of FAK followed by either Myc or FAK blot. Only Pdgfb-iCreert1;FAKfl/fl;R26FAKWT/WT hearts showed Myc expression indicating the presence of chicken FAK in this tissue, while the amount of immunoprecipitated FAK is similar between Pdgfb-iCreert2;FAKfl/fl;R26FAKWT/WT and Pdgfb-iCreert1;FAKfl/fl;R26FAKWT/WT lysates. Lysates from the immunoprecipitation were used for blots using an anti-HSC70 antibody as loading control. Scale bar 5 25 mm.
are so low that we cannot detect the GFP-signal in tissue or cells. Generation of Point-Mutant FAK Knockin/Knockout Mice SacII linearized vectors were electroporated into mouse ES cells from hybrid 129S6.C57BL6J mice. After
positive selection of neomycin resistant-clones, correctly targeted ES cell clones were screened by Southern Blot analysis using Probe A (369 bp) that lies inside the Rosa26 locus upstream of the targeted region (external 50 probe) and identifies an 11.5 kb band in the wild-type allele and 3.8 kb band in the targeted allele of EcoRV digested samples; and a 5.5 kb band in the wild-
POINT-MUTANT FAK KNOCKIN MICE
type allele and a 8.4 kb band in the targeted allele for AvrII digested samples. Positive ES clones were expanded and DNA extracted and Southern blots performed to confirm homologous recombination using Probe A and also Probe B (762 bp) that lies in the eGFP sequence in the targeting construct, and recognizes a 9 kb band in EcoRV-digested samples and a 4.2 kb band in AvrII-digested DNA. Probe sequences. Probe A—50 R26 external probe is a 369 bp fragment, which was isolated from the pTopoRosa26pr21 (provided by David Stevenson, Beatson Institute, Glasgow, UK), contains an ampicillinresistant cassette, and is generated by EcoRI digestion (highlighted in red below). GAATTCGCCCTTGAGATAGGAACTGGAAAACCAGAG GAGAGGCGTTCAGGAAGATTATGGAGGGGAGGACTGG GCCCCCACGAGCGACCAGAGTTGTCACAAGGCCGCAA GAACAGGGGAGGTGGGGGGCTCAGGGACAGAAAAAA AAGTATGTGTATTTTGAGAGCAGGGTTGGGAGGCCTC TCCTGAAAAGGGTATAAACGTGGAGTAGGCAATACCCA GGCAAAAAGGGGAGACCAGAGTAGGGGGAGGGGAAG AGTCCTGACCCAGGGAAGACATTAAAAAGGTAGTGGG GTCGACTAGATGAAGGAGAGCCTTTCTCTCTGGGCAA GAGCGGTGCAATGGTGTGTAAAGGTAGCTGAGAAGGG CGAATTC. Probe B—GFP internal probe is a 752 bp fragment isolated from the pEGFP-N2 vector, contains a kanamycin-resistant cassette, and is generated by double digestion with BamHI and NotI (highlighted in red below). GGATCCACCGGCCGGTCGCCACCATGGTGAGCAAG GGCGAGGAGCTGTTCACCGGGGTGGTGCCCATCCTG GTCGAGCTGGACGGCGACGTAAACGGCCACAAGTTC AGCGTGTCCGGCGAGGGCGAGGGCGATGCCACCTAC GGCAACTGACCCTGAAGTTCATCTGCACCACCGGCAA GCTGCCCGTGCCCTGGCCCACCCTCGTGACCACCCT GACCTACGGCGTGCAGTGCTTCAGCCGCTACCCCGAC CACATGAAGCAGCACGACTTCTTCAAGTCCGCCATGC CCGAAGGCTACGTCCAGGAGCGCACCATCTTCTTCAA GGACGACGGCAACTACAAGACCCGCGCCGAGGTGAA GTTCGAGGGCGACACCCTGGTGAACCGCATCGAGCT GAAGGGCATCGACTTCAAGGAGGACGGCAACATCCT GGGGCACAAGCTGGAGTACAACTACAACAGCCACAAC GTCTATATCATGGCCGACAAGCAGAAGAACGGCATCA AGGTGAACTTCAAGATCCGCCACAACATCGAGGACGG CAGCGTGCAGCTCGCCGACCACTACCAGCAGAACACC CCCATCGGCGACGGCCCCGTGCTGCTGCCCGACAAC CACTACCTGAGCACCCAGTCCGCCCTGAGCAAAGACC CCAACGAGAAGCGCGATCACATGGTCCTGCTGGAGTT CGTGACCGCCGCCGGGATCACTCTCGGCATGGACGA GCTGTACAAGTAAAGCGGCCGC. Chimeric mice from each mutant line were generated by microinjection of selected positive ES cells into C57BL/6J blastocysts. Following injection, embryos were transferred into plugged pseudopregnant foster mice, which gave birth 18 days later. Germline transmis-
913
sion was confirmed by Southern blot analysis, using probes A and B, of the progeny that resulted from breeding each mutant chimera with C57BL/6J females. Progeny from the same chimera with the correct targeting was chosen as the founder mice for each pointmutant colony. R26FAKKI/1 mice were intercrossed to generate homozygous knockin mice (R26FAKKI/KI). mice with PdgfbIntercrossing R26FAKKI/KI ert fl/fl mice generated Pdgf-iCreert1;FAKfl/fl; iCre 1;FAK R26FAKKI/KI mice to provide a validation system of the inducible knockin/knockout system in a single selected cell type (here, Pdgfb-positive endothelial cells) in adult mice. These new transgenic lines will be available upon acceptance of the manuscript. Genotyping PCR Genotyping for Pdgfb-iCreert;FAKfloxed allele and Rosa26 targeting was performed using DNA isolated from ear snips using the following protocols: Pdgfb-iCreert Primer sequences: (Pdgfb-iCreert F) 50 —GCCGCCGGGATCACTCTC—30 (Pdgfb-iCreert R) 50 —CCAGCCGCCGTCGCAACT—30 Reaction conditions: 94 C for 3 min followed by 34 cycles of 94 C for 30 s, 57.5 C for 45 s, 72 C for 1 min, with a final step at 72 C for 10 min. Expected band size 443 bp. FAK floxed Primer sequences: (LP2 F) 50 —ATTGTGCTATACTCACATTTGGA—30 , (LP2 R2) 50 —TTAATAAGACCAGAGGACTCAGC—30 , (LP2 R1) 50 —GGAAGAAGCTTGTATACTGTATG—30 . Reaction conditions: 94 C for 3 min followed by 35 cycles of 94 C for 1 min, 56 C for 45 s, and 70 C for 1 min, with a final step at 70 C for 7 min. Expected band sizes 429 base pairs for wild-type FAK locus and 697 base pairs band for floxed FAK. Rosa 26/FAK knockin Primer sequences: (Rosa F) 50 —GTTATCAGTAAGGGAGCTGCAGTGG—30 , (Rosa R1) 50 —GGCGGATCACAAGCAATAATAACC–30 , (Rosa R2) 50 —AAGACCGCGAAGAGTTTGTCCTC–30 . Reaction conditions: 94 C for 3 min followed by 35 cycles of 94 C for 30 s, 59.5 C for 30 s and 72 C for 30 s, with a final step at 72 C for 10 min. Expected band sizes 415 base pairs for wild-type and 302 base pairs for targeted Rosa26 locus. Real-Time PCR Immortalized endothelial cells were used to extract RNA using an RNA extraction kit (Qiagen). Reverse transcription was performed using the high-capacity cDNA reverse transcription kit. cDNA was used for the realtime PCR taqman reactions. Custom made primers and probes specific for the mouse and chicken transcripts
914
TAVORA ET AL
were used. GAPDH was included in the same reaction as a loading control. All reagents were from Applied Biosystems. Chicken FAK primers and probe sequences: Forward primer: CAACAGCAAGAGATGGAAGAAG ATC: Probe: ACGATTCCTGGTAATGAA; Reverse primer: 50 -CCGTCCTCCCGTTCAATG-30 . Mouse FAK primers and probe sequences: Forward primer: GGCGTTGCCATCAATACCA Probe: AAGGCATGCGGACACA Reverse primer: GGTGTATGTGTCTTCCTCATCGAT Endothelial Cell Preparation and Tamoxifen Treatment Endothelial cells were prepared, cultured, and tamoxifen treated as described previously (Tavora et al., 2010). Cells were infected with a polyoma middle T retrovirus on days 3 and 4 after isolation, for 4 h each time. Hydroxytamoxifen (500 nM) was added on day 4. The cells were kept under 500 nM hydroxytamoxifen pressure for the duration of the culture. They were tested for deletion of the mouse and the expression of the chicken FAK mRNA, or Myc expression at least 2 weeks after tamoxifen treatment started. Please note that the timing of knockout/knockin induction will likely change according to the Cre line being used and the system being tested. Treatment of Mice with Tamoxifen Mice were administered tamoxifen for 5 consecutive days (3 mg per day). During which time mice were placed onto Clearing diet (16% protein rodent diet, cat. number T.2016.112, Harlan Labs) and thereafter fed with tamoxifen containing diet (base diet standard with Tamoxifen400 added; TD.55125-2016, Harlan Labs) until the experimental endpoint. Please note that the timing of knockout/knockin induction will likely change according to the Cre line being used and the system being tested. Isolation of RIPA Lysates from Hearts Whole mouse hearts were collected 22 days after the first tamoxifen gavage (see above). Mouse hearts were frozen in dry ice and kept in 280 C. They were then crushed using a pestle and mortar in the presence of liquid nitrogen. RIPA lysis buffer (1 ml) was used and the lysates were syringed and span at 10,000g for 10 min. Clear supernatants were removed and were kept frozen. Immunoprecipitation and Western Blotting The immunoprecipitation Dynabeads protein G kit from Invitrogen was used according to the manufacturer’s instructions. A rabbit polyclonal FAK antibody (2 mg; clone C20, cat. number sc558 from Santa Cruz) was
incubated with the beads for 10 min on a rotor at room temperature. The beads–antibody complex was then washed and incubated overnight at 4 C with RIPA lysates from immortalized endothelial cells or mouse hearts. The eluted protein was loaded on 3–8% Tris acetate gels that were ran at 150 V and transferred at 25 V for 1.5 h using nitrocellulose membrane. The remaining supernatant was used in Western blots to confirm equal levels of loading control (HSC70, sc-7298 Santa Cruz). The membrane was blocked with 5% milk TBST solution for 1 h and incubated overnight with a mouse monoclonal anti-Myc antibody (clone 9E10 cat. number ab32 from Abcam) or a mouse monoclonal anti-FAK antibody (cat. number 610087, BD). The membranes were washed and incubated with an HRP-labeled anti-mouse antibody for 1 h at room temperature and the blots were developed using ECL reagent. RIPA cell lysates were used for total FAK Western blots as described above. Immunofluorescence of Cells in Culture Cells were fixed in 4% paraformaldehyde (PFA) for 10 min at room temperature and washed three times in PBS. Cells were permeabilized with 0.5% NP40/PBS for 10 min, washed three times in PBS, and blocked with 0.1% BSA/0.2% Triton X-100 for 10 min at room temperature. Primary antibodies were diluted in PBS and incubated for 1 h at room temperature: mouse anti-FAK antibody (1:100; BD Biosciences 610087) and rabbit anti-Myc tag antibody (1:10,000; Abcam ab 9106). Cells were washed three times and incubated with the corresponding secondary antibodies: anti-mouse-FITC (1:100; Invitrogen F2761) and anti-rabbit-AlexaFluor 555 (1:100; Invitrogen A31572) for 45 min at room temperature. After three washes in PBS, coverslips were mounted in ProLong Gold Antifade reagent with DAPI (Molecular Probes, Invitrogen). All pictures were taken with a confocal microscope (Zeiss LSM 710). ACKNOWLEDGMENTS Authors thank Margaret Frame (University of Edinburgh) and David Stevenson (formally Beatson Institute UK) for their helpful advice in the initial design of the knockin mice; Ian Rosewell (Cancer Research UK, London Research Institute, UK) for technical expertise in the ES cell work and generation of chimeras; Shankar Srinivas (University of Oxford); Garry Saunders, Colin Wren, Colin Pegrum (Cancer Research UK, London Research Institute, UK), Bruce Williams, and Julie Holdworth (Barts Cancer Institute, Queen Mary University of London, UK) for their excellent assistance in the management of the mice. LITERATURE CITED Abouzahr-Rifai S, Hasmim M, Boukerche H, Hamelin J, Janji B, Jalil A, Kieda C, Mami-Chouaib F, Bertoglio J,
POINT-MUTANT FAK KNOCKIN MICE
Chouaib S. 2008. Resistance of tumor cells to cytolytic T lymphocytes involves Rho-GTPases and focal adhesion kinase activation. J Biol Chem 283:31665– 31672. Abu-Ghazaleh R, Kabir J, Jia H, Lobo M, Zachary I. 2001. Src mediates stimulation by vascular endothelial growth factor of the phosphorylation of focal adhesion kinase at tyrosine 861, and migration and antiapoptosis in endothelial cells. Biochem J 360:255– 264. Chen XL, Nam JO, Jean C, Lawson C, Walsh CT, Goka E, Lim ST, Tomar A, Tancioni I, Uryu S, Guan JL, Acevedo LM, Weis SM, Cheresh DA, Schlaepfer DD. 2012. VEGF-induced vascular permeability is mediated by FAK. Dev Cell 22:146–157. Claxton S, Kostourou V, Jadeja S, Chambon P, HodivalaDilke K, Fruttiger M. 2008. Efficient, inducible Crerecombinase activation in vascular endothelium. Genesis 46:74–80. Halder J, Lin YG, Merritt WM, Spannuth WA, Nick AM, Honda T, Kamat AA, Han LY, Kim TJ, Lu C, Tari AM, Bornmann W, Fernandez A, Lopez-Berestein G, Sood AK. 2007. Therapeutic efficacy of a novel focal adhesion kinase inhibitor TAE226 in ovarian carcinoma. Cancer Res 67:10976–10983. Hellstrom M, Kalen M, Lindahl P, Abramsson A, Betsholtz C. 1999. Role of PDGF-B and PDGFR-beta in recruitment of vascular smooth muscle cells and pericytes during embryonic blood vessel formation in the mouse. Development 126:3047–3055. Ilic D, Furuta Y, Kanazawa S, Takeda N, Sobue K, Nakatsuji N, Nomura S, Fujimoto J, Okada M, Yamamoto T. 1995. Reduced cell motility and enhanced focal adhesion contact formation in cells from FAK-deficient mice. Nature 377:539–544. Lechertier T, Hodivala-Dilke K. 2012. Focal adhesion kinase and tumour angiogenesis. J Pathol 226:404–412. Lie PP, Mruk DD, Mok KW, Su L, Lee WM, Cheng CY. 2012. Focal adhesion kinase-Tyr407 and -Tyr397 exhibit antagonistic effects on blood-testis barrier dynamics in the rat. Proc Natl Acad Sci USA 109: 12562–12567. Lim ST, Chen XL, Tomar A, Miller NL, Yoo J, Schlaepfer DD. 2010. Knock-in mutation reveals an essential role for focal adhesion kinase activity in blood ves-
915
sel morphogenesis and cell motility-polarity but not cell proliferation. J Biol Chem 285:21526–21536. Luo M, Guan JL. 2010. Focal adhesion kinase: A prominent determinant in breast cancer initiation, progression and metastasis. Cancer Lett 289:127–139. Mao X, Fujiwara Y, Chapdelaine A, Yang H, Orkin SH. 2001. Activation of EGFP expression by Cremediated excision in a new ROSA26 reporter mouse strain. Blood 97:324–326. Mitra SK, Schlaepfer DD. 2006. Integrin-regulated FAKSrc signaling in normal and cancer cells. Curr Opin Cell Biol 18:516–523. Parsons JT, Slack-Davis J, Tilghman R, Roberts WG. 2008. Focal adhesion kinase: Targeting adhesion signaling pathways for therapeutic intervention. Clin Cancer Res 14:627–632. Schaller MD, Otey CA, Hildebrand JD, Parsons JT. 1995. Focal adhesion kinase and paxillin bind to peptides mimicking beta integrin cytoplasmic domains. J Cell Biol 130:1181–1187. Shen TL, Park AY, Alcaraz A, Peng X, Jang I, Koni P, Flavell RA, Gu H, Guan JL. 2005. Conditional knockout of focal adhesion kinase in endothelial cells reveals its role in angiogenesis and vascular development in late embryogenesis. J Cell Biol 169:941– 952. Shi Y, Pontrello CG, DeFea KA, Reichardt LF, Ethell IM. 2009. Focal adhesion kinase acts downstream of EphB receptors to maintain mature dendritic spines by regulating cofilin activity. J Neurosci 29:8129– 8142. Soriano P. 1999. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat Genet 21:70–71. Tavora B, Batista S, Reynolds LE, Jadeja S, Robinson S, Kostourou V, Hart I, Fruttiger M, Parsons M, HodivalaDilke KM. 2010. Endothelial FAK is required for tumour angiogenesis. EMBO Mol Med 2:516–528. Zhao JH, Reiske H, Guan JL. 1998. Regulation of the cell cycle by focal adhesion kinase. J Cell Biol 143: 1997–2008. Zhao X, Peng X, Sun S, Park AY, Guan JL. 2010. Role of kinase-independent and -dependent functions of FAK in endothelial cell survival and barrier function during embryonic development. J Cell Biol 189: 955–965.
genesis 52:907–915 (2014)
TECHNOLOGY REPORT
Generation of Point-Mutant FAK Knockin Mice B. Tavora,1 S. Batista,1 A.N. Alexopoulou,1 V. Kostourou,2 I. Fernandez,1 S.D. Robinson,3 D.M. Lees,1 B. Serrels,4 and K. Hodivala-Dilke1* 1
Adhesion and Angiogenesis Laboratory, Centre for Tumour Biology, Barts Cancer Institute, -A CR-UK Centre of Excellence, Queen Mary University of London, London, United Kingdom
2
Vascular Adhesion Lab, BSRC Al. Fleming, Athens, Greece
3
School of Biological Sciences, University of East Anglia, Norwich, United Kingdom
4
Edinburgh Cancer Research UK Centre, Institute of Genetics and Molecular Medicine, University of Edinburgh, Western General Hospital, Edinburgh, United Kingdom
Received 17 December 2013; Revised 18 September 2014; Accepted 18 September 2014
Summary: Focal adhesion kinase is a non-receptor protein tyrosine kinase with signaling functions downstream of integrins and growth factor receptors. In addition to its role in adhesion, migration, and proliferation it also has non-kinase scaffolding functions in the nucleus. Focal adhesion kinase (FAK) activation involves the following: (1) ligand bound growth factors or clustered integrins activate FAK kinase domain; (2) FAK autophosphorylates tyrosine (Y) 397; (3) Src binds pY397 and phosphorylates FAK at various other sites including Y861; (4) downstream signaling of activated FAK elicits changes in cellular behavior. Although many studies have demonstrated roles for the kinase domain, Y397 and Y861 sites, in vitro much less is known about their functions in vivo. Here, we report the generation of a series of FAK-mutant knockin mice where mutant FAK, either kinase dead, non-phosphorylatable mutants Y397F and Y861F, or mutant Y397E—containing a phosphomimetic site that results in a constitutive active Y397, can be expressed in a Cre inducible fashion driven by the ROSA26 promoter. In future studies, intercrossing these mice with FAKflox/flox mice and inducible cre-expressing mice will enable the in vivo study of mutant FAK function in the absence of endogenous FAK in a spatially and temporally regulated fashion within the whole organism. genesis 52:907–915, C 2014 Wiley Periodicals, Inc. 2014. V Key words: focal adhesion kinase; mutants; mice
Focal adhesion kinase (FAK) is one of the central players in integrin and growth factor receptor mediated signaling and has also been reported to have scaffolding
functions. FAK is a 125-kDa protein with a central tyrosine kinase domain flanked by FERM and FAT domains. FAK is maintained in an inactive conformation in which the FERM domain is bound to the kinase domain. This impairs access to the tyrosine 397 residue (Y397) of FAK, thus leading to auto-inhibition. Integrin or growth factor receptor activation initiates a signal that releases the auto-inhibitory interaction between the FERM and kinase domains, leading to kinase activation and subsequent autophosphorylation of Y397 (Lechertier and Hodivala-Dilke, 2012; Luo and Guan, 2010; Mitra and Schlaepfer, 2006; Parsons et al., 2008; Schaller et al., 1995). Phosphorylation of Y397 provides an Sh2binding site allowing Src to interact with FAK. This interaction stimulates Src to transphosphorylate FAK on a number of key tyrosine residues, including FAK– Y861. Phosphorylation at FAK–Y861 allows p130Cas binding, activating the PI3-kinase signaling pathway. Despite this apparently clear pathway of events, some Additional Supporting Information may be found in the online version of this article. * Correspondence to: K. Hodivala-Dilke, Adhesion and Angiogenesis Laboratory, Centre for Tumour Biology, Barts Cancer Institute, -A CR-UK Centre of Excellence, Queen Mary University of London, Charterhouse Square, London, UK. E-mail: [email protected] B.T., S.B., and A.N.A. contributed equally to this work. Contract grant sponsor: Cancer Research UK, Contract grant number: C8218/A12007; Contract grant sponsor: Breast Cancer Campaign, Contract grant number: 2010MayPR39; Contract grant sponsor: The Association of International Cancer Research, Contract grant number: 12–1068 Published online 20 September 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/dvg.22823
908
TAVORA ET AL
results have demonstrated unexpected pathways for FAK function. For example, growth factor stimulation can induce a Src-dependent increase in pY861 levels but not pY397 levels (Abu-Ghazaleh et al., 2001); FAK knockdown or deletion can increase invadopodia formation, while still impairing invasive cell migration (Ilic et al., 1995; Shen et al., 2005); and re-expression of the kinase dead mutant of FAK (K454R) can restore normal invadopodia formation while mutations in FAK–Y397 do not. Thus, understanding how these different residues of FAK control cell behavior both in vitro and in vivo become important. FAK can be transphosphorylated and thus overexpression of mutant FAK in cells or mice where endogenous FAK is still present makes it impossible to demonstrate the effect of the FAK mutation precisely. To overcome this issue, we have developed a series of Cre inducible mutant FAK-knockin mice which when crossed with FAK-floxed mice will generate mice where endogenous FAK is knocked-out and mutant FAK simultaneously knocked in driven by the ROSA 26 promoter. Other groups have produced kinase defective FAK mutant knockin mice by making point mutations in the FAK gene (Lim et al., 2010; Zhao et al., 2010). These mutations might affect not only downstream signaling pathways but also the regulation of FAK gene transcription (i.e., controlled by the endogenous functional promoter). In contrast, our knockin/knockout system is able to dissect the effect of FAK mutations in signaling without influencing FAK transcription as endogenous FAK is genetically ablated. In addition, our system has the advantage over some of being used with inducible Cre lines. RESULTS Using chicken FAK cDNA constructs we have generated knockin mutants of FAK where: the Y397 site cannot be phosphorylated (Y397F) or mimics a constitutively phosphorylated form (Y397E); the Y861 site cannot be phosphorylated (Y861F); or the kinase domain is disabled whereby the ATP binding site is mutated (lysine (AAA) in the position 454 to arginine (AGA). In addition, double mutants of FAK Y397E/kinase dead (KD), which generates a constitutively active Y397 site with an inactive kinase domain, were generated to assess kinase function independently of Y397 phosphorylation and possibly Src binding. The wild type (WT) chicken FAK construct was used as a negative control (Supporting Information Fig. S1a). Many of these mutants have been used in in vitro studies and their functions validated (Abouzahr-Rifai et al., 2008; Lie et al., 2012; Shi et al., 2009; Zhao et al., 1998). Mutant chicken-FAK constructs (preceded by a STOP sequence flanked by loxP sites) were targeted to the ubiquitous Rosa26 locus to generate a Cre-inducible sys-
tem for mutant FAK expression (Supporting Information Fig. S1b). Targeting the Rosa26 locus, with similar constructs, has been used to induce the expression of various cDNAs in various cell types after Cre-mediated recombination (Halder et al., 2007; Mao et al., 2001; Soriano, 1999). For this purpose, the following cloning steps were required: (1) FAK constructs were cloned into pBluescript KS to allow for subsequent cloning steps; (2) these vectors were inserted into pBigT downstream of a STOP sequence flanked by loxP sites; and finally cloned into the pRosa26-1 vector to allow homologous recombination into the Rosa26 locus (Supporting Information Fig. S1b). FAK WT, Y397F, Y397E, KD, Y397E/KD, and Y861F constructs were electroporated into hybrid 129S6.C57BL6J ES cells and neomycin-resistant clones were selected for DNA extraction. Southern blot screening of ES cell clones was performed using EcoRV and AvrII digested DNA with the 50 R26 probe (Probe A), an external probe to the targeting vector that lies 50 of the Rosa26 locus (Soriano, 1999). EcoRV digested DNA probed with Probe A revealed both a WT (11 kb) and a targeted band (4 kb—indicative of homologous recombination in the Rosa 26 locus) whereas AvrII digested DNA probed with probe A shows a WT (5.5 kb) and a targeted band (8.4 kb). These ES cell clones were expanded for validation screens using Probe B, a probe internal to the targeting vector. EcoRV digested DNA screened using internal Probe B revealed a 9 kb band and AvrII digested DNA showed a 4.2 kb band and also confirmed homologous recombination (Supporting Information Fig. S2a,b). Confirmed positive ES clones, containing more than 70% of cells with normal chromosome counts (data not shown), were injected into blastocysts. Following injection, embryos were transferred into pseudopregnant foster mice, which gave birth 18 days later. High percentage chimeras (95–100%) were bred to C57/BL6 mice to generate offspring. Tail snip DNA was genotyped by Southern blot analysis for homologous recombination using Probes A and B. Mice that displayed both WT and targeted band with Probe A and targeted band with internal Probe B were chosen as founders for each FAK-mutant knockin (R26FAKKI) colony (Supporting Information Fig. S2c). Founder R26FAKKI/1 were bred together to generate homozygous R26FAKKI/KI mice that only reveal a 4 kb targeted band by Southern Blot in EcoRV digested DNA from tail snips of these animals (Fig. 1a). Sequencing mutation sites in DNA extracted from each homozygous FAKknockin mutant mouse lines confirmed the persistence of mutations in vivo (Fig. 1b). R26FAKKI/KI mice were then bred with cell-type specific Creert;FAKfl/fl mice (Tavora et al., 2010) to generate mice with tamoxifen-inducible mutant FAK-knockin and endogenous FAK-knockout in selected cell types (Fig. 2a). As an example, tamoxifen-inducible endothelial
POINT-MUTANT FAK KNOCKIN MICE 1/1
FIG. 1. Generation of FAK-mutant knockin mice. (a) Southern blot analysis of EcoRV-digested DNA extracted from tail snips of WT/WT, 397F/397F, 397E/397E, KD/KD, 397E/KD/397E/KD, and 861F/861F homozygous mice. Homologous recombination was confirmed in all mutations with Probe A. (b) Examples of sequencing of mutation sites in DNA extracted from tails of the same progeny. Tyrosine 397 is replaced by a phenylalanine (TTT) or glutamic acid (GAG) generating Y397F and Y397E, respectively. Lysine 454 is replaced by an arginine (AGA) generating a kinase dead (KD K454R). Tyrosine 861 is replaced by a phenylalanine (TTT) generating Y861F.
specific mutant-FAK-knockin and endogenous FAKknockout mice have been generated (Pdgfb-iCreert;FAKfl/fl;R26FAKKI/KI). Pdgfb (platelet-derived growth factor b) is produced predominantly by endothelial cells (Hellstrom et al., 1999). The efficiency of tamoxifeninduced Cre recombinase activity under the control of Pdgfb promoter (Pdgfb-iCreert mice) has been tested using Rosa26-lacZ reporter crosses and cre activity was shown to be endothelial specific (Claxton et al., 2008). Mice were genotyped using PCR and examples of these are given in Figure 2b. Pdgfb-iCreert PCR shows mice positive (Pdgfb-iCreert1) or negative (Pdgfb-iCreert2) for Creert. FAK-floxed PCR shows mice with both (FAKfl/ fl ) or one (FAKfl/1) FAK floxed alleles and WT mice non-
909
floxed (FAK ). Rosa 26 PCR shows mice homozygous (R26FAKKI/KI) and heterozygous (R26FAKKI/1) for Rosa 26 targeting and mice non-targeted in the Rosa 26 locus (R26FAK1/1) (Fig. 2b). We have isolated, cultured, and immortalized endothelial cells from Pdgfb-iCreert1;FAKfl/fl;R26FAKWT/WT and Pdgfb-iCreert2;FAKfl/fl;R26FAKWT/WT mice that were grown for a minimum of 2 weeks in the presence of 4-hydroxytamoxifen (OHT) to induce the recombination of the loxP sites in vitro to generate endothelial cells with the following genotypes, respectively, msFAKKO;chFAKKIWT and msFAKWT. Real-time PCR shows that OHT-treatment induces the deletion of endogenous mouse FAK and the simultaneous expression of the chicken FAK at the mRNA level (Fig. 3a). Immunoprecipitation with an antibody that recognizes both mouse and chicken FAK followed by blotting with an antibody that recognizes the Myc-tag of the chicken protein shows that chicken FAK is only expressed in the msFAKKO;chFAKKIWT cells (Fig. 3b). Transient Cre transfection in vitro may increase chFAK levels (data not shown). Western blot analysis of total FAK protein expression levels, using an antibody that recognizes both mouse and chicken FAK, shows that total FAK levels in msFAKKO;chFAKKIWT endothelial cell lysates are slightly less but comparable to the levels expressed in msFAKWT cells (Fig. 3c,d). In addition, immunofluorescence analysis of FAK and Myc was performed in cultured cells to further examine the expression pattern of knockin Myc-tagged FAK in endothelial cells. Results showed that although FAK was detectable in msFAKWT endothelial cells Myc was not detectable (Fig. 4a). In contrast, FAK and Myc colocalized, especially in focal contact sites, in cultured msFAKKO;chFAKKIWT endothelial cells (Fig. 4b). Furthermore, in vivo validation of Myc-tagged chFAK knockin expression was performed using whole heart lysates enriched in endothelial cells (Chen et al., 2012) from PDGFb-iCreert2;FAKfl/fl;R26FAKWT/WT and PDGFbiCreert1;FAKfl/fl;R26FAKWT/WT mice treated with tamoxifen for 22 days. Immunoprecipitation of FAK followed by Western blotting for Myc only showed a signal in the PDGFb-iCreert1;FAKfl/fl;R26FAKWT/WT whole heart samples. In contrast, the total levels of FAK immunoprecipitated were similar in both genotypes as shown by an IP for FAK followed by Western blotting for FAK (Fig. 4c). These data suggest strongly that the Myc-tagged FAK knockin is being expressed in whole tissue. In summary, we have generated inducible pointmutant FAK knockin–knockout mice that permit a temporal control of the knockin/knockout as well as tissuespecific activation by crossing with different Cre lines. This constitutes a powerful tool that will allow more directed studies on the mechanism of FAK regulation in different cell types in vivo.
910
TAVORA ET AL
FIG. 2. Generation of Cre-specific mouse FAK knockout/mutant-chicken FAK-knockin strategy. (a) Schematic of Cre-specific mouse FAK knockout/mutant-chicken FAK knockin generation. Cell type-specific CreERT;FAKfl/fl mice can be crossed with mutant chicken FAKknockin lines (R26FAKKI/KI) to generate mice where tamoxifen induction leads to the simultaneous excision of endogenous mouse FAK (msFAK) and of the STOP codon that precedes both alleles of mutant chFAK-cMyc resulting in the simultaneous deletion of mouse endogenous FAK and induction of homozygous expression of the mutant chFAK-cMyc in selected cell types. (b) Examples of PCR genotyping performed using DNA extracted from ear snips. Pdgfb-iCreert PCR shows mice positive (Pdgfb-iCreert1) or negative (Pdgfb-iCreert2) for CreER in the Pdgfb locus; FAK-floxed PCR shows mice with both FAK-floxed alleles (FAKfl/fl), WT mice non-floxed (FAK1/1), and mice heterozygous for the floxed allele (FAKfl/1); and Rosa 26 PCR shows mice with both (R26FAKKI/KI) or just one (R26FAKKI/1) of the alleles targeted in Rosa 26 locus with chicken FAK or WT mice non-targeted in the Rosa26 locus (R26FAK1/1).
MATERIALS AND METHODS Targeting Vector Construction pWZL Hygro vector containing FAKWT, Y397F, Y397E, KD and Y397E/KD inserts, all of which carrying a myc tag and Rcas vector containing the Y861F insert were subcloned into pBlueScript KS2 (Clontech). FAK constructs, originally in pWZL Hygro, were digested using the BamHI and SalI restriction enzymes, while the Y861F construct was digested using the BamHI and EcoRV restriction enzymes prior to insertion into pBlueScript KS2. These were then transferred from pBluescript KS2 into pBigT vector (gift from Shanhar
Srinivas) which contains a neomycin selection cassette followed by a triple polyadenylation sequence (transcriptional stop) preceded by a neomycin selection cassette and flanked by loxP sites. The pBigT vector was previously modified by the insertion of the IRES-GFP sequence in the XhoI–NotI sites. The pBigT was digested sequentially with SalI and NheI restriction enzymes, while the FAK constructs in pBlueScript KS2 were sequentially digested with Sal I and SpeI. The generated SpeI and NheI compatible restriction sites allowed the directional cloning of the SalI—SpeI FAK mutant fragments upstream of the IRES-GFP. Finally, PacI and AscI were used to subclone a 7.8 kb FAK-IRES
POINT-MUTANT FAK KNOCKIN MICE
911
FIG. 3. Loss of mouse FAK and expression of mutant chicken FAK in endothelial cells isolated from Pdgfb-iCreert;FAKfl/fl;R26FAKWT/WT mice. Endothelial cells were isolated from the lungs of Pdgfb-iCreert1;FAKfl/fl;R26FAKWT/WT and control Pdgfb-iCreert2;FAKfl/fl;R26FAKWT/WT mice, immortalized and treated with 4-hydroxytamoxifen (OHT) in vitro to generate msFAKKO;chFAKKIWT and msFAKWT, respectively. (a) Real-time PCR using mouse-specific or chicken-specific primers. Bar charts show that mRNA levels of endogenous mouse FAK are decreased and mutant chicken FAK expressed in msFAKKO;chFAKKIWT endothelial cells while msFAKWT cells express only mouse FAK. Data represent mean of two experiments performed in duplicate. (b) RIPA lysates from endothelial cells were used for immunoprecipitation of FAK followed by either Myc or FAK blot. Only msFAKKO;chFAKKIWT cells show Myc expression indicating the presence of chicken FAK in these cells, while the amount of immunoprecipitated FAK is similar between msFAKWT and msFAKKO;chFAKKIWT. Lysates from the immunoprecipitation were used for blots using an anti-HSC70 antibody as loading control. (c) Total endothelial cell lysates were used for Western blotting with an anti-FAK and anti-HSC70 antibodies. (d) Mean densitometric analysis from two endothelial cell preps.
GFP construct, including the lox-neomycin-STOP-lox region, into pRosa26-1 (gift from Shanhar Srinivas) which contains the necessary homologous regions for recombination into the Rosa26 genomic locus (1.2 kb
on the 50 and 4.3 kb on the 30 Rosa 26). All cloning steps were verified by DNA sequencing. Please note, that despite the presence of the IRES-GFP insert, analysis has indicated that if it is working, the levels of GFP
912
TAVORA ET AL
FIG. 4. Reporter expression in vitro and in vivo. Endothelial cells were isolated from the lungs of Pdgfb-iCreert1;FAKfl/fl;R26FAKWT/WT and control Pdgfb-iCreert2;FAKfl/fl;R26FAKWT/WT mice, were immortalized, and were treated with 4-hydroxytamoxifen (OHT) in vitro to generate msFAKKO;chFAKKIWT and msFAKWT cells, respectively. (a) Immunofluorescence staining for FAK (green) and Myc (red) shows no Myc expression in msFAKWT cells. (b) In contrast, immunofluorescence staining for FAK (green) and Myc (red) shows colocalization at focal contacts in msFAKKO;chFAKKIWT and provides a demonstration of the knockin expression in cultured cells. Arrows, regions of colocalization of FAK and c-Myc in vitro. (c) RIPA lysates from hearts were used for immunoprecipitation of FAK followed by either Myc or FAK blot. Only Pdgfb-iCreert1;FAKfl/fl;R26FAKWT/WT hearts showed Myc expression indicating the presence of chicken FAK in this tissue, while the amount of immunoprecipitated FAK is similar between Pdgfb-iCreert2;FAKfl/fl;R26FAKWT/WT and Pdgfb-iCreert1;FAKfl/fl;R26FAKWT/WT lysates. Lysates from the immunoprecipitation were used for blots using an anti-HSC70 antibody as loading control. Scale bar 5 25 mm.
are so low that we cannot detect the GFP-signal in tissue or cells. Generation of Point-Mutant FAK Knockin/Knockout Mice SacII linearized vectors were electroporated into mouse ES cells from hybrid 129S6.C57BL6J mice. After
positive selection of neomycin resistant-clones, correctly targeted ES cell clones were screened by Southern Blot analysis using Probe A (369 bp) that lies inside the Rosa26 locus upstream of the targeted region (external 50 probe) and identifies an 11.5 kb band in the wild-type allele and 3.8 kb band in the targeted allele of EcoRV digested samples; and a 5.5 kb band in the wild-
POINT-MUTANT FAK KNOCKIN MICE
type allele and a 8.4 kb band in the targeted allele for AvrII digested samples. Positive ES clones were expanded and DNA extracted and Southern blots performed to confirm homologous recombination using Probe A and also Probe B (762 bp) that lies in the eGFP sequence in the targeting construct, and recognizes a 9 kb band in EcoRV-digested samples and a 4.2 kb band in AvrII-digested DNA. Probe sequences. Probe A—50 R26 external probe is a 369 bp fragment, which was isolated from the pTopoRosa26pr21 (provided by David Stevenson, Beatson Institute, Glasgow, UK), contains an ampicillinresistant cassette, and is generated by EcoRI digestion (highlighted in red below). GAATTCGCCCTTGAGATAGGAACTGGAAAACCAGAG GAGAGGCGTTCAGGAAGATTATGGAGGGGAGGACTGG GCCCCCACGAGCGACCAGAGTTGTCACAAGGCCGCAA GAACAGGGGAGGTGGGGGGCTCAGGGACAGAAAAAA AAGTATGTGTATTTTGAGAGCAGGGTTGGGAGGCCTC TCCTGAAAAGGGTATAAACGTGGAGTAGGCAATACCCA GGCAAAAAGGGGAGACCAGAGTAGGGGGAGGGGAAG AGTCCTGACCCAGGGAAGACATTAAAAAGGTAGTGGG GTCGACTAGATGAAGGAGAGCCTTTCTCTCTGGGCAA GAGCGGTGCAATGGTGTGTAAAGGTAGCTGAGAAGGG CGAATTC. Probe B—GFP internal probe is a 752 bp fragment isolated from the pEGFP-N2 vector, contains a kanamycin-resistant cassette, and is generated by double digestion with BamHI and NotI (highlighted in red below). GGATCCACCGGCCGGTCGCCACCATGGTGAGCAAG GGCGAGGAGCTGTTCACCGGGGTGGTGCCCATCCTG GTCGAGCTGGACGGCGACGTAAACGGCCACAAGTTC AGCGTGTCCGGCGAGGGCGAGGGCGATGCCACCTAC GGCAACTGACCCTGAAGTTCATCTGCACCACCGGCAA GCTGCCCGTGCCCTGGCCCACCCTCGTGACCACCCT GACCTACGGCGTGCAGTGCTTCAGCCGCTACCCCGAC CACATGAAGCAGCACGACTTCTTCAAGTCCGCCATGC CCGAAGGCTACGTCCAGGAGCGCACCATCTTCTTCAA GGACGACGGCAACTACAAGACCCGCGCCGAGGTGAA GTTCGAGGGCGACACCCTGGTGAACCGCATCGAGCT GAAGGGCATCGACTTCAAGGAGGACGGCAACATCCT GGGGCACAAGCTGGAGTACAACTACAACAGCCACAAC GTCTATATCATGGCCGACAAGCAGAAGAACGGCATCA AGGTGAACTTCAAGATCCGCCACAACATCGAGGACGG CAGCGTGCAGCTCGCCGACCACTACCAGCAGAACACC CCCATCGGCGACGGCCCCGTGCTGCTGCCCGACAAC CACTACCTGAGCACCCAGTCCGCCCTGAGCAAAGACC CCAACGAGAAGCGCGATCACATGGTCCTGCTGGAGTT CGTGACCGCCGCCGGGATCACTCTCGGCATGGACGA GCTGTACAAGTAAAGCGGCCGC. Chimeric mice from each mutant line were generated by microinjection of selected positive ES cells into C57BL/6J blastocysts. Following injection, embryos were transferred into plugged pseudopregnant foster mice, which gave birth 18 days later. Germline transmis-
913
sion was confirmed by Southern blot analysis, using probes A and B, of the progeny that resulted from breeding each mutant chimera with C57BL/6J females. Progeny from the same chimera with the correct targeting was chosen as the founder mice for each pointmutant colony. R26FAKKI/1 mice were intercrossed to generate homozygous knockin mice (R26FAKKI/KI). mice with PdgfbIntercrossing R26FAKKI/KI ert fl/fl mice generated Pdgf-iCreert1;FAKfl/fl; iCre 1;FAK R26FAKKI/KI mice to provide a validation system of the inducible knockin/knockout system in a single selected cell type (here, Pdgfb-positive endothelial cells) in adult mice. These new transgenic lines will be available upon acceptance of the manuscript. Genotyping PCR Genotyping for Pdgfb-iCreert;FAKfloxed allele and Rosa26 targeting was performed using DNA isolated from ear snips using the following protocols: Pdgfb-iCreert Primer sequences: (Pdgfb-iCreert F) 50 —GCCGCCGGGATCACTCTC—30 (Pdgfb-iCreert R) 50 —CCAGCCGCCGTCGCAACT—30 Reaction conditions: 94 C for 3 min followed by 34 cycles of 94 C for 30 s, 57.5 C for 45 s, 72 C for 1 min, with a final step at 72 C for 10 min. Expected band size 443 bp. FAK floxed Primer sequences: (LP2 F) 50 —ATTGTGCTATACTCACATTTGGA—30 , (LP2 R2) 50 —TTAATAAGACCAGAGGACTCAGC—30 , (LP2 R1) 50 —GGAAGAAGCTTGTATACTGTATG—30 . Reaction conditions: 94 C for 3 min followed by 35 cycles of 94 C for 1 min, 56 C for 45 s, and 70 C for 1 min, with a final step at 70 C for 7 min. Expected band sizes 429 base pairs for wild-type FAK locus and 697 base pairs band for floxed FAK. Rosa 26/FAK knockin Primer sequences: (Rosa F) 50 —GTTATCAGTAAGGGAGCTGCAGTGG—30 , (Rosa R1) 50 —GGCGGATCACAAGCAATAATAACC–30 , (Rosa R2) 50 —AAGACCGCGAAGAGTTTGTCCTC–30 . Reaction conditions: 94 C for 3 min followed by 35 cycles of 94 C for 30 s, 59.5 C for 30 s and 72 C for 30 s, with a final step at 72 C for 10 min. Expected band sizes 415 base pairs for wild-type and 302 base pairs for targeted Rosa26 locus. Real-Time PCR Immortalized endothelial cells were used to extract RNA using an RNA extraction kit (Qiagen). Reverse transcription was performed using the high-capacity cDNA reverse transcription kit. cDNA was used for the realtime PCR taqman reactions. Custom made primers and probes specific for the mouse and chicken transcripts
914
TAVORA ET AL
were used. GAPDH was included in the same reaction as a loading control. All reagents were from Applied Biosystems. Chicken FAK primers and probe sequences: Forward primer: CAACAGCAAGAGATGGAAGAAG ATC: Probe: ACGATTCCTGGTAATGAA; Reverse primer: 50 -CCGTCCTCCCGTTCAATG-30 . Mouse FAK primers and probe sequences: Forward primer: GGCGTTGCCATCAATACCA Probe: AAGGCATGCGGACACA Reverse primer: GGTGTATGTGTCTTCCTCATCGAT Endothelial Cell Preparation and Tamoxifen Treatment Endothelial cells were prepared, cultured, and tamoxifen treated as described previously (Tavora et al., 2010). Cells were infected with a polyoma middle T retrovirus on days 3 and 4 after isolation, for 4 h each time. Hydroxytamoxifen (500 nM) was added on day 4. The cells were kept under 500 nM hydroxytamoxifen pressure for the duration of the culture. They were tested for deletion of the mouse and the expression of the chicken FAK mRNA, or Myc expression at least 2 weeks after tamoxifen treatment started. Please note that the timing of knockout/knockin induction will likely change according to the Cre line being used and the system being tested. Treatment of Mice with Tamoxifen Mice were administered tamoxifen for 5 consecutive days (3 mg per day). During which time mice were placed onto Clearing diet (16% protein rodent diet, cat. number T.2016.112, Harlan Labs) and thereafter fed with tamoxifen containing diet (base diet standard with Tamoxifen400 added; TD.55125-2016, Harlan Labs) until the experimental endpoint. Please note that the timing of knockout/knockin induction will likely change according to the Cre line being used and the system being tested. Isolation of RIPA Lysates from Hearts Whole mouse hearts were collected 22 days after the first tamoxifen gavage (see above). Mouse hearts were frozen in dry ice and kept in 280 C. They were then crushed using a pestle and mortar in the presence of liquid nitrogen. RIPA lysis buffer (1 ml) was used and the lysates were syringed and span at 10,000g for 10 min. Clear supernatants were removed and were kept frozen. Immunoprecipitation and Western Blotting The immunoprecipitation Dynabeads protein G kit from Invitrogen was used according to the manufacturer’s instructions. A rabbit polyclonal FAK antibody (2 mg; clone C20, cat. number sc558 from Santa Cruz) was
incubated with the beads for 10 min on a rotor at room temperature. The beads–antibody complex was then washed and incubated overnight at 4 C with RIPA lysates from immortalized endothelial cells or mouse hearts. The eluted protein was loaded on 3–8% Tris acetate gels that were ran at 150 V and transferred at 25 V for 1.5 h using nitrocellulose membrane. The remaining supernatant was used in Western blots to confirm equal levels of loading control (HSC70, sc-7298 Santa Cruz). The membrane was blocked with 5% milk TBST solution for 1 h and incubated overnight with a mouse monoclonal anti-Myc antibody (clone 9E10 cat. number ab32 from Abcam) or a mouse monoclonal anti-FAK antibody (cat. number 610087, BD). The membranes were washed and incubated with an HRP-labeled anti-mouse antibody for 1 h at room temperature and the blots were developed using ECL reagent. RIPA cell lysates were used for total FAK Western blots as described above. Immunofluorescence of Cells in Culture Cells were fixed in 4% paraformaldehyde (PFA) for 10 min at room temperature and washed three times in PBS. Cells were permeabilized with 0.5% NP40/PBS for 10 min, washed three times in PBS, and blocked with 0.1% BSA/0.2% Triton X-100 for 10 min at room temperature. Primary antibodies were diluted in PBS and incubated for 1 h at room temperature: mouse anti-FAK antibody (1:100; BD Biosciences 610087) and rabbit anti-Myc tag antibody (1:10,000; Abcam ab 9106). Cells were washed three times and incubated with the corresponding secondary antibodies: anti-mouse-FITC (1:100; Invitrogen F2761) and anti-rabbit-AlexaFluor 555 (1:100; Invitrogen A31572) for 45 min at room temperature. After three washes in PBS, coverslips were mounted in ProLong Gold Antifade reagent with DAPI (Molecular Probes, Invitrogen). All pictures were taken with a confocal microscope (Zeiss LSM 710). ACKNOWLEDGMENTS Authors thank Margaret Frame (University of Edinburgh) and David Stevenson (formally Beatson Institute UK) for their helpful advice in the initial design of the knockin mice; Ian Rosewell (Cancer Research UK, London Research Institute, UK) for technical expertise in the ES cell work and generation of chimeras; Shankar Srinivas (University of Oxford); Garry Saunders, Colin Wren, Colin Pegrum (Cancer Research UK, London Research Institute, UK), Bruce Williams, and Julie Holdworth (Barts Cancer Institute, Queen Mary University of London, UK) for their excellent assistance in the management of the mice. LITERATURE CITED Abouzahr-Rifai S, Hasmim M, Boukerche H, Hamelin J, Janji B, Jalil A, Kieda C, Mami-Chouaib F, Bertoglio J,
POINT-MUTANT FAK KNOCKIN MICE
Chouaib S. 2008. Resistance of tumor cells to cytolytic T lymphocytes involves Rho-GTPases and focal adhesion kinase activation. J Biol Chem 283:31665– 31672. Abu-Ghazaleh R, Kabir J, Jia H, Lobo M, Zachary I. 2001. Src mediates stimulation by vascular endothelial growth factor of the phosphorylation of focal adhesion kinase at tyrosine 861, and migration and antiapoptosis in endothelial cells. Biochem J 360:255– 264. Chen XL, Nam JO, Jean C, Lawson C, Walsh CT, Goka E, Lim ST, Tomar A, Tancioni I, Uryu S, Guan JL, Acevedo LM, Weis SM, Cheresh DA, Schlaepfer DD. 2012. VEGF-induced vascular permeability is mediated by FAK. Dev Cell 22:146–157. Claxton S, Kostourou V, Jadeja S, Chambon P, HodivalaDilke K, Fruttiger M. 2008. Efficient, inducible Crerecombinase activation in vascular endothelium. Genesis 46:74–80. Halder J, Lin YG, Merritt WM, Spannuth WA, Nick AM, Honda T, Kamat AA, Han LY, Kim TJ, Lu C, Tari AM, Bornmann W, Fernandez A, Lopez-Berestein G, Sood AK. 2007. Therapeutic efficacy of a novel focal adhesion kinase inhibitor TAE226 in ovarian carcinoma. Cancer Res 67:10976–10983. Hellstrom M, Kalen M, Lindahl P, Abramsson A, Betsholtz C. 1999. Role of PDGF-B and PDGFR-beta in recruitment of vascular smooth muscle cells and pericytes during embryonic blood vessel formation in the mouse. Development 126:3047–3055. Ilic D, Furuta Y, Kanazawa S, Takeda N, Sobue K, Nakatsuji N, Nomura S, Fujimoto J, Okada M, Yamamoto T. 1995. Reduced cell motility and enhanced focal adhesion contact formation in cells from FAK-deficient mice. Nature 377:539–544. Lechertier T, Hodivala-Dilke K. 2012. Focal adhesion kinase and tumour angiogenesis. J Pathol 226:404–412. Lie PP, Mruk DD, Mok KW, Su L, Lee WM, Cheng CY. 2012. Focal adhesion kinase-Tyr407 and -Tyr397 exhibit antagonistic effects on blood-testis barrier dynamics in the rat. Proc Natl Acad Sci USA 109: 12562–12567. Lim ST, Chen XL, Tomar A, Miller NL, Yoo J, Schlaepfer DD. 2010. Knock-in mutation reveals an essential role for focal adhesion kinase activity in blood ves-
915
sel morphogenesis and cell motility-polarity but not cell proliferation. J Biol Chem 285:21526–21536. Luo M, Guan JL. 2010. Focal adhesion kinase: A prominent determinant in breast cancer initiation, progression and metastasis. Cancer Lett 289:127–139. Mao X, Fujiwara Y, Chapdelaine A, Yang H, Orkin SH. 2001. Activation of EGFP expression by Cremediated excision in a new ROSA26 reporter mouse strain. Blood 97:324–326. Mitra SK, Schlaepfer DD. 2006. Integrin-regulated FAKSrc signaling in normal and cancer cells. Curr Opin Cell Biol 18:516–523. Parsons JT, Slack-Davis J, Tilghman R, Roberts WG. 2008. Focal adhesion kinase: Targeting adhesion signaling pathways for therapeutic intervention. Clin Cancer Res 14:627–632. Schaller MD, Otey CA, Hildebrand JD, Parsons JT. 1995. Focal adhesion kinase and paxillin bind to peptides mimicking beta integrin cytoplasmic domains. J Cell Biol 130:1181–1187. Shen TL, Park AY, Alcaraz A, Peng X, Jang I, Koni P, Flavell RA, Gu H, Guan JL. 2005. Conditional knockout of focal adhesion kinase in endothelial cells reveals its role in angiogenesis and vascular development in late embryogenesis. J Cell Biol 169:941– 952. Shi Y, Pontrello CG, DeFea KA, Reichardt LF, Ethell IM. 2009. Focal adhesion kinase acts downstream of EphB receptors to maintain mature dendritic spines by regulating cofilin activity. J Neurosci 29:8129– 8142. Soriano P. 1999. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat Genet 21:70–71. Tavora B, Batista S, Reynolds LE, Jadeja S, Robinson S, Kostourou V, Hart I, Fruttiger M, Parsons M, HodivalaDilke KM. 2010. Endothelial FAK is required for tumour angiogenesis. EMBO Mol Med 2:516–528. Zhao JH, Reiske H, Guan JL. 1998. Regulation of the cell cycle by focal adhesion kinase. J Cell Biol 143: 1997–2008. Zhao X, Peng X, Sun S, Park AY, Guan JL. 2010. Role of kinase-independent and -dependent functions of FAK in endothelial cell survival and barrier function during embryonic development. J Cell Biol 189: 955–965.
