Using Flp-Recombinase to Characterize Expansion ofWnt1-Expressing Neural Progenitors in the Mouse

DEVELOPMENTAL BIOLOGY 201, 57– 65 (1998) ARTICLE NO. DB988971

Using Flp-Recombinase to Characterize Expansion of Wnt1-Expressing Neural Progenitors in the Mouse Susan M. Dymecki*,1 and Henry Tomasiewicz† *Department of Genetics, Harvard Medical School, 200 Longwood Avenue, Boston, Massachusetts 02115-5701; and †Department of Cell Biology, Emory University School of Medicine, Atlanta, Georgia 30322-3030

Here we demonstrate how a Flp recombinase-based tagging system can be used to link temporally distinct developmental events in the mouse. By directly following Flp-mediated DNA rearrangements we have analyzed the adult expansion of embryonic neural progenitors which transiently express the signaling factor Wnt1. We report Wnt1 promoter activity in embryonic cells that give rise to aspects of the adult midbrain, cerebellum, spinal cord, and dorsal root ganglia. These findings show that cells transiently expressing Wnt1 play more than an inductive role during early brain regionalization, giving rise to distinct adult brain regions as well as neural crest derivatives. Moreover, these results reveal two new features of the Flp–FRT system: First, Flp(F70L) can effectively recombine target sites (FRTs) placed in an endogenous locus in a variety of tissues in vivo, despite previous in vitro evidence of thermolability; and second, Flp(F70L) action can be predictably and tightly regulated in the mouse embryo, making it suitable for fate mapping applications. A further advantage of the Flp–FRT system is that marked lineages can ultimately be combined with germline mutations and deficiencies currently being generated using the Cre–loxP recombination system—in this way it should be possible to analyze mutant gene activities directly for their effect on cell fate. © 1998 Academic Press Key Words: Flp recombinase; Wnt1 descendants; fate map; neural development.

INTRODUCTION Site-specific recombinases can delete DNA lying between directly repeated target sites engineered into the genomes of plants, flies, or mice (Lakso et al., 1992; Orban et al., 1992; Kilby et al., 1993; Dymecki, 1996a) (Fig. 1a). Because such deletions are stably inherited by progeny cells without requiring continued recombinase activity, descendants of recombinase-expressing cells are genetically marked. Therefore, regardless of cell migration or morphogenetic movements, a link can be forged between two temporally separated developmental events— early and transient recombinase expression in embryonic progenitor cells and later contribution of progeny cells to various adult tissues (Fig. 1b). This fate-mapping paradigm (O’Gorman et al., 1991) has recently been exploited in the mouse using the Cre-loxP 1 To whom correspondence and reprint requests should be addressed at present address: Department of Genetics, Harvard Medical School, 200 Longwood Ave., Boston, MA 02115. Fax: (617) 432-4811. E-mail: [email protected].

0012-1606/98 $25.00 Copyright © 1998 by Academic Press All rights of reproduction in any form reserved.

system; Zinyk et al. (1998) mapped the adult fate of En2expressing neural progenitors which originated from the midbrain– hindbrain constriction during embryonic days 9 –12 (Zinyk et al., 1998). Here we show that the sitespecific recombinase Flp can be similarly employed to fate map embryonic cell populations; more specifically we have used Flp to analyze the expansion of those neural progenitors which transiently express the signaling molecule Wnt1 (Fig. 1b). Development of the Flp–FRT system to study cell deployment during embryogenesis should represent an additional valuable fate-mapping tool, as Flp-marked progenitor/progeny cells can be studied in the context of Cre-induced mutations (Gu et al., 1994; Rossant and Nagy, 1995) or deficiencies (Ramirez-Solis et al., 1995), yielding high-resolution studies of gene function. To track or collate the adult expansion of cells which transiently expressed Wnt1, we have used the tm1Cwr allele (Tomasiewicz et al., 1993) of Ncam as a Flp substrate (Fig. 1c). Exon 18 of tm1Cwr has been replaced with a b-actin::neo cassette flanked by directly repeated Flp re-

57

58

Dymecki and Tomasiewicz

combinase target sites (FRTs) such that Flp-mediated deletion of neo can be readily detected by PCR or Southern hybridization. This allele is also developmentally neutral when carried over the wild-type (wt) allele (Tomasiewicz et al., 1993), allowing the DNA rearrangement to serve as a benign lineage tag. Because the tm1Cwr harbors two FRT sites we refer to the allele as FRT2; we denote the product of Flp-mediated excisional recombination as D, for “deleted.” To track Wnt1 descendants, we generated mice expressing a Wnt1::Flp transgene and crossed them to FRT2 mice to yield compound mice (FRT2/1; Wnt1::Flp) in which Flp could act on the FRT2 target allele; following activation of Wnt1 expression and Flp, cell populations which had undergone excision were identified in dissected embryonic and adult tissues and their initial distribution and later expansion were determined.

MATERIALS AND METHODS Molecular Detection of Flp-Induced Mosaicism To analyze mosaicism, compound (FRT2/1; Flp) embryos and adults were first identified by PCR genotyping of yolk-sac, limb bud, and/or tail lysates essentially as described (Dymecki, 1996a). DNA isolates from various tissues of compound animals were then subjected to PCR or Southern blot detection of FRT2, D, and wt. Genotyping primers and amplification conditions were as follows: hACTB::Flp and Wnt1::Flp, primer 24, 59-CTAATG T T G T G G G A A A T T G G A G C - 39 , a n d 2 5 , 5 9 - C T C G A G GATAACTTGTTTATTGC-39, cycling at 94°C for 1 min, 66°C for 1 min, and 72°C for 1 min; FRT2, primer 91, 59CTGTGCTCGACGTTGTCACTGAAGC-39; and 92, 59CGGCAAGCAGGCATCGCCATGGGTC-39, cycling at 94°C for 1 min, 60°C for 1 min, and 72°C for 1 min. Primers to detect target gene deletions were 74, 59-CAGTGTTGCTAAGACCACTTAGT39; and 78, 59-CGGCTCCTGGCCAAAGGTGGAAG-39; amplification conditions were 40 cycles of 95°C for 1 min, 60°C for 1 min, and 72°C for 3.5 min. Southern blot analyses involved subjecting genomic DNA to BamHI digestion, gel electrophoresis, immobilization to nitrocellulose, and hybridization to radiolabeled DNA probe 1 (random primed (Feinberg and Vogelstein, 1993) to specific activity of 2–5 3 108 cpm/mg; see Fig. 1c). Quantification of radioactivity in specific bands was performed with a Molecular Dynamics PhosphorImager. Percentages of D alleles were calculated by subtracting background radioactivity from the activity measured in the FRT2 and D bands, adding the remaining activity and setting to 100%, and calculating the proportion of D alleles.

Transcript Detection For RNA detection, transgenic embryos were first identified by PCR of yolk-sac DNA (Dymecki, 1996a) and then subjected to whole mount in situ hybridization using single-strand digoxigeninUTP-labeled RNA probes (Wilkinson, 1992). For Northern blot and RT–PCR analyses, fresh tissue was homogenized in 6 M guanidinium isothiocyanate and RNA isolated using acid:phenol (Chomczynski and Sacchi, 1987). Total cellular RNA (20 mg) was either separated and assayed for hybridization to Flp sequence (probe 3) or reverse transcribed (1 mg), amplified, and detected by Southern

hybridization. Flp cDNAs were synthesized (Superscript preamplification system for first strand cDNA synthesis, GIBCO/BRL) using the gene-specific primer 43, 59-CGGATCAACGTTCTTAATATCGC-39; alpha 4 tubulin cDNAs, using random hexamers. Amplification (40 cycles of 95°C for 1 min, 52°C for 1 min, and 72°C for 1.5 min) was carried using the following primers: Wnt1::Flp, primer 29, 59-GACAGCAACCACAGTCGTCAG-39; and primer 72, 59-CCCAAGCAGGAATCAATTTC-39; alpha-4tubulin, 59-GAGGCGGCCGCTATAATGAAGCCACAGGTG-39 and 59-GAGGCGGCCGCACAGAGTCAACCAACTCAG-39 (Freytag and Geddes, 1992). Flp-specific products were detected by Southern hybridization to 32 P-labeled oligo 73 (59-GGAACAGCAATCAAGAGAGC-39).

Flp Transgenics To generate the Wnt1::Flp transgene, the 2-kb SalI fragment from pFlp(F70L) (Dymecki, 1996b) was inserted into the EcoRV site of pWEXP2 (Echelard et al., 1993). To produce transgenic mice, transgenes were purified away from plasmid sequences and injected into fertilized eggs from B6SJLF1 3 B6SJLF1 mice (Hogan et al., 1994).

RESULTS The Target Allele FRT2 Serves as an Effective Flp Substrate and Heritable Tag To show that FRT2 was an effective substrate for Flpmediated recombination and the proposed fate mapping experiments, we first introduced Flp by crossing to a transgenic mouse strain that expresses Flp under the control of the broadly expressed human b -actin promoter, TgN(hACTB::Flp) (Dymecki, 1996a). Southern analysis of DNA isolates from F1 compound animals (FRT2/1; hACTB::Flp) revealed D alleles in ;75% of the cells in heart and skeletal muscle, 60% in kidney, 30% in brain (Fig. 1d, lanes 7–10), and ;100% of tail cells (Fig. 1d, lane 13). These percentages were found to correlate directly with Flp transcript levels for each tissue, Flp mRNA being abundant in heart and skeletal muscle and low in liver (data not shown and (Dymecki, 1996a)). Notably, the efficiency of Flpmediated recombination of the single copy FRT2 locus (e.g., ;75% in skeletal muscle) was found to be better than our previously reported results (Dymecki, 1996a) using a multicopy transgene array as substrate (e.g., only ;30% in skeletal muscle). We also detected recombination in germ cell lineages, because stable D/FRT2 animals were produced in the F2 generation independent of further Flp activity (representative example Fig. 1d, lane 17). Moreover, both FRT2 alleles can undergo efficient recombination in the same cell; this conclusion stems from our detecting principally D alleles in tail cells (Fig. 1d, lanes 15 and 16) isolated from a number of F2 (FRT2/FRT2; hACTB::Flp) littermates. Together these results indicate that the FRT2 allele is accessible to recombinase activity in germline as well as somatic cells, and establish somatic in vivo Flp-mediated recombination between direct FRT repeats placed in an endogenous mouse gene.

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.

59

Using Flp to Study Wnt1 Descendants

FIG. 1. Strategy for using Flp-mediated deletions to map cell fate. (a) Recombination between direct FRT repeats results in deletion of intervening DNA. Black triangle represents a 48-bp FRT site (Dymecki, 1996b); black line, chromosomal DNA. Excised DNA is circularized and lost. (b) Binary scheme for targeting Flp-induced mosaicism to map fate of Wnt1-expressing cells. Each grid represents a mouse, and is cartooned developing from an early embryo (T1) to adult (T4). The left column illustrates a mouse line expressing Flp in the restricted and transient pattern of endogenous Wnt1 (small black squares); the middle column (light gray), a mouse line harboring an engineered FRT-containing target gene. Fate mapping involves crossing these two stocks to produce progeny heterozygous for the target and which carry Wnt1::Flp, column on right. Flp-induced mosaicism (1D /FRT ; blue/light gray) is detected in DNA isolates from dissected tissues and is used to link Wnt1 promoter activity in 1 progenitor cells (black squares) to a pattern of fates these cells and their descendants adopt in the adult (blue squares). (c) Southern hybridization and PCR strategies to distinguish FRT2, D, and wild-type (wt) alleles. BamHI (B) fragments containing exons 16 –19 (gray rectangles) and adjacent genomic regions are shown. Hybridization probes are represented by numbered lines; PCR oligonucleotides by half arrows. In the target FRT2 allele, exon 18 has been replaced by a 3 kb b-actin::neo gene flanked by two direct repeats of a 48-bp FRT site (black triangles) (Tomasiewicz et al., 1993). Upon BamHI digestion the FRT2, wt, and D alleles yield fragments of 5.9, 5.3, and 3 kb, respectively; PCR amplification using primers 74 and 78 yield products 3.6, 3, and 0.7 kb. The neo-specific primers, 91 and 92, amplify only FRT2. Although referred to as D for convenience, genetic nomenclature dictates that the recombined target allele be symbolized as Ncamtm3C1W. (d) Detection of mosaicism following broad expression of Flp using a hACTB::Flp transgene. Southern blot analysis using probe 1 on BamHI-digested genomic tissue DNA (10 mg). Lanes 1– 6, DNA isolated from brain, heart, muscle, kidney, liver, and testes, respectively, from an FRT2/1 mouse; lanes 7–13, DNA isolated from brain, heart, muscle, kidney, liver, testes, and tail, respectively, from an F1 compound (FRT2/1; hACTB::Flp) littermate; lanes 14 –18, DNA extracted from tails of intercross (FRT2/1; hACTB::Flp 3 FRT2/1; hACTB::Flp) progeny. Fragments derived from the FRT2, wt, and D alleles are indicated. Plus (1) sign denotes presence of the hACTB::Flp transgene; open arrow (.) denotes nonspecific hybridization independent of Flp. Deleted (D) alleles were not detected in the absence of Flp. The wt allele serves as a loading control. Lanes 15 and 16 reveal homozygous gene deletions in F2 progeny which inherited hACTB::Flp; lane 17, germline transmission of D allele independent of Flp.

Flp Expression Can Be Restricted to a Wnt1 Pattern Having confirmed that the FRT2 allele serves as a good Flp substrate in a wide range of cell types, we investi-

gated whether similar gene deletions could be produced in a restricted set of cells following targeted and transient expression of Flp in the embryo. Using transcriptional regulatory sequences from Wnt1 (Echelard et al., 1993,

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.

60

Dymecki and Tomasiewicz

FIG. 2. Transgenic mice express Flp in a Wnt1 pattern. (a) Structure of Wnt1::Flp transgene. SalI fragment from pWEXP2 containing Wnt1 genomic DNA (Echelard et al., 1993) is shown in black; open rectangles, Flp coding and SV40 polyadenylation sequence (Dymecki, 1996b; O’Gorman et al., 1991); dashed line, intron. Probes used for Northern and in situ hybridization of Flp mRNAs are indicated by bars 3 and

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.

61

Using Flp to Study Wnt1 Descendants

1994), we generated seven independent transgenic mouse lines; three showed Flp expression in the pattern of endogenous Wnt1 (Bally-Cuif et al., 1992; Wilkinson et al., 1987) and four were completely devoid of expression. Characterizing two of the positive Wnt1::Flp lines by whole mount in situ hybridization, Northern analyses, and RT–PCR showed that Flp transcripts were initially restricted to the embryonic mesencephalon, dorsal myelencephalon, and spinal cord, and by 11.5 dpc were also observed in the rhombic lip and caudal diencephalon (Fig. 2). Flp mRNA was not detected outside the developing nervous system nor in adult tissues. Expression of Flp under the control of these Wnt1 promoter/enhancer sequences therefore mirrored, with high fidelity, the characteristic spatial and temporal expression pattern of Wnt1; we speculate, however, that Flp mRNA may be present at lower levels than endogenous Wnt1 because a 5- to 10-fold longer histochemical development was required during in situ detection. These findings show that Flp can be predictably regulated by heterologous sequences without interference by sequences within the Flp transgene itself; consequently, most standard transgenic or knock-in methods can likely be used to express Flp.

Regulated Gene Deletion Following Transient and Restricted Wnt1::Flp Expression To determine whether Flp activity (like Flp mRNA) was restricted to the transcript-positive neural tube, we looked directly for Flp-mediated recombination events in embryos (10 dpc) from matings of FRT2/1 females with Wnt1::Flp males. Recombination was detected using flanking PCR primers which readily distinguish the deleted allele (D) from the target (FRT2) and wt alleles (primers 74 and 78, Fig. 1c). Paralleling the Wnt1::Flp expression pattern, gene deletion events were found to be restricted to neural tube-containing tissue isolated from compound (FRT2/1; Wnt1::Flp) em-

bryos (Figs. 3a and 3b); recombined D alleles were not observed in limb buds or viscera. These results provide a contrast to published observations with some Cre transgenics where the pattern of recombinase activity exceeds the transcript or reporter profile (Betz et al., 1996). The consequence in those cases has been recombination in unexpected tissues.

Wnt1 Descendants Expand to Populate Distinct Regions of the Adult Nervous System Next we assessed whether this Flp-induced mosaicism in the embryo could be detected later in the adult. Deleted alleles detected in compound (FRT2/1; Wnt1::Flp) adults should reflect recombination descendants of embryonic Wnt1::Flp-positive cells since Wnt1::Flp transcripts were not detected in adult tissues when assayed using a sensitive RT–PCR approach (Fig. 2g). Southern analysis of DNA isolates from compound adults demonstrated that recombined (D) alleles were present in neural tissue (Fig. 3c), including the cerebellum (lane 5, cortex; lane 6, deep nuclei and anterior vermal lobules of cortex), midbrain (lane 7), spinal cord (lane 8), and dorsal root ganglia (lane 9), with the greatest proportion of recombined alleles in the cerebellum. This distribution of D alleles was confirmed using a second independent Wnt1::Flp transgenic line (Fig. 3d), indicating that the observed mosaicism reflects Wnt1 promoter activity and not the chromosomal context of a given Wnt1::Flp transgene. Our detection of Wnt1-lineages in the adult spinal cord and dorsal root ganglia is consistent with avian-derived fate maps for dorsal neuroectoderm: Chimeric studies (Le Douarin, 1982) and vital dye labeling (Bronner-Fraser and Fraser, 1989) show that some dorsally located neural progenitors remain in the neural tube to form parts of the brain and spinal cord, while other cells differentiate as neural crest, migrating away from the neural tube to form a variety of cell types including dorsal root ganglia (DRG), sympathetic

4, respectively; bar 5, probe to detect endogenous Wnt1. PCR oligonucleotides are represented by half arrows. (b– e) Whole mount in situ hybridization to Wnt1::Flp (b,d) or endogenous Wnt1 (c,e) transcripts. Representative 8.5–9.0 and 11.5 dpc embryos are shown; open arrows (.) point to the upper rhombic lip. Flp transcripts required at least 5-fold longer histochemical development than Wnt1 during in situ detection. Identical results were obtained with both transgenic lines, TgN(Wnt1::Flp)5666 and TgN(Wnt1::Flp)6831. Mesencephalon (mes), metencephalon (met), myelencephalon (mye), spinal cord (sc). (f) RNA (20 mg) blot analysis using probe 3 to detect Flp mRNA isolated from Wnt1::Flp.6831 adult tissues. Upper panel, lanes: 1, control RNA from brain of hACTB::Flp transgenic mouse (Dymecki, 1996a); 2, olfactory bulb; 3, cerebrum and midbrain; 4, cerebellum; 5, brain stem; 6, spinal cord; 7, testes; 8, thymus; 9, spleen; 10, heart; 11, muscle; 12, small intestine; 13, large intestine; 14, liver; 15, lung; 16, kidney; 17, adrenal glands; 18, skin. Lower panel shows EtBr staining to document RNA loading. (g) RT–PCR analysis of Flp mRNA isolated from Wnt1::Flp.6831 whole embryos or dissected adult tissues. Top panel, Wnt1::Flp cDNAs were synthesized from total RNA (1 mg) using primer 43, amplified with 29 and 72, and products hybridized to 32P-labeled primer 73. Lanes: 1, Wnt1::Flp whole embryo (11.5 dpc); 2, 10-fold dilution of Wnt1::Flp embryo RNA into RNA isolated from wild-type (Flp-negative) embryonic fibroblasts (EF); 3, 100-fold dilution into EF RNA; 4, EF (faint signal is bleed-through from lane 3); 5, adult cerebrum and midbrain; 6, cerebellum; 7, brain stem; 8, spinal cord; 9, testes; 10, thymus; 11, spleen; 12, heart; 13, muscle; 14, small intestine; 15, large intestine; 16, liver; 17, lung; 18, kidney; 19, skin. Open arrow (.) indicates amplification from either unspliced mRNA or contaminating genomic DNA. Lower panel, negative image of ethidium-stained amplification products of alpha-4-tubulin (Freytag and Geddes, 1992) cDNAs synthesized from same samples as above. Analysis of line TgN(Wnt-1::Flp)5666 showed similar results (data not shown).

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.

62

Dymecki and Tomasiewicz

ganglia (SG), enteric neurons, Schwann cells surrounding peripheral nerves and neuromuscular junctions, adrenal chromaffin cells, facial cartilage and bone, and melanocytes in the skin. To test whether Wnt1 lineages give rise to neural crest derivatives other than the spinal ganglia identified here by Southern analysis, we screened the same DNA isolates using a more sensitive PCR assay and detected additional D alleles in muscle and skin (Fig. 3e, lanes 11 and 17). Since Wnt1::Flp transcripts were not detected in these tissues (Figs. 2f and 2g), we propose by analogy to the avian fate maps that the recombined D alleles detected in muscle reside in neural crest-derived Schwann cells concentrated at neuromuscular junctions, and those found in skin mark neural crest-derived melanocytes. We were unable to detect D alleles in intestinal tissue, which contains enteric neurons (Fig. 3c, lane 13; Figs. 3d and 3e, lanes 12 and 13), or in adrenal tissue (data not shown). One interpretation is that only some neural crest derivatives share a Wnt1-expressing precursor; it is, however, possible that the PCR assay is not sufficiently sensitive to identify these particular neural crest derivatives within bulk tissue samples. Control reactions demonstrated that this PCR assay was capable of detecting a single copy of D among at least 30,000 FRT2 alleles (data not shown). Given the previously reported results with Cre (Betz et al., 1996) mentioned above, a concern here is that some of the D alleles we have detected in compound (FRT2/1; Wnt1::Flp) adults may, in part, result from ectopic Flp expression. This is unlikely for two reasons. First, stochastic ectopic Flp expression in the embryo would result in different adult tissue profiles of D alleles with each cross. This was not observed because the D profile was quite reproducible. Second, whether driving expression using hACTB or Wnt1 sequences, the extent of Flp recombination always correlated directly with the detected levels of Flp mRNA. If ectopic Flp expression contributed significantly to the detected D alleles in the adult, then the transcript and recombination profiles would not be expected to match.

DISCUSSION Our findings suggest that Wnt1-expressing cells and their descendants expand to populate aspects of the adult midbrain, cerebellum, spinal cord, dorsal root ganglia (summarized in Fig. 4), and likely other neural neural crest derivatives such as melanocytes and Schwann cells. Indeed, while identifying the cis-regulatory sequences governing Wnt1 expression, Echelard et al. (1994) constructed a Wnt1::lacZ transgene that not only labeled cells of the dorsal neural tube, but also due to perdurance of b-galactosidase activity transiently (;24 h) marked the early migration of neural crest out of the neural tube to form spinal ganglia. Our mosaic analysis extends these findings into the adult. Because Wnt1 is broadly expressed in the midbrain neural plate, we were surprised to detect a lower percentage of D alleles in the adult midbrain compared with the cerebellum

(Fig. 3c, lanes 7 and 5, respectively; Fig. 3d, lanes 6 and 5). This may simply reflect the level of Flp transcripts in each primordium since the mid- to hindbrain ratio of Flp mRNA (Fig. 2b) was lower than the ratio of endogenous Wnt1 mRNA (Fig. 2c). It is also possible that a portion of Wnt1expressing cells or their descendants migrate out of the developing midbrain. Indeed, cell grafting and labeling experiments show that midbrain neural crest migrates into the facial mesenchyme and mandibular arch (Echelard et al., 1994; Noden, 1983; Trainor and Tam, 1995). Finding Wnt1 descendants in adult cerebellum was of particular interest, given that the cerebellar primordium lies on both sides of the mid– hindbrain constriction, and, therefore, arises from two classically defined neuromeres— the caudal mesencephalon and the rostral rhombencephalon (Alvarez Otero et al., 1993; Hallonet and Le Douarin, 1993; Hallonet et al., 1990; Martinez and Alvarado-Mallart, 1989). Results from recent gene expression studies together with chick/quail grafting experiments, however, support the hypothesis that during early neural tube regionalization (8.5–9.5 dpc in the mouse and stage HH10 in the chick) the interneuromeric boundary separating the primordia of mesencephalon and cerebellum actually coincides with the posterior boundary of Otx2 expression, which in fact lies rostral to the so-called “mid– hindbrain constriction” (Ang et al., 1994; Millet et al., 1996). It is not until later (10.5 dpc in the mouse and stage HH18 in the chick) that the caudal limit of Otx2 expression (Ang et al., 1994), as well as the Wnt1-positive ring (Bally-Cuif et al., 1992; Wilkinson et al., 1987), parallel the mid– hindbrain constriction. The cerebellar primordium, therefore, appears to extend (rostral to caudal) from the posterior boundary of Otx2 expression (Millet et al., 1996), through the entire length of the metencephalon, to include at least part of rhombomere 2 (Marin and Puelles, 1995). Within this newly defined primordium, Wnt1 transcripts are first seen in 3– 4 rows of cells lying immediately caudal to the posterior limit of Otx2 expression (Bally-Cuif et al., 1995) and in the dorsal edge of rhombomere 2 (8.5–14.5 dpc); later within this primordium, Wnt1 mRNA becomes more pronounced, being expressed in the upper rhombic lip of the metencephalon (11.5–14.5 dpc) (Bally-Cuif et al., 1992, 1995; Wilkinson et al., 1987) (arrowhead in Figs. 2d and 2e). As described earlier, a similar pattern of Flp expression was seen in each of our Wnt1::Flp transgenic strains. Precursors to most cerebellar neurons arise throughout this entire cerebellar field; granule layer neurons, in contrast, undergo a restricted developmental scheme, arising solely from the upper rhombic lip, seemingly coincident with Wnt1-expressing cells. The upper rhombic lip (also referred to as the germinal trigone) has been shown to give rise to a pool of mitotic precursor cells which stream rostrally up the lip and onto the surface of the forming cerebellar cortex where they establish a secondary cerebellar neuroepithelium called the external granular layer (Altman and Bayer, 1997; Alvarez Otero et al., 1993; Hallonet and Le Douarin, 1993; Hatten and Heintz, 1995). Cells of

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.

Using Flp to Study Wnt1 Descendants

63

FIG. 3. Wnt1::Flp-induced mosaicism detected in embryonic neural tube and adult brain. (a) PCR identification of compound (FRT2/1; Wnt1::Flp) 10 dpc embryos from two independent crosses. Upper panel, detection of FRT2 alleles using neo-specific primers 91 and 92 (see Fig. 1c); lower panel, detection of Wnt1::Flp alleles using primers 24 and 25 (see Fig. 2a). Plus (1) signs mark compound embryos. Control lanes: 1, no DNA; 2, litter 2 parental NcamFRT2/1 tail DNA; 3, litter 1 parental FRT2/1 tail DNA; 4, parental Wnt1::Flp.6831 tail DNA. A neo tag present in exon 4 of the pWEXP2 expression cassette (Echelard et al., 1993), and therefore our Wnt1::Flp transgene, accounts for the amplification seen in lane 4, upper panel. (b) PCR detection of recombined alleles is restricted to neural tube-containing tissue. Amplification using primer pair 74/78 yields products of 3.6, 3, and 0.7 kb, for FRT2, wt, and D, respectively. Upper panel, PCR assay on neural tube tissue dissected from embryos in a. Products reflecting wt and D are marked; the larger product (3.6 kb) representing FRT2 is not efficiently amplified under these conditions and therefore not shown. Lower panel, PCR assay on limb buds and viscera. Control lanes: 1, no DNA; 2, litter 2 parental FRT2/1 tail DNA; 3, litter 1 parental FRT2/1 tail DNA; 4, tail DNA from positive control compound (FRT2/1; hACTB::Flp) mouse. Sagittal illustration of embryo is on left. (c) Detection of Flp-induced mosaicism in DNA isolates from a compound adult (FRT2/1; Wnt1::Flp.6831). Southern blot analysis using probe 1 on BamHI-digested genomic tissue DNA (10 mg) showing D alleles (3-kb fragment) only. Lanes: 1, positive control DNA isolated from brain of compound mouse (FRT2/1; hACTB::Flp; as in Fig. 1d, lane 7); 2, negative control from brain of FRT2/1 sibling; 3, olfactory bulb (ob); 4, cerebral hemispheres, thalamus, and hypothalamus (ch); 5, cerebellar (cb) cortex (c); 6, region containing cerebellar deep nuclei (dn); 7, midbrain (mb); 8, spinal cord (sc); 9, dorsal root ganglia (drg); 10, testes; 11, heart; 12, muscle; 13, small intestine; 14, liver; 15, lung; 16, kidney; 17, skin; 18, thymus; 19, spleen; 20, tail. (d) Southern detection of Flp-induced mosaicism using an independently derived compound adult (FRT2/1; Wnt1::Flp.5666). Lanes: 1, positive control as in c; 2, negative control as in c; 3, blank; 4, ch; 5, cb (pooled cortex and deep nuclei); 6, mb; 7, brain stem (bs); 8, sc plus drg; 9, spleen; 10, heart; 11, muscle; 12, small intestine; 13, large intestine; 14, liver; 15, lung; 16, kidney; 17, skin; 18, uterus. (e) PCR/Southern assay detects additional D alleles in muscle (m) and skin (sk). DNA (0.5 mg) from d was amplified using primers 74/78, followed by Southern hybridization with 32P-labeled FRT-specific probe 2 (see Fig. 1c). Inadvertently, brain stem (bs) DNA was not added to PCR reaction in lane 7, therefore repeated in lane 19. The 0.7-kb product specific to D is marked on the right. Samples from tail, olfactory bulb, adrenal, and ovary were negative by this assay (data not shown). Crosses performed using the independent line, TgN(Wnt1::Flp)6831, showed a similar pattern of mosaicism by PCR (data not shown). Sagittal illustration of adult mouse brain is shown on left, with dashed lines representing planes of dissection.

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.

64

Dymecki and Tomasiewicz

FIG. 4. Schematic of adult neural tissues populated by Wnt1 descendants. Left, embryonic Wnt1 expression pattern marked in gray; right, adult brain regions populated by Wnt1 descendants shaded gray. Mesencephalon (mes), metencephalon (met), myelencephalon (mye), olfactory bulb (ob), cerebral hemispheres (ch), midbrain (mb), cerebellar cortex (cb), spinal cord (sc), and dorsal root ganglia (drg).

this layer then migrate inward to form an internal layer of mature granule neurons. Because the rhombic lip appears to be the major Wnt1-positive (and Wnt1::Flp-positive) region within the cerebellar primordium, it is possible that the Wnt1 descendants detected in the adult cerebellar cortex may in part contribute to granular layer neurons.

mouse development, we are currently producing an in situ reporter that can be irreversibly “switched-on” by Flp so that these Wnt1-expressing progenitors and their descendant cells can be visualized at single cell resolution. Not only will this reporter strain provide high-resolution maps of recombination descendants, but it will also permit a more precise determination of the in vivo kinetics of Flp recombination in the embryo. Although certainly more powerful, such a reporter is less straightforward because its permanent activation by Flp not only depends on a sitespecific recombination event (as exploited here), but also on promoter/enhancer sequences which must constitutively drive reporter expression as cell differentiation and development proceeds. While the Cre–loxP recombination system functions well in the mouse (Kilby et al., 1993; Zinyk et al., 1998), we suggest that the Flp–FRT system is a viable and perhaps advantageous choice for fate-mapping studies in mice for two reasons: First, its dose–response properties allow tight regulation of recombinase activity at the transcriptional level, as shown here, and second, it should be possible to combine Flp-marked lineages with the many mutations (Gu et al., 1993; Rossant and Nagy, 1995) and deficiencies (Ramirez-Solis et al., 1995) currently being generated using Cre. In this way, gene activities can be assessed directly for their effect on cell fate, providing much higher resolution studies of gene function and cell-fate decisions.

ACKNOWLEDGMENTS CONCLUSIONS In summary, we have shown how the Flp–FRT recombination system can be tightly regulated in the mouse embryo to produce restricted and heritable gene deletions in response to transient recombinase expression. This property makes Flp suitable for fate mapping progenitor populations in the mouse, similar to the Flp-based methodologies originally established in Drosophila (Golic and Lindquist, 1989; Struhl and Basler, 1993). Moreover, we have shown effective Flp action at an endogenous locus in somatic tissues in vivo despite previous in vitro evidence that Flp is thermolabile (Buchholz et al., 1996). We then used this system to track a population of Wnt1 descendants into the adult midbrain, cerebellum, spinal cord, and dorsal root ganglia. These data suggest that Wnt1-expressing cells not only play a transient inductive role during early brain regionalization as previously established (Bally-Cuif and Wassef, 1995; Joyner, 1996), but also that these cells expand to populate distinct adult brain regions and at least some neural crest derivatives. Identification of Wnt1 descendants in mouse cerebellum, in particular, suggests a role for the Wnt1expressing cells, themselves, in the growth and regionalization of the cerebellum. Having demonstrated that the Flp–FRT system can be used to follow the expansion of progenitor cells during

We thank Andrew McMahon for plasmids pWEXP2, pSP72MT34DXhoISalCOOH, S. Hachenberg for technical assistance, A. Pinder for oligonucleotide synthesis, A. Fire, C. M. Fan, M. Halpern, N. Seeds, C. Thompson, and anonymous reviewers for helpful suggestions, and D. Threadgill for “introducing” us to each others’ work. Transgenic mice were generated in conjunction with the National Institute of Child Health and Human Development transgenic mouse development facility, in support of DNX, Inc. This work was supported by grants from the NIH (S.M.D.; H. T. and J. Wood), the John Merck Fund (S.M.D.), and the Helen Hay Whitney Foundation (S.M.D.).

REFERENCES Altman, J., and Bayer, S. A. (1997). “Development of the Cerebellar System.” CRC Press, Boca Raton, FL. Alvarez Otero, R., Sotelo, C., and Alvarado-Mallart, R.-M. (1993). Chick/quail chimeras with partial cerebellar grafts: An analysis of the origin and migration of cerebellar cells. J. Comp. Neurol. 333, 597– 615. Ang, S.-L., Conlon, R. A., Jin, O., and Rossant, J. (1994). Positive and negative signals from mesoderm regulate the expression of mouse Otx2 in ectoderm explants. Development 120, 2979 –2989. Bally-Cuif, L., Alvarado-Mallart, R.-M., Darnell, D., and Wassef, M. (1992). Relationship between Wnt-1 and En-2 expression domains during early development of normal and ectopic metmesencephalon. Development 115, 999 –1009.

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.

65

Using Flp to Study Wnt1 Descendants

Bally-Cuif, L., Cholley, B., and Wassef, M. (1995). Involvement of Wnt-1 in the formation of the mes/metencephalic boundary. Mech. Dev. 53, 23–34. Bally-Cuif, L., and Wassef, M. (1995). Determination events in the nervous system of the vertebrate embryo. Curr. Opin. Genet. Dev. 5, 450 – 458. Betz, U. A. K., Voßhenrich, C. A. J., Rajewsky, K., and Muller, W. (1996). Bypass of lethality with mosaic mice generated by CreloxP-mediated recombination. Curr. Biol. 6, 1307–1316. Bronner-Fraser, M., and Fraser, S. E. (1989). Developmental potential of avian trunk neural crest cells in situ. Neuron 3, 755–766. Buchholz, F., Ringrose, L., Angrand, P.-O., Rossi, F., and Stewart, A. F. (1996). Different thermostabilities of FLP and Cre recombinases: Implications for applied site-specific recombination. Nucleic Acids Res. 24, 4256 – 4262. Chomczynski, P., and Sacchi, N. (1987). Single-step method of RNA isolation by acid guanidinium thiocyanate–phenol– chloroform extraction. Anal. Biochem. 162, 156 –159. Dymecki, S. M. (1996a). Flp recombinase promotes site-specific DNA recombination in embryonic stem cells and transgenic mice. Proc. Natl. Acad. Sci. USA 93, 6191– 6196. Dymecki, S. M. (1996b). A modular set of Flp, FRT and lacZ fusion vectors for manipulating genes by site-specific recombination. Gene 171, 197–201. Echelard, Y., Epstein, D. J., St-Jacques, B., Shen, L., Mohler, J., McMahon, J. A., and McMahon, A. P. (1993). Sonic Hedgehog, a member of a family of putative signaling molecules, is implicated in the regulation of CNS polarity. Cell 75, 1417–1430. Echelard, Y., Vassileva, G., and McMahon, A. (1994). cis-acting regulatory sequences governing Wnt-1 expression in the developing mouse CNS. Development 120, 2213–2224. Feinberg, A., and Vogelstein, B. (1993). A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132, 6 –13. Freytag, S., and Geddes, T. (1992). Reciprocal regulation of adipogenesis by Myc and C/EBPa. Science 256, 379 –382. Golic, K. G., and Lindquist, S. (1989). The FLP recombinase of yeast catalyzes site-specific recombination in the Drosophila genome. Cell 59, 499 –509. Gu, H., Marth, J. D., Orban, P. C., Mossmann, H., and Rajewsky, K. (1994). Deletion of a DNA polymerase b gene segment in T cells using cell type-specific gene targeting. Science 265, 103–106. Hallonet, M. E. R., and Le Douarin, N. M. (1993). Tracing neuroepithelial cells of the mesencephalic and metencephalic alar plates during cerebellar ontogeny in quail– chick chimaeras. Eur. J. Neurosci. 5, 1145–1155. Hallonet, M. E. R., Teillet, M.-A., and Le Douarin, N. M. (1990). A new approach to the development of the cerebellum provided by the quail– chick marker system. Development 108, 19 –31. Hatten, M. E., and Heintz, N. (1995). Mechanisms of neural patterning and specification in the developing cerebellum. Annu. Rev. Neurosci. 18, 385– 408. Hogan, B., Beddington, R., Costantini, F., and Lacy, E. (1994). “Manipulating the Mouse Embryo.” Cold Spring Harbor Laboratory Press, Plainview, NY. Joyner, A. L. (1996). Engrailed, Wnt and Pax genes regulated midbrain– hindbrain development. Trends Genet. 12, 15–20.

Kilby, N. J., Snaith, M. R., and Murray, J. A. H. (1993). Site-specific recombinases: Tools for genome engineering. Trends Genet. 9, 413– 421. Lakso, M., Sauer, B., Mosinger, B., Jr., Lee, E. J., Manning, R. W., Yu, S. H., et al. (1992). Targeted oncogene activation by site-specific recombination in transgenic mice. Proc. Natl. Acad. Sci. USA 89, 6232– 6236. Le Douarin, N. M. (1982). “The Neural Crest.” Cambridge Univ. Press, Cambridge. Marin, F., and Puelles, L. (1995). Morphological fate of rhombomere in quail/chick chimeras: A segmental analysis of hindbrain nuclei. Eur. J. Neurosci. 7, 1714 –1738. Martinez, S., and Alvarado-Mallart, R.-M. (1989). Rostral cerebellum originates from the caudal portion of the so-called ‘mesencephalic’ vesicle: A study using chick/quail chimeras. Eur. J. Neurosci. 1, 549 –560. Millet, S., Bloch-Gallego, E., Simeone, A., and Alvarado-Mallart, R.-M. (1996). The caudal limit of Otx2 gene expression as a marker of the midbrain/hindbrain boundary: A study using in situ hybridization and chick/quail homotopic grafts. Development 122, 3785–3797. Noden, D. M. (1983). The role of neural crest in patterning of avian cranial skeletal, connective and muscle tissues. Dev. Biol. 96, 144 –165. Orban, P. C., Chui, D., and Marth, J. D. (1992). Tissue- and site-specific DNA recombination in transgenic mice. Proc. Natl. Acad. Sci. USA 89, 6861– 6865. O’Gorman, S., Fox, D. T., and Wahl, G. M. (1991). Recombinasemediated gene activation and site-specific integration in mammalian cells. Science 251, 1351–1355. Ramirez-Solis, R., Liu, P., and Bradley, A. (1995). Chromosome engineering in mice. Nature 378, 720 –724. Rossant, J., and Nagy, A. (1995). Genome engineering: the new mouse genetics. Nat. Med. 1, 592–594. Struhl, G., and Basler, K. (1993). Organizing activity of wingless protein in Drosophila. Cell 72, 527–540. Tomasiewicz, H., Ono, K., Yee, D., Thompson, C., Goridis, C., Rutishauser, U., and Magnuson, T. (1993). Genetic deletion of a neural cell adhesion molecule variant (N-CAM-180) produces distinct defects in the central nervous system. Neuron 11, 1163–1174. Trainor, P. A., and Tam, P. P. L. (1995). Cranial paraxial mesoderm and neural crest cells of the mouse embryo: Co-distribution in the craniofacial mesenchyme but distinct segregation in branchial arches. Development 121, 2569 –2582. Wilkinson, D. G. (1992). “In Situ Hybridization: A Practical Approach.” IRL, Oxford. Wilkinson, D. G., Bailes, J. A., and McMahon, A. P. (1987). Expression of the proto-oncogene int-1 is restricted to specific neural cells in the developing mouse embryo. Cell 50, 79 – 88. Zinyk, D., Mercer, E., Harris, E., Anderson, D., and Joyner, A. (1998). Fate mapping of the mouse midbrain– hindbrain constriction using a site-specific recombination system. Curr. Biol. 8, 665– 668. Received for publication April 9, 1998 Revised June 1, 1998 Accepted June 1, 1998

Copyright © 1998 by Academic Press. All rights of reproduction in any form reserved.