Generation of Lrp6 conditional gene-targeting mouse line for modeling and dissecting multiple birth defects/congenital anomalies


  • Chengji J. Zhou,

    Corresponding author
    1. Department of Cell Biology and Human Anatomy, University of California, Davis, School of Medicine; Institute for Pediatric Regenerative Medicine, Shriners Hospitals for Children-Northern California, Sacramento, California
    • Room 602B, Shriners Hospital, 2425 Stockton Blvd., Sacramento, CA 95817
    Search for more papers by this author
  • Ya-Zhou Wang,

    1. Department of Cell Biology and Human Anatomy, University of California, Davis, School of Medicine; Institute for Pediatric Regenerative Medicine, Shriners Hospitals for Children-Northern California, Sacramento, California
    Search for more papers by this author
  • Takashi Yamagami,

    1. Department of Cell Biology and Human Anatomy, University of California, Davis, School of Medicine; Institute for Pediatric Regenerative Medicine, Shriners Hospitals for Children-Northern California, Sacramento, California
    Search for more papers by this author
  • Tianyu Zhao,

    1. Department of Cell Biology and Human Anatomy, University of California, Davis, School of Medicine; Institute for Pediatric Regenerative Medicine, Shriners Hospitals for Children-Northern California, Sacramento, California
    Search for more papers by this author
  • Lanying Song,

    1. Department of Cell Biology and Human Anatomy, University of California, Davis, School of Medicine; Institute for Pediatric Regenerative Medicine, Shriners Hospitals for Children-Northern California, Sacramento, California
    Search for more papers by this author
  • Kai Wang

    1. Department of Cell Biology and Human Anatomy, University of California, Davis, School of Medicine; Institute for Pediatric Regenerative Medicine, Shriners Hospitals for Children-Northern California, Sacramento, California
    Search for more papers by this author


Lrp6 is a key coreceptor in the canonical Wnt pathway that is widely involved in tissue/organ morphogenesis. We generated a loxP-floxed Lrp6 mouse line. Crossing with a general Cre deleter, we obtained the Lrp6-floxdel mice, in which the loxP-floxed exon 2 of Lrp6 gene has been deleted ubiquitously. The homozygotes of Lrp6-floxdel mice reproduced typical defects as seen in the conventional Lrp6-deficient mice, such as defects in eye, limb, and neural tube, and die around birth. We also found new phenotypes including cleft palate and agenesis of external genitalia in the Lrp6-floxdel mice. In addition, the Lrp6-deficient embryos are known to be defective in other systems and internal organs including the heart and brain. Thus, by selectively crossing with a lineage-specific or inducible Cre mouse line, the Lrp6 conditional gene-targeting mice will allow us to model specific types of birth defects for mechanism and prevention studies. Developmental Dynamics 239:318–326, 2010. © 2009 Wiley-Liss, Inc.


Lrp6, a single-span transmembrane protein, together with its vertebrate subfamily member Lrp5 and the fly ortholog Arrow, is known as the coreceptor for canonical Wnt/β-catenin signaling and has key roles in development and disease (He et al.,2004; Logan and Nusse,2004). Lrp6 has functional redundancy with Lrp5 in gastrulation (Kelly et al.,2004). Human Lrp5 mutations are associated with familial osteoporosis, high bone density syndromes, and ocular disorders (Gong et al.,2001), and Lrp5-deficient mice are viable and fertile (Kato et al.,2002). In contrast, Lrp6 mutant mice are embryonic or perinatal lethal with broad defects (Pinson et al.,2000), including abnormalities in embryonic brain, neural tube, and limbs. In addition, mutations in the human Lrp6 gene have recently been implicated in late-onset Alzheimer's disease, and familial coronary artery disease associated with diabetes and osteoporosis (De Ferrari et al.,2007; Mani et al.,2007). These implications suggest a crucial but pleiotropic role of Lrp6 in developmental disorders or genetic disease. We have previously demonstrated that Lrp6 mutants have multiple defects in forebrain structures and may be implicated in the pathogenesis of lissencephaly, schizophrenia, and related mental retardation (Zhou et al.,2004a,b,2006). Recently we have further found that Lrp6 is responsible for ocular birth defects including microphthalmia, coloboma, and dorsoventral neuroretina patterning defects (Zhou et al.,2008). Additionally, Lrp6 mutants have severe defects in other systems, particularly in embryonic hearts (Bryja et al.,2009; Song et al.,2009b).

Because of the significance of Lrp6 in organogenesis and the association of Lrp6 mutants with multiple birth defects, we have for the first time, generated a loxP-floxed Lrp6 mutant mouse line for conditional gene-targeting analyses. We obtained the Lrp6-floxdel mice with the CMV-Cre to delete the Lrp6 gene ubiquitously. We show the evidence of the identical phenotypes in external organs of the homozygous Lrp6-floxdel mice as seen in conventional Lrp6 mutant mice. We also report new phenotypes including the cleft palate and agenesis of external genitalia in Lrp6-floxdel mice. Combined with tissue-specific or inducible Cre mouse lines, the conditional Lrp6 mouse line will allow us to model and dissect specific types of birth defects for mechanisms and prevention studies.


Generation of the loxP-floxed Lrp6 Mice and Lrp6-floxdel Mice With Characteristic External Organ Defects

We designed a targeting construct by flanking exon 2 of mouse Lrp6 with two loxP sites (Fig. 1A), which will cause a frameshift in the mRNA reading frame by Cre deletion. Crossing with CMV-Cre mice, we obtained the exon 2 deleted Lrp6-floxdel mice. We detected neither exon 2 mRNAs nor Lrp6 proteins by Northern blot, reverse transcriptase-polymerase chain reactions (RT-PCR), and Western blot in the homozygotes (KO) of Lrp6-floxdel embryos accordingly (Fig. 1D–F). We selectively examined Lrp6-floxdel KOs (Lrp6floxdel/floxdel or floxdel−/−) and their littermate controls (heterozygous Lrp6floxdel/+ or wild-type Lrp6+/+) at embryonic day (E) 11.5 and E16.5. The external organ defects of the mutant embryos were identical to what was observed in the conventional knockout Lrp6βgeo/βgeo mice that were generated by a gene-trap approach (Pinson et al.,2000; Zhou et al.,2008). The external organ defects (Fig. 2) include a full penetrance of tail truncation, limb disorders (milder in forelimb and severer in hindlimb), ocular defects, and spina bifida. Exencephaly was also detectable in some but not all homozygous embryos (Fig. 2C, k4). By crossing heterozygotes of Lrp6-floxdel and Lrp6-βgeo mice, the same defects were seen in the compound mutant embryos (Lrp6floxdel/βgeo; Fig. 2E). These results demonstrate the successful generation of both Lrp6-flox mice and Lrp6-floxdel mice through the conditional gene-targeting approach in this study.

Figure 1.

Generation of the Lrp6 conditional gene-targeting mice and Lrp6-floxdel mice. A: Schematic illustration of the Lrp6 conditional gene-targeting construct, wild-type allele, the exon 2-floxed allele, and the exon 2-deleted floxdel allele. The Frt-flanked Neo cassette was deleted by crossing with ROSA-Flp1 mice. Subsequently, the loxP-flanked exon 2 was deleted by crossing with CMV-Cre mice, and thus generating the Lrp6-floxdel allele. The positions of polymerase chain reaction (PCR) genotyping primers F1, F2, and R1 were indicated by arrows. B: PCR genotyping of the Lrp6-flox mice by primer pair F1/R1 revealed a 232-bp wild-type band and a 411-bp flox band. C: PCR genotyping of the Lrp6-floxdel mice revealed a 232-bp wild-type band and a 413-bp floxdel band. D: Northern blots with the digoxigenin (DIG) -labeled exon 2 cRNA probes revealed that a 9 kb band of Lrp6 transcripts in the wild-type (WT) and heterozygote (Het) but not in the homozygote (KO) of Lrp6-floxdel samples. Note the significantly reduced amount (approximately 50% compared with the wild-type) of Lrp6 RNAs in the heterozygote. The left panel shows the equal loading amounts and the approximate RNA size indicated by 18s and 28s ribosomal RNA bands. E: Reverse transcriptase PCR (RT-PCR) confirmed the deletion of exon 2 transcripts in Lrp6-floxdel homozygotes. The negative controls were performed without reverse transcription step (RT−) for each sample. F: Western blots detected a 179-kDa Lrp6 protein in either wild-type or heterozygous mice but not in the Lrp6-floxdel homozygotes (KO1 and KO2). The positive control of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) immunoblots was performed on the same membrane.

Figure 2.

Reproduced external organ defects in the homozygous Lrp6floxdel/floxdel or compound heterozygous Lrp6floxdel/βgeo mice as seen in conventional Lrp6βgeo/βgeo mice. A: Sagittal body view of a homozygous Lrp6floxdel/floxdel mouse embryo at embryonic day (E) 11.5 with external defects including small eye (arrow), hypoplastic limb buds especially the hindlimb bud (black dashed line), no tail, and open caudal neural tube (asterisk). B: Sagittal body view of a littermate control or wild-type (wt) mouse embryo at E11.5 with intact developing eye (e), forelimb bud (fl), hindlimb bud (hl), tail (t), and closed caudal neural tube (asterisk). C: Sagittal body views of four homozygous Lrp6floxdel/floxdel mouse embryos (k1∼k4) with full penetrance of spina bifida (sb), incomplete penetrance of exencephaly (ex in k4), varied severity of eye defects (arrows) and limb defects. D: Sagittal body view of a littermate control mouse embryo at E16.5. E: The compound heterozygote of Lrp6floxdel/βgeo embryos exhibited the identical defects in external organs as seen in either Lrp6floxdel/floxdel or Lrp6βgeo/βgeo homozygous embryos.

Ocular Defects in Lrp6-floxdel Homozygotes

The Lrp6-floxdel homozygous embryos exhibited severe ocular defects including open eyes, microphthalmia (small eyes), and ocular coloboma (missing retina or defective retina closure in the ventral eye; Fig. 3). This is identical to what we reported in the conventional Lrp6-βgeo mutant eyes (Zhou et al.,2008); the ocular coloboma in the ventral eye likely being caused by the defective dorsoventral neuroretina patterning, particularly the absence of Bmp4 and Raldh1 gene expression in the dorsal neuroretina and the dorsal shifting of ventral neuroretina domains during early eye development of the Lrp6-floxdel mice (data not shown). The varied severity of microphthalmia (Fig. 3B–D) suggests the influence of environmental factors and/or genetic backgrounds in regulation of early eye development. The open eyes suggest a role of Lrp6 in periocular tissue morphogenesis.

Figure 3.

Ocular defects of Lrp6-floxdel mice at embryonic day (E) 16.5. A: Sagittal body view of a normal eye at E16.5. The retina is evident by retinal pigment epithelium. The surface epithelium covering the lens and retina is still closed at this age. B–D: Sagittal body views of three E16.5 mutants with full penetrance of open eyes (dashed circles) and ocular coloboma in the ventral eye region (arrows), and the varied severity of microphthalmia (mild in k1, medium in k2, and most severe in k3).

Limb and Digital Defects in Lrp6-floxdel Homozygotes

The Lrp6-floxdel homozygous embryos exhibited severe limb defects. Particularly, the mutant forelimbs showed a full penetrance of defects in digits with oligodactyly (frequently three to four digits) and a low penetrance of polydactyly (Fig. 4A). The forearms of the mutants were twisted and/or shorter in length compared with the normal arms of the littermate controls. The digital tendons, which were normally located only in the ventral site of wild-type digits (Fig. 4B), were duplicated in the dorsal site of the Lrp6-floxdel KOs (Fig. 4C). The mutant hindlimb defects were more severe with a complete absence of the distal structures (Fig. 5). Some mutants also showed a typical sirenomelus in which the legs were fused together proximally (Fig. 5F). Similar phenotypes were also observed in the conventional Lrp6βgeo/βgeo or compound Lrp6floxdel/βgeo mutant embryos (Fig. 2E, and data not shown).

Figure 4.

Forelimb and digital defects of Lrp6-floxdel mice at embryonic day (E) 16.5. A: Dorsal view of an intact forelimb showing a normally developed forearm (asterisk) and five digits (t, g, m, r, l, equally to thumb, grooming, middle, ring, and little fingers in humans). Varied defects in the forearm (asterisks) and digits are seen in four mutants (k1–k4) with full penetrance of oligodactyly and low penetrance of polydactyly (arrow). Number 1–4 only indicate the quantity but not the identity of digits in the mutants. B,C: Hematoxylin and eosin-stained transverse sections of the forelimb digits show the ventral tendon (arrows) and the duplicated tendon in the dorsal region (arrowheads in C) of the mutant at E16.5.

Figure 5.

Severe defects in hindlimb, external genitalia, and caudal neural tube of Lrp6-floxdel mice. A,B: Sagittal (A) and ventral (B) caudal body views of a control embryo at E16.5 show normally developed hindlimbs, tail, and external genitalia (arrow in B). CF: Accordingly sagittal (C,D,F) or ventral (E) caudal body views of the mutants show absence of tail, spina bifida (sb), hypoplastic hindlimbs (asterisks) that have no digits, and complete absence of external genitalia (arrows indicated in E,F). The mutant k4 also shows a typical sirenomelus in which the legs are fused together (asterisk in F).

Agenesis of External Genitalia in Lrp6-floxdel Homozygotes

The Lrp6-floxdel homozygous embryos showed severe defects in the urogenital system. Specifically, the external genitalia were completely absent in the mutant embryos (Fig. 5). The agenesis of external genitalia was not reported in the conventional Lrp6βgeo/βgeo embryos. Therefore, we also examined both Lrp6βgeo/βgeo and Lrp6floxdel/βgeo embryos and found the identical agenesis of external genitalia in these mutant embryos (data not shown), suggesting important roles of Lrp6 in external genitalia development.

Lip and Palatal Defects in Lrp6-floxdel Homozygotes

Recently, we found that the conventional Lrp6βgeo/βgeo knockout embryos exhibited characteristic cleft lip and palate, and that the Lrp6-mediated Wnt signaling pathway is essential for orofacial development through coordinated regulation of several important downstream factors (Song et al.,2009a). Indeed, the Lrp6-floxdel homozygous embryos were also defective in fusion of both upper and lower lips as well as palates (Fig. 6). These mutant embryos also showed tongue protrusion from the mouth. The severity of clef lip in these Lrp6-floxdel mutant embryos was varied, suggesting the influence of environmental factors and/or genetic backgrounds in orofacial morphogenesis.

Figure 6.

Lip and palatal defects and tongue protrusion in Lrp6-floxdel mice. A: Front facial views of a control (wt) and four mutant (k1–k4) mouse embryos at embryonic day (E) 16.5. The control embryo shows that both upper and lower lips are fused with minor incomplete fusion in the midline of the upper lip (white dashed line) at this age. Note that the tongue is not seen in this normal condition. The mutant embryos show full penetrance of tongue (t) protruding (red dashed lines) from the mouth, cleft midline of lower lips (black dashed lines), and varied clefts in upper lips (white dashed lines). B: Hematoxylin and eosin-stained coronal orofacial sections show the fused palate (arrows) in the normal control (left panels) and the cleft palate in the mutant embryo (floxdel−/−; right panels) at E16.5.


Lrp6 Mutant Mouse Lines in Comparison to β-Catenin Conditional Mutant Mouse Lines

In this study, we demonstrated that most types of the external organ defects in the Lrp6-floxdel homozygous mice were identical to those seen in the conventional Lrp6-βgeo homozygotes. In our conditional gene-targeting mice, the 394-bp exon 2 of Lrp6 gene was floxed by two loxP sites, which caused a frameshift by Cre deletion resulting in severe but identical defects. The mouse Lrp6 gene was located in chromosome 6. The known protein-coding transcript (NM_008514.3) consisted of 23 exons (that encoded a transcript of 9,368 base pairs with the translation of 1,613 amino acids). Thus, the genomic deletion of exon 2 only allowed translation of a short nonfunctional residue (18 amino acids) encoded by the exon 1 of Lrp6.

By contrast, the conventional Lrp6-βgeo mutant mouse line was generated by a gene-trap approach, in which the βgeo (fused β-galatosidase and neomycine) reporter cassette joined to the first 321 amino acids of the Lrp6 protein in-frame (Pinson et al.,2000). No Lrp6 mRNA was detected from the Lrp6-βgeo homozygous embryos and embryonic fibroblasts by Northern blot (Pinson et al.,2000). The lack of Lrp6 mRNA and severe defects in the conventional Lrp6 mutant mice suggest that the residual first 321 amino acids is not functional in vivo, and that no functional splice variants of Lrp6 may exist.

Of interest, two spontaneous point mutant mouse lines of Lrp6 were identified. The crooked tail mice with exencephaly (cranial but not spinal neural tube defects) were found to have a single nucleotide substitution in exon 7, which resulted in a G494D conversion in the second YWTD-EGF repeat domain of Lrp6 (Carter et al.,2005). In the ringelschwanz mice with spina bifida (the spinal neural tube defects) and skeletal defects, a missense mutation was found, resulting in a R886W transition in the Dkk-binding region of Lrp6 (Kokubu et al.,2004). In contrast to the severe and broad defects with embryonic lethality seen in Lrp6-βgeo and Lrp6-floxdel mice, these spontaneous Lrp6 mutant mice could survive partially or fully after birth with milder and more limited organ defects. The region- or organ-specific birth defects in the mouse models were linked with point mutation sites in the Lrp6 gene, suggesting that Lrp6 may also play similar roles in human genetic diseases.

Lrp6 plays a key role in the canonical Wnt/β-catenin signaling pathway. Three conditional mutant mouse lines of β-catenin have been generated for loss- and gain-of-function analyses (Harada et al.,1999; Brault et al.,2001; Huelsken et al.,2001; Grigoryan et al.,2008). Nevertheless, the use of the Lrp6 conditional mutant mouse line will provide us with significantly distinct implications in comparison to the β-catenin conditional mutant mouse lines. In general, the β-catenin conditional mutants will have even more severe and broader defects than seen in the Lrp6 conditional mutants. This is largely because β-catenin has dual roles in Wnt signaling and cell adhesion (Brembeck et al.,2006). β-Catenin null mice die during gastrulation (Haegel et al.,1995), while Lrp6-null mice can survive around birth with broad but unique organ defects. Thus, the Lrp6 conditional mutant mouse line can be used specifically for modeling and dissecting birth defects/congenital diseases.

Disease Connection of the Lrp6 Conditional Mutant Mouse Model and the Phenotypic Influence of the Genetic Background

Each year approximately 3% of babies are born with physical or mental abnormalities resulting from birth defects or developmental disorders (Canfield et al.,2006). The leading categories of birth defects are congenital cardiovascular defects (1 in 115 births), musculoskeletal defects (1 in 130 births), genital and urinary tract disorders (1 in 135 births), nervous system and eye defects (1 in 235 births), cleft lip/palate (1 in 930 births), and spina bifida (1 in 2,000 births; estimated incidences by March of Dimes Perinatal Data Center, 2000). Most formats of these common birth defects are seen in Lrp6 mutant mice.

Particularly in this study, we observed the severe cleft palate and less dominant cleft lip in Lrp6-floxdel mice. Cleft lip with or without cleft palate (CLP) is a result of the failure of fusion in the lip and/or roof of the mouth during early embryonic development with complex but largely unknown etiology (Schutte and Murray,1999; Cobourne,2004). Although CLP is common in humans but is rare in mice, in which many mutant mouse lines exhibit only cleft lip and with incomplete penetrance. Thus, the Lrp6 mutant mouse will be a valuable genetic model for CLP study. We have recently found the severe CLP in the conventional Lrp6-βgeo mice with full penetrance (Song et al.,2009a). The cleft lip in Lrp6-floxdel mutants is less severe than seen in Lrp6-βgeo mice. Different strain backgrounds in these mutant mouse lines may cause the varied severity of cleft lip. We used FLP1 and CMV-Cre mouse lines to delete the FRT-flanked neo cassette and the loxP-flanked exon 2 in the Lrp6-loxP-Neo-FRT mice, respectively. The FLP1 transgenic construct was injected into fertilized eggs from B6SJLF2 mice, and then crossed with C57BL/6 mice (Rodriguez et al.,2000). The CMV-Cre mice were derived from BALB/c-1 background and backcrossed to the C57BL/6 background (Schwenk et al.,1995). The conventional Lrp6-βgeo mice were maintained in C57BL/6 background with early lethality of some homozygous embryos which died between E9.5 to E11.5 (Song et al.,2009b). Therefore, we backcrossed and maintained the Lrp6-floxdel mice in a compound background of C57BL/6 and CD1, which increased the survival rate of the Lrp6-floxdel homozygous embryos during the late gestation with expected Mendelian rations. However, all of the non-C57BL/6 backgrounds, particularly the CD1 background, may contribute to the reduced severity of cleft lip in Lrp6-floxdel mice.

In addition, it has been reported that exencephaly, the cranial neural tube defect, occurred in approximately half of the Lrp6-βgeo homozygous mice at C57BL/6 background (Pinson et al.,2000). We noted a low occurrence rate (approximately 1 in 10) of the exencephaly in Lrp6-floxdel homozygotes at the compound background of C57BL/6 mixed with CD1. By contrast, we observed full penetrant spina bifida (the caudal neural tube defects) in either Lrp6-βgeo or Lrp6-floxdel homozygotes regardless of their different strain backgrounds. These results suggest that the phenotypic influence of the mouse strain background is limited to particular, but not all fundamental defects existed in Lrp6-deficient mice.

We also detected full penetrant agenesis of external genitalia with no visible structures in either Lrp6-floxdel or Lrp6-βgeo homozygous embryos. Recently, β-catenin has been shown to play important roles in all three germ layer-derived endodermal, mesenchymal, and ectodermal lineage cells during early development of external genitalia (Lin et al.,2008). These results suggest that Lrp6 may act upstream of β-catenin in the canonical Wnt signaling pathway to regulate the external genitalia morphogenesis. Another possibility of the agenesis of external genitalia in Lrp6-deficience mice may be the indirect cause of the truncation of caudal body axis by loss-of-function of canonical Wnt signaling. The severe disruption of the caudal body axis in Lrp6 null embryos has been suggested to resemble the phenotypes observed in Wnt3a null or the hypomorphic vestigial tail mice which lack many posterior structures and can be affected by Wnt3a dosage and genetic background (Greco et al.,1996; Pinson et al.,2000). There are no previous reports of external genitalia defects in either Wnt3a or Lrp6 mutant mice. However, the external genitalia are likely still developed, by re-examining the published data of Wnt3a/vestigial tail mutant mice (Fig. 4B in Greco et al.,1996 and Fig. 3c in Pinson et al.,2000), or the compound mutant mice of vestigial tail and Lrp6-βgeo mice (Fig. 3b,d in Pinson et al.,2000). These data suggest that Lrp6 and its mediated canonical Wnt signaling may play a direct role in genitalia development, which is independent from their roles in caudal neural tube development.

The successful generation of the conditional Lrp6-flox mice in this study provides a significant opportunity to dissect the tissue- or cell lineage-specific roles of Lrp6-mediated signaling pathways during organogenesis and related birth defects. Many Cre transgenic mouse lines have been used to dissect the cell lineage-specific roles of β-catenin during development (Grigoryan et al.,2008). These Cre mouse lines can be crossed with our Lrp6-flox mice to investigate the distinct and common roles of Lrp6 and β-catenin in development and related congenital defects.


Generation of Lrp6-flox Mice

Conditional gene-targeting construct.

An approximate 10-kb region of the Lrp6 gene used to construct the targeting vector was first subcloned from a positively identified C57BL/6 BAC clone into a pSP72 backbone vector (Promega). The region was designed by the following strategies. The short homologous arm extends approximately 1.9 kb 3′ to exon 2 of Lrp6 (Fig. 1A). The 6.6-kb-long homologous arm ends 5′ of exon 2. The loxP/FRT flanked Neo cassette (driven by the PGK promoter) is inserted on the 3′ side of exon 2 and the single loxP site is inserted at the 5′ side of exon 2. The target region is 1.7 kb and includes exon 2 of Lrp6 gene. The targeting vector is confirmed by enzymatic restriction and sequencing analysis.

ES cell electroporation, screening, and verification.

Ten micrograms of the targeting construct was linearized by NotI and then transfected by electroporation into C57BL/6 embryonic stem cells. After selection in G418 antibiotic, surviving clones were expanded for PCR screening to identify recombinant ES cell clones. Primers A1, A2, and A3 were designed downstream to the short homology arm (SA) outside the region used to generate the targeting construct (Supp. Fig. S1A, which is available online). PCR reactions using A1, A2, or A3 with the LAN1 primer at the 5′ end of the Neo cassette amplified 2.2, 2.3, and 2.3 kb fragments, respectively. The control PCR reaction was done using primer pair AT1 and AT2, which is at the 5′ end of the SA inside the region used to create the targeting construct. This amplified a 2.0-kb band. Individual clones were screened with A1/LAN1 primers. Recombinant clones were identified by a 2.2-kb PCR fragment (Supp. Fig. S1B). PCR reaction controls were done with screening and internal primers. The selected positive clones were further verified by sequencing (Supp. Fig. S1C,D) and a long range PCR using the primer pair TZ1f and TZ1r. The latter generated a 6.3-kb band of wild-type Lrp6, an 8.3-kb band of recombinates with the targeting cassette, and a 4.6-kb floxdel band after Cre deletion (Supp. Fig. S1E).

Microinjection and chimera mice.

Four confirmed clones (no. 182, 213, 251, and 341; Supp. Fig. S1B) were microinjected into epiblasts for making the mice, and three of them (except no. 341) generated chimeras with high percentage (90–100%). The chimera mice were mated with C57BL/6 mice to generate the heterozygous Lrp6-loxP-Neo-FRT mice.

Neo-cassette deletion and PCR genotyping for Lrp6-flox mice.

We crossed the Lrp6-loxP-Neo-FRT mice with the FLP mice ((TgN(ACTFLPe)9205Dym mice from the Jackson Laboratory) to delete the FRT-flanked neo cassette. The primers for PCR genotyping of the loxP-floxed Lrp6 mice are designed as follow. Lrp6-f1, 5′-TGTGGGTAATGGACACGAGA-3′; Lrp6-r1, 5′-GAAACTAACCAGGCCCAAAG-3′. This primer pair generates a 232 bp wild-type band and a 411 flox band (Fig. 1B).

Generation and PCR Genotyping of Lrp6-floxdel Mice

The Lrp6-floxdel mice were generated by crossing Lrp6-flox mice with CMV-Cre mice (Jackson Laboratory). The female and male heterozygous Lrp6-floxdel mice were bred to generate the homozygous Lrp6-floxdel embryos. The same primer pair Lrp6-f1/r1 was used for the wild-type allele PCR that generates a 232 bp band. A forward primer, Lrp6-floxdel-f2 (5′-GTTGTCCACATTTGGGTTGA-3′) was used to amplify the mutant/floxdel allele with the reverse primer Lrp6-r1. This primer pair generated a 413 mutant band for the Lrp6-floxdel allele (Fig. 1C).

Animal Housing and Sampling

The heterozygous Lrp6-floxdel mice were maintained in a mixed strain background of C57/BL/6J and CD1. The mice were housed in the vivarium of the UC Davis Medical School (Sacramento, CA). Pregnant, timed-mated mice were killed by cervical dislocation after being anesthetized with isoflurane. The embryos were dissected into 4% paraformaldehyde and kept in 4°C. Embryos at E11.5 and E16.5 were photographed and analyzed in this study. The day of conception is designated E0. All research procedures using mice were approved by the UC Davis Animal Care and Use Committee and conformed to NIH guidelines.

Northern Blot and RT-PCR

Total RNAs were isolated from E14.5 whole embryos of the Lrp6-floxdel KO and littermate control using Trizol reagent (Invitrogen). A total of 10 μg of total RNAs of each sample were loaded and separated by electrophoresis in a formaldehyde gel. RNA gel pictures were taken to show the equal loading amounts (Fig. 1D) after the electrophoresis. After gel washing, the RNA was blotted onto a nylon membrane overnight by capillary blotting. The membrane was ultraviolet cross-linked and prehybridized in digoxigenin (DIG) -Easy-Hyb buffer (Roche Applied Science) at 65°C for 2 hr. For the hybridization, DIG-labeled RNA probes targeting the exon 2 (297–648 bp) of Lrp6 was added into the hybridization buffer to a final concentration of 200 ng/ml. The hybridization was carried out at 65°C for 16 hr. The membrane was then washed and incubated with AP-conjugated anti-DIG antibodies (1:10,000; Roche), and subsequently detected with CDP-Star (Roche) and exposed to X-ray films for 30 to 60 min.

The RT-PCR was performed with the same RNA samples described above. The cDNA was made from the reverse transcription (RT) of 1 μg of total RNA for each sample. The RT without reverse transcriptase was done for each sample as the negative control. The PCR was carried out by a primer pair of the forward (5′-CAGACGGGACTTGAGATTGG-3′) and the reverse (5′-GCAATAGCTCTGGGTTGATCC-3′) from exon 2 of Lrp6, which generated a 352-bp cDNA band. The PCR for the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was performed as the positive control.


Wild-type, heterozygous, and Lrp6-floxdel homozygous embryos were collected at E12.5 and homogenized in the RIPA solution (Santa Cruz Biotechnology, Santa Cruz, CA), which contains a complete protease inhibitor cocktail. The total protein concentration of the lysate was quantified using the BCA kit (Thermo Scientific, Waltham, MA) according to the manufacturer's instruction. Thirty to 35 μg of proteins were separated by electrophoresis on a sodium dodecyl sulfate-polyacrylamide gel electrophoresis gel, and then transferred to the polyvinylidene difluoride membrane. The membrane was blocked with 5% milk solution and incubated with the primary antibody for Lrp6 (rabbit anti-Lrp6, Abcam, Cambridge, MA) with a dilution of 1:800. After 1-hr incubation at room temperature or overnight at 4°C, the membrane was washed and incubated with the horseradish peroxidase (HRP) -conjugated secondary antibody against rabbit IgG (1:4,000, GE healthcare, Giles, UK). The signal detection was achieved using the Immobilon Western Chemiluminescent HRP substrate kit (Millipore, Billerica, MA) according to the manufacturer's protocol. The rabbit anti-GAPDH antibody (1:4,000, Abcam) was used for the control blotting.


The authors thank Lisa Aronov for advice and technical support in generating the Lrp6-flox mice, William Skarnes for Lrp6-βgeo mice, the journal reviewers and editors of Developmental Dynamics for constructive comments and suggestions for the paper revision, and Carol Morse and Michelle Tu for reading this manuscript.