当前位置: 首页 > 期刊 > 《核酸研究》 > 2005年第16期 > 正文
编号:11370790
Expression and regulation of the Msx1 natural antisense transcript dur
http://www.100md.com 《核酸研究医学期刊》
     Laboratoire de Biologie Oro-Faciale et Pathologie, INSERM U 0714, Centre de recherches biomédicales des Cordeliers 15-21, rue de l'Ecole de Médecine, 75270 Paris Cedex 6, France 1Department of Craniofacial Development, Dental Institute, King's College London Floor 28 Guy's Tower, Guy's Hospital, London SE1 9RT, UK 2Department of Pediatric Dentistry, Dental School, University of Texas Health Science Center at San Antonio San Antonio, TX, USA 3Unité de Génétique Moléculaire de la Morphogenèse, Institut Pasteur, CNRS URA 2578 25, rue du Dr Roux, 75724 Paris Cedex 15, France

    *To whom correspondence should be addressed.; Tel: +33 1 43 26 94 96; Fax: +33 1 44 07 14 21; Email: biol_odonto_fr@yahoo.com

    ABSTRACT

    Bidirectional transcription, leading to the expression of an antisense (AS) RNA partially complementary to the protein coding sense (S) RNA, is an emerging subject in mammals and has been associated with various processes such as RNA interference, imprinting and transcription inhibition. Homeobox genes do not escape this bidirectional transcription, raising the possibility that such AS transcription occurs during embryonic development and may be involved in the complexity of regulation of homeobox gene expression. According to the importance of the Msx1 homeobox gene function in craniofacial development, especially in tooth development, the expression and regulation of its recently identified AS transcripts were investigated in vivo in mouse from E9.5 embryo to newborn, and compared with the S transcript and the encoded protein expression pattern and regulation. The spatial and temporal expression patterns of S, AS transcripts and protein are consistent with a role of AS RNA in the regulation of Msx1 expression in timely controlled developmental sites. Epithelial–mesenchymal interactions were shown to control the spatial organization of S and also AS RNA expression during early patterning of incisors and molars in the odontogenic mesenchyme. To conclude, this study clearly identifies the Msx1 AS RNA involvement during tooth development and evidences a new degree of complexity in craniofacial developmental biology: the implication of endogenous AS RNAs.

    INTRODUCTION

    Until very recently, the significance of natural antisense (AS) transcripts was underestimated. AS transcripts in prokaryotes have been known for at least two decades (1,2) while a limited number of cases have been reported in eukaryotes. Recent data have established their importance in eukaryotes (3–5). Indeed, these transcripts are involved in gene expression regulation (5) as exemplified by RNA interference, presently applied in numerous knockdown strategies (6). Natural AS transcripts are grouped into two classes: cis AS transcripts which are transcribed with the sense (S) transcripts from a unique gene locus, and trans AS transcripts which are transcribed from a different locus. Rare data are available on the physiological impact of AS RNAs during development. For example, in Caenorhabditis elegans let4 controls the timing of post-embryonic cell division and fate . In mouse, miR196a negatively regulates Hoxb8 and restricts Hox homeogene expression pattern (8). Interestingly, these two AS RNAs belong to the specific subclass of small RNAs also named microRNA (21–23 nt).

    Recent studies have highlighted another subclass of AS RNAs: the long cis AS (containing >100 bp). In silico studies based on expressed sequence tag database have predicted that they may constitute 15% of the mouse genome (9) and >20% of the human genome (4). Long cis AS transcripts have been shown to exert regulatory functions on protein expression at various levels such as epigenetic imprinting, RNA maturation, edition and translation inhibition (5,10). However, the functional data on AS transcripts have been essentially generated in vitro. Very few long AS transcripts have been analyzed in the specific context of mammalian development. Some expression patterns have been reported: for instance, Hoxa11 AS and S transcripts show complementary expression territories in developing limbs (11).

    The mouse Msx1 locus has been shown previously to undergo a bidirectional convergent and overlapping transcription (12). Msx1 AS transcript belongs to the long cis-AS subclass (2184 nt, in mouse), and was proposed to control protein expression. This assertion was based on (i) in vitro evidence of protein down-regulation by the AS transcript, and (ii) in vivo expression patterns in newborn mouse osteoblasts (12). Indeed, in the mandibular bone Msx1 AS transcript evidenced a complementary expression pattern to the Msx1 S transcript. More specifically, the Msx1 AS transcript is expressed in the latest stage of osteoblast maturation (osteocytes) whereas Msx1 S transcript is expressed in preosteoblasts and osteoblasts (12). According to these observations, it was hypothesized that Msx1 AS transcript expression is instrumental in the regulation of osteoblastic cell differentiation. It would control the progressive reduction of Msx1 homeoprotein expression and therefore induce the interdependent cell cycle exit and differentiation (12). This assertion is supported by several in vitro and in vivo data on Msx1 homeoprotein. In vitro, Msx1 overexpression blocks myoblasts differentiation (13), induces myotubes dedifferentiation into myoblasts and promotes transdifferentiation into osteoblasts (14). Consistently, in vivo, Msx1 expression is reversely correlated with cell differentiation progression (15,16).

    During embryonic development, Msx1 is instrumental in various systems, notably in the craniofacial complex where it acts as a transcriptional repressor (17–21). Msx1 expression is observed early in development in neural crest cells and their derivatives, including first branchial arch ectomesenchymal cells devoted to tooth formation (17,22,23). In mice, the first evidence of tooth morphogenesis occurs at embryonic day 11.5 (E11.5) as a thickening of the dental epithelium, called the dental lamina (24). However, before this morphological event and as early as E10.5, the respective incisor and molar fields are already determined. Oral epithelium induces a site-specific combination of homeobox genes expressions in the subjacent dental mesenchyme, and these combinations are organized as an ‘odontogenic homeobox code’ (25–27). Msx1 has been shown to be a key element within this code, specifying the incisor presumptive region at E10.5 (25,28). After E11.5, tooth morphogenesis is already initiated and the dental lamina epithelium progressively invades the subjacent mesenchyme and forms the dental bud (E12.5–E13.5). Thereafter, dental development progresses through the cap (E14.5) and bell stages (E16.5), characterized by tooth-specific morphogenesis and cell differentiation. Tooth morphogenesis is orchestrated by successive organization centers, called primary and secondary enamel knots (29). This overall process of tooth morphogenesis is driven by sequential and reciprocal interactions between dental epithelium and mesenchyme . Msx1 plays a central role in these epithelial–mesenchymal interactions as evidenced by molar tooth development arrest at the bud stage in mice lacking Msx1 (23,32,33).

    Finally, during tooth development, Msx1 is important for two major processes, firstly, the dental field determination and secondly, the tooth morphogenesis. Msx1 is, therefore, an exemplary model to analyze cell–cell communication leading early determination and subsequent morphogenesis in mammals. Based on this conclusion and on the recent discovery of the endogenous AS transcript of Msx1 (12), the present study aimed to delineate the timing and impact of this AS transcript in these established developmental cascades; firstly, to determine whether Msx1 AS transcript is expressed prior to the late terminal differentiation stages (12), and secondly, whether Msx1 AS transcript is instrumental in epithelial–mesenchymal interactions leading to the two distinct steps of odontogenesis, namely, tooth field determination (E10.5) and tooth morphogenesis (E11.5–E16.5). Using different approaches we have proposed a novel role for Msx1 AS in tooth development.

    MATERIALS AND METHODS

    Embryo collection

    Timed matings were set up such that noon of the day on which vaginal plugs were detected was considered as embryonic day 0.5 (E0.5). Wild-type mouse embryos used for in situ hybridization and explants cultures were collected from matings of CD-1 mice (Elevage Dépré, Saint Doulchard). Msx1/nLacZ heterozygous embryos and Dlx2/LacZ transgenic embryos were collected from CD-1 females crossed with Msx1/nLacZ heterozygous males (23) and Dlx2/LacZ transgenic males (34), respectively. Pregnant mice were sacrificed by cervical dislocation and the embryos collected from E9.5 to E18.5. Newborns, 1- and 2-day-old mice were sacrificed by head cutting.

    Msx1/nLacZ and Dlx2/LacZ embryos whole mount ?-galactosidase staining

    ?-galactosidase activity was evidenced by standard histochemical methods as described previously (35). Briefly, the embryos were harvested, fixed for 30 min in a 1x phosphate-buffered saline (PBS) containing 4% paraformaldehyde (PFA), washed in 1x PBS and finally stained overnight at 32°C in a PBS 1x solution containing 5 mM potassium ferricyanide, 5 mM potassium ferrocyanide, 2 mM magnesium chloride, 0.1% Nonidet P-40 and 400 μg/ml of X-Gal substrate (all products from Sigma, Saint Quentin Fallavier). The embryos were then fixed overnight in the same fixative solution and rinsed in PBS 1x before being either photographed, wax embedded and sectioned, or used for whole mount in situ hybridization.

    In situ hybridization experiments

    In situ hybridization experiments were performed for both Msx1 transcripts and for Dlx2 and Barx1 S transcripts. Four probes were designed to detect Msx1 transcripts as presented in Figure 2. Two probes, E2A and E2B, are localized in Msx1 exon 2 and are complementary for the AS and S transcripts, respectively (Figure 2). These probes were generated as described previously (12) from a SphI/SphI Msx1 exon 2 960 bp DNA fragment cloned in a bluescript plasmid (18). The E1 probe, complementary to the S transcript, was obtained by EcoRI digestion and T7 transcription of a bluescript plasmid containing the XbaI/EcoRI 368 bp fragment of Msx1 exon 1 (Figure 2). This fragment was obtained by PCR amplification. The I probe, complementary to the AS transcript, was obtained by EcoRI digestion and T7 transcription of a bluescript plasmid containing the SmaI/SmaI 935 bp fragment of Msx1 intron (Figure 2). The Dlx2 probe was generated as described previously from a bluescript derivative (named E61) (35). The Barx1 probe was generated, using EcoRI digestion and T7 transcription, from a bluescript vector containing a 951 bp Barx1 DNA fragment (kindly gifted by J. -F. Brunet).

    Figure 2 Schematic representation of the two Msx1 transcripts and position of the four probes used in situ for hybridizations. Msx1 S and AS transcripts organization in regard of the Msx1 gene structure (accession number S73812 ) is schematically presented. The Msx1 S (top) and AS (bottom) transcripts are shown. The S transcript correspond to exons 1 and 2 sequences. The AS transcript sequence is complementary to the exon2 and a major part of the intron. Positions of probes used for hybridization are given. Green and red arrows, respectively, represent S and AS promoter.

    Whole mount digoxigenin-labeled hybridizations were carried out as described previously (35). Briefly, after proteinase K and triethanolamine/acetic anhydride pretreatments, samples were fixed in 4% PFA before overnight hybridization at 65°C with probe at 1 μg/ml. After post-hybridization washes, digoxigenin immunodetection was realized with an antidigoxygenin antibody coupled to alkaline phosphatase (AP) (Roche Diagnostics, Meylan) diluted at 1/2000. The AP enzymatic activity was finally revealed using NBT/BCIP solution (Roche) according to the manufacturer's instructions.

    Radioactive (35S-UTP) in situ hybridizations were performed on 7 μm paraffin frontal sections as described previously (36). Sample sections floated onto polylysine (Sigma) coated slides were pretreated with proteinase K (Sigma) and 0.25% (vol/vol) acetic anhydride (Sigma) to reduce the background. Hybridization was carried out overnight in a humidified chamber at 60°C. The slides were then washed twice at high stringency (20 min at 60°C in 2x SSC, 50% formamide and 10 mM DTT), and treated with 40 mg/ml RNAse A (Sigma) for 15 min at 37°C to remove any non-specifically bound probe. The high stringency washes were repeated, followed by a further wash at 60°C in 0.1x SSC and 10 mM DTT. Sections were then washed in 0.1x SSC at room temperature and dehydrated in 300 mM ammonium acetate (Sigma) in 70%, 95% and absolute ethanol. The slides were air-dried and dipped in hypercoat LM-1 emulsion (Amersham-Biosciences, Orsay) and stored at 4°C in a light-tight box for 6–8 weeks. They were finally developed using Kodak D19 and fixed in Kodak UNIFIX (Kodak).

    Digoxigenin-labeled in situ hybridizations were performed on 10 μm cryosections. Sections were hybridized overnight at 65°C in a humidified chamber with 1 mg/ml of probe in hybridization buffer. The slides were washed twice in 5x SSC, 50% formamide at 65°C, then in maleic acid buffer (MAB) at room temperature (all products were obtained from Sigma). They were incubated in MAB with 2% blocking reagent (Roche), 2% lamb serum (Gibco, Cergy-Pontoise) and a 1/500 dilution of antidigoxigenin antibody coupled with AP (Roche) at room temperature. The digoxigenin-labeled probes were detected as described previously.

    Explants cultures

    Mandible explants were dissected out from embryos under a stereomicroscope and, if necessary, treated with 2 U/ml Dispase (GibcoBRL) in order to separate the epithelium from mesenchyme. Intact mandibles or isolated mesenchyme were cultured as described previously (37). Briefly, all explants were cultured at 37°C, 5% CO2 and 40% O2 on 0.1 μm Nuclepore filters (Millipore) in a Trowell type organ culture dish containing DMEM (GibcoBRL) supplemented with 10% fetal calf serum (FCS; Gibco) and 1% penicillin–streptomycin (Gibco). After 12 h of in vitro culture, tissues were treated for 2 min with 100% ice cold methanol, fixed overnight in 4% PFA at 4°C, and processed for whole mount in situ hybridization analysis.

    Recombinant protein bead implantation assays were done as described previously (38). Heparin acrylic beads (Sigma) were incubated in Fibroblast Growth Factor 8 (FGF8) protein (1 mg/μl; R&D). Affi-gel blue beads (BioRad) were incubated in Bone Morphogenetic Protein 4 (BMP4) protein (100 ng/μl; R&D). About 100 beads were washed with 1x PBS and soaked with 10 μl of growth factor solution for 1 h at 37°C for BMP4 and overnight at 4°C for FGF8. The beads were placed on top of the whole mandible or on top of isolated mesenchyme using fine forceps. Explants with beads were cultured for 12 h as described previously.

    Cell culture and transfection assays

    MD10H1 cell line was obtained by large T antigen immortalization of E18 mouse embryo first molar mesenchyme (39). These cells were used for transfection assays in triplicate using Exgen transfection system according to the manufacturer's protocol (Euromedex, Mundolsheim). Briefly, MD10H1 cells were plated out in -MEM supplemented with 10% of FCS (GibcoBRL) at 5 x 104 cells/cm2. After 24 h, cells were transfected with 0.5 μg of either Dlx2 or Barx1 expression vectors (provided by S. Harris and P. T. Sharpe, respectively) or with 0.5 μg of Msx1 AS expression vectors. The latter one was generated by PCR amplification from a DNA template with the following primers (5'-TTA CAT CCT GGT GGT CTG AG-3' and 5'-CCA GCA TGC ACC CTA CGC AA-3'). The amplified fragment corresponds to the entire AS transcript. Control transfection experiments were performed using empty expression plasmids. After 24 h, the medium was removed, the cells rinsed with 1x PBS and used for RNA analysis.

    RNA analysis

    RNA from the transfected cells, from the heads dissected from E9.5 to E18.5 embryos and from newborn to 2-day-old mice were extracted using TriReagent (Euromedex) as described previously (40) and used for RT–PCR. RT was carried out on 1 μg of total RNA with Superscript II using oligo(dT) primer according to the manufacturer's protocol (Invitrogen). The PCR was performed in 20 μl with 2 μl of the RT reaction and 10 pmol of the following primers for 30 cycles: Msx1 S (forward primer: 5'-TCC TCA AGC TGC CAG AAG AT-3'; reverse primer: 5'-TCA GGT GGT ACA TGC TGT AG-3'), Msx1 AS (forward primer: 5'-TTA TGT CCA CCT GCC CTT TC-3'; reverse primer: 5'-GGG CCC AAA GGA TTA TTG TT-3'), Dlx2 (forward primer: 5'-TCC TAC CAG TAC CAA GCC A-3'; reverse primer: 5'-AAG CAC AAG GTG GAG AAG C-3'), Barx1 (forward primer: 5'-CCA TGC CCG GCC CCG CAG GCG CAT C-3'; reverse primer: 5'-GAA TTC AGT CCT CGC AAT TTC GG-3') and GAPDH (forward primer: 5'-TTC CAG TAT GAT TCC ACT CA-3'; reverse primer: 5'-CTG TAG CCA TAT TCA TTG TG-3'). In all analysis, RNA without RT served as control.

    RESULTS

    First evidence of Msx1 AS transcript expression during development

    The first evidence for endogenous Msx1 AS transcripts was shown previously in postnatal mice (12). In order to determine if Msx1 AS was expressed during embryonic development and in early postnatal life, semi-quantitative RT–PCR was performed on RNA from E9.5 to E18.5 embryos and from day 0 to day 2 newborn mice. These experiments showed that both Msx1 transcripts were detected as early as E9.5 and throughout development until postnatal day 2, suggesting a potential role in embryogenesis (Figure 1).

    Figure 1 Msx1 S and AS transcripts detection from E9.5 to 2 days post natal. The two Msx1 transcripts were detected by semi-quantitative RT–PCR during this entire period.

    Msx1 S and AS transcripts and protein expression from E9.5 to E11.5

    Expression of both S and AS Msx1 transcripts was analyzed during early embryonic development by whole mount in situ hybridization. The protein expression was analyzed at these same stages using Msx1/LacZ heterozygous embryos, in which the LacZ knock in enabled protein expression followed indirectly (23).

    The Msx1 AS transcript was described previously as a 2.1 kb long transcript which contains sequence overlapping the entire exon 2 and 1012 bp of the intron of Msx1 gene (Figure 2). Both the S and AS Msx1 transcripts are complementary in exon 2 sequence. A set of four probes (E1, I, E2A and E2B) was used for in situ hybridization experiments. The S transcript probe (E1) was nested in the Msx1 exon 1 since this sequence is specific to the S transcript (Figure 2). The AS transcript probe (I) was nested in Msx1 intron since this sequence is specific to the AS transcript (Figure 2). With the original Msx1 vector (18) containing part of exon 2 sequence, two probes were generated. E2A probe was complementary for the AS transcript whereas E2B probe was complementary for the S transcript (Figure 2). For all experiments the four probes were used. In all experiments, with the exception of E13.5–E16.5, detected expression using both S probes produced the same pattern of expression, as also seen with both the AS probes.

    At E9.5, the Msx1 AS transcript was not detected in mandibulary and maxillary regions (Figure 3B and E) but appeared to be expressed in brain. In contrast, the S transcript (Figure 3A and D) and protein (Figure 3C and F) were detected in the distal parts of the mandible primordium. In the facial region, Msx1 AS transcript expression was first detected at E10.5 (Figure 3H and K). At this stage, both Msx1 transcripts exhibited an interesting opposite pattern in the mandibular arch. Msx1 AS transcript was detected in the proximal part of mandible (Figure 3H and K) whereas Msx1 S transcript (Figure 3G and J) and protein (Figure 3I and L) were still detected in the distal part of mandible. Msx1 AS transcript was detected in other anatomical sites such as the future brain and limb buds (Figure 3H). At E11.5, in the facial region, the territories of both Msx1 transcripts and protein appeared to be enlarged and overlapping, corresponding to almost all the oral part of mandible (Figure 3M–R). Two opposite gradients of Msx1 S and AS transcript expression patterns were observed. Msx1 S transcript expression level appeared higher in the distal region (Figure 3P) and reversely the Msx1 AS transcript was expressed more significantly in the proximal regions (Figure 3Q). Protein expression also appeared to be expressed at a higher level in the distal region (Figure 3R). Msx1 AS transcript was still detected in the forming brain and in the limb buds at this stage, (Figure 3N).

    Figure 3 Msx1 S and AS transcripts and Msx1 protein expression from E9.5 to E11.5 in mouse embryos. Both Msx1 transcripts are detected by whole mount in situ hybridization on wild-type mice embryos. Msx1 protein expression is revealed by the ?-galactosidase activity using Msx1-nLacZ heterozygous mouse embryos. (A–F) At E9.5, the Msx1 AS transcript is not detected in the first branchial arch in contrast to the S transcript and the protein. (G–L) At E10.5, both S and AS Msx1 transcripts are detected in complementary regions in the mandible. The Msx1 S transcript and the protein are detected in the distal part of the mandible and the Msx1 AS transcript in the proximal parts. (M–R) At E11.5, both Msx1 transcripts are detected in progressively overlapping areas in mandible outlining two reverse gradients, proximal-distal for the Msx1 AS transcript and distal-proximal for the Msx1 S transcript and the protein. The bar scale represent 800 μm for whole embryos, and 500 μm for the mandible.

    Comparative expression of Dlx2, Barx1, Msx1 S and AS transcripts at E10.5 in presumptive molar and incisor fields

    The expression of Dlx2, Barx1 and both Msx1 transcripts were comparatively analyzed in the presumptive incisor and molar fields of E10.5 embryo mandible by whole mount in situ hybridization. The Msx1 AS transcript was detected in the proximal regions of the E10.5 mandibular arch (Figure 4A) whereas the Msx1 S transcript was detected in the distal part of the mandible (Figure 4C) adjacent to the Dlx2 epithelial expression (blue in Figure 4C and D). The mesenchymal expression of Dlx2 (Figure 4B) and Barx1 (Figure 4D) was evidenced in the proximal area, as was the Msx1 AS expression. However, the Barx1 mesenchymal expression pattern appeared more restricted than that of the Dlx2 and Msx1 AS. In the epithelium, the Msx1 AS transcript was not detected in the distal region of Dlx2 epithelial expression (Figure 4B and blue LacZ staining in Figure 4C and D).

    Figure 4 Analysis at E10.5 of Msx1 AS transcript potential integration in mandibular dental field determination. (A–D) Comparison of the Msx1 AS transcript expression pattern with Dlx2 and Barx1 transcripts in E10.5 mandible. The Msx1 AS transcript (A), the Dlx2 transcript (B) and the Barx1 transcript (D) were detected in the same proximal regions of the mesenchyme by whole mount in situ hybridization whereas the Msx1 S transcript (C) was detected in the distal part of mandible. In contrast, in the epithelium where Dlx2 expression is seen distally there is no associated expression of either Msx1 AS or Barx1 transcripts expression . (E–K) The Msx1 AS transcript mesenchymal expression pattern in E10.5 mandible is controlled by epithelial signals. (E) The Msx1 AS transcript expression was absent when the mandible explant is cultured without oral epithelium. (F) The Msx1 AS transcript mesenchymal expression was restored around an FGF8 soaked bead as was the Msx1 S transcript expression (G). Msx1 AS transcript expression was not restored around BMP4 or BSA soaked beads (H and I). In contrast to the AS transcript, the Msx1 S transcript expression increased around the BMP4 soaked beads (J). No inhibitory effect on Msx1 AS transcript expression is observed when BMP4 soaked beads are put in presence of epithelium (K). (L) The Msx1 AS transcription activation by Dlx2. MD10H1 cells were transfected with either an empty vector, a Dlx2 expression vector, or an Msx1 AS transcript expression vector. RNA extracted from these cells was reverse-transcribed and cDNA analyzed by PCR. Dlx2 overexpression appears to be able to induce a moderate increase of the Msx1 AS transcript synthesis whereas Msx1 AS transcript overexpression had no effect on either Dlx2 or Barx1 expressions. (M) Schematic representation of the regulation of both Msx1 transcripts by epithelial BMP4 and FGF8 signaling. In proximal regions of the E10.5 mandible, FGF8 activate Msx1 AS transcript expression directly or/and potentially through Dlx2 activation. The AS transcript is suspected to inhibit Msx1 homeoprotein expression in these proximal regions though a mechanism that remains to be elucidated. In the distal region, BMP4 activates Msx1 S transcript and protein expression and inhibits Dlx2 expression. No direct effect of BMP4 on Msx1 AS transcript expression was evidenced.

    FGF8 regulates Msx1 AS transcript expression in mandible explant culture

    In order to determine whether Msx1 AS transcript expression would be integrated in the epithelial signaling pathways that establish the morphogenetic fields in E10.5 mandible, epithelial induction of Msx1 AS transcript mesenchymal expression was evaluated in mandibular explants cultures. Interestingly, 12 h after the removal of mandibular arch, a loss of Msx1 AS transcript expression was observed in the mandibular explant (Figure 4E). To test the potential induction of Msx1 AS transcript expression by epithelium signals, BMP4 and FGF8 soaked beads were applied to the mandibular mesenchyme explants. After 12 h of culture, Msx1 AS transcript expression was detected around the FGF8 beads (Figure 4F), as was Msx1 S transcript (Figure 4G), whereas BMP4 or BSA beads had no effect (Figure 4H and I). In contrast to the AS transcript, the Msx1 S transcript expression was increased by BMP4 soaked beads (Figure 4J). To explore a potential inhibitory effect of BMP4 on the Msx1 AS transcript expression, mandibular explants were cultured for 12 h with BMP4 soaked beads in the presence of epithelium (Figure 4K). No significant change was observed in the Msx1 AS transcript expression (Figure 4K).

    Impact of Dlx2 and Barx1 overexpression on Msx1 AS transcript expression in dental cells

    As Dlx2 and Barx1 are coexpressed with Msx1 AS transcript in the proximal parts of E10.5 mouse embryo mandible, a potential regulation between these genes was analyzed in vitro in undifferentiated mesenchymal MD10-H1 cells. In order to examine whether the Msx1 AS transcript could be under the control of Dlx2 and Barx1 in the mesenchyme, the expression level of this transcript was analyzed after Dlx2 or Barx1 overexpression. The semi-quantitative RT–PCR results showed a small but reproducible increase in Msx1 AS transcript steady state levels which was observed 24 h after the transfection with a Dlx2 vector (Figure 4L). Quantitative analyses of Msx1 AS transcript expression increase after Dlx2 overexpression were realized using two techniques: real-time PCR (Roche) and classical PCR fragments quantification on BET gel using Optimate 5.21 (Visiomic). Both techniques showed a moderate stimulation to 1.2 times that of Msx1 AS transcript steady state expression. The impact of Dlx2 on Msx1 AS transcript expression in these experiments is probably underestimated according to the low transfection efficiency obtained for MD10-H1 cells (only 35%). Barx1 transcript expression level was not affected by Dlx2 overexpression (Figure 4L), nor Dlx2 expression affected by Barx1 overexpression (data not shown). The controlled overexpression of Msx1 AS transcript showed no effect on either Dlx2 or Barx1 expression levels (Figure 4L).

    Msx1 S, AS transcripts and protein expression during tooth morphogenesis

    All in situ hybridization experiments were carried out with the panel of four different probes (Figure 2). From E9.5 to E12.5, similar results were obtained with the two probes for Msx1 AS transcript. The two probes for Msx1 S transcript also gave identical results. At E11.5, both Msx1 transcripts and the protein were detected in similar regions of the dental mesenchyme surrounding the dental lamina (Figure 5A–C). At E12.5, both Msx1 transcripts and the protein were still detected in the same anatomical sites, more precisely in the mesenchyme surrounding the growing tooth buds (Figure 5D–F). At E13.5, the probes revealed different expression patterns. The Msx1 S transcript was detected in the dental mesenchyme (Figure 5G and H) with both probes (E1 and E2B) and unexpectedly in the dental epithelium with the E1 probe (Figure 5G). The Msx1 AS transcript was detected, with both AS probes, in the dental epithelium (Figure 5I and J), and also, at a low level, in the dental mesenchyme. At this stage, the protein was expressed only in the dental mesenchyme (Figure 5K). At E16.5, the Msx1 S transcript was detected by the two probes in the dental mesenchyme (Figure 5L and M). At this stage, as observed previously at E13.5, an epithelial expression of Msx1 S transcript was revealed only by the E1 probe (Figure 5L). The Msx1 AS transcript was detected at E16.5 only in the dental epithelium with whatever probes used (Figure 5N and O). No protein expression was detectable at this stage (Figure 5P). In newborns, before the first sign of odontoblast differentiation, the S and AS transcripts were detected with E1 (Figure 5Q) and I (Figure 5T) probes, respectively, whereas the exon two probes (Figure 5R and S) gave no signal.

    Figure 5 Msx1 S and AS transcript detection and Msx1 protein expression during mouse tooth development from E11.5 to birth. Both transcripts of Msx1 were detected by in situ hybridization in wild-type mouse embryonic serial sections in comparison with LacZ expression in Msx1/LacZ hereterozygotes. (A–C) At E11.5, both Msx1 transcripts were detected whatever the probes used in the dental mesenchyme as was LacZ. (D–F) At E12.5, both Msx1 transcripts were detected whatever the probes used along with LacZ in the dental mesenchyme but in a more restricted area than at E11.5, surrounding the tooth buds. (G–K) At E13.5, the expression patterns of the both Msx1 transcripts were more complex. Using E1 probe, the Msx1 S transcript was detected in both dental mesenchyme and epithelium (G). Using E2B probe, the Msx1 S transcript is only detected in dental mesenchyme surrounding the tooth bud (H). Using E2A and I probes, the Msx1 AS transcript is detected in both dental epithelium and mesenchyme (I and J). LacZ expression was only detected in the dental mesenchyme surrounding the tooth bud and the epithelium was clearly devoid (K). (L–P) At E16.5, the Msx1 S transcript was detected in dental epithelium and mesenchyme with the E1 probe whereas it was only detected in mesenchyme with E2B probe. Msx1 AS transcript detection was restricted to the dental epithelium independently of the probe used. No Msx1 fusion protein expression was detected at this stage. (Q–T) At birth, the Msx1 S transcript was detected in dental epithelium and mesenchyme only with the E1 probe. The Msx1 AS transcript was detected in both dental tissues only with the I probes. (U) Experimental control (Ctrl) corresponding to the use of unlabeled probe. (Md: mandible; Mx: maxilla).

    Msx1 S, AS transcripts and protein expression during early limb bud development

    Msx1 transcripts and protein expression were comparatively analyzed during anterior limb bud development, from E9.5 to E14.5, by whole mount in situ hybridization and ?-galactosidase staining (Figure 6). At E9.5, Msx1 S transcript and protein were detected in the entire limb bud whereas Msx1 AS transcript was absent (Figure 6A–C). At E10.5, similarly to the observed pattern in the mandible, both Msx1 transcripts were detected with complementary expression patterns. Msx1 S transcript and protein were detected in the area bordering the apical ectodermal ridge whereas the Msx1 AS transcript was detected in a more proximal region, but immediately adjacent to the Msx1 S transcript positive region (Figure 3G–I and Figure 6D–F). At E11.5, the Msx1 AS transcript was still located in an area adjacent to the Msx1 S transcript region (Figure 3M and N and Figure 6G and H). At E14.5, the Msx1 AS transcript was detected in the proximal part of the interdigital regions (Figure 6J), whereas the protein was expressed with a proximal-distal gradient in the same interdigital areas and also in tip of digits (Figure 6K).

    Figure 6 Msx1 S and AS transcripts detection and Msx1 protein expression from E9.5 to E14.5 in mouse anterior limb bud. Both transcripts of Msx1 were detected by whole mount in situ hybridization in wild-type mouse embryos. Msx1 protein expression was detected by the ?-galactosidase activity in Msx1-nLacZ heterozygote mouse embryos. The Msx1 AS transcript was detected in limb bud at E10.5 and E11.5 but not at E9.5 (B, E and H). Its detections territories edged the Msx1 S transcript and protein detection territories (A, C, D, F, G and I). At E14.5, Msx1 AS transcript and protein show complementary expression pattern (J and K). The bar scale represent 200 μm.

    DISCUSSION

    The Msh-like homeobox genes, orthologous to the Muscle Segment Homeobox gene of Drosophila melanogaster, constitute an ancient family described in numerous species ranging from coelenterates to mammals (20). Inside this family, a variable number of members have been reported for different species, corresponding to evolutionary duplications (20). These genes exhibit a simple genomic structure with two exons separated by a single intron (41). The Msx1 gene appears to be the most conserved gene within the Msx family of mammals and avians (20). Interestingly, a bidirectional overlapping and convergent transcription was identified for this Msx1 gene in mouse and human and might also be conserved in at least three other species: rat, bovine and chicken. Such a possibility was suggested by the existence of a conserved identical 60 bp sequence containing a TATA box and corresponding to the AS promoter region (12). According to the importance of Msx1 homeoprotein during development and the preliminary in vitro observation that the Msx1 AS transcript has the ability to interfere with Msx1 homeoprotein expression (12), the investigation of the physiological significance of the bidirectional transcription of the Msx1 locus during development appeared to be important. The experimental strategy was to investigate the presence of the Msx1 AS transcript during mouse embryogenesis by RT–PCR, to analyze its expression pattern in comparison with the Msx1 S transcript and the protein in tooth and other significant developing systems where Msx1 signaling pathways are instrumental (42–44), and finally to analyze its potential regulation by epithelial signals.

    Evidence for Msx1 AS transcript expression during embryonic development

    The Msx1 AS transcript was originally identified in newborn and young adult mice (12), in association with skeletal growth. Msx1 AS transcript was proposed to be instrumental postnatally in terminal differentiation, by inhibiting Msx1 protein expression (12,45). Indeed, Msx1 homeoprotein expressing cells were considered as undifferentiated cells, based on in vitro observations that Msx1 (i) inhibits master gene expression such as MyoD (46) and Runx2 (12), (ii) modulates the expression of cell cycle determinants such as cyclin D1 (47) and p19(INK4d) (48), (iii) induces dedifferentiation of myotubes (14), and (iv) is specifically present in progenitor cell niches in adults (49). Interestingly, the initial analysis of Msx1 AS transcript expression during antenatal mouse development (Figure 1) revealed that, similar to the S transcript, the Msx1 AS transcript was already expressed at E9.5 before the initial steps of skeleton morphogenesis and cell differentiation, and continuously up to birth. These data suggest that the Msx1 bidirectional convergent and overlapping transcription may not only be linked to the control of cell terminal differentiation but also implicated in more complex processes during development, such as dental and skeletal morphogenesis.

    In the past, Msx1 gene expression has been documented in numerous developing organs and systems, notably, heart (18,50), brain (51), limb (50), mammary glands (52), hair follicles (43,50) and teeth (22,28,33). Due to the detailed analysis of the role of Msx1 during odontogenesis, experimental studies of Msx1 AS transcript were carried out in this developmental model system. In addition, Msx1 protein expression was compared with the AS transcript throughout tooth morphogenesis stages and showed a continuous expression until late bell stage (E16.5). The results suggest that, during tooth development, Msx1 AS transcript may not only be involved in cell terminal differentiation control as proposed previously (12), but also in tooth initiation and subsequent morphogenesis through the control of Msx1 homeoprotein expression and function.

    The most important result obtained here was that both S and AS Msx1 transcripts were present during embryonic development. Indeed, it appeared that during development, when an organ expressed Msx1 S transcript and protein, it also expressed the AS transcript. This observation was not only verified in tooth but also in bone (12), hair follicles (data not shown), neural tube (data not shown) and limb bud (Figure 6).

    Msx1 AS transcript and tooth initiation

    Msx1 (S transcript) has been proposed to be an important factor in the odontogenic homeobox code proposed to define the different tooth fields at E10.5 in mouse prior to the morphological appearance of tooth formation (25–27). This homeobox code is based on combinations of homeobox gene expression patterning territories in the first branchial arch neural crest derived mesenchyme (37,53–56). In this homeobox code, Msx1 may define the incisor region (28) whereas Barx1 in combination with Dlx2 specifies molar region (27,57,58). This assertion is exemplified in Msx1 knock out mice which exhibit no initiation of incisor development in contrast to molars where development is blocked later at the bud stages (32). To understand further the means of spatially restricting the expression of the Msx1 homeoprotein, the role of Msx1 AS transcript in the control of protein expression was considered. Our data showed that at E9.5 Msx1 AS transcript was not expressed in the mandibular arch in contrast to the S transcript, whereas at E10.5 Msx1 AS transcript was detected and presented an expression pattern complementary to the S transcript and protein. Such a result suggests that Msx1 AS transcript could participate to Msx1 homeoprotein expression restriction within the distal region of the mandible. At this stage, some homeobox containing gene expression patterns in the mesenchyme are driven by epithelial signaling (28,31,33,34,37,54,59,60). Two major signaling molecules implicated are growth factors BMP4 and FGF8. They are secreted by the distal and proximal parts of the mandibular oral epithelium, respectively. Effects of these diffusible growth factors on Msx1, Barx1 and Dlx2 expression have been established by application of protein soaked beads in mandibular explant cultures. These experiments show that mesenchymal expression of Barx1 and Dlx2 is induced by FGF8 while inhibited by BMP4 (34,53,61). In contrast, Msx1 S transcript appears to be induced around both FGF8 and BMP4 soaked beads (28,33,37,53) raising the question of Msx1 S transcript expression control (inhibition) in the mandible proximal parts in vivo. Interestingly, at E10.5, Msx1 AS transcript showed an expression pattern in the mesenchyme of the mandible similar to those of Barx1 and Dlx2 (Figure 4A, B and D) and in opposition to Msx1 S transcript and homeoprotein expression pattern (Figure 4C). The presently reported expression of Msx1 AS transcript lead to propose a hypothesis: the AS RNA would inhibit S RNA and protein expression in the proximal parts of the mandible by a mechanism that still remains to be elucidated but that could occur at transcriptional level according to the absence of S transcript detection. The present soak-beads experiments provided evidence that the mesenchymal expression of Msx1 AS transcript at E10.5 is indeed responsive to FGF8 and not BMP4 (Figure 4F, I and K). Interestingly, the Msx1 AS transcript activation by FGF8 at E10.5 appears to be correlated to an absence of Msx1 S transcript response to this growth factor, while beads experiments suggest that Msx1 S transcript should be activated (Figure 4G). In vitro overexpression experiments in mesenchymal MD10-H1 dental cells (39) show that Msx1 AS transcript expression may be regulated by Dlx2 but not by Barx1 (Figure 4L), suggesting that Dlx-2 could act as an intermediary in the pathway between FGF8 and Msx1 AS. A schematic representation of FGF8 and BMP4 differential regulation of both Msx1 transcripts mesenchymal expression in the E10.5 mandible is presented in Figure 4M.

    Finally, Msx1 AS transcript appeared to be inserted inside the homeobox code at E10.5 through its joint expression in the proximal region with Barx1 and Dlx2. At this developmental stage, the obtained data suggest that Msx1 AS transcript would spatially limit Msx1 S transcript and protein expression areas. The exclusive Msx1 AS transcript expression might result in the complete inhibition of Msx1 S transcript expression by a mechanism that remains to be elucidated.

    Msx1 AS transcript and tooth morphogenesis

    During tooth morphogenesis, Msx1 (S transcript) was shown to be expressed in the mesenchyme with a progressive restriction, from E11.5 to E12.5, to the dental mesenchyme surrounding both incisor and molar buds (22,62,63) (Figure 5). The importance of epithelial signaling, in particular BMP4 expression, in such a restriction has been shown and corresponds to a crucial step in the tooth morphogenesis process (31,33). Moreover, the importance of Msx1 homeoprotein expression in dental mesenchyme is evidenced by the observation of molar development arrest at the bud stage (E13.5) in Msx1 null mutant mice (32). Recombination experiments between wild-type and Msx1 null mutant dental tissues showed that Msx1 mesenchymal expression is transiently necessary at the early bud stage (33) mainly to induce BMP4 expression in the dental mesenchyme as validated by BMP4 complementation experiment using transgenic animals (64,65). Indeed, the important function of Msx1 homeoprotein in tooth morphogenesis is its implication in epithelial–mesenchymal interactions and the question of the AS transcript part in this function control is raised from this point of view. Our results showed that Msx1 AS transcript was continuously expressed during tooth morphogenesis with a stage-specific pattern. At the dental lamina stage (E11.5), the AS transcript was detected jointly with the S transcript in the mesenchyme and by E12.5, its expression was progressively restricted to the mesenchyme area surrounding the epithelial invagination, as reported previously for the S transcript. At the following stages (cap and bell stages), the Msx1 S and AS transcripts expression patterns diverge showing a complementary in the mesenchyme. The S transcript was expressed in the dental cells which will differentiate into odontoblasts and pulp cells whereas the AS transcript was expressed in cells which will give rise to the follicular sac and alveolar bone. However, the function associated to such a distribution of the two transcripts remains unclear. Concerning Msx1 S and AS transcript expression in dental epithelium (detected only with certain probes), the situation is even more complex and discussed below. Indeed, according to our data whatever be the stage of tooth morphogenesis Msx1 homeoprotein expression was restricted to the mesenchyme. Finally, our result establishes that during tooth morphogenesis, Msx1 AS transcript showed a dynamic expression pattern relative to the S transcript and the protein suggesting its implication in the protein localization and expression level.

    In conclusion, regarding Msx1 transcripts and relative protein expression during tooth morphogenesis, five distinct situations were identified depending on the developmental stages, areas and tissues: no expression (at E11.5–E12.5 in the epithelium), exclusive expression of S transcript (at E14.5 and E16.5 in dental mesenchyme), exclusive expression of AS transcript (at E16.5 in epithelial regions such as stellate reticulum and stratum intermedium), coexpression of S and AS (i) with all probes (E11.5–E12.5) or (ii) with certain probes (at E13.5, E16.5 and D0). The differences observed with the set of probes used could be explained by the complementarities of S and AS exon 2 sequences. Indeed, the exclusive results obtained with S and AS specific probes suggest that at E13.5, E16.5 and D0 the overlapping sequence from both transcripts are not accessible to exon two probes. So, we hypothesized that S and AS transcripts could physically interfere and impair riboprobe hybridization in this area. Functionally, the five situations should be considered through the narrow window of AS RNA knowledge. Several transcriptional and post-transcriptional events are candidates for the inhibition of Msx1 homeoprotein. These events have been recently reviewed and classified in four categories (5). Transcriptional interference and S gene silencing ends with exclusive expression of AS RNA. Posttranscriptional mechanisms (RNA masking and degradation) are those which involve S–AS coexpression and finely tuned ratios of S/AS transcripts or modifications in RNAs overlapping sequences. According to coexpression or distinct expression patterns of both Msx1 transcripts, the bidirectional convergent and overlapping transcription system would function differently. In the case of distinct expression pattern of S and AS transcripts, transcriptional interference or gene silencing could be implied, whereas, in the case of coexpression, RNA masking and degradation could be involved (5,66,67). Hence two types of ratio threshold could be defined. A macrothresholding when only one kind of Msx1 transcript is expressed (E9.5–E10.5) and a microthresholding (E13.5–E16.5-D0) when the two transcripts are coexpressed. Further complementary studies at the cellular level would be necessary to clearly establish the presumably distinct mechanisms.

    Natural antisense transcripts: a growing list

    The Msx1 AS transcript is not an isolated case of large natural AS RNA. Bidirectional convergent and overlapping transcription is an emerging field opening a previously unsuspected level of complexity in protein expression regulation (5). AS transcripts have been associated with different processes during development including parental imprinting (68), developmental clock control (7) and transcriptional regulation (5). The precise molecular modalities of such a control are not clarified. However, in vitro experiments have established the importance of the relative ratio of the two transcripts. This has been shown previously for Msx1 by in vivo observations of the two transcripts distribution during osteoblast differentiation process (12).

    Interestingly, Msx1 is not the only gene implicated in maxillofacial and limb development showing a bidirectional convergent and overlapping transcription. Thus, homeobox containing genes belonging to the Dlx gene family have also AS transcripts. This is the case for Dlx1 (69) and Dlx6 (67) genes. AS transcripts were also described for other divergent homeobox genes as Otx2 (70) and members of the Hox clusters such as HoxA11 (11) and HoxD3 (71), raising the question of the general importance of bidirectional convergent and overlapping transcription during development. In vitro studies have established that (i) Msx and Dlx homeoproteins can form heterodimers through their homeodomains and consequently inhibit each other's transcriptional activity (21) (F. Lezot, unpublished data) and (ii) that Dlx5 (12) and Dlx2 (this study) homeoproteins can differentially modulated Msx1 AS transcript expression, and thus Msx1 homeoprotein expression. These in vitro observations strongly suggest that understanding AS transcript integration in developmental signaling pathways will be a difficult, but necessary challenge for the next decade. Interestingly, 9 out of 248 genes shown to be involved in tooth development (indexed on http://bite-it.helsinki.fi) have so far been shown to have AS transcripts. Moreover, computer analysis of available human and mouse genome sequences have shown that 15 and 20% of the mouse and human genes can have AS transcript, respectively (4,9). Finally it appears that the importance of AS transcripts is largely underestimated concerning development and more precisely during tooth development. ‘Making sense with antisense’ may be operative for the understanding of morphogenetic gradients in developmental biology.

    ACKNOWLEDGEMENTS

    We thank Drs Isabelle Miletich, Agamemnon Grigoriadis and Stephanie Ricaud for discussions and critical reading of the manuscript. This work was supported by grants of INSERM U714 laboratory, ‘ATC Viellissement 2002 INSERM’ grant to F.L. and ‘GIS Maladie Rare INSERM’ grant no. A02177HS to A.B. A.E.C. was a recipient of fellowship from MENRT (Ministère de la Recherche). Funding to pay the Open Access publication charges for this article was provided by INSERM U714.

    REFERENCES

    Tomizawa, J., Itoh, T., Selzer, G., Som, T. (1981) Inhibition of ColE1 RNA primer formation by a plasmid-specified small RNA Proc. Natl Acad. Sci. USA, 78, 1421–1425 .

    Stougaard, P., Molin, S., Nordstrom, K. (1981) RNAs involved in copy-number control and incompatibility of plasmid R1 Proc. Natl Acad. Sci. USA, 78, 6008–6012 .

    Vanhee-Brossollet, C. and Vaquero, C. (1998) Do natural antisense transcripts make sense in eukaryotes? Gene, 211, 1–9 .

    Chen, J., Sun, M., Kent, W.J., Huang, X., Xie, H., Wang, W., Zhou, G., Shi, R.Z., Rowley, J.D. (2004) Over 20% of human transcripts might form sense-antisense pairs Nucleic Acids Res., 32, 4812–4820 .

    Lavorgna, G., Dahary, D., Lehner, B., Sorek, R., Sanderson, C.M., Casari, G. (2004) In search of antisense Trends Biochem. Sci., 29, 88–94 .

    Hannon, G.J. (2002) RNA interference Nature, 418, 244–251 .

    Pasquinelli, A.E. and Ruvkun, G. (2002) Control of developmental timing by microRNAs and their targets Annu. Rev. Cell. Dev. Biol., 18, 495–513 .

    Mansfield, J.H., Harfe, B.D., Nissen, R., Obenauer, J., Srineel, J., Chaudhuri, A., Farzan-Kashani, R., Zuker, M., Pasquinelli, A.E., Ruvkun, G., et al. (2004) MicroRNA-responsive ‘sensor’ transgenes uncover Hox-like and other developmentally regulated patterns of vertebrate microRNA expression Nature Genet., 36, 1079–1083 .

    Kiyosawa, H., Yamanaka, I., Osato, N., Kondo, S., Hayashizaki, Y. (2003) Antisense transcripts with FANTOM2 clone set and their implications for gene regulation Genome Res., 13, 1324–1334 .

    Kumar, M. and Carmichael, G.G. (1998) Antisense RNA: function and fate of duplex RNA in cells of higher eukaryotes Microbiol. Mol. Biol. Rev., 62, 1415–1434 .

    Hsieh-Li, H.M., Witte, D.P., Weinstein, M., Branford, W., Li, H., Small, K., Potter, S.S. (1995) Hoxa 11 structure, extensive antisense transcription, and function in male and female fertility Development, 121, 1373–1385 .

    Blin-Wakkach, C., Lezot, F., Ghoul-Mazgar, S., Hotton, D., Monteiro, S., Teillaud, C., Pibouin, L., Orestes-Cardoso, S., Papagerakis, P., Macdougall, M., et al. (2001) Endogenous Msx1 antisense transcript: in vivo and in vitro evidences, structure, and potential involvement in skeleton development in mammals Proc. Natl Acad. Sci. USA, 98, 7336–7341 .

    Song, K., Wang, Y., Sassoon, D. (1992) Expression of Hox-7.1 in myoblasts inhibits terminal differentiation and induces cell transformation Nature, 360, 477–481 .

    Odelberg, S.J., Kollhoff, A., Keating, M.T. (2000) Dedifferentiation of mammalian myotubes induced by Msx1 Cell, 103, 1099–1109 .

    Houzelstein, D., Auda-Boucher, G., Cheraud, Y., Rouaud, T., Blanc, I., Tajbakhsh, S., Buckingham, M.E., Fontaine-Perus, J., Robert, B. (1999) The homeobox gene Msx1 is expressed in a subset of somites, and in muscle progenitor cells migrating into the forelimb Development, 126, 2689–2701 .

    Houzelstein, D., Cheraud, Y., Auda-Boucher, G., Fontaine-Perus, J., Robert, B. (2000) The expression of the homeobox gene Msx1 reveals two populations of dermal progenitor cells originating from the somites Development, 127, 2155–2164 .

    Hill, R.E., Jones, P.F., Rees, A.R., Sime, C.M., Justice, M.J., Copeland, N.G., Jenkins, N.A., Graham, E., Davidson, D.R. (1989) A new family of mouse homeo box-containing genes: molecular structure, chromosomal location, and developmental expression of Hox-7.1 Genes Dev, 3, 26–37 .

    Robert, B., Sassoon, D., Jacq, B., Gehring, W., Buckingham, M. (1989) Hox-7, a mouse homeobox gene with a novel pattern of expression during embryogenesis EMBO J., 8, 91–100 .

    Catron, K.M., Zhang, H., Marshall, S.C., Inostroza, J.A., Wilson, J.M., Abate, C. (1995) Transcriptional repression by Msx-1 does not require homeodomain DNA-binding sites Mol. Cell. Biol., 15, 861–871 .

    Davidson, D. (1995) The function and evolution of Msx genes: pointers and paradoxes Trends Genet., 11, 405–411 .

    Zhang, H., Wang, H., Siavolino, P., Iler, N., Shen, M.M., Abate-Shen, C. (1997) Heterodimerization of Msx1 and Dlx homeoproteins results in functional antagonism Mol. Cell. Biol., 17, 2920–2932 .

    MacKenzie, A., Leeming, G.L., Jowett, A.K., Ferguson, M.W., Sharpe, P.T. (1991) The homeobox gene Hox 7.1 has specific regional and temporal expression patterns during early murine craniofacial embryogenesis, especially tooth development in vivo and in vitro Development, 111, 269–285 .

    Houzelstein, D., Cohen, A., Buckingham, M.E., Robert, B. (1997) Insertional mutation of the mouse Msx1 homeobox gene by an nlacZ reporter gene Mech. Dev., 65, 123–133 .

    Thesleff, I. (2003) Epithelial–mesenchymal signalling regulating tooth morphogenesis J. Cell Sci., 116, 1647–1648 .

    Sharpe, P.T. (1995) Homeobox genes and orofacial development Connect. Tissue Res., 32, 17–25 .

    Thomas, B.L. and Sharpe, P.T. (1998) Patterning of the murine dentition by homeobox genes Eur. J. Oral Sci., 106, Suppl. 1, 48–54 .

    Tucker, A. and Sharpe, P.T. (2004) The cutting-edge of mammalian development; how the embryo makes teeth Nature Rev. Genet., 5, 499–508 .

    Tucker, A.S., Al Khamis, A., Sharpe, P.T. (1998) Interactions between Bmp-4 and Msx-1 act to restrict gene expression to odontogenic mesenchyme Dev. Dyn., 212, 533–539 .

    Thesleff, I., Keranen, S., Jernvall, J. (2001) Enamel knots as signaling centers linking tooth morphogenesis and odontoblast differentiation Adv. Dent. Res., 15, 14–18 .

    Thesleff, I. and Sharpe, P.T. (1997) Signalling networks regulating dental development Mech. Dev., 67, 111–123 .

    Jernvall, J. and Thesleff, I. (2000) Reiterative signaling and patterning during mammalian tooth morphogenesis Mech. Dev., 92, 19–29 .

    Satokata, I. and Maas, R. (1994) Msx1 deficient mice exhibit cleft palate and abnormalities of craniofacial and tooth development Nature Genet., 6, 348–356 .

    Chen, Y., Bei, M., Woo, I., Satokata, I., Maas, R. (1996) Msx1 controls inductive signaling in mammalian tooth morphogenesis Development, 122, 3035–3044 .

    Thomas, B.L., Liu, J.K., Rubenstein, J.L., Sharpe, P.T. (2000) Independent regulation of Dlx2 expression in the epithelium and mesenchyme of the first branchial arch Development, 127, 217–224 .

    Lezot, F., Thomas, B., Hotton, D., Forest, N., Orestes-Cardoso, S., Robert, B., Sharpe, P.T., Berdal, A. (2000) Biomineralization, life-time of odontogenic cells and differential expression of the two homeobox genes MSX-1 and DLX-2 in transgenic mice J. Bone Miner. Res., 15, 430–441 .

    Hotton, D., Mauro, N., Lezot, F., Forest, N., Berdal, A. (1999) Differential expression and activity of tissue-nonspecific alkaline phosphatase (TNAP) in rat odontogenic cells in vivo J. Histochem. Cytochem., 47, 1541–1552 .

    Ferguson, C.A., Tucker, A.S., Sharpe, P.T. (2000) Temporospatial cell interactions regulating mandibular and maxillary arch patterning Development, 127, 403–412 .

    Mitsiadis, T.A., Angeli, I., James, C., Lendahl, U., Sharpe, P.T. (2003) Role of Islet1 in the patterning of murine dentition Development, 130, 4451–4460 .

    MacDougall, M., Thiemann, F., Ta, H., Hsu, P., Chen, L.S., Snead, M.L. (1995) Temperature sensitive simian virus 40 large T antigen immortalization of murine odontoblast cell cultures: establishment of clonal odontoblast cell line Connect. Tissue Res., 33, 97–103 .

    Lezot, F., Descroix, V., Mesbah, M., Hotton, D., Blin, C., Papagerakis, P., Mauro, N., Kato, S., MacDougall, M., Sharpe, P.T., et al. (2002) Cross-talk between Msx/Dlx homeobox genes and vitamin D during tooth mineralization Connect. Tissue Res., 43, 509–514 .

    Kuzuoka, M., Takahashi, T., Guron, C., Raghow, R. (1994) Murine homeobox-containing gene, Msx-1: analysis of genomic organization, promoter structure, and potential autoregulatory cis-acting elements Genomics, 21, 85–91 .

    Brown, J.M., Wedden, S.E., Millburn, G.H., Robson, L.G., Hill, R.E., Davidson, D.R., Tickle, C. (1993) Experimental analysis of the control of expression of the homeobox-gene Msx-1 in the developing limb and face Development, 119, 41–48 .

    Reginelli, A.D., Wang, Y.Q., Sassoon, D., Muneoka, K. (1995) Digit tip regeneration correlates with regions of Msx1 (Hox 7) expression in fetal and newborn mice Development, 121, 1065–1076 .

    Lallemand, Y., Nicola, M.-A., Ramos, C., Bach, A., Saint Cloment, C., Robert, B. (2005) Analysis of Msx1; Msx2 double mutants reveals multiple roles for Msx genes in limb development Development, 132, 3003–3014 .

    Berdal, A., Lezot, F., Pibouin, L., Hotton, D., Ghoul-Mazgar, S., Teillaud, C., Robert, B., MacDougall, M., Blin, C. (2002) Msx1 homeogene antisense mRNA in mouse dental and bone cells Connect. Tissue Res., 43, 148–152 .

    Woloshin, P., Song, K., Degnin, C., Killary, A.M., Goldhamer, D.J., Sassoon, D., Thayer, M.J. (1995) MSX1 inhibits myoD expression in fibroblast x 10T1/2 cell hybrids Cell, 82, 611–620 .

    Hu, G., Lee, H., Price, S.M., Shen, M.M., Abate-Shen, C. (2001) Msx homeobox genes inhibit differentiation through upregulation of cyclin D1 Development, 128, 2373–2384 .

    Han, J., Ito, Y., Yeo, J.Y., Sucov, H.M., Maas, R., Chai, Y. (2003) Cranial neural crest-derived mesenchymal proliferation is regulated by Msx1-mediated p19(INK4d) expression during odontogenesis Dev. Biol., 261, 183–196 .

    Orestes-Cardoso, S.M., Nefussi, J.R., Hotton, D., Mesbah, M., Orestes-Cardoso, M.D., Robert, B., Berdal, A. (2001) Postnatal Msx1 expression pattern in craniofacial, axial, and appendicular skeleton of transgenic mice from the first week until the second year Dev. Dyn., 221, 1–13 .

    Lyons, G.E., Houzelstein, D., Sassoon, D., Robert, B., Buckingham, M.E. (1992) Multiple sites of Hox-7 expression during mouse embryogenesis: comparison with retinoic acid receptor mRNA localization Mol. Reprod. Dev., 32, 303–314 .

    MacKenzie, A., Ferguson, M.W., Sharpe, P.T. (1991) Hox-7 expression during murine craniofacial development Development, 113, 601–611 .

    Phippard, D.J., Weber-Hall, S.J., Sharpe, P.T., Naylor, M.S., Jayatalake, H., Maas, R., Woo, I., Roberts-Clark, D., Francis-West, P.H., Liu, Y.H., et al. (1996) Regulation of Msx-1, Msx-2, Bmp-2 and Bmp-4 during foetal and postnatal mammary gland development Development, 122, 2729–2737 .

    Tucker, A.S., Matthews, K.L., Sharpe, P.T. (1998) Transformation of tooth type induced by inhibition of BMP signaling Science, 282, 1136–1138 .

    Peters, H. and Balling, R. (1999) Teeth. Where and how to make them Trends Genet., 15, 59–65 .

    McCollum, M. and Sharpe, P.T. (2001) Evolution and development of teeth J. Anat., 199, 153–159 .

    Cobourne, M.T. and Sharpe, P.T. (2003) Tooth and jaw: molecular mechanisms of patterning in the first branchial arch Arch. Oral Biol., 48, 1–14 .

    Tissier-Seta, J.P., Mark, M., Mattei, M.G., Goridis, C., Brunet, J.F. (1995) Barx1, a new mouse homeodomain transcription factor expressed in cranio-facial ectomesenchyme and the stomach Mech. Dev., 51, 3–15 .

    Thomas, B.L., Tucker, A.S., Qui, M., Ferguson, C.A., Hardcastle, Z., Rubenstein, J.L., Sharpe, P.T. (1997) Role of Dlx-1 and Dlx-2 genes in patterning of the murine dentition Development, 124, 4811–4818 .

    Bei, M. and Maas, R. (1998) FGFs and BMP4 induce both Msx1-independent and Msx1-dependent signaling pathways in early tooth development Development, 125, 4325–4333 .

    Tucker, A.S., Yamada, G., Grigoriou, M., Pachnis, V., Sharpe, P.T. (1999) Fgf-8 determines rostral-caudal polarity in the first branchial arch Development, 126, 51–61 .

    Park, B.K., Sperber, S.M., Choudhury, A., Ghanem, N., Hatch, G.T., Sharpe, P.T., Thomas, B.L., Ekker, M. (2004) Intergenic enhancers with distinct activities regulate Dlx gene expression in the mesenchyme of the branchial arches Dev. Biol., 268, 532–545 .

    MacKenzie, A., Ferguson, M.W., Sharpe, P.T. (1992) Expression patterns of the homeobox gene, Hox-8, in the mouse embryo suggest a role in specifying tooth initiation and shape Development, 115, 403–420 .

    Jowett, A.K., Vainio, S., Ferguson, M.W.J., Sharpe, P.T., Thesleff, I. (1993) Epithelial–mesenchymal interactions are required for msx1 and msx2 gene expression in the developing murine tooth Development, 117, 461–470 .

    Bei, M., Kratochwil, K., Maas, R.L. (2000) BMP4 rescues a non-cell-autonomous function of Msx1 in tooth development Development, 127, 4711–4718 .

    Zhao, X., Zhang, Z., Song, Y., Zhang, X., Zhang, Y., Hu, Y., Fromm, S.H., Chen, Y. (2000) Transgenically ectopic expression of Bmp4 to the Msx1 mutant dental mesenchyme restores downstream gene expression but represses Shh and Bmp2 in the enamel knot of wild type tooth germ Mech. Dev., 99, 29–38 .

    Farrell, C.M. and Lukens, L.N. (1995) Naturally occurring antisense transcripts are present in chick embryo chondrocytes simultaneously with the down-regulation of the alpha 1 (I) collagen gene J. Biol. Chem., 270, 3400–3408 .

    Liu, J.K., Ghattas, I., Liu, S., Chen, S., Rubenstein, J.L. (1997) Dlx genes encode DNA-binding proteins that are expressed in an overlapping and sequential pattern during basal ganglia differentiation Dev. Dyn., 210, 498–512 .

    Delaval, K. and Feil, R. (2004) Epigenetic regulation of mammalian genomic imprinting Curr. Opin. Genet. Dev., 14, 188–195 .

    McGuinness, T., Porteus, M.H., Smiga, S., Bulfone, A., Kingsley, C., Qiu, M., Liu, J.K., Long, J.E., Xu, D., Rubenstein, J.L. (1996) Sequence, organization, and transcription of the Dlx-1 and Dlx-2 locus Genomics, 35, 473–485 .

    Alfano, G., Vitiello, C., Caccioppoli, C., Caramico, T., Carola, A., Szego, M.J., McInnes, R.R., Auricchio, A., Banfi, S. (2005) Natural antisense transcripts (NATs) associated with genes involved in eye development Hum. Mol. Genet., 14, 913–923 .

    Bedford, M., Arman, E., Orr-Urtreger, A., Lonai, P. (1995) Analysis of the Hoxd-3 gene: structure and localization of its sense and natural antisense transcripts DNA Cell Biol., 14, 295–304 .(Amélie E. Coudert, Laurence Pibouin, Bri)