QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 23 March 2005
doi: 10.1242/dev.01773

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Related articles in Development Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Grifone, R. Articles by Maire, P. Search for Related Content PubMed PubMed Citation Articles by Grifone, R. Articles by Maire, P.

Six1 and Six4 homeoproteins are required for Pax3 and Mrf expression during myogenesis in the mouse embryo

Raphaelle Grifone¹, Josiane Demignon^1,*, Christophe Houbron^2,*, Evelyne Souil³, Claire Niro¹, Mary J. Seller⁴, Ghislaine Hamard² and Pascal Maire^1,

¹ Département Génétique, Développement et Pathologie Moléculaire, Institut Cochin – INSERM 567, CNRS UMR 8104, Université Paris V, 24 Rue du Faubourg Saint Jacques 75014 Paris, France
² Plateforme de recombinaison homologue, Institut Cochin – INSERM 567, CNRS UMR 8104, Université Paris V, 24 Rue du Faubourg Saint Jacques 75014 Paris, France
³ Plateforme d'histologie, Institut Cochin – INSERM 567, CNRS UMR 8104, Université Paris V, 24 Rue du Faubourg Saint Jacques 75014 Paris, France
⁴ Division of Medical and Molecular Genetics, Guy's Hospital, London SE1 9RT, UK

View larger version (47K): [in a new window]   Fig. 1. Targeted disruption of the Six4 gene. (A) Schematic representation of the wild-type Six4-Six1 locus, with the restriction sites used in the present study. The 4.2 kb 5' and 3.1 kb 3' homologous fragments are indicated by double headed arrows. The Six4-loxP targeting fragment is shown, and the two potential targeted Six4 mutant alleles after Cre excision are depicted below. Schematic representations are not to scale. (B) Southern blot analysis of genomic DNA digested by PstI and hybridized with a 5' external probe (above) and a 3' external probe (below), before Cre recombination leading to a Six4hygrogfp allele, or after Cre recombination leading to a Six4gfp allele. (C) Phenotype of wild type (left) and Six1–/–Six4–/– (right) newborn mice.

View larger version (72K): [in a new window]   Fig. 2. Skeletal defects in Six1–/–Six4–/– newborn mice as revealed by Alizarin Red (bones) and Alcian Blue (cartilage) staining. (A,D,F) Six1+/–Six4+/– fetuses, (B,C,E,G) Six1–/–Six4–/– fetuses. Lateral and ventral view (A-C) of the trunk skeleton, showing truncation of the ribs, fusion and branching (black arrow in B). The sternum is shortened (black arrow in C). (D,E) Lateral view at the head level showing a shortened mandible bone (*), shortened squamosal bone devoid of retrotympanic process (arrow) and lack of Meckel's cartilage (arrowhead) in the dKO fetuses, and complete absence of ear structure (**). (F,G) Dorsal view of hands of forelimbs at the left or hindlimbs at the right of Six1+/–Six4+/– (F) and Six1–/–Six4–/– fetuses (G) showing a clinodactily in the dKO. The fifth finger is curved (arrows in G). (H) Human DNA samples from one individual with a 14q22 deletion (1), and two control DNA samples (2,3). DNA was digested with BamHI and hybridized with SIX1 cDNA (a, left panel) or SIX5 cDNA (c, right panel), or digested with HindIII and hybridized with SIX4 cDNA (b, middle panel). M.W., standard molecular weights. The signal observed for SIX1 and SIX4 in the individual with the 14q22 deletion is half that one observed in the control DNA, while the signal for SIX5 is comparable in all samples.

View larger version (112K): [in a new window]   Fig. 3. Absence of most skeletal muscles in Six1–/–Six4–/– E18.5 fetuses (A-L) and in Six1–/–Six4–/– E13.5 embryos (M-V). (A,C,E,G,I,K,M,O,Q,S,U) Six1–/+Six4–/+ animals; (B,D,F,H,J,L,N,P,R,T,V) Six1–/–Six4–/– animals. General view of E18.5 heterozygous (A) and double dKO (B) fetuses, as revealed by X-gal staining. Skeletal muscles are absent from the limbs, abdominal ventral muscles are missing and, at the shoulder and hip levels, most muscles are lacking. X-gal Eosin staining on transversal sections (C-H) or immunostaining performed with fast-type myosin heavy chain antibody My32 (I-L) confirmed the general axial muscle hypoplasia and that limbs have no muscle. (C,D) Shoulder level, with the humerus (hu), thymus (thy) and vertebra (v) indicated. Most muscles are lacking in the dKO fetuses, except epaxial muscles, which appear incorrectly shaped, near the vertebral column. (E,F) Hip level revealing deep back muscles in the control and dKO fetus. (G,H) Hip level revealing that proximal muscles around the femur (fe) are lacking in the dKO fetus, and that ventral-most muscles are also lacking (see enlargement of the boxed ventral region. int, intestine. (I,J) Forelimb and (K,L) hindlimb levels, revealing the absence of muscles in the dKO fetuses. r, radius; u, ulna; t, tibia; f, fibula. General view of E13.5 heterozygous (M) and dKO (N) fetuses, as revealed by X-gal staining. Abdominal ventral muscles are missing and, at the shoulder level, most muscles are lacking. Transverse sections (O-V) revealing desmin expression after immunostaining. (O,P) Eye level, showing that most muscles are absent in the dKO embryo. Eo, extraocular muscles; pt, pterigoid muscles; ca, capitalis muscles; nt, neural tube (Q,R) Tongue level, revealing absence of tongue muscle (to) in the dKO embryo. Ma, masseter. (S,T) Forelimb level, revealing absence of any desmin-positive cells in the forelimb of dKO embryo. fl, forelimb; he, heart; db, deep back muscles; st, spinotrapezius; cm, cutaneus maximus; ii, intercostals interni; bb, biceps brachii, pe, pectoralis; tt, transversus thoracis. (U,V) Diaphragm level, showing that ventral muscles and diaphragm are lacking in dKO embryos. li, liver; dia, diaphragm; sm, scalenus medius; ie, intercostales externi; ra, rectus abdominis; ta, transversus abdominis.

View larger version (89K): [in a new window]   Fig. 4. X-Gal and GFP expression in heterozygous (A,C,E,G,I,K,M,O,Q,S,U,W,Y) and homozygous mutant animals (B,D,F,H,J,L,N,P,R,T,V,X,Z) at E9.5 (A-D), E10.5 (E-P), E11.5 (Q-V) and E12.5 embryos (W-Z), revealing Six4-GFP expression and Six1-lacZ expression. (A-D) Six1 expression is detected in the head mesenchyme (asterisk in A), otic vesicle (white arrow in A), branchial arches (black arrow in A) and pharyngeal clefts (white arrows in C) in heterozygous embryos. In dKO embryos, Six1 expression at the head level is severely reduced (asterisk in B) and lost in the otic vesicle (white arrow in B); although mesodermal Six1 expression is still detected (D), most Six1 expression is lost in pharyngeal pouch endoderm and surface ectoderm in dKO embryos. Fusion between the first and second branchial arches is observed at this stage (white arrow in D). At the thoracic level, the nephrogenic chord expression of Six1 is detected in the heterozygote (black arrowhead in A) but not detected in the dKO embryo (black arrowhead in B). At the somitic level, Six1 is expressed at comparable levels in the heterozygous (A) and homozygous (B) animals. (E-P) Six1 is still highly expressed in the ventral otic vesicle (white arrow in E,G) and trigeminal ganglion (white arrowhead in E,G) in heterozygotes, but not in the dKO embryos (white arrow and white arrowhead, respectively, in F and H). (G,H) Enlargement of control (G) and KO (H) embryos at the head level. (I,J) Enlargement of control (I) and KO (J) embryo at the interlimb level. At the somitic interlimb level, Six1 expression is detected in the myotome (E,I). In the dKO, Six1 expression is more diffuse, while still expressed in myotomes (F,J). (K-P) Vibratome sections. At the hindlimb level (K,L), Six1 is detected in the nephrogenic chord (arrow in K) but is not in the dKO, while many ß-gal-positive cells invade this ventral region of the embryo (L). Six1-positive cells enter the limb bud in the heterozygous embryos (K), but not in the dKO embryo, where somitic ß-gal-positive cells are not confined to the somite but are found more ventromedially (L). At the interlimb level (M,N) Six1-positive cells are diffuse and do not extend ventrally in the dKO embryo (N), while they invade ventral region in the heterozygous embryos (M). Six1 expression in DRG is preserved, but a few ß-gal-positive cells are also seen in the neural tube in the dKO (N,P; data not shown). At the forelimb level (O,P), Six1-positive cells are present in the limb bud in the heterozygous embryos (O), while in dKO embryos most cells accumulate less ventrally, and are not found in the limb bud, but are also detected beneath the ectoderm more medially (P). (Q-V) Six1 (Q, R) and Six4 (S-V) are mainly detected in limbs, and somites, where they are co-expressed in the DRG, myotomes, and ventral (black arrows) and dorsal dermomyotomal lips. In dKO, somites still express these two genes (R,T). Myotomes are disorganized, their ventral extensions are reduced (white double headed arrow), and GFP and ß-gal are no longer detected in the ventral lips (black arrows). A population of GFP and ß-gal-positive cells is found in the limbs of dKO embryos but they are not myogenic cells (Bonnin et al., 2005). (W-Z) Six1 is expressed at higher levels in heterozygous (W,Y) than in homozygous mutant embryos (X,Z). At the thoracic level, ventral extension of the dermomyotome, normally marked by Six1 expression, is lost in dKO embryos (white arrows). Posterior expression in the limbs is maintained.

View larger version (81K): [in a new window]   Fig. 5. Increased apoptosis in Six1–/–Six4–/– embryos. Sections (10 µm) of heterozygous (A,C) and dKO (B,D) embryos at E10.5 at the hindlimb level, hybridized with antibodies revealing ß-gal protein (A,B) and activated caspase 3 (C,D), showing activated caspase 3-positive cells only in dKO embryos (white arrow in D). Sections (10 µm) of heterozygous (E) or dKO (F) embryos at E12.5 at the hindlimb level revealed with antibodies against ß-gal protein (in red) and activated caspase 3 (in green), and showing apoptosis of ß-gal-positive cells in dKO embryos that are rerouted ventrally (white arrow in F). Sections (10 µm) of heterozygous (G) and dKO (H) embryos at E10.5 at the first branchial arch level hybridized with antibodies revealing ß-gal protein (in red) and activated caspase 3 (in green), and showing apoptosis of ß-gal-positive cells in dKO embryos.

View larger version (87K): [in a new window]   Fig. 6. Detection of Pax3 (part I), Met (part II), Lbx-1 (part III) and Pax7 (part IV) by whole-mount in situ hybridization on E10.5 heterozygous and dKO embryos. (Part I) Pax3 expression in heterozygous (A,C,E,G,I,K,M,O,Q) and dKO embryos (B,D,F,H,J,L,N,P,R). Pax3 is still expressed in the dKO (B). At the caudal level, in youngest somites Pax3 expression is initiated similarly in both control and dKO embryo (C,D) and its dorsoventral expression domain is comparable, as revealed by vibratome sections (E,F). At the hindlimb level, ventral expression of Pax3 is perturbed (G-J) and no Pax3 myogenic precursor delaminate to reach the limb bud. At the interlimb level (K-N), Pax3 expression is restricted to the posterior lip in the dKO embryo (arrow in L); in ventral (arrow in N) and dorsal lips, expression is progressively lost (L,N) when compared with wild-type expression (K,M). More rostrally at the forelimb level Pax3 is only detected in the caudal lip of dKO embryos (P), and completely lost in anterior, dorsal and ventral lips (P,R) when compared with heterozygous embryos (O,Q). No myogenic precursor is detected in the limb bud (R). Dorsal neural tube expression of Pax3 is conserved in dKO embryos along the rostrocaudal level (F,J,N,R). In occipital somites, Pax3 expression is very low in dKO embryos. (Part II) Comparison of Met expression in heterozygous embryos (A,C,E,G), with its caudal expression in the top right-hand corner of A, and in dKO embryos (B,D,F,H), with its caudal expression in the bottom left-hand corner in B. Although Met expression is correct in the two first caudal somites of dKO embryos, most expression is lost in more rostral somites, and ectodermal expression is detected at the interlimb level in dKO embryos (arrow in B). Transverse sections through embryos at hindlimb level (C,D), interlimb levels (E,F) and forelimb bud levels (G,H). At the hindlimb level, Met can be detected in dorsal and ventral dermomyotomal lips (D) and few cells are detected in the hindlimb bud – in caudal-most areas of this limb bud (arrow in D); however, Met expression is lost more rostrally from these hindlimb somites. The central somitic expression of Met is faint and restricted (H) at the forelimb level (arrowhead). Epaxial Met expression (white arrow in A) is lost in dKO embryos (white arrow in B). (Part III) Lbx1 expression in wild-type (A,C,E) or KO (B,D,F) embryos. Lbx1 expressed at the hypoglossal chord (arrow in C and D) and forelimb bud levels (asterisks in C and D) is abrogated in dKO embryos but still faintly detected in a few somites at the hindlimb level (arrowhead in F). (G) Bandshift assays performed with recombinant Eya1, Six1 or Six1+Eya1 proteins on a potential MEF3 site present in the promoter of mouse Lbx1 gene showing formation of a MEF3Lbx-Six1 or a MEF3Lbx-Six1-Eya1 DNA protein complex. (Part IV) Pax7 expression in heterozygous (A,C,E,G,I) and KO (B,D,F,H,J) embryos. Pax7 is expressed in dermomyotome and its expression pattern is mainly unchanged in KO somites (C,D) at the hindlimb level, as revealed by vibratome sections (E,F), and at the interlimb level (G,H). At the forelimb bud level, we noticed that dorsoventral extension of the expression domain of Pax7 was reduced (I,J).

View larger version (72K): [in a new window]   Fig. 7. Disruption of ventral and dorsal dermomyotomal lips in Six1–/–Six4–/– embryos at caudal (A-D), hindlimb (E-N), thoracic (O-R) and forelimb (S-X) levels. Section (10 µm) of heterozygous (A,B,E,F,I,J,O,P,S,T,W) and dKO (C,D,G,H,K,L,M,N,Q,R,U,V,X) E10.5 embryos hybridized with antibodies revealing Pax3 protein in red (A,C,E,G,M,O,Q,S,U), ß-gal protein in green (B,D,F,H,J,L,N,P,R,T,V), desmin (J,L) and laminin (I,K,W,X). Pax3 expression is similarly expressed in rostral somites of both heterozygous and dKO embryos, and is co-expressed with Six1 (compare C,D with A,B). At the hindlimb level in the dKO embryos, we checked desmin, laminin, Six1 and Pax3 expression in sacral somites facing the limb bud (G,H,K,L) and in more rostral somites facing the limb bud (M,N). Laminin and desmin expression is found in the myotomes in Six1-expressing cells on both wild-type and dKO sections (white arrows in I-L). At this hindlimb level, the myotome of dKO embryos is not disorganized. More rostrally, Pax3 expression is progressively lost ventrally in lumbar somites at the hindlimb level, while most cells are Six1 positive (compare G,H with M,N). At the thoracic level, Pax3 expression is restricted in the medial aspect of the dermomyotome of dKO embryos (Q), while Six1 gene expression is found in the myotomes (R). More rostrally, at the forelimb bud level, Six1-positive/Pax3-negative cells are disorganized and have lost their identity; they also fail to enter the limb bud in the double KO (compare S,T with U,V). At this level, the myotome of dKO embryos is disorganized, as revealed by the low laminin expression (white arrow in X), when compared with laminin expression in control embryos (white arrow in W).

View larger version (82K): [in a new window]   Fig. 8. (Part I) Myf5 expression in E9.25 (20 somites stage) in heterozygous (A) and dKO (B) embryos. Myf5 is expressed in the epaxial somites of heterozygous embryos, and more faintly in epaxialrostral somites in dKO embryos. (Part II) Myogenin expression in E9.5 embryos in heterozygous (A) and dKO (B) embryos. A few positive myogenin-expressing cells are detected in the most rostral somites of dKO embryos (arrows in B). (Part III) Myogenin expression in E10.5 heterozygous (A,C,E) and dKO (B,D,F) embryos. Vibratome sections at the interlimb level (C,D) and forelimb level (E,F) show a strong decrease of myogenin expression that is faintly detected in the epaxial region of dKO embryos, hypaxial extension expression domain being lost. (Part IV) Myod1 expression in E10.5 heterozygous (A,C,E,G) and dKO (B,D,F,H) embryos. Most Myod1 expression is lost in Six1–/–Six4–/– embryos, and remaining expression can be visualized at the interlimb or forelimb levels in the central and hypaxial somite (F,H). (Part V) Myf5 expression in heterozygous (A,C,E,G) and in dKO (B,D,F,H) E10.5 embryos. Myf5 is mainly detected in caudal lips of homozygous dKO animals at interlimb levels (B, upper right), ventral and dorsal lip expression being lost (compare A with B). This expression loss is well detected on vibratome sections at hindlimb (C,D), interlimb (E,F) and forelimb (G,H) levels. In the dKO, at the forelimb level the myotome is formed only of a central region expressing Myf5 (H). In the top right-hand corners of A and B, magnification of interlimb Myf5 expression can be seen that is restricted to the caudal region of the somites in the dKO embryo (B). (Part VI) At E10.5, MRF4 expression is lost in Six1–/–Six4–/– embryos (B), when compared with control hetererozygous expression (A). (Part VII) Myod1 expression in heterozygous (A,C,E,G) and in Six1–/–Six4–/– (B,D,F,H) E11.5 embryos. A low diffuse Myod1 expression in interlimb somites of dKO embryos is detected (B) with restricted ventral extension (B, and enlargement in the top right-hand corner) when compared with Myod1 expression in heterozygous embryos (A, enlargement in the top right-hand corner). Vibratome sections (C-H) showing a faint Myod1 expression at hindlimb level, with most dorsal and ventral extension being lost (C,D). At the interlimb level, Myod1 expression is restricted to the central region of the myotome, dorsal and ventral expression is lost (E,F). A low myotomal expression is detected at the forelimb level (G,H).

View larger version (80K): [in a new window]   Fig. 9. Whole-mount in situ hybridization with Fgf4 (part I), Fgf6 (part II), Fgf8 (part III), Fgfr4 (part IV), scleraxis (parts V,VI) on heterozygous or dKO E10.5 embryos and vibratome sections. (Part I) Fgf4 expression in heterozygous (A,C,E) or dKO (B,D,F) embryos, showing a decrease of Fgf4 somitic expression, AER expression is maintained (A,B). At the interlimb level (C-F), Fgf4 expression is restricted to the central region of the myotome. (Part II) Fgf6 expression in heterozygous (A,C) or dKO (B,D) embryos, showing a severe reduction of Fgf6 in the somites (A,B) of the dKO embryo. Vibratome sections at the interlimb level reveal a strong myotomal expression in the heterozygous embryo (C), while a low signal can be detected in the dKO embryo (D). (Part III) Fgf8 expression in heterozygous (A,D,F) or dKO (B,C,E,G) embryos showing that somitic Fgf8 expression is maintained in the dKO embryos. Fgf8 is expressed in the AER, presomitic mesoderm, otic placode and branchial arches in wild-type or heterozygous embryo (A,D,F). At the somitic level, we detect Fgf8 mRNA, mainly in the caudal lip and more faintly in rostral lip (A,D), allowing Fgf8 to accumulate preferentially in the central myotome (F). In the dKO embryos, presomitic, AER and somitic expression (C,E,G) are preserved, while branchial arches expression (arrows in A and B) is lost (B). (Part IV) Fgfr4 expression in heterozygous (A) or dKO (B) embryos at the interlimb level showing a decrease of Fgfr4 expression in Six1–/–Six4–/– embryos. (Part V) Scleraxis expression in heterozygous (A) or dKO (B) embryos. Scleraxis is expressed in tendon progenitors at the limb level (A), this expression is preserved in the dKO embryo (B), while early somitic syndetomal expression of scleraxis is lost in the dKO embryo (B). (Part VI) Scleraxis is not expressed in Six1 somitic cell population. Double Six1-lacZ (blue) and scleraxis (purple) heterozygous embryo showing the different cell populations of the somites, which express exclusively scleraxis (white arrow) or Six1 (black arrow).