Tamoxifen-inducible NaV1.8-CreERT2 recombinase activity in nociceptive neurons of dorsal root ganglia

TECHNOLOGY REPORT

Tamoxifen-Inducible NaV1.8-CreERT2 Recombinase Activityin Nociceptive Neurons of Dorsal Root GangliaJing Zhao,1 Mohammed A. Nassar,1 Isabella Gavazzi,2 and John N. Wood1*1Molecular Nociception Group, Department of Biology, University College London, London, United Kingdom2Centre for Neuroscience Research, King’s College London, Guy’s Campus, London, United Kingdom

Received 14 March 2006; Revised 7 June 2006; Accepted 14 June 2006

Summary: To explore the function of genes expressed inadult mouse nociceptive neurons, we generated hetero-zygous knock-in mice expressing the tamoxifen-induci-ble Cre recombinase construct CreERT2 downstream ofthe NaV1.8 promoter. CreERT2 encodes a Cre recombi-nase (Cre) fused to a mutant estrogen ligand-bindingdomain (ERT2) that requires the presence of tamoxifenfor activity. We have previously shown that heterozy-gous NaV1.8-Cre mice will delete loxP flanked genesspecifically in nociceptive sensory neurons from embry-onic day 14. We therefore used the same strategy of ho-mologous recombination and mouse generation, substi-tuting the Cre cassette with CreERT2. No functional Crerecombinase activity was found in CreERT2 micecrossed with reporter mice in the absence of tamoxifen.We found that, as with NaV1.8-Cre mice, functional Crerecombinase was present in nociceptive sensory neu-rons after tamoxifen induction in vivo. However, the per-centage of dorsal root ganglion (DRG) neurons express-ing functional Cre activity was much reduced (<10% ofthe number found in the NaV1.8-Cre mouse). We alsoexamined Cre recombinase activity in sensory neuronsin culture. After treatment with 1 lM tamoxifen for 48 h,15% of DRG neurons showed Cre activity. NaV1.8-CreERT2 animals may thus be useful for single cell stud-ies of the functional consequences of gene ablation inculture, but are unlikely to be useful for behavioral stud-ies. genesis 44:364–371, 2006. VVC 2006 Wiley-Liss, Inc.

Key words: NaV1.8-CreERT2; pain and nociception; tamo-xifen inducible; ROSA26 reporter; behavior; DRG

INTRODUCTION

Gene targeting in the mouse has provided remarkableadvances in the understanding of gene function. How-ever, some inherent limitations of global ablation, suchas perinatal lethality or developmental compensatorymechanisms, may prevent the appearance of phenotypicchanges in the genetically manipulated adult animal. Atissue-specific and inducible gene targeting system hasbeen developed to circumvent these problems (Metzgerand Chambon, 2001).

Cre recombinase from bacteriophage P1 recognizes34-bp loxP sites and catalyzes molecular recombination,so that DNA segments flanked with loxP sites will beexcised or inverted (Abremski and Hoess, 1984). TheCre-loxP recombinase system allows mice expressing tis-sue-specific Cre recombinase to be crossed with micecontaining floxed genes to produce tissue-specific nullmutants (Le and Sauer, 2001; Nagy, 2000). Cre-ER recom-binase was generated by the fusion of Cre to the ligand-binding domain of the estrogen receptor (ER). The Cre-ER recombinase activity is induced by the presence ofan estrogen agonist that allows nuclear entry to therecombinase by disrupting interactions with Hsp-90(Metzger and Chambon, 2001). The CreERT2 recombi-nase is a fusion protein that contains three mutations inthe human ER, so that the complex is efficiently acti-vated by the synthetic estrogen-like agonist tamoxifen,but not by endogenous estrogens (Feil et al., 1997; Indraet al., 1999). A number of groups have demonstrated thatinducible tissue-specific recombination systems are apowerful tool for gene targeting (Feil et al., 1997; Hayashiand McMahon, 2002; Leone et al., 2003).

Recent studies have shown that hundreds of genes aredysregulated in pain states in peripheral sensory neu-rons (Battaglia et al., 2003; Wang et al., 2002; Woodet al., 2002). It is difficult to identify the function ofthese genes if there are no pharmacological blockers orif global null mutants die during development. Tissue-specific or time-specific gene deletion provides a usefulway to address these questions. Previously, we generatedNaV1.8-Cre recombinase mice (Stirling et al., 2005)expressing Cre recombinase driven by the NaV1.8 pro-moter. The heterozygous NaV1.8-Cre mice have success-fully been used for gene deletion in damage-sensing pe-

* Correspondence to: John N. Wood, Molecular Nociception Group,

Department of Biology, University College London, London WO1E 6BT,

United Kingdom.

E-mail: [email protected]

Contract grant sponsors: MRC; BBSRC; and Wellcome TrustPublished online in Wiley InterScience (www.interscience.wiley.com).

DOI: 10.1002/dvg.20224

' 2006 Wiley-Liss, Inc. genesis 44:364–371 (2006)

ripheral neurons (nociceptors) (Nassar et al., 2004).NaV1.8 is a voltage-gated sodium channel expressed onlyin a subset of sensory neurons of which more than 85%are nociceptors (Akopian et al., 1996; Djouhri et al.,2003). Cre expression began at embryonic day 14 insmall diameter neurons in dorsal root, trigeminal andnodose ganglia, but was absent in nonneuronal or CNStissues of NaV1.8-Cre mice. Pain behavior in response tomechanical or thermal stimuli, and in acute, inflamma-tory, and neuropathic pain is also normal in this animal(Stirling et al., 2005). In order to allow gene deletion inthe same nociceptive neurons in adult animals, we haveadapted the NaV1.8-Cre approach to make a tamoxifen-activatable Cre mouse line. Here, we report the genera-tion of mice expression CreERT2 recombinase under thecontrol of NaV1.8 promoter.

We generated the NaV1.8-CreERT2 mice based on theNaV1.8-Cre construct (Stirling et al., 2005). Figure 1a

demonstrates the structure of the NaV1.8 wild-type(WT) allele, the NaV1.8-CreERT2 targeting construct andthe NaV1.8-CreERT2 targeted allele with and without theneomycin selection cassette (neoR). After electropora-tion in 129/Sv embryonic stem cells, 209 clones that sur-vived selection with Geneticin were screened usingSouthern blotting, and 41 correctly targeted clones wereidentified. Three positive cell lines were used to gener-ate transgenic mice, and germ line transmitting chimeraewere generated. F1 animals were crossed to the FLPeRmice (Farley et al., 2000) to delete the neomycin cas-sette. Southern blot analysis of genomic DNA fromNaV1.8-CreERT2 mice with 30 arm external probe is pre-sented in Figure 1b. The digestion of murine genomicDNA with the restriction enzyme BamHI generates aDNA fragment of 6.7 kb from the endogenous NaV1.8gene, an 8.5 kb band from NaV1.8-CreERT2 transgenemouse with neomycin cassette, and a 7.2 kb band fromNaV1.8-CreERT2 after removal of neomycin cassette.Heterozygous NaV1.8-CreERT2 mice were also analysedby PCR with six different PCR primer sets. The positionand the product size are showed in Figure 2a. The WTband (438 bp, A-B), NaV1.8-Cre band (461 bp, A-C), andERT2 band (824 bp, D-E) were amplified from thegenomic DNA of heterozygous NaV1.8-CreERT2 (þneoR)and NaV1.8-CreERT2 (�neoR) mice. The neomycin cas-sette chimera band (350 bp, F-B) was amplified fromNaV1.8-CreERT2 (þneoR) mice. Only the chimera stop

FIG. 1. Generation of NaV1.8-CreERT2 mice. a: Diagram of thenative NaV1.8 allele and NaV1.8-CreERT2 targeting construct.CreERT2 followed by NaV1.8 30UTR (U), replaced the initiator methi-onine of NaV1.8. Polyadenylation signals (S) ensured cessation oftranscription at this point. The construct also contained a FRT-flanked (triangle) neomycin positive selection cassette (neoR) and aHSK-TK negative selection cassette (TK). BamHI site and theexpected sizes of the resulting DNA fragments are indicated. b:Analysis of genomic DNA by Southern blot. Southern blot withBamHI and external probe confirms correct targeting and excisionof neomycin cassette. The 6.7-kb WT band is seen in all lanes, andthe 8.5- and 7.2-kb bands represent the targeted allele before andafter excision of neomycin cassette, respectively.

FIG. 2. Analysis of NaV1.8-CreERT2 mice by PCR. a: Structure ofthe native NaV1.8 allele, NaV1.8-CreERT2(þneoR) allele, andNaV1.8-CreERT2 allele. The location of PCR primers and theexpected sizes of fragment are indicated. b: PCR was used todetect the presence of NaV1.8-CreERT2(þneoR) and NaV1.8-CreERT2.

365TAMOXIFEN-INDUCIBLE Cre RECOMBINASE ACTIVITY IN DRG

genesis DOI 10.1002/dvg

band (280 bp, G-B) was found in NaV1.8-CreERT2 mice(Fig. 2b). The PCR results indicate that the CreERT2 wasintegrated as a whole, without deletions, into the in-tended location.

To test the efficiency of tamoxifen-induced Cre recom-binase activity in dorsal root ganglion (DRG), NaV1.8-CreERT2 mice were crossed with the reporter lineROSA26R (Soriano, 1999). The expression of functionalCre was examined in DRG of NaV1.8-CreERT2/ROSA26Rmice. Cre recombinase excises the floxed polyadenyl-ation sites that block synthesis of functional b-galactosi-dase, so that functional Cre expression can be detectedwith X-gal staining (Soriano, 1999). After tamoxifen treat-ment with intraperitoneal injection (ip), the DRG sec-

tions were stained with X-gal. No blue cells were found inmice not treated with tamoxifen (Fig. 3a), demonstratingthat there is no tamoxifen-independent Cre activity in theNaV1.8-CreERT2 mice. The test NaV1.8-CreERT2/ROSA26Rmice were injected with 2 mg of tamoxifen per day, for5 days each week. X-gal staining shows that the positive cells(blue cells) appeared in DRG section after 5 days of tamox-ifen treatment. The number of positive cells increasedwith the time of tamoxifen treatment. There are 0–1(1 week), 0–3 (2 weeks, Fig. 3b), 4–6 (4 weeks, Fig. 3c),and 6–12 (6 weeks, Fig. 3d) positive cells in each DRGsection. This is much less than in our previous NaV1.8-Cre mice, where positive X-gal staining was found in 68%of sensory neurons (Stirling et al., 2005). To characterize

FIG. 3. Tamoxifen induces re-combination in the DRG of adultNaV1.8-CreERT2 mice. AdultNaV1.8-CreERT2/ROSA26R micewere injected (ip) once a day with2 mg of tamoxifen for (a) 0, (b) 2,(c) 4, and (d) 6 weeks. After tamox-ifen treatment, functional Cre ac-tivity was assessed using X-galstaining of DRG. Neutral red wasused for counterstaining. Scalebar = 50 lm.

FIG. 4. Colocalization of b-galactivity and peripherin in DRGfrom NaV1.8-CreERT2/ROSA26Rmice. a: Sections were first co-stained with anti-peripherin (green)and anti-N200 (red). b: After immu-nohistochemistry, sections werestained with X-gal. Scale bar =50 lm.

366 ZHAO ET AL.


the types of sensory neurons expressing Cre recombinasewithin the DRG of NaV1.8-CreERT2/ROSA26R mice, label-ing of DRG sections using antibodies directed againstperipherin (Fig. 4a, in green), neurofilament-200 (Fig. 4a, inred) was performed with immunohistochemistry. Cre acti-vity can be directly visualized with X-gal staining in thesame sections after immunohistochemistry (Fig. 4b). Over-laying Figure 4a to Figure 4b, the result shows that most ofthe blue cells were peripherin-positive and N200-negativecells (see arrowheads in Fig. 4a,b). This indicates that themajority of Cre expression is likely to be in nociceptors.

We further investigated the ability of tamoxifen toinduce recombination in cultured DRG neurons. Thecultured DRG neurons prepared from NaV1.8-CreERT2/

ROSA26R adults were exposed to varying concentra-tions (1 nM to 10 lM) of 4-OH tamoxifen (4-OHT) ingrowth medium for 2 days. After 4-OHT treatment, thecultured DRG neurons were fixed and stained with X-gal. The results show that there were no blue cells incontrol (without 4-OHT), 1.9% 6 0.2% (1 nM 4-OHT),5.3% 6 2.2% (10 nM 4-OHT), 12.3% 6 3.8% (100 nM 4-OHT), 15.0% 6 5.7% (1 lM 4-OHT), and 12.1% 6 1.9%(10 lM of 4-OHT) (Fig. 5e). Tamoxifen-induced recombi-nation was also tested in cultured DRG neurons at differ-ent times (1–5 days, Fig. 5f). No blue cells were found incontrol cultures. A small increase (1.4% 6 0.3%) in Creactivity over 24 h and a large increase (15.3%6 0.7%) over2 days was observed after the initiation of 1 lM 4-OHT

FIG. 5. 4-OHT induces efficientrecombination in cultured DRGneurons from NaV1.8-CreERT2/ROSA26R mice. DRG neuronswere isolated from adult NaV1.8-CreERT2/ROSA26R mice andtreated with 1 lM of 4-OHT for (a)0 day, (b) 1 day, (c) 3 days, and (d)5 days. After 4-OHT treatment,the cultured cells were stainedwith X-gal. Scale bar = 50 lm.Cultured DRG neurons were treatedwith 4-OHT in different doses (e)and with 1 lM of 4-OHT for differ-ent durations (f). The b-galactosi-dase activity was determined bydirect counting of neurons after X-gal staining. Each point is themean of triplicate measurements,with standard deviation indicatedby the error bars.



treatment (Fig. 5f). The images were taken at 0 day(Fig. 5a), 1 day (Fig. 5b), 3 day (Fig. 5c), and 5 day (Fig. 5d).

To characterize the types of sensory neurons express-ing Cre, histochemical labeling of cultured DRG neuronsusing antibodies directed against peripherin (Fig. 6a, ingreen), neurofilament-200 (Fig. 6a, in red), and X-galstaining (Fig. 6b) were performed. Overlaying Figure 6ato Figure 6b, the result showed that most of the bluecells were peripherin-positive and N200-negtive cells(see arrowheads in Fig. 6a,b). This also indicates that amajority of Cre expression is, as in the DRG sections, insmall neurons, the majority of which are nociceptors.

We have generated a NaV1.8-CreERT2 mouse line thatexhibits the functional expression of a tamoxifen-induci-ble Cre transgene in DRG: addition of tamoxifen or itsderivative 4-OHT results in gene recombination in adultmice and in cell culture. However, the percentage ofDRG neurons expressing functional Cre activity wasmuch reduced (<10% of the number found in theNaV1.8-Cre mouse). There are two possible explanationsfor the low percentage of functional Cre expression.One possibility is that high level expression of Hsp90occurs in DRG. Another possibility is that only low con-centrations of tamoxifen are found in the DRG on sys-temic treatment. To address the first possibility, we usedan Hsp90 inhibitor. In the tamoxifen-inducible Crerecombination system, Cre-mediated recombination isprevented by the ER-dependent cytoplasmic sequestra-tion of Cre by Hsp90 (Mattioni et al., 1994; Picard,1994). Tamoxifen disrupts the interaction with Hsp90and allows access of CreERT2 into the nucleus and theinitiation of recombination. Hsp90 is a highly abundantand essential cytosolic protein, and the expression levelof Hsp90 increases on environmental stress. Hsp90 func-tions as a molecular chaperone by binding to various cel-lular proteins and supporting the proper folding, stabil-ity, and function of target proteins (Whitesell and Lind-quist, 2005). Geldanamycin, an ansamycin-derivativebenzoquinone compound, specifically binds and inhibitsthe molecular chaperone, Hsp90 (Miyata, 2005). A dras-tic reduction in the basal level of Hsp90 was found inthe brain of goldfish after administration of geldanamy-cin. Geldanamycin (1 lM) has been used together withtamoxifen on NaV1.8-CreERT2/ROSA26R mice (2 weeks,ip) and on DRG-cultured neurons (3 days). Unfortu-

nately, no significant improvement in the frequency ofrecombination was found (data not shown).

To test the second possibility, we examined the Creactivity in cultured DRG neurons of NaV1.8-CreERT2/ROSA26R mice which have been treated with tamoxifen(ip) for 6 weeks. The result shows that 5.4% 6 1.8% neu-rons were positively stained in control cells (without 4-OHT in media, Fig. 7a,c) and 14.4% 6 1.9% neuronswere positively stained in the test cells (1 lM 4-OHT for2 days, Fig. 7b,c). This result indicates that a high con-centration of tamoxifen could induce an increase of therecombination in cultured DRG neurons, even thoughthe NaV1.8-CreERT2/ROSA26R mice have been treatedwith a long-term tamoxifen application in vivo. Theresult shows that a significant increase (threefold) ofrecombination was found after 4-OHT treatment. Thissuggests that a suboptimal dose of tamoxifen is presentin the DRG neurons and that the frequency of recombi-nation could increase if the concentration of tamoxifenis higher in DRG. Unfortunately, the animals displaytoxic symptoms if the dosage is more than 3 mg/day.

After 48 h of 4-OHT treatment, functional Crerecombinase was expressed in 15% of cultured DRGneurons, which are almost exclusively peripherin-posi-tive cells. Thus, NaV1.8-CreERT2 mice can be poten-tially used to delete genes in a subset of adult nocicep-tive neurons either in vivo or in culture. However,given the fact that a large percentage of nociceptiveneurons did not exhibit functional Cre expression, it isunlikely that these mice will be useful for behavioralstudies, because of a mosaic pattern of gene deletionin nociceptors. In contrast, inducible gene deletion incultured DRG neurons is likely to be very useful forelectrophysiological studies, where the contribution ofindividual gene products to neuronal excitability canbe examined at the single cell level.

MATERIALS AND METHODS

Generation of Transgenic Construct

To engineer pNaV1.8-CreERT2, the CreERT2 codingsequence from pCreERT2 (Feil et al., 1997) (a gift fromProf. Chambon) was subcloned first between the EcoRIand SacI sites of plasmid pUC18 (Roche). And then the

FIG. 6. Colocalization of b-galactivity and peripherin in DRG-cul-tured neurons from NaV1.8-CreERT2/ROSA26R mice. DRG-cul-tured neurons were treated with 1lM of 4-OHT for 2 days. After 4-OHT treatment, the cultured neu-rons were first co-stained with anti-peripherin (green, a) and anti-N200(red, a). They were then stainedwith X-gal (b). Scale bar = 50 lm.

368 ZHAO ET AL.


CreERT2 fragment was subcloned between AgeI and XhoI(SalI in pUC18-CreERT2) of plasmid pBS-NaV1.8-Cre (Stir-ling et al., 2005) to generate pBS-NaV1.8-CreERT2. The ori-entation of the gene was confirmed by sequencing.

Transgenic Mouse Production

The targeting construct was linearized by NotI and elec-troporated into 129/Sv embryonic stem cells, drugselected using geneticin, and transgenic mice generatedby standard means (Joyner, 2000). After establishinggerm-line transmission, the neomycin cassette wasexcised by crossing to a universal Flp deletor mouse(Farley et al., 2000) to generate the NaV1.8-CreERT2mouse line.

Southern Blot Analysis

The genomic DNA was extracted from tail (Sambrookand Russell, 2001). A 30 arm external probe (709 bp),which was amplified from C57/Bl6 genomic DNA withprimers (50-GCTTGAGTTGAAACGCGGAGG-30 and 50-CACCTCGCTGCTGATGGAGA-30), was used to detectthe NaV1.8 allele. Southern blot hybridization of BamHI-

digested genomic DNA with the 30 probe yielded a 6.7kb WT band, an 8.5 kb targeted band (þneoR), and a 7.2kb targeted band (Fig. 1a).

PCR Analysis

The genotyping for the NaV1.8-CreERT2 mice was per-formed with NaV1.8, Cre, ERT2, and stop primers.Primer sequences were as follows:A, NaV1.8 sense primer (50-TGTAGATGGACTGCAGAG-GATGGA-30)B, NaV1.8 antisense primer (50-TTACCCGGTGTGTGC-TGTAGAAAG-30)C, Cre antisense primer (50-AAATGTTGCTGGATAGT-TTTTACTGCC-30)D, ERT2 sense primer (50-CAAGCCCGCTCATGATCAAA-30)E, ERT2 antisense primer (50-GTG GCT TTG GTC CGTCTC CT-30)F, neoR sense primer (50-GCGTCACCTTAATATGCGAA-GTGG-30)G, Stop sense primer (50-TCATGCATAATAAAATATCTT-TATTTTCAT-30)

FIG. 7. 4-OHT increases recom-bination in cultured DRG neuronsfrom NaV1.8-CreERT2/ROSA26Rmice pretreated with tamoxifen.After injection (ip) with tamoxifendaily (2 mg/day) for 6 weeks, DRGneurons were isolated from adultNaV1.8-CreERT2/ROSA26R mice.The cultured DRG neurons werefixed and stained with X-gal aftertreatment without (a) and with (b)1 lM of 4-OHT for 2 days. Scalebar = 25 lm. The b-galactosidaseactivity was determined by directcounting of neurons after X-galstaining (c). Each point is themean of six measurements withstandard deviation shown.



The product sizes are as follows: A-B: 438 bp, A-C: 461bp, D-E: 824 bp, F-B: 350 bp, G-B: 280 bp (Fig. 2). Prod-ucts were electrophoresed on 1.2% agarose gels stainedwith ethidium bromide and photographed.

Cross-Breeding of Mice

Heterozygous NaV1.8-CreERT2 mice were crossed tohomozygous ROSA26R (Soriano, 1999). Offspring weregenotyped by PCR for NaV1.8-CreERT2 and ROSA26R al-leles (Soriano, 1999). The mice that carried NaV1.8-CreERT2 allele were used in further experiments.

Tamoxifen Treatment

Tamoxifen (TM, Tamoxifen free base, Sigma T5648) wasdissolved in sunflower oil as described (Metzger andChambon, 2001). Six to 8-week old NaV1.8-CreERT2/ROSA26R mice were injected (ip) once a day with vehi-cle or with 2 mg TM for 5 consecutive days per week.Two days after treatment, the animals were euthanized,and DRGs were collected for further analysis. 4-Hydroxy-tamoxifen (4-OHT, Sigma H7904) was dissolved in etha-nol. 4-OHT (2 M) was stored at �208C and used in fur-ther experiments.

DRG Neurons Culture

DRG neurons were cultured from 6–8-week old NaV1.8-CreERT2/ROSA26R mice. Animals were killed by inhala-tion of a rising concentration of CO2 followed by cervi-cal dislocation, and 30–40 DRGs were dissected fromeach. Ganglia were digested in collagenase (Type XI, 0.6mg/ml, Sigma), dispase (3.0 mg/ml, Sigma), and glucose(1.8 mg/ml) in Ca2þ, Mg2þ free PBS for 40 min prior tomechnical trituration. Cells were then resupended inDulbecco’s modified Eagle’s medium (Gibco) containing10% fetal bovine serum (Gibco), 10,000 i.u./ml penicil-lin–streptomycin (Gibco), and 100 ng/ml nerve growthfactor (Sigma), and plated on 13-mm cover slips coatedwith poly-L-lysine.

Immunohistochemistry Analysis

After treatment with TM, the DRGs isolated were imme-diately frozen in OCT (O.C.T. Compounds, BDH) on dryice. 12 lM cryosections were dried at room temperaturefor 30 min and then fixed with 4% PBS-buffered parafor-maldehyde solution for 10 min on ice. After three washesin PBS, sections were incubated for 30 min in 10% goatserum diluted in PBS containing 0.3% Triton X-100(PBST), and for 1 h at room temperature in a 1:500 dilu-tion of an anti peripherin monoclonal antibody (P5117,Sigma) and 1:200 dilution of an anti N-200 polyclonalantibody (N4142, Sigma). Following three washes inPBST, the sections were incubated for 1 h in a 1:500 dilu-tion of Alexa Fluor 488 goat anti-mouse IgG (A-11017,Molecular Probes) and 1:1,000 dilution of Alexa Fluor594 goat anti-rabbit IgG (A-11037, Molecular Probes).After three washes in PBST, the sections were mounted

in CITIFlour solution and analysed using a fluorescentmicroscope.

X-gal Staining

After fixation with 4% PBS-buffered paraformaldehyde orafter immunohistochemistry, DRG sections were washedthree times with PBS. And then incubated overnight inX-gal solution at 358C. For cultured DRG neurons, theslides were fixed with 2% PBS-buffered formaldehydecontaining 0.25% glutaraldehyde for 15 min at room tem-perature, and then stained with X-gal solution at 358Covernight. Sections or slides (cultured DRG neurons)were counterstained with 1% neutral red, dehydratedwith ethanol, and cleared with Histo-ClearII andmounted with DePex mounting medium (BDH). The X-gel staining in freshly frozen sections is usually punctatein appearance, but homogeneous in perfused DRG sec-tions and in cultured DRG neurons.

ACKNOWLEDGMENTS

We thank Dr. Pierre Chambon for kindly providing thepCreERT2 plasmid, The Darwin Transgenic Mouse CoreFacility of Baylor College of Medicine for help with trans-genic mouse preparation, Dr. Nic Kessaris for providingthe ROSA26 reporter line, and Mr. Bjarke Abrahamsenfor help with tamoxifen treatment.

LITERATURE CITED

Abremski K, Hoess R. 1984. Bacteriophage P1 site-specific recombina-tion. Purification and properties of the Cre recombinase protein. JBiol Chem 259:1509–1514.

Akopian AN, Sivilotti L, Wood JN. 1996. A tetrodotoxin-resistant volt-age-gated sodium channel expressed by sensory neurons. Nature379:257–262.

Battaglia AA, Sehayek K, Grist J, McMahon SB, Gavazzi I. 2003. EphBreceptors and ephrin-B ligands regulate spinal sensory connectiv-ity and modulate pain processing. Nat Neurosci 6:339, 340.

Djouhri L, Fang X, Okuse K, Wood JN, Berry CM, Lawson SN. 2003.The TTX-resistant sodium channel Nav1.8 (SNS/PN3): Expressionand correlation with membrane properties in rat nociceptive pri-mary afferent neurons. J Physiol 550:739–752.

Farley FW, Soriano P, Steffen LS, Dymecki SM. 2000. Widespread recombi-nase expression using FLPeR (flipper) mice. Genesis 28:106–110.

Feil R, Wagner J, Metzger D, Chambon P. 1997. Regulation of Crerecombinase activity by mutated estrogen receptor ligand-bindingdomains. Biochem Biophys Res Commun 237:752–757.

Hayashi S, McMahon AP. 2002. Efficient recombination in diverse tis-sues by a tamoxifen-inducible form of Cre: A tool for temporallyregulated gene activation/inactivation in the mouse. Dev Biol244:305–318.

Indra AK, Warot X, Brocard J, Bornert JM, Xiao JH, Chambon P, MetzgerD. 1999. Temporally-controlled site-specific mutagenesis in the ba-sal layer of the epidermis: Comparison of the recombinase activityof the tamoxifen-inducible Cre-ER(T) and Cre-ER(T2) recombi-nases. Nucleic Acids Res 27:4324–4327.

Joyner AL. 2000. Gene targeting. Oxford: Oxford University Press.Le Y, Sauer B. 2001. Conditional gene knockout using Cre recombinase.

Mol Biotechnol 17:269–275.Leone DP, Genoud S, Atanasoski S, Grausenburger R, Berger P, Metzger

D, Macklin WB, Chambon P, Suter U. 2003. Tamoxifen-inducibleglia-specific Cre mice for somatic mutagenesis in oligodendrocytesand Schwann cells. Mol Cell Neurosci 22:430–440.

370 ZHAO ET AL.


Mattioni T, Louvion JF, Picard D. 1994. Regulation of protein activities byfusion to steroid binding domains. Methods Cell Biol 43 (Part A):335–352.

Metzger D, Chambon P. 2001. Site- and time-specific gene targeting inthe mouse. Methods 24:71–80.

Miyata Y. 2005. Hsp90 inhibitor geldanamycin and its derivatives as novelcancer chemotherapeutic agents. Curr Pharm Des 11:1131–1138.

Nagy A. 2000. Cre recombinase: The universal reagent for genome tai-loring. Genesis 26:99–109.

Nassar MA, Stirling LC, Forlani G, Baker MD, Matthews EA, DickensonAH, Wood JN. 2004. Nociceptor-specific gene deletion reveals amajor role for Nav1.7 (PN1) in acute and inflammatory pain. ProcNatl Acad Sci USA 101:12706–12711.

Picard D. Regulation of protein function through expression of chimae-ric proteins. 1994. Curr Opin Biotechnol 5:511–515.

Sambrook J, Russell DW. 2001. Molecular cloning. A laborartory man-ual, 3rd ed. New York: Cold Spring Harbor Laboratory Press.

Soriano P. 1999. Generalized lacZ expression with the ROSA26 Cre re-porter strain. Nat Genet 21:70, 71.

Stirling LC, Forlani G, Baker MD, Wood JN, Matthews EA, Dicken-son AH, Nassar MA. 2005. Nociceptor-specific gene deletionusing heterozygous Na(V)1.8-Cre recombinase mice. Pain 113:27–36.

Wang H, Sun H, Della PK, Benz RJ, Xu J, Gerhold DL, Holder DJ, KoblanKS. 2002. Chronic neuropathic pain is accompanied by globalchanges in gene expression and shares pathobiology with neuro-degenerative diseases. Neuroscience 114:529–546.

Whitesell L, Lindquist SL. 2005. HSP90 and the chaperoning of cancer.Nat Rev Cancer 5:761–772.

Wood JN, Akopian AN, Baker M, Ding Y, Geoghegan F, Nassar M, Malik-Hall M, Okuse K, Poon L, Ravenall S, Sukumaran M, Souslova V.2002. Sodium channels in primary sensory neurons: Relationshipto pain states. Novartis Found Symp 241:159–168.



Documents

Tamoxifen-inducible NaV1.8-CreERT2 recombinase activity in nociceptive neurons of dorsal root ganglia