Dorsal telencephalon-specific expression of Cre recombinase in PAC transgenic mice

TECHNOLOGY REPORT

Dorsal Telencephalon-Specific Expression of CreRecombinase in PAC Transgenic MiceTakuji Iwasato,1,2* Ryochi Nomura,2 Reiko Ando,2,4 Toshio Ikeda,2 Mika Tanaka,2,3 andShigeyoshi Itohara2*1PRESTO, Japan Science and Technology Agency, Saitama, Japan2Laboratory for Behavioral Genetics, Brain Science Institute (BSI), Riken Saitama, Japan3Research Resources Center, Brain Science Institute (BSI), Riken Saitama, Japan4Brain Science and Life Technology Research Foundation, Tokyo, Japan

Received 5 August 2003; Accepted 3 November 2003

Summary: The ability to restrict gene expression or dis-ruption to specific regions of the brain would enhanceunderstanding of the molecular basis for brain develop-ment and function. For this purpose, brain region-re-stricted promoters are essential. Here we report theisolation of a DNA fragment containing the Emx1 genepromoter, which is responsible for dorsal telencepha-lon-specific expression. The Cre recombinase gene wasinserted into a mouse PAC (P1-derived artificial chromo-some) Emx1-locus clone (PAC-Emx1#1 clone) and uti-lized to generate three transgenic mouse lines. In allthree lines, especially Tg3, Cre-mediated recombinationwas highly restricted to Emx1-expressing cell lineages,from embryonic stages to adulthood. Immunohisto-chemical analyses showed that Cre protein is expressedin the dorsal telencephalon in all three lines in adulthood.Thus, the PAC-Emx1#1 clone contains essentially allregulatory elements necessary for Emx1 gene expres-sion. Our results suggest that Emx1-Cre Tg3 mice andthe PAC-Emx1#1 clone constitute powerful tools for dor-sal telencephalon-specific gene manipulation. genesis38:130–138, 2004. © 2004 Wiley-Liss, Inc.

Key words: P1-derived artificial chromosome; Cre/loxPrecombination; Emx1 promoter; central nervous system

The cerebral cortex is a region of the brain that isimportant for higher brain functions. Its developmentand function depend on its connectivity with other re-gions of the brain (Iwasato et al., 1997; Schlaggar andO’Leary, 1991; Sharma et al., 2000). Therefore, the abil-ity to manipulate genes in a cerebral cortex-restrictedmanner may be necessary to dissect out the molecularand cellular mechanisms running in the cerebral cortexduring cortical development and operation from thoserunning in other brain regions, including the striatum,thalamus, and brainstem.

Transgenic mice are a powerful model system forstudying the molecular mechanisms of brain functionand development specific to particular regions of thebrain and/or cell-types (Huang et al., 1999; Ichise et al.,

2000). Proper study of these mechanisms requires theuse of region- and/or cell type-specific promoters, suchas the CaMKII promoter specific to excitatory neurons inthe postnatal forebrain (Mayford et al., 1995) and the L7promoter specific to the cerebellar Purkinje neurons(Oberdick et al., 1990). Since no promoter has beenavailable for the dorsal telencephalon, however, the abil-ity to utilize transgenic mice to study the developmentor function of this region of the brain has been limited.The promoter of Emx1, a homeobox gene, is an idealcandidate for the purpose, because its expression isrestricted to the dorsal telencephalon from embryonicstages to adulthood (Gulisano et al., 1996; Simeone etal., 1992).

To isolate a DNA fragment containing transcriptionalregulatory elements for the Emx1 gene expression, weexamined a mouse PAC Emx1-locus clone (PAC-Emx1#1) (Fig. 1). The Cre recombinase gene was in-serted into the PAC-Emx1#1 clone and utilized to gener-ate three transgenic mouse lines (Tg1, Tg2, and Tg3).Cre recombinase is a unique reporter of promoter activ-ity. Using Cre/loxP recombination reporters, such asCAG-CAT-Z (Sakai and Miyazaki, 1997) and R26R mice(Soriano, 1999), cell lineages in which the promoter hasbeen activated transiently or permanently can be de-tected. In addition, immunostaining with an anti-Creantibody can detect cells in which the promoter is active

* Correspondence to: Takuji Iwasato, BSI, Riken, 2-1 Hirosawa Wako-shi,Saitama 351-0198, Japan. E-mail: [email protected] or Shigeyoshi Ito-hara, BSI, Riken, 2-1 Hirosawa Wako-shi, Saitama 351-0198, Japan.

E-mail: [email protected] grant sponsors: Grant-in-Aid for Scientific Research-C, Japan

Society for the Promotion of Science, Grant-in-Aid for Scientific Research onPriority Areas-A, Ministry of Education, Culture, Sports, Science, and Tech-nology (to T.I.).

DOI: 10.1002/gene.20009

© 2004 Wiley-Liss, Inc. genesis 38:130–138 (2004)

at particular developmental stages (Nakazawa et al.,2002).

To characterize the patterns of Cre/loxP recombina-tion (the promoter activity in cell lineages), we crossedeach of our three transgenic lines with CAG-CAT-Z re-combination-reporter mice to obtain double transgenic(Emx1-Cre Tg/CAG-CAT-Z) mice. Brains of adult mice(up to 18-month-old) were removed and thick (400 �m)coronal sections from the olfactory bulb to the anteriorspinal cord were stained for �-galactosidase activity(Figs. 2, 3). This strategy allowed us to examine all brainregions for Cre-mediated recombination with high sen-sitivity. We compared the staining intensity in Tg/CAG-CAT-Z mice with that in CAG-�-Z-positive control mice(see Materials and Methods). We detected dense LacZstaining, indicative of many cells with Cre-mediated re-combination, in the olfactory bulb, neocortex, piriformcortex, hippocampus, and amygdala in each of Tg3/CAG-CAT-Z mice (n � 4). In contrast, there were noLacZ-positive cells in other regions of the brain, includ-ing the striatum, thalamus, hypothalamus, brainstem,cerebellum, and spinal cord. Tg1/CAG-CAT-Z (n � 4)

and Tg2/CAG-CAT-Z (n � 2) mice showed staining pat-terns similar to those of Tg3/CAG-CAT-Z mice, althoughseveral differences were also observed.

To determine whether there are developmentalchanges in recombination patterns, we examined Tg/CAG-CAT-Z mice at postnatal days (P)7–8 by X-gal stain-ing of coronal or parasagittal slices (400 �m-thick) (datanot shown). At this stage of development all Tg1/CAG-CAT-Z (n � 1), Tg2/CAG-CAT-Z (n � 5), and Tg3/CAG-CAT-Z (n � 4) mice showed strong staining in the dorsaltelencephalon. In the Tg1/CAG-CAT-Z brain, we ob-served staining in the inferior colliculus. In the Tg2/CAG-CAT-Z brain, staining of the olfactory bulb was weak andstaining of the thalamus and cerebellum was not de-tected yet. Using whole-mount X-gal staining, we exam-ined Tg3/CAG-CAT-Z embryos at embryonic days(E)8.5–17 (n � 14) and found that Cre-mediated recom-bination in the dorsal telencephalon was first detectableat E10 (Fig. 4). These results were similar to those foronset of endogenous Emx1 expression (Simeone et al.,1992).

FIG. 1. PAC modification andgeneration of Emx1-Cre trans-genic mice. A: The transgenicconstruct. A nuclear localizationsignal (NLS)-Cre-poly (A) signal(pA) cassette was inserted, in thesense orientation, immediately 5�to the Emx1 translational initiationsite (ATG) into the PAC-Emx1#1clone. B: Restriction enzyme di-gestions of PAC-Emx1#1 and twoindependent transgenic con-structs (PAC-Cre-A and -B). Ar-rowheads show bands whosesizes were expected to be shiftedby homologous recombination inbacteria. Other bands of PAC-Emx1-Cre#1, PAC-Cre-A, andPAC-Cre-B had similar sizes,suggesting no major deletion dur-ing homologous recombination inthe bacteria. C: Pulsed-field gelelectrophoresis showing that thelinearized transgenic construct(PAC-Cre-A) was intact just priorto injection into pronuclei.

131CORTEX-SPECIFIC PROMOTER, CRE/LOXP RECOMBINATION

To determine the recombination specificities outsidethe brain, we dissected various body parts from Tg3/CAG-CAT-Z mice (n � 2) at P7 and stained them for X-gal(data not shown). We detected various staining intensi-ties in the thymus, stomach, testis, and some regions ofthe skin, including the limbs and tail. In contrast, therewas no staining of the heart, lung, liver, spleen, muscle,or bladder. In CAG-�-Z-positive control mice at P7, all ofthese organs were stained, suggesting that CAG-CAT-Zreporter system is suitable for detecting recombinationin these tissues. We further found that some (not all)cells in germline had Cre-mediated recombination in Tg3male (data not shown). Recombination outside the brainwas also detected in Emx1-Cre knockin (KI) mice (Iwa-sato et al., 2000), although these animals were not ana-lyzed extensively.

To characterize the recombination patterns at the cel-lular level, thin (8–10 �m) coronal brain slices of Tg1/

CAG-CAT-Z (n � 6), Tg2/CAG-CAT-Z (n � 7), and Tg3/CAG-CAT-Z (n � 6) mice (P7 to 13-month-old) wereanalyzed. X-gal/anti-GABA-antibody double and X-gal/an-ti-GABA-antibody/Nissl triple staining revealed that, inthe neocortex and hippocampus of all three lines ofmice, most GABA-positive (inhibitory) neurons wereLacZ-negative, whereas most GABA-negative (excitatory)neurons were LacZ-positive (Fig. 5). Thus, throughoutthe life span of all three lines Cre-mediated recombina-tion is specific to excitatory neurons. These results areconsistent with those of Emx1-Cre KI mice (Fig. 5D)(Gorski et al., 2002; Iwasato et al., 2000), as well as withthe cell type specificities of endogenous Emx1 expres-sion (Chan et al., 2001).

To determine whether Cre-mediated recombinationoccurred in astrocytes, we stained brain slices of Tg1/CAG-CAT-Z and Tg3/CAG-CAT-Z mice with X-gal and ananti-GFAP antibody. In both transgenic lines, most GFAP-

FIG. 2. Dorsal telencephalon-specific Cre-mediated recombination in transgenic mouse brains. Cre-mediated recombination was assayedby X-gal staining of 400-�m-thick coronal sections of adult Tg1/CAG-CAT-Z (Tg1), Tg2/CAG-CAT-Z (Tg2), Tg3/CAG-CAT-Z (Tg3),knockin/CAG-CAT-Z (KI), CAG-�-Z, and CAG-CAT-Z mice. A–C: All Tg1, Tg2, and Tg3 mice showed dense staining, indicative ofrecombination, in the neocortex (cx), piriform cortex (pir), and hippocampus (hip). Tg1 and Tg3 amygdala were stained in somecompartments, such as the basolateral complex (BL), while others, such as the MePD nuclei, were not stained. In Tg2, the amygdalashowed much weaker staining than the other two transgenic lines, whereas there were a few and some positive cells in the thalamus (th)(arrowhead) and cerebellar granule cell layer (data not shown), respectively. D: In KI mice, there were some positive cells in the thalamus,cerebellum, brainstem, etc. (data not shown). E: In CAG-�-Z mice (positive control, PC), all areas of the brain were LacZ-positive. F: InCAG-CAT-Z (negative control, NC) mice, there was no staining in any area of the brain. str, striatum; ht, hypothalamus; fi, fimbria. Scalebar � 2 mm.

132 IWASATO ET AL.

positive astrocytes in the dentate gyrus (DG) molecularlayer and in neocortex layer I were LacZ-positive (datanot shown). Thus, in addition to excitatory neurons,astrocytes (GFAP-positive cells) showed Cre-mediatedrecombination in the dorsal telencephalon of Emx1-Cretransgenic mice. This finding is consistent with the re-

sults with our (data not shown) and others’ (Gorski etal., 2002) Emx1-Cre KI mice.

Furthermore, we utilized R26R (Soriano, 1999), an-other recombination-reporter mouse line, to determinethe reproducibility of the tight regulation of Cre-medi-ated recombination observed in Tg3 mice. In Tg3/R26R

FIG. 3. X-gal staining of coronal sections (400-�m-thick) of Tg3/CAG-CAT-Z (18-month-old) (A), and Tg1/CAG-CAT-Z (14-month-old) (B)mice. In Tg1/CAG-CAT-Z mice, staining was detected in the mammillary nuclei (mn) of the hypothalamus, in the pontine nuclei (pn), andin the inferior colliculus (ic). Ob, olfactory bulb; str, striatum; sp, septum; cx, neocortex; pir, piriform cortex; hip, hippocampus; am,amygdala; cc, corpus callosum; th, thalamus; cb, cerebellum; bs, brainstem; sc, spinal cord. Scale bar � 2 mm.

133CORTEX-SPECIFIC PROMOTER, CRE/LOXP RECOMBINATION

mice (n � 3; 2–4-month-old), Cre-mediated recombina-tion in the brain was restricted to the cerebral cortex,hippocampus, olfactory bulb, and amygdala (Fig. 6), aresult consistent with those observed in Tg3/CAG-CAT-Zmice. In addition, most of the LacZ-negative neurons inthe cerebral cortex and hippocampus were GABA-posi-tive inhibitory neurons. With whole-mount X-gal stainingof E12.5 Tg3/R26R embryos (n � 2), we found thatCre-mediated recombination occurred at early embry-onic stages in R26R allele as in CAG-CAT-Z allele (Fig. 4).

Finally, we assayed Cre protein expression in the adultdorsal telencephalon of Emx1-Cre Tg1, Tg2, Tg3, and KImice (Fig. 7). Using an anti-Cre antibody, we found thatall cortical layers (II–VI) and the piriform cortex werestained densely and uniformly in Tg1 mice (n � 3,4.5–10-month-old). In the hippocampus, granule cells inthe DG were strongly stained, whereas CA1 and CA3pyramidal cells were stained relatively weakly. In con-trast, the oriens layers, stratum radiatum, and DG molec-ular layer in the hippocampus had no Cre-positive cells.In the amygdala, the basolateral complex showed highlevels of expression, while no expression was detectedin the MePD nuclei. These patterns were almost identicalto those in homozygous (n � 8, 1–8-month-old) and

heterozygous (n � 3, 3–5-month-old) KI mice. In Tg2(n � 2, 4.5- and 7.5-month-old) and Tg3 (n � 2, 4.5- and7-month-old) mice, we observed overall expression pat-terns similar to those in Tg1 mice. However, Cre expres-sion in the ventral cerebral cortex and amygdala wasmuch weaker in Tg2 than in Tg1, Tg3, and KI mice. Inthe Tg2 and Tg3 neocortex, Cre expression was de-tected in all layers (II–VI), but expression levels were noteven. Weak Cre protein expression in some layers of Tg2and Tg3 cortex might be sufficient for Cre function,because Cre-mediated recombination was detected uni-formly in cortical layers even in Tg2 and Tg3 mice (Figs.2, 3). Alternatively, transient Cre expression in earlierdevelopmental stages in these two lines may make uni-form Cre-mediated recombination in cortical layers.Thus, although there were slight differences among thetransgenic and KI lines, in all lines the patterns of Creprotein expression as a whole were similar to that ofendogenous Emx1 gene expression (Gulisano et al.,1996; Simeone et al., 1992).

Taken together, our data demonstrate that the 135-kbPAC-Emx1#1 fragment contains virtually all essentialtranscriptional-regulatory elements for dorsal telenceph-alon-specific Emx1 gene expression. Transgenic mouselines generated by others (Guo et al., 2000c) using ashort (11 kb) Emx1 promoter had transgene expressionpatterns completely different from those of endogenousEmx1 gene. Therefore, regulatory elements for Emx1gene expression appear to be widespread within the135-kb PAC-Emx1#1 fragment.

Our three transgenic lines, especially Tg3, demon-strated extremely strict region- and cell type-specificityin Cre-mediated recombination from embryonic to agedstages. In Tg3 mice, these rigid specificities were ob-served at two distinct target loxP loci of CAG-CAT-Z andR26R mice. Therefore, using Tg3 mice may enable ma-nipulation of any target gene in a similarly strict dorsaltelencephalon-specific manner. We previously gener-ated Emx1-Cre KI mice and showed that Cre-mediatedrecombination occurs only in the dorsal telencephalonbetween stages E11.5 and P7 (Iwasato et al., 2000). Atlater developmental stages, such as 3-month-old, how-ever, the CAG-CAT-Z reporter system detected Cre-me-diated recombination in parts of the thalamus, brain-stem, and cerebellum (Fig. 2D, data not shown). InEmx1-Cre KI mice, the pgk promoter inserted in theEmx1 locus appears to have caused ectopic Cre-medi-ated recombination in the brain. This ectopic recombi-nation disappeared completely in an Emx1-Cre KI vari-ant in which the pgk-neo selection marker gene hadbeen deleted (data not shown). Thus, Tg3 line as well asthe Emx1-Cre KI variant has an advantage in analyzingadult brain function.

Tg3 mice have additional advantages over KI mice. Inmost Emx1-Cre KI mouse lines (Guo et al., 2000b; Iwa-sato et al., 2000), one out of two endogenous Emx1genes is disrupted by insertion of the Cre gene. Becausea half dosage of Emx1 expression might be insufficientfor some mechanisms (Guo et al., 2000a; Qiu et al.,

FIG. 4. Whole-mount X-gal staining of Tg3/CAG-CAT-Z (A–C) andTg3/R26R (D) embryos at E8.5 (A), E9.5 (B), E10 (C), and E12.5 (D).Cre-mediated recombination in the dorsal telencephalon (arrowheads)was first detectable at E10. Recombination was also detected inpart of the nonneural ectoderm (arrows). Scale bars � 0.5 mm.

134 IWASATO ET AL.

1996; Yoshida et al., 1997), Tg3 mice may provide amore suitable model for studying the functions of mole-cules that may be closely related to Emx1 function. Inaddition, KI mice were generated in a 129 genetic back-ground (Guo et al., 2000b; Iwasato et al., 2000), whereasTg3 mice were generated in a C57BL/6 genetic back-ground. Pure C57BL/6 genetic background is more ad-vantageous in many fields of research, including learningand memory.

In summary, using the PAC-Emx1#1 fragment, wehave generated transgenic mouse lines (especially Tg3)that demonstrate lifelong expression of Cre recombinasein a pattern highly specific to the dorsal telencephalon.Emx1-Cre Tg3 mice may enable the manipulation of anytarget gene in a manner restricted to the dorsal telen-cephalon. Furthermore, the PAC-Emx1#1 fragment willallow us to express any gene of interest in the dorsaltelencephalon by the transgenic mouse approach.

MATERIALS AND METHODS

PAC ModificationAn RPCI-21 PAC mouse genomic library (Roswell Park

Cancer Institute, Buffalo, NY) was screened with an

Emx1 probe and several positive clones were analyzedby restriction enzyme digestion, pulsed-field gel electro-phoresis (FIGE Mapper: Bio-Rad, Hercules, CA), andSouthern hybridization. We chose one clone (PAC-Emx1#1), which contained about 95 kb upstream and 40kb downstream of the NotI site immediately 5� to theEmx1 translational initiation site. The NLS-Cre-poly(A)cassette was inserted into the NotI site in the PAC-Emx1#1 clone by homologous recombination in bacteria(Yang et al., 1997). The first and second homologousrecombination events were confirmed by Southern blothybridization.

Generation of Transgenic Founder Mice

Two independent homologous recombinants (PAC-Cre-A and B) were digested with NotI. The PAC-Cre-Afragment was separated from the vector sequence bypulsed-field gel electrophoresis in 1% SeaPlaque agarose(BMA, Portland, ME) in TAE buffer and purified by �-aga-rase digestion and with Ultra Free-MC UFC3LTK25 (Mil-lipore, Billerica, MA). The PAC-Cre-B fragment was sep-arated by electrophoresis in 0.5% SeaPlaque agarose inTAE and purified by extraction with phenol (threetimes), phenol/chloroform, and chloroform, followed by

FIG. 5. Cre-mediated recombination is restricted to excitatory neurons in the cerebral cortex of Emx1-Cre transgenic and KI mice.X-gal/anti-GABA-antibody/Nissl triple staining of the adult barrel cortex (10-�m-thick) of Tg1/CAG-CAT-Z (A), Tg2/CAG-CAT-Z (B),Tg3/CAG-CAT-Z (C), KI/CAG-CAT-Z (D), and CAG-�-Z positive control (E) mice. A–D: GABA-negative (excitatory) neurons (purple) had blueparticles, indicative of Cre-mediated recombination, while GABA-containing (inhibitory) neurons (brown, arrows) had no blue particles. E:LacZ-staining of both GABA-positive (arrows) and GABA-negative neurons in CAG-�-Z mice, confirming that the CAG-CAT-Z reportersystem can detect recombination in both excitatory and inhibitory neurons. Scale bar � 100 �m.

135CORTEX-SPECIFIC PROMOTER, CRE/LOXP RECOMBINATION

ethanol precipitation. Microinjection of the PAC-Cre-Afragment (1 ng/�l) into the pronuclei of zygotes ofC57BL/6 inbred mice generated two transgenic mouselines (Tg1 and Tg2), whereas microinjection of the PAC-Cre-B fragment (2 ng/�l) generated one line (Tg3).Southern blot hybridization using a Cre probe revealedthat Tg1 and Tg2 mice each have 2–3 copies of thetransgene, whereas Tg3 mice have 7–9 copies of thistransgene (data not shown). Transgenic mice were geno-typed by Southern hybridization and/or PCR with theCre primers (5�-ACCTGATGGACATGTTCAGGGATCG-3�and 5�-TCCGGTTATTCAACTTGCACCATGC-3�; productsize, 108 bp).

The experimental procedures and housing conditionsfor animals were approved by the Institute’s AnimalExperimental Committee and all animals were cared forand treated humanely in accordance with the Institu-tional Guidelines for Experiments using Animals.

Histology

Whole embryos and the brains of young (�P8) micewere fixed in 10% formalin in 0.1M sodium phosphatebuffer (PB) (pH 7.4) for 30 min on ice. Adult mice wereperfused intracardially with cold fixative. Brains wereembedded in 2% agarose in 0.1M PB and cut into 400-�m-thick sections with a Micro-slicer (Dosaka, Kyoto,

Japan). Whole embryos and brain slices were stainedovernight (CAG-CAT-Z mice) or for 6 h (R26R mice) inX-gal solution (5 mM K3FeCN6, 5 mM K4FeCN6, 2 mMMgCl2, 0.02% NP-40, 0.01% Na-deoxycholate, 1 mg/mlX-gal in 0.1M PB) at 37°C. In another set of experiments,mice were perfused and the brains were postfixed for4 h with 4% paraformaldehyde (PFA) in 0.1M PB andequilibrated in 30% sucrose overnight at 4°C. Eight or 10�m-thick sections were cut in a cryostat and mountedonto slides. Slides were stained with X-gal overnight(CAG-CAT-Z mice) or for 30 min (R26R mice) at 37°C.The tissue sections were subsequently incubated withantibodies against GABA (Sigma, St. Louis, MO; A2052,1:2,000) or GFAP (Progen GF12.24, 1:500). CAG-�-Zmice were generated by crossing CAG-CAT-Z mice with“deleter” mice to remove the “loxP-CAT-loxP” fragmentin the germline (Iwasato et al., 2000).

Cre immunohistochemistry was performed as de-scribed previously (Nakazawa et al., 2002) with somemodifications. Briefly, 50-�m-thick coronal sectionswere permeabilized by incubation in 50% ethanol/PBSfor 30 min and then incubated in 3% hydrogen peroxidein PBS for 10 min to inactivate endogenous peroxidases.Sections were preincubated in 10% normal goat serum(NGS) in TNB buffer (0.1M Tris 7.5, 0.15M NaCl, 0.5%blocking reagent (TSA kit; NEN, Boston, MA)) for 30 min

FIG. 6. Dorsal telencephalon-specific Cre-mediated recombination patterns in R26R reporter mice. A–F: X-gal/Nissl double staining ofcoronal sections (10-�m-thick) of adult Tg3/R26R mice. G: X-gal/anti-GABA-antibody/Nissl triple staining of the barrel cortex (10-�m-thick)of adult Tg3/R26R mice. Most LacZ-negative cells were GABAergic (brown, arrows). Cx, neocortex; Str, striatum; Am, amygdala; Th,thalamus; Hip, hippocampus; IC, inferior colliculus; Cb, cerebellum; BS, brainstem. Scale bar � 2 mm (A–F) and 62.5 �m (G).

136 IWASATO ET AL.

FIG. 7. Dorsal-telencephalon-specific Cre protein expression in three lines of transgenic mice. Coronal sections (50-�m-thick) of Tg1(7-month-old), Tg2 (7.5-month-old), Tg3 (7-month-old), homozygous KI (4-month-old), and wildtype (WT) mice were stained with an anti-Creantibody. Cre protein expression was observed in layers II–VI of the neocortex in all Tg1, Tg2, Tg3, and KI mice, although there were slightdifferences in expression patterns. In the hippocampus, there was higher expression in the dentate gyrus (DG) than in the CA1 and CA3in all four transgenic lines. In the amygdala, Tg1, Tg3, and KI mice had strong expression in the basolateral complex (BL) and basomedialnucleus (BM), but no expression in the MePD nucleus; whereas Cre expression in the Tg2 amygdala was weak. Or, oriens layer; rad, stratumradiatum; mol, molecular layer; pir, piriform cortex. Scale bar � 250 �m (left and middle columns), 500 �m (right column).

and incubated in anti-Cre antibody (BABCO, 1:2,000) inTNB buffer in a cold room overnight. After incubation inbiotinylated goat antirabbit IgG (H�L) (1:200)/3% NGSin TNB buffer at room temperature for 1 h, sections wereincubated in StrABC/HRP (Vectastain Elite PK-6101, Vec-tor, Burlingame, CA) for 1 h at room temperature andsubsequently in Biotinylated Tyramide Amplification Re-agent (1:50 in 1X Amplification diluent, TSA kit) for 10min at room temperature. After incubation in SA-HPR(1:100 in TNB: TSA kit) for 30 min at room temperature,signals were detected by 5-min treatment in DAB (SK-4100, Vector).

ACKNOWLEDGMENTS

We thank Dr. J.-I. Miyazaki for CAG-CAT-Z mice; Dr. P.Soriano for R26R mice; Drs. X.W. Yang and N. Heintz forthe pSv.RecA vector; Drs. K. Nakazawa and M. Takemurafor protocols of Cre and GFAP immunohistochemistry,respectively; Ms. N. Yoshida, Ms. Y. Onodera, Mr. Y.Taguchi, and Ms. M. Terasawa for technical assistance;Drs. H. Kanki and S. Nishimura for helpful discussionand/or comments on the manuscript; and the ResearchResources Center of BSI for help in mouse care.

LITERATURE CITED

Chan CH, Godinho LN, Thomaidou D, Tan SS, Gulisano M, ParnavelasJG. 2001. Emx1 is a marker for pyramidal neurons of the cerebralcortex. Cereb Cortex 11:1191–1198.

Gorski JA, Talley T, Qiu M, Puelles L, Rubenstein JL, Jones KR. 2002.Cortical excitatory neurons and glia, but not GABAergic neurons,are produced in the Emx1-expressing lineage. J Neurosci 22:6309–6314.

Gulisano M, Broccoli V, Pardini C, Boncinelli E. 1996. Emx1 and Emx2show different patterns of expression during proliferation anddifferentiation of the developing cerebral cortex in the mouse. EurJ Neurosci 8:1037–1050.

Guo H, Christoff JM, Campos VE, Jin XL, Li Y. 2000a. Normal corpuscallosum in Emx1 mutant mice with C57BL/6 background. Bio-chem Biophys Res Commun 276:649–653.

Guo H, Hong S, Jin XL, Chen RS, Avasthi PP, Tu YT, Ivanco TL, Li Y.2000b. Specificity and efficiency of Cre-mediated recombinationin Emx1-Cre knock-in mice. Biochem Biophys Res Commun 273:661–665.

Guo H, Mao C, Jin XL, Wang H, Tu YT, Avasthi PP, Li Y. 2000c.Cre-mediated cerebellum- and hippocampus-restricted gene mu-

tation in mouse brain. Biochem Biophys Res Commun 269:149–154.

Huang ZJ, Kirkwood A, Pizzorusso T, Porciatti V, Morales B, Bear MF,Maffei L, Tonegawa S. 1999. BDNF regulates the maturation ofinhibition and the critical period of plasticity in mouse visualcortex. Cell 98:739–755.

Ichise T, Kano M, Hashimoto K, Yanagihara D, Nakao K, Shigemoto R,Katsuki M, Aiba A. 2000. mGluR1 in cerebellar Purkinje cellsessential for long-term depression, synapse elimination, and motorcoordination. Science 288:1832–1835.

Iwasato T, Erzurumlu RS, Huerta PT, Chen DF, Sasaoka T, Ulupinar E,Tonegawa S. 1997. NMDA receptor-dependent refinement of so-matotopic maps. Neuron 19:1201–1210.

Iwasato T, Datwani A, Wolf AM, Nishiyama H, Taguchi Y, Tonegawa S,Knopfel T, Erzurumlu RS, Itohara S. 2000. Cortex-restricted dis-ruption of NMDAR1 impairs neuronal patterns in the barrel cor-tex. Nature 406:726–731.

Mayford M, Wang J, Kandel ER, O’Dell TJ. 1995. CaMKII regulates thefrequency-response function of hippocampal synapses for theproduction of both LTD and LTP. Cell 81:891–904.

Nakazawa K, Quirk MC, Chitwood RA, Watanabe M, Yeckel MF, SunLD, Kato A, Carr CA, Johnston D, Wilson MA, Tonegawa S. 2002.Requirement for hippocampal CA3 NMDA receptors in associativememory recall. Science 297:211–218.

Oberdick J, Smeyne RJ, Mann JR, Zackson S, Morgan JI. 1990. Apromoter that drives transgene expression in cerebellar Purkinjeand retinal bipolar neurons. Science 248:223–226.

Qiu M, Anderson S, Chen S, Meneses JJ, Hevner R, Kuwana E, PedersenRA, Rubenstein JL. 1996. Mutation of the Emx-1 homeobox genedisrupts the corpus callosum. Dev Biol 178:174–178.

Sakai K, Miyazaki J. 1997. A transgenic mouse line that retains Crerecombinase activity in mature oocytes irrespective of the cretransgene transmission. Biochem Biophys Res Commun 237:318–324.

Schlaggar BL, O’Leary DD. 1991. Potential of visual cortex to developan array of functional units unique to somatosensory cortex.Science 252:1556–1560.

Sharma J, Angelucci A, Sur M. 2000. Induction of visual orientationmodules in auditory cortex. Nature 404:841–847.

Simeone A, Gulisano M, Acampora D, Stornaiuolo A, Rambaldi M,Boncinelli E. 1992. Two vertebrate homeobox genes related to theDrosophila empty spiracles gene are expressed in the embryoniccerebral cortex. EMBO J 11:2541–2550.

Soriano P. 1999. Generalized lacZ expression with the ROSA26 Crereporter strain. Nat Genet 21:70–71.

Yang XW, Model P, Heintz N. 1997. Homologous recombination basedmodification in Escherichia coli and germline transmission intransgenic mice of a bacterial artificial chromosome. Nat Biotech-nol 15:859–865.

Yoshida M, Suda Y, Matsuo I, Miyamoto N, Takeda N, Kuratani S,Aizawa S. 1997. Emx1 and Emx2 functions in development ofdorsal telencephalon. Development 124:101–111.

138 IWASATO ET AL.