PITX2 gain-of-function induced defects in mouse forelimb development
© Holmberg et al; licensee BioMed Central Ltd. 2008
Received: 03 October 2007
Accepted: 29 February 2008
Published: 29 February 2008
Skip to main content
© Holmberg et al; licensee BioMed Central Ltd. 2008
Received: 03 October 2007
Accepted: 29 February 2008
Published: 29 February 2008
Limb development and patterning originate from a complex interplay between the skeletal elements, tendons, and muscles of the limb. One of the genes involved in patterning of limb muscles is the homeobox transcription factor Pitx2 but its role in forelimb development is uncharacterized. Pitx2 is expressed in the majority of premature presumptive forelimb musculature at embryonic day 12.5 and then maintained throughout embryogenesis to adult skeletal muscle.
To further study the role of Pitx2 in forelimb development we have generated transgenic mice that exhibit a pulse of PITX2 over-expression at embryonic day 13.5 and 14.5 in the developing forelimb mesenchyme. These mice exhibit a distal misplacement of the biceps brachii insertion during embryogenesis, which twists the forelimb musculature resulting in severe skeletal malformations. The skeletal malformations have some similarities to the forearm deformities present in Leri-Weill dyschondrosteosis.
Taken together, the tendon, muscle, and bone anomalies further support a role of Pitx2 in forelimb development and may also shed light on the interaction between the skeletal elements and muscles of the limb during embryogenesis.
Limb development and patterning originate from close interactions between tendon, cartilage and muscle precursor cells. Mouse forelimb development is first evident at about embryonic day (E) 9.5. Approximately 24 h later myogenic cells are identified at the base of the forelimb and at E11.5 the first hint of humerus is apparent . At E14.5 a miniature model of the forelimb has been formed. Limb muscle precursors migrate from the lateral part of the somites into the limb bud where they undergo final differentiation. Among the transcription factors involved in early myogenesis are Pax3 and Lbx1 whose expression precedes the expression of myogenic regulatory factors (MRFs). MRFs belong to the MyoD family of basic helix-loop-helix factors. In mammals there are four such factors: Myf5, MyoD, myogenin, and MRF4 . Little is known about the mechanisms whereby these genes regulate limb muscle development. The homeobox transcription factor PITX2 was originally identified as one of the genes responsible for Axenfeld-Rieger syndrome, mainly affecting eyes, teeth, and abdominal organs [3, 4]. Pitx2 is expressed in a subset of Pax3+ limb muscle precursors already at E10.5. By E12.5 Pitx2 is expressed in all limb musculature and persists until adulthood . Still, Pitx2 null mutants form nearly all muscle anlagen even though several of these muscle anlagen are distorted, coupled with malformation of the part of the body to which they attach . Muscles attach to bone through tendons. Limb tendon cells originate from lateral plate mesoderm and tendon progenitor cells are regionalized in the dorsal and ventral areas of the limb where they are mixed with muscle progenitor cells [7, 8]. Compared to other mesodermal tissues, such as blood vessels, cartilage, bone, and muscles, very little is known about the early formation and role of tendons during development. Among the transcription factors identified in developing tendons are scleraxis, Eya1, Eya2, Six1, and Six2 of which Eya1, Eya2, Six1 and Six2 are also expressed in limb muscle precursors [7, 9–11]. This parallel expression in myoblast precursors of somite origin and in mesenchymal cells derived from the lateral plate may ensure correct and concerted migration of the two cell types . To further study the role of Pitx2 in forelimb development we have generated mice that exhibit a brief pulse of PITX2 over-expression in the forelimb mesenchyme from E13.5 to E14.5. The expression is driven by mouse keratocan (Kera) 5'-flanking sequence, which has been used previously to achieve over-expression of PITX2 in the cornea . The construct was termed Ktcn-PITX2. Keratocan is one of three major components of the extracellular keratan sulfate proteoglycans present mainly in vertebrate corneal stroma but also expressed in non-ocular tissues such as skeletal muscle and tendon [14, 15]. The Kera gene is expressed in limbs of mouse embryos at E13.5 and E14.5 [16, 17]. This is the first report of PITX2 over-expression in the forelimb. The Ktcn-PITX2 mice exhibit PITX2 over-expression in the anterior forelimb mesenchyme extending from the humerus to the radius. The cells over-expressing PITX2 are of non-myogenic origin and co-express Six2. As Six2 is involved in tendon development, we hypothesize that the observed expression disturbs correct muscle insertion, which in the Ktcn-PITX2 mouse leads to a random left-right distal misplacement of the biceps brachii insertion. This in turn results in a 180 degrees twist of the forelimb musculature. The muscle and tendon anomalies also lead to severe skeletal malformations consisting of a shortened, thickened and malformed humerus, a bowed ulna and a deformed radius. These skeletal malformations have some similarities to the pathogenesis of Leri-Weill dyschondrosteosis, which is characterized by disproportionate short stature and a characteristic curving of the radius, known as the Madelung deformity . In conclusion, these findings may increase our understanding about the role of Pitx2 in limb development and on the interactions between muscle, tendon, and bone during development.
Distribution of phenotypes in 1-month-old wild type and Ktcn-PITX2 mice.
Pitx2 acts in hindlimb development together with Pitx1 where the former specifies laterality and the latter determines hindlimb identity . During forelimb development Pitx2 is expressed in almost all muscle anlagen from embryogenesis until adulthood [4, 5, 22–24]. We show here for the first time that over-expression of PITX2 in non-myogenic cells of the developing forelimb results in a distal misplacement of the biceps brachii insertion, which gives rise to severe bone malformations, randomly affecting left or right forelimb. The bone anomalies include a diminished deltoid tuberosity, a shortened, thickened and malformed humerus. In addition, it causes the radius to approach the elbow via a perpendicular bend, which in turn limits the elbow movement, and forces a dorsally dislocated ulna to curve over radius to reach the carpals. The phenotype worsens two to three weeks after birth, possibly due to the increased load on the forelimbs as the mouse starts to walk. On the other hand we frequently detected mice with both forelimbs affected from early on. Also, mice with only one affected leg never developed any phenotype on the supporting leg. The skeletal malformations have some similarities to the pathogenesis of Leri-Weill dyschondrosteosis, characterized by disproportionate short stature, curving of the radius, and subluxation of the distal end of the ulna. Mutations in the SHOX gene have been shown to cause the disorder. However, for 27% of the cases, no causative gene has been identified . Pitx2 is expressed in the left lateral plate mesoderm and acts as an effector for left-right asymmetry in mesoderm derived organs such as lungs . Also, Pitx1-/- embryos show left-right asymmetry in the severity of a hindlimb phenotype due to redundancy between Pitx1 and Pitx2 . Still, the limbs are symmetrical. It has previously been shown that function of Pitx2 is dosage-sensitive [13, 27, 28]. Hence, the more plausible explanation for the random left-right forelimb anomalies of the Ktcn-PITX2 mice could be minor differences in PITX2 expression levels between left and right. Pitx2 expression in limbs is restricted to muscle lineages and it has recently been shown that Pitx2 is one of the most complete muscle markers . Immunohistochemistry on wild type forelimb sections demonstrate co-expression of Pitx2 and other muscle markers such as MyoD, myogenin, and myosin. In contrast, the cells over-expressing PITX2 in the Ktcn-PITX2 forelimbs do not express any of these muscle markers. Based on the observed tendon anomalies in the Ktcn-PITX2 forelimbs we stained for Six2, normally expressed in developing tendons. Co-expression of PITX2 and Six2 could indicate that the cells over-expressing PITX2 are involved in tendon development and positioning. The factors regulating correct attachments of tendons to bone and the time of commitment of tendon precursors to a tendon cell fate in vertebrates are largely unknown . However, reports have shown that myotubes can induce tendon primordial into individual tendons and, moreover, if tendon precursors are removed, myotubes attach ectopically . It is possible that the observed expression of Six2 in the cells over-expressing PITX2 interferes with the signals coordinating correct tendon insertion. Based on the observed co-expression of PITX2 and Six2 in our mouse model it is intriguing to note that both transcription factors are expressed during other developmental processes. Neural crest cells that migrate to the periocular mesenchyme during eye development express both Pitx2 and Six2 [31, 32]. In addition, Eya1, another limb tendon marker, is expressed in developing anterior chamber structures of the eye including the iris, ciliary structures, and cornea, tissues that also express Pitx2 . It would be interesting to further study a possible relation between Pitx2, Six2, and Eya1. Notably, Pitx1 is closely related to Pitx2 and normally involved in and restricted to hindlimb development. When ectopically expressed in the forelimbs it induces transformation and translocation of specific muscles, tendons, and bones of the forelimb . Finally, the temporal overlap of muscle and tendon formation with the over-expression of PITX2 makes the Ktcn-PITX2 mouse an appropriate animal model to study the development of the musculoskeletal system.
Over-expression of PITX2 during mouse forelimb development results in severe tendon, muscle, and bone anomalies. These observations further support a role of Pitx2 in forelimb development. This animal model may also be valuable for future studies on the interaction between the skeletal elements and muscles of the limb during embryogenesis and it could provide information on new regulatory pathways involving PITX2.
Transgenic animals were previously described . The local ethics committee for animal research has approved of our experiments (Lund district). The mice were housed and treated according to guidelines of national and local animal ethics guidelines at the BMC animal facilities at Lund University, Lund.
We used Trizol (InVitrogen, Carlsbad, CA) and RNeasy MinElute Clean-up (Qiagen, Valencia, CA) kits to isolate total RNA from limbs. Tissues were whole forelimbs frozen in liquid nitrogen, grinded on dry ice, and further homogenized in Trizol with syringes. RNA was purified with phenol/chloroform, precipitated, and treated with DNA-free (Ambion, Austin, TX) to remove contaminating DNA. Reverse transcription was performed using SuperscriptII (Gibco/BRL/InVitrogen, Carlsbad, CA). We used a mixture of random hexamer and oligo-dT primers (Promega, Madison, WI) according to the recommendations of the manufacturer. We used an amplicon for ribosomal binding protein 18 (Rps18) as a loading control. PITX2A and Rps18 were amplified as described .
Limbs were fixed in 4% formaldehyde in PBS and then embedded in paraffin according to standard procedures. Sections of 5 μm were mounted on slides. Sections for immunohistochemistry were deparaffinized and boiled in 0.1 M sodium citrate and 0.05% Tween 20 for 15 minutes prior to over night primary antibody staining. Antibodies were against PITX2A , cyclin D2 (Abcam, Cambridge, UK), myosin (My32), MyoD1 (SIGMA, Saint Louis, MO), Pax3, myogenin (F5D) (Developmental Studies Hybridoma Bank, Iowa City, IA), and Six2 (Affinity Bioreagents, Golden, CO). We used goat anti-rabbit Alexafluor 546, goat anti-mouse 488, and rabbit-anti-goat Alexafluor 546 as secondary antibodies (Molecular Probes, Eugene, OR). Slides were mounted in Permafluor (Beckman-Coulter, Fullerton, CA).
We used the In situ cell death detection kit fluorescein (Roche, Indianapolis, IN), to test apoptotic activity on 5 μm paraffin sections according to the recommendations of the manufacturer. Slides were mounted in Permafluor (Beckman-Coulter) and viewed in a fluorescence microscope.
Embryos and newborn animals were skinned, eviscerated and fixed in 95% ethanol for one week. The specimens were then placed in acetone for one week, stained in 1 volume 0,1% filtered alizarin red S (SIGMA), 1 volume 0,3% filtered alcian blue 8GX (SIGMA), 1 volume 99.8% acetic acid, and 17 volumes 70% ethanol for one week, cleared in 1% KOH, destained in 20% glycerol + 1% KOH, and transferred through a graded glycerol series and stored in 100% glycerol.
The mice were sacrificed before X-ray exposure. A clinical Arco Ceil, Arcoma, Generator CPI Indico 100, and Canon CXDI-31 Portable DR System were used. Exposure settings were: 46 kV, 1.6 mAs.
We thank Dr. Lena Persson-Feld and staff at Lund University Animal Care facilities; Dr. Madeleine Durbeej-Hjalt and Mikael Akerlund for helpful discussions. Supported by the Swedish Research Council (to T.A.H) and Craafordska Stiftelsen (to T.A.H).
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.