Skip to main content
NCBI home page
Journal of Anatomy logoLink to Journal of Anatomy
. 2013 May 16;223(1):1–13. doi: 10.1111/joa.12060

Molecular analysis of regulative events in the developing chick limb

Chris Mahony 1, Neil Vargesson 1,✉
PMCID: PMC4487758  PMID: 23678942

Abstract

The developing chick limb has the remarkable ability to regulate for the loss of large amounts of mesenchyme and maintain a normal limb pattern in early (Hamburger and Hamilton Stage 19; E3) limbs. How the limb can regulate for tissue loss and why this ability is lost as development proceeds (after Hamburger and Hamilton Stage 21; E3.5) is unclear. We have investigated the origins of cells involved in regulative processes and, for the first time, the molecular changes occurring, and find striking differences between developmental time points just 0.5 days apart. We demonstrate that subtle changes in cell dispersal and cell proliferation occur in HH St21 limbs but not in HH St19 limbs and also demonstrate that there is no net replacement of removed tissue at either HH St21 or St19. We further show that changes in the Fgf8/Shh/Bmp4/Gremlin signaling pathway together with the appearance of distal Hox gene activation coincide with the limbs' ability to regulate for large tissue loss. We also demonstrate that following small tissue loss, limbs can regulate for missing tissue to produce normal pattern with no net replacement of missing tissue, as seen in limbs following large tissue loss. Our results indicate the regulative ability of the limb is not due to changes in cell proliferation, cell lineage nor replacement of the missing tissue – regulative ability is reliant upon the signaling environment remaining.

Keywords: cell proliferation, fatemaps, mesenchyme removal, phocomelia, proximo-distal axis, thalidomide, Bmp4, Fgf8, Gremlin, Hox genes, Shh

Introduction

The chick limb forms around HH St17 (day 2.5) of development (Hamburger & Hamilton, 1951). The limb grows out from the flank in a proximal to distal manner under the control of the apical ectodermal ridge (AER). Thus, the humerus/femur forms before the radius/tibia and ulna/fibula, followed finally by the digits (Saunders, 1948; Tabin & Wolpert, 2007; Towers & Tickle, 2009). The major molecules controlling these processes are the FGFs from the AER and Shh from the zone of polarising activity (ZPA) in the posterior mesenchyme (Tickle, 1999; Tabin & Wolpert, 2007; Towers & Tickle, 2009; Vargesson, 2009; Zeller, 2010). FGF and Shh expression is tightly controlled by Gremlin1 and Bmp4, via feedback loops, to coordinate limb outgrowth (Zeller et al. 2009; Probst et al. 2011). Collectively these signals induce the expression of multiple genes to pattern the limb. The main markers of the proximal, medial and distal limb regions are Meis1, Hoxa11 and Hoxa13, respectively (Galloway et al. 2009).

Limb development is a highly dynamic process. At early stages (HH St19; Hamburger and Hamilton Stage 19; E3) (Hamburger & Hamilton, 1951) of development, the limb bud has the remarkable capacity to pattern normally and regulate for loss of large and small blocks of mesenchyme (Zwilling, 1956; Amprino & Camosso, 1958; Barasa, 1964; Amprino, 1965; Stark & Searls, 1974; Pautou, 1977; Summerbell, 1977; Hornbruch, 1980; Hayamizu et al. 1994). The chick wing bud has also been shown to be able to regulate for the loss of the distal limb tip, when an exogenous source of FGF protein is present (Taylor et al. 1994; Kostakopoulou et al. 1996, 1997). Further evidence of the limbs' regulative ability comes from transplantation of additional mesenchyme into early limb buds. Even mesenchyme transplanted from the leg into the wing, has little impact on limb development pattern (Yallup & Hinchliffe, 1983; Krabbenhoft & Fallon, 1989; Wyngaarden & Hopyan, 2008). However, from HH St21 (E3.5), the ability of the limb bud to regulate for large tissue loss or addition of tissue is gradually lost, resulting in skeletal defects such as loss of the radius and ulna and phocomelia (Stark & Searls, 1974; Summerbell & Lewis, 1975; Summerbell, 1977; Hornbruch, 1980; Yallup & Hinchliffe, 1983; Cooper et al. 2011; Roselló-Díez & Torres, 2011; Roselló-Díez et al. 2011; Ozpolat et al. 2012). Limbs, however, retain the ability to regulate for the loss of small amounts of proximal tissue up to around HH St26 (E6) (Ozpolat et al. 2012).

The mechanisms underlying the developing chick limb's ability to produce a normal pattern, despite large tissue loss, at HH St19 is poorly understood. Interpretation of these results has suggested that cells may be intercalating, or changing fate in response to a change in signalling environment (Summerbell, 1977; Hornbruch, 1980; Yallup & Hinchliffe, 1983). However, these theories have not been fully investigated and the cellular and molecular mechanisms that underpin regulation remain unclear. Shedding light on the molecules involved in this process may provide insight into the mechanisms controlling proximo-distal limb outgrowth.

We want to uncover the cellular and molecular mechanisms maintaining or re-establishing the proximal to distal (P-D) axis in HH St19 limbs and to identify the temporal change(s) occurring at HH St21 where such re-establishment of pattern is lost. Here we have investigated how developing limbs regulate for loss of mesenchyme at different time points by studying effects upon cell dispersal, proliferation and, for the first time, molecular markers of limb patterning. Somewhat surprisingly, following large mesenchymal loss at HH St19 and small tissue loss at HH St21, we demonstrate that no changes in cell dispersal patterns, cell proliferation or gene expression domains were observed. The limbs compensate for the tissue loss and re-establish pattern. However, following loss of large mesenchyme at HH St21, significant differences in response are observed. We observe a change in cell dispersal patterns in the posterior proximal mesenchyme, changes in Fgf8/Shh/Bmp4/Gremlin signalling activity, together with the concomitant activation of distal Hox genes which correlate with the limbs' ability to regulate for large mesenchyme loss.

Materials and methods

Embryology

Fertilised White Leghorn chicken embryos were purchased from Henry Stewart, Herefordshire, UK, and stored for up to 1 week in a cooled incubator at 12 °C. For experimentation, embryos were incubated at 37 °C. The eggs were staged according to Hamburger & Hamilton (1951). We have labelled the limb digits according to new experimental findings indicating avian digits are 1, 2 and 3 and not, as convention previously thought, 2, 3 and 4 (Tamura et al. 2011; Towers et al. 2011; Wang et al. 2011).

Mesenchyme removal

Mesenchyme was removed from intact wing buds in ovo by gently scoring the outline of a square of the desired size onto the dorsal surface of the limb bud with a sharp tungsten needle. Care was taken to leave the apical ridge intact. The outline of the square was then cut deeply so that the tungsten needle pierced the ventral surface. Small blocks of mesenchyme up to 400 μm in length and 400 μm in width in central mesenchyme were removed from HH St21 limbs. Large blocks up to 600 μm in length and 1000 μm in width in central and distal mesenchyme were removed in HH St21 embryos. Large blocks up to 1000 μm in length and 400 μm in width in central and distal mesenchyme were removed in HH St19 embryos. The block of mesenchyme was removed, leaving a limb with an area of central mesenchyme missing. Embryos were imaged 2 h following operation and classified into AER damaged, AER intact and correct manipulation, and incorrect manipulation (incorrect sizes were removed or healing did not occur spontaneously). Only AER intact and correct manipulations were used. Embryos were either fixed immediately or incubated for 24 h, 48 h or 7 days before fixation. Post-operation limb and digit lengths were measured using Adobe photoshop. Statistical comparisons were made using the Mann-Whitney U-test.

Pressure injection of DiA

DiA (Molecular Probes) was administered as described previously (Vargesson et al. 1997). Cells in the right limb bud were labelled, and measurements were taken of the dot size and position at the time of administration and after the incubation period. The average initial size of the DiA injected dot was 25–50 μm (n = 35). After 48 h the limbs were fixed in 4% paraformaldehyde (PFA) for 24 h and photographed, under a coverslip with a solution of 70% glycerol/29% phosphate-buffered saline (PBS)/1% Dabco, using a Nikon SMZ1500 stereo dissecting microscope with fluorescence attachment. Whole mounts were photographed using a Nikon DS-5 digital camera. DiA length measurements were carried out using Adobe photoshop. Statistical comparisons were made using the Mann-Whitney U-test.

Processing and staining for cartilage

For cartilage staining, embryos were incubated until day 7, fixed in 5% trichloroacetic acid overnight, rinsed in 70% alcohol for 5 min, then washed twice in acid alcohol (1% concentrated hydrochloric acid in 70% alcohol) for 10 min. Embryos were stained with Alcian green or Alcian blue (0.1% in acid alcohol) for 6 h and rinsed in acid alcohol overnight, dehydrated in absolute alcohol before being cleared in Methyl salicylate (Sigma) and photographed.

Whole mount in situ hybridisation

The whole mount in situ hybridisation method has been described previously (Vargesson et al. 1997; Vargesson & Laufer, 2009). Probes for Fgf8, Gremlin, Shh, Hoxa11 and Hoxa13 were gifts from Dr Ed Laufer and probes for Meis1 gifts from Dr Miguel Torres. Embryos were visualised and photographed using a Nikon SMZ1500 microscope with a Nikon DS-5 digital camera.

Cell proliferation analysis

After mesenchyme removal and incubation, embryos were dissected in PBS and fixed in 4% PFA overnight at 4 °C. Embryos were then placed in 30% sucrose solution in PBS overnight and subsequently embedded in CryoMBed (Bright) for cryosectioning. Sections 10 μm thick were cut on a cryostat, allowed to air dry, and stored at 20 °C before staining to reveal mitotic cells. The tissue sections were blocked in 10% goat serum in 0.2% Triton X-100 PBS for 90 min at room temperature before incubation in primary antibody {anti-phospho-histone H3 [Ser-10, rabbit polyclonal IgG (Millipore)] diluted 1 : 100 in block solution} overnight at 4 °C. The following day, the slides were washed several times in PBS and incubated in secondary antibody (goat-anti-rabbit-Cy3 diluted 1 : 500 in 1% goat serum/PBS) for 2 h at room temperature and in the dark. At the end of the incubation, the slides were washed in PBS, mounted in Vectashield mounting medium for fluorescence (Vector Laboratories) and pictures were taken using Nikon SMZ1500 stereo dissecting microscope with fluorescence attachment. The number of fluorescent cells was counted per limb bud in each section.

Results

Changes in limb pattern following removal of large amounts of mesenchyme from the St21 limb bud but not the St19 limb bud

To investigate the basis of the regulatory capacity of the developing limb, we removed large blocks of mesenchyme from intact developing limbs in ovo at HH St19 (E3) and HH St21 (E3.5). Cartilaginous progenitors as distinguished by expression of Sox9 are not present at either of these stages, indicating the tissue has not started to differentiate at these stages (Supporting Information Fig. S1; Healy et al. 1999).

We compared two conditions; large mesenchyme block removal (including overlying ectoderm) at HH St21 (600 μm wide × 1000 μm long) and at HH St19 (400 μm wide × 1000 μm long); these dimensions represented the largest area of limb tissue that was mechanically possibly to remove, leaving the AER intact. All manipulations left the AER intact and the mesenchyme healed within 2 h (Fig.1B', C'). After large tissue removal at HH St21 and HH St19 the remaining proximal and distal tissue was juxtaposed. Following large tissue removal, the distance between the wound margin and the AER was measured, leaving 92 ± 3.4 μm (n = 15) and 86 ± 6.8 μm (n = 6) from the distal wound edge to the AER in HH St21 and 19 limbs, respectively.

Fig 1.

Fig 1

Open in a new tab

Removal of large amounts of limb mesenchyme results in shorter but normally patterned limbs or phocomelia depending on the extent of tissue loss and the timepoint of manipulation. (A,B) Schematic diagrams of HH St21 limbs, showing (A) control limb bud, (B) limb bud depicting the position and size of large mesenchymal removal (600 × 1000 μm) and (C) schematic diagram of an HH St19 limb showing the position and size of the large tissue removal (400  × 1000 μm). (B'-C) Typical phenotype following 2 h incubation. (A',B″,C″) Alcian Blue cartilage stained whole-mounts of limbs at 9 days of development stained in an unoperated control wing (A') or following removal of (B″) large blocks at HH St21, indicating loss of proximal elements (C″) and large blocks at HH St19, indicating normal pattern, but reduced in length. (D) Graph outlining the relative decrease in total limb length and of individual skeletal elements at 9 days of development following removal of large blocks of mesenchyme. Data are presented as the percentage decrease in length relative to unoperated, contralateral control limbs at day 9. H,  humerus; r,  radius; u,  ulna; 1,  digit 1; 2,  digit 2; 3,  digit 3. * P < 0.05, ** P < 0.01, *** P < 0.005, **** P < 0.001 (Mann-Whitney U-test). Scale bars: 500 μm.

HH St21

In HH St21 limbs the large blocks of mesenchyme removed (Fig.1B) were fated to contribute to the distal part of the humerus, radius, ulna and the proximal wrist (Vargesson et al. 1997). Removal of large blocks of tissue resulted in severe patterning defects (n = 31/31). In many cases the limbs consisted of a hand plate only, which in itself was reduced in length (n = 14/31; Fig.1B″). In a further 6/31 cases just digits were seen articulating with either a coracoid or severely shortened humerus (Fig.1B″). We never observed ectopic digits in such limbs but the resulting digits did exhibit patterning errors ranging from reduction of digit number (Fig.1B″). We also observed in 11/31 cases a limb with shortened proximal structures and normal digits. Thus, following large mesenchyme removal, severe pattern defects and phocomelia were observed; little if any regulation was seen and all skeletal elements were decreased significantly in length (Fig.1D). The most severely affected elements were the humerus (57.1% decrease) and the radius/ulna (60.98% reduced), whilst the handplate was the least affected (36.4% decrease) (Fig.1D). A hand plate was always formed (n = 31/31).

HH St19

At HH St19 the mesenchyme removed contains tissue normally fated to become humerus, radius/ulna and wrist (Vargesson et al. 1997; Sato et al. 2007). In contrast to HH St21 embryos, the HH St19 limbs have an outstanding ability to regulate for tissue loss, producing a normally patterned limb, with reduced length (Fig.1C–C″). In 6/7 cases the limbs contained all proximal to distal elements with 1/7 case giving no humerus. On average there was a reduction in total limb length by 35% (Fig.1D). There was a reduction in humerus length by 34%, radius/ulna length by 26% and hand-plate length by 26%.

To investigate the regulative processes underlying these phenomena as well as shed light on why the limb loses the regulative ability we embarked on cellular and molecular analysis looking at cell dispersal, cell proliferation and gene expression.

Cell dispersal patterns are unchanged following removal of large blocks of limb mesenchyme except in the posterior proximal mesenchyme

To investigate whether changes in cell dispersal and behaviour contribute to regulative abilities of the limb, DiA was injected into three sites (injection sites x, y, z) bordering the wound in control (Fig.2A,B) and operated limbs (Fig.2C,D). To quantify the degree of cell dispersal we measured the extent the labelled cells had spread from their position along the proximo-distal axis of the limb. To summarise findings, typical control and operated limbs were overlaid in Adobe photoshop and the mean cell dispersal in the PD axis of cells was measured and drawn onto limbs (Fig.3A-C); the values used are shown in Fig.3C.

Fig 2.

Fig 2

Open in a new tab

Patterns of cell dispersal in regulating limbs. Cell labelling analysis with the fluorescent lipophilic dye DiA was carried out to investigate changes in cell dispersal in regulating limbs. Control, unoperated, limbs at (A) HH St21 and (B) HH St19. DiA was used to label small groups of cells at specific locations (injection point x, y, z) around the prospective wound margin. Images showing extent of DiA labelling in HH St21 limbs following removal of (C) large block of mesenchyme at HH St21 and (D) large block of mesenchyme from HH St19 limbs. Dispersal patterns of cells in the operated limbs correlate well with cell dispersal in unoperated limbs, except in HH St21 limbs (C) at injection point y, which exhibits an increased dispersal pattern toward the distal limb. Scale bars: 500 μm.

Fig 3.

Fig 3

Open in a new tab

Quantification and summary of the extent of cell dispersal in control and operated limbs. (A,B) Figures summarising cell dispersal patterns in control and tissue-removed limbs at HH St21 (A) and HH St19 (B). (C) Cell dispersal patterns were measured using Adobe photoshop and a representative cell dispersal pattern was drawn onto an idealised limb outline for each injection site for each experiment. Mean cell dispersal patterns as shown in (C) indicate no significant change in cell dispersal or position in HH St19 large block removals. However, in HH St21 limbs (A) there is a change in cell dispersal position (A, injection point y, arrowhead). However, all other dispersal patterns remain equivalent to the size of the limb, indicating that the tissue removed is not replaced and cell dispersal patterns proximal to and distal to the wound site do not change global dispersal rates to replace the tissue. Scale bar: 500 μm. Red colour, injection point x; black colour, injection point y; blue colour, injection point z.

HH St21

We investigated the extent of cell dispersal following large tissue removal at HH St21. Limbs were analysed 48 h after tissue removal (Fig.2C). No significant increase in cell dispersal was found at either position x (471.4 ± 34.2 μm, n = 9 for control limbs vs. 611 ± 34.9 μm, n = 8 for operated limbs) or position z (916.9 ± 54.9 μm, n = 9 for control limbs vs. 737 ± 49.2 μm, n = 10). However, in posterior proximal cells (position y) there was a significant increase in dispersal (Figs2C and 3A; 820 ± 67.8 μm, n = 8 for control limbs vs. 966.4 ± 78.7 μm, n = 8 for operated limbs; P < 0.05).

HH St19

We next investigated the extent of cell dispersal following large tissue removal at HH St19. Limbs were analysed 48 h after tissue removal (Figs2D and 3B). We found no significant increase in the extent of cell dispersal at position x (508.6 ± 3.2 μm, n = 7 for control limbs vs. 462.5 ± 47.3 μm, n = 8 for operated limbs; Figs2D and 3B), position y (800.1 ± 85.2 μm; control n = 6, operated n = 6; Figs2D and 3B) or position z (1080.2 ± 96.9 μm, n = 5 for control limbs vs. 1033 ± 57.2 μm, n = 6; Figs2D and 3B).

These results suggest that following loss of large amounts of limb tissue at both HH St19 and St21 the remaining cells are unable to replace this tissue or compensate for its loss through marked changes in cell dispersal, despite proximal and distal tissue signals being juxtaposed. In HH St19 limbs, pattern is restored, whereas in HH St21 it is not. How can limb buds recover pattern after loss of extensive mesenchyme? And why do limbs lose this ability as development proceeds?

Global cell proliferation is unaltered in regulating limbs

The operated limbs do not appear to be grossly changing cell dispersal patterns in order to regulate for tissue loss. To determine whether rescue of the limb pattern is due to changes in cell proliferation, we cryosectioned limbs and stained for a marker of mitosis (H3-phosphohistone) 24 h after the operation as limbs were in the process of recovering from the tissue loss. No significant changes in cell proliferation were apparent compared with control tissue from a similar region of the limb when experiments were carried out at HH St19. However, we observed a small, but significant increase in cell proliferation in the posterior of the wound site in HH St21 limbs (Fig.4A-C; n = 3 per condition). These data indicate that there appears to be increased proliferation of cells in the posterior proximal region of the HH St21 regulating limb, which fits with the altered dispersal patterns in this region in response to loss of the tissue (Fig.4A-C).

Fig 4.

Fig 4

Open in a new tab

Cell proliferation is largely unchanged at 24 h in regulating limbs. (A,B) Time 0 manipulations at HH St21 and St19, indicating extent of tissue removed. (A′,A″,B′,B″) Cryosections of HH St21 and HH St19 control and operated limbs at 24 h after large tissue removal. Sections were antibody stained for anti-H3-phosphohistone. (C) Graphs of the mean (± SEM) number of mitotic cells per μm² in control and operated limbs, NS, P > 0.05, *P < 0.05 (Mann-Whitney U-test). No significance difference in mitotic cell number was observed between the control limb section or the operated limb section, except in the posterior half of HH St21 operated limbs (asterisk). Scale bar: 300 μm.

Somewhat surprisingly we did not detect an increase in cell proliferation or dispersal in limbs following tissue removal in the tissue now abutting the AER.

Regulative ability for large tissue loss is linked to appearance of distal Hox genes and changes in A-P signalling

Effects on molecular markers in the regulating limb following tissue loss have not been reported. Here we selected a range of molecular markers for the proximo-distal axis and anterior-posterior axis and observed effects of tissue removal upon gene expression.

Proximo-distal axis gene expression analysis

We next studied whether changes in PD gene expression occurred following tissue removal, which could shed light on the limbs' ability to regulate for tissue loss. We studied the expression patterns of proximal (Meis1), medial (Hoxa11) and distal (Hoxa13) limb markers (Galloway et al. 2009) at time 0 and 48 h after tissue manipulation. These time points were selected as we wanted to identify the signalling environment present at the time of the experiment and then analyse resulting changes in expression as limb patterning was occurring. Embryos were operated in ovo, then immediately fixed for gene expression analysis or incubated for a further 48 h before fixation (Fig.5). Results were summarised in gene expression maps. These were constructed by calculating the average expression area of each gene and drawing a representative expression pattern onto typical limb examples (Fig.5D-D′, H-H″).

Fig 5.

Fig 5

Open in a new tab

Expression patterns of proximal-distal genes in control and in limbs 48 h after tissue removal. Whole-mount RNA in situ hybridisation patterns for markers of the proximal limb region Meis1, medial limb region Hoxa11 and distal limb region Hoxa13. (A-D) Time 0 gene expression analysis for HH St21 limbs at the time of tissue removal. (A′-D′) Time 0 gene expression analysis for HH St19 limbs indicating expression of markers at time of tissue removal. (E-H) Expression in unoperated control limbs. (E′-H″) Gene expression patterns following large block removal in HH St21 (E′-H′) and St19 limbs (E″-H″). Note the absence of Hoxa13 expression in HH St19 limbs (C′; asterisk) and the near complete loss of Hoxa11 expression in limbs 48 h after large tissue removal at HH St21 (F'; arrows). The effects of limb tissue removal on PD gene expression are summarised by measuring the average expression area of each gene at time 0 and 48 h later and overlaying onto typical limbs for each condition, indicating the loss of the medial Hoxa11 gene expression, but normal proximal and distal gene expression in HHSt21 limbs following large tissue removal (D–D′,H–H″). Scale bar: 500 μm.

Time 0 analysis

In normal, unoperated HH St21 limbs, Meis1, Hoxa11 and Hoxa13 (n = 3/3 for each gene) were expressed proximally, medially and distally, respectively, in every case (Fig.5A-D). Removal of large blocks of mesenchyme from HH St21 limbs removed tissue that expresses a small amount of the Meis1 and Hoxa13 domains and most of the Hoxa11 region (Fig.5A-D).

In contrast in HH St19 limbs, Hoxa13 expression is not present and we never observed Hoxa13 expression at HH St19 (n = 5/5) (Fig.5A′-D′). Following tissue removal in HH St19 limbs the tissue block removed contains a small amount of Meis1 (Fig.5A′) and much of the Hoxa11 region (Fig.5B′) (n = 4/4 for each gene).

48 h analysis

Following removal of large blocks of mesenchyme at HH St21 and 48 h incubation we find that Meis1 (n = 7/7) and Hoxa13 (n = 5/5) expression was not altered significantly (Fig.5E′,G′) compared with stage-matched unoperated controls (Fig.5E-G). In contrast, expression of Hoxa11 was decreased significantly (n = 3/6) and often ablated (n = 3/6) in the operated limb (Fig.5F). Following large mesenchyme loss at HH St21, almost all the Hoxa11 expression domain is missing at time 0 (Fig.5B) but Meis1 (Fig.5A) and distal Hoxa13 (Fig.5C) expression are expressed normally. After 48 h following the operation Meis1 and Hoxa13 (Fig.5E′-G′) are expressed in normal domains; however, Hoxa11 is absent or is expressed in a very small domain (Fig.5F′).

In HH St19 limbs we observed that Hoxa13 expression was never seen in unoperated or operated limbs at the time of experiment (Fig.5C′). The operation removes part of the Meis1 expression domain and the majority of the Hoxa11 expression domain (Fig.5A,B). After 48 h incubation, we observed that resulting limbs all expressed Meis1, Hoxa11 and Hoxa13 in their normal areas of expression (n = 3/3 for each gene; Fig.5E″-G″). These results suggest that in HH St19 limbs following tissue removal the resulting wound heals and limb tissue can continue to normally specify the remaining tissue and activate more distal genes to establish a normal pattern.

Fgf-Shh-Bmp4-Grm expression analysis

We also investigated expression patterns of Fgf8, Bmp4, Gremlin, and Shh signaling, due to their involvement in the initiation, propagation and termination of limb development (Zeller et al. 2009; Probst et al. 2011). We looked at the expression of these markers 24  and 48 h following large tissue block removal at HH St21 and St19 (Fig.6).

Fig 6.

Fig 6

Open in a new tab

Limb anterior-posterior gene expression is transiently modulated following tissue removal, dependent on timing of tissue loss. Whole-mount RNA in situ hybridisation patterns for markers of the antero-posterior axis, Fgf8, Shh, Gremlin and Bmp4, following large tissue removal and 24 h incubation. (A-H) HH St19 limbs. (I-P) HH St21. (Q-T) HH St21 limbs following large tissue removal and 48 h incubation. Inset boxes represent control limbs at 48 h Fgf8 (Q′) and Shh (R′). Expression patterns of the genes appear normal and unchanged in limbs operated at HH St19 despite limb reductions (A-H). However, in limbs operated at HH St21, upregulation of Fgf8 is seen after 24 h (black arrow in M; limb photographed at oblique angle to observe AER) which is maintained at 48 h (Q, when compared with stage-matched control Q'). Expression of Shh is also upregulated at 24 h (white arrows, J,N) and 48 h (R,R') after operation. Gremlin expression is changed 24 h post-operation from a proximal domain and a weak distal domain (K) to a single domain in the central mesenchyme (white arrow, O). However, Gremlin expression is gone at 48 h (S) just as in stage-matched, unoperated limbs. Bmp4 expression is downregulated 24 h post-operation (white arrows, L,P) and is gone at 48 h, as seen in stage-matched, unoperated limbs. Despite the transient upregulation of Fgf8 and Shh, ectopic digits or a rescue of limb pattern was never seen. Scale bar: 500 μm.

Following large block removal from HH St21 limbs and 24 h incubation there was marked upregulation of Fgf8 (Fig.6I, M, n = 4/4), Shh (Fig.6J, N, n = 6/6) and Gremlin (Fig.6K,O, n = 3/3) and a loss of Bmp4 in the posterior mesenchyme (Fig.6L,P, n = 3/3). This effect was transient, never giving rise to ectopic digits. Following 48 h incubation Gremlin and Bmp4 (n = 3 for both gene; Fig.6S,T) were both expressed in normal domains. Shh was only weakly expressed at 48 h in control limbs (Fig.6R′, n = 3/3) but, in contrast, Shh was still upregulated in the operated limb until 48 h (n = 5/5; Fig.6R). Fgf8 upregulation was maintained at 48 h (n = 3; Fig.6Q) in the operated limbs but was only weakly expressed in control limbs (Fig.6Q′; n = 3/3) and was expressed normally by 72 h (n = 3/3). Despite the transiently raised levels of Shh and Fgf8 in operated limbs we never observed ectopic digits or a rescue of the limb pattern.

In contrast, in HH St19 limbs following large mesenchyme block removal, we found no change in Fgf8, Gremlin, Bmp4 or Shh gene expression at 24 h (Fig.6A-H, n = 3 for each gene) and 48 h post-operation (n = 3 for each gene).

Small block removal

It is known that limbs can regulate well following loss of small tissue blocks up to HH St21 and have some ability to do so until HH St26 (Stark & Searls, 1974; Hornbruch, 1980; Yallup & Hinchliffe, 1983; Ozpolat et al. 2012). We wondered whether the regulative mechanisms controlling loss of small tissue blocks were similar to or different from events underlying large tissue removal regulation. We removed small blocks of mesenchyme from proximal regions of the limb at HH St21; this tissue was fated to become mainly humerus and radius/ulna (Vargesson et al. 1997). Limbs were analysed for cartilage patterns (Fig.7A), cell proliferation (Fig.7B), cell dispersal (Fig.7C,D) and gene expression (Fig.7E-H). At day 9 of development, limbs were patterned normally (Fig.1A-A″′) but were 28.8% smaller in length than the unoperated contralateral limbs (Fig.7A″′). All elements were reduced in length as follows: humerus, 35.1% decrease in length; ulna, 27.0% decrease; handplate, 21.1% decrease (n = 7/7; Fig.7A″), never giving missing elements or digits, but with shorter skeletal elements proximally and distally. Strikingly, we found that such limbs do not replace the missing tissue, have no change in cell dispersal patterns or changes in gene expression in the P-D axis (Meis1, Hoxa11, Hoxa13; n = 3/3 in each case) or A-P axis (Fgf8 and Shh; n = 3/3 for each case) and were expressed normally 48 h after operation or, as in the case of Shh, which is no longer normally expressed in control limbs, expression is no longer seen in operated limbs (Fig.7E-J′).

Fig 7.

Fig 7

Open in a new tab

Following removal of small blocks of mesenchyme, limbs pattern normally, but without observable changes in cell dispersal, cell proliferation or gene expression. (A-A″′) Schematic diagrams of HH St21 limbs showing (A) removal of small block of mesenchyme, (A′) the healed tissue 2 h later, and (A″) the final cartilage pattern at E 9.0. Scale bars: 500 μm. (A″′) Graph outlining the relative decrease in total limb length and of individual skeletal elements at 9 days of development following removal of small blocks of mesenchyme. Data are presented as the percentage decrease in length relative to unoperated, contralateral control limbs at day 9. H,  humerus; r,  radius; u,  ulna; 1,  digit 1; 2,  digit 2; 3,  digit 3. *P < 0.05, **P < 0.01, ***P < 0.005, ****P < 0.001 (Mann-Whitney U-test). (B-B′) Cryosections of HH St21 limbs at 24 h after small tissue removal. Sections were antibody stained for anti-H3-phosphohistone. Scale bars: 300 μm. (B″) Graph of the mean (± SEM) number of mitotic cells per μm² in control and operated limbs, NS, P < 0.05 (Mann-Whitney U-test). (C) Cell labelling analysis with DiA in limbs at HH St21 following small block removal at specific locations (injection point x, y, z) around the prospective wound margin. Scale bars: 500 μm. (D) The mean cell dispersal pattern was drawn onto an idealised limb outline for each injection site for each experiment. (E-J′) Whole-mount RNA in situ hybridisation patterns for Meis1, Hoxa11, Hoxa13, Shh and Fgf8 in control and limbs following small block removal at HH St21. Note expression domains are unchanged or exactly the same as unoperated controls. Shh expression is no longer expressed in control limbs at 48 h post-operation and is not seen in operated limbs either. Scale bars: 500 μm.

Discussion

The ability of the developing chick limb to regulate for small and large tissue loss and produce a normal pattern (albeit shorter in length) has fascinated developmental biologists for many years. Why the limb loses the ability to regulate for large tissue loss is equally interesting and understanding this process could shed light on normal limb development and on the regenerative abilities of the limb. Here we have revisited this issue using a cellular and molecular biology approach in intact and in vivo limb buds.

HH St19 limbs can pattern normally following large tissue removal, yet HH St21 limbs fail to do so, resulting in severely truncated limbs as well as phocomelia. Cartilage precursor cells, as marked by Sox9 expression (Healy et al. 1999; Fig. S1), are not present at either of the stages we used in this study. This indicates the changes in limb pattern are not due to removing differentiating cartilage progenitors from the forming limb.

Regulative events following large tissue loss do not involve replacement of missing tissue or global changes in cell dispersal

With the exception of the posterior proximal mesenchyme in HH St21 limbs, following large tissue removal, there is no net change in the patterns of cell dispersal or proliferation following tissue loss (Figs4). In operated HH St21 limbs given the transient upregulation of Shh and Fgf8, and the resulting close proximity of the ZPA to proximal tissue, cells are likely to disperse more. However, given that the other dispersal and proliferation patterns are unchanged, we conclude that there is no net replacement of tissue in either HH St19 or HH St21 limbs. This is further corroborated by the loss of Hoxa11 expression and its failure to be reestablished following tissue removal in HH St21 limbs. Our in vivo data demonstrate that limbs cannot compensate for loss of large amounts of tissue through changes in cell dispersal patterns.

Changes in Fgf8, Shh, Gremlin and Bmp4 expression and activation of distal Hox gene expression is concomitant with the loss of regulative ability between HH St19 and St21

We undertook a study looking for molecular changes that may underlie the limbs' ability to regulate for tissue loss.

We observed a lack of Hoxa13 expression in HH St19 limbs (which regulate for large tissue loss) but the presence of Hoxa13 in HH St21 limbs (which fail to regulate for large tissue loss) (Fig.8A-C). We never saw an expansion or change of Hoxa13 expression in HH St19 or HH St21 limbs following tissue removal. This leads to the interesting correlation that the appearance of the distal Hox genes may be involved in the limbs' regulative ability. We hypothesise that in HH St19 limbs following tissue loss, signalling between the distal AER and mesenchyme recovers. Limb outgrowth recovers and some reestablishment of tissue that will express Hoxa11 can occur until Hoxa13 expression is established. Hoxa13 expression signals distal fate and a shorter limb with normal pattern results (Fig.8A-C). In contrast, in HH St21 limbs Hoxa13 is expressed already in the distal mesenchyme at the time of the experiment. According to the theory of posterior prevalence, once more 5′ Hox genes are expressed these then dominate the proximal Hox genes, thus cells can not now re-express the proximal Hox genes (Duboule & Morata, 1994). It is known that loss of Hoxa13 expression results in ectopic Hoxa11 expression, indicating that Hoxa13 represses Hoxa11 expression (Post & Innis, 1999). It is also known that Hoxa13 expression and function does not require Hoxa11 expression (Davis et al. 1995), thus in HH St21 limbs due to the expression of Hoxa13, the Hoxa11-expressing domain cannot be reformed, resulting in limb defects including phocomelia (Fig.8B).

Fig 8.

Fig 8

Open in a new tab

Model of Hox gene involvement in limb regulation. Summary figures of proximo-distal gene expression domains at time 0, 2 h post-operation, 48 h post-operation and showing resulting cartilage pattern 9 days after operation. (A) HH St21 control limbs; (B) HH St 21 large block removal limbs; (C) HH St19 large block removal limbs. Gene expression patterns were measured using Adobe photoshop and a representative gene expression pattern was drawn onto an idealised limb outline for each experiment at time 0 and 48 h. An idealised figure of the healed limb 2 h after operation is also included. The typical cartilage patterns resulting from each experiment are also drawn. h, humerus; r/u, radius/ulna; 1,2,3 represent digits 1, 2, 3. (A) In an HH St21 limb, domains of Meis1, Hoxa11 and Hoxa13 are clearly defined. (B) In contrast, following large tissue removal at HH St21 (time 0) the majority of the Hoxa11 domain is missing and a large amount of the Meis1 is missing. Hoxa13 domain is relatively normal. By 48 h, Hoxa11 expression is hardly seen and the limb is short. By day 9 the resulting limb is phocomelic-like. (C) At HH St19 there is no Hoxa13 expression and despite a similar amount of tissue being removed, a normal limb pattern is produced. We hypothesise that tissue removal at HH St19 before the distal Hoxa13 expression is expressed allows the limb tissue to recover and as Fgf and Shh signaling are normal in such limbs, the limb continues to grow out and Hoxa11 expression occurs in the remaining cell populations, expanding the domain. As soon as Hoxa13 expression is activated, the distal domain of the limb forms – this provides some explanation for the normal, albeit reduced in length, limb pattern. The theory of posterior prevalence would seem to be at work in this model where in contrast to the situation at HH St19 in HH St21 limbs Hoxa13 expression is already present; tissue that is lost can not be repatterned with Hoxa11 expression as more proximal Hox genes cannot be reactivated. Scale bar: 500 μm.

We looked at the expression patterns of other genes involved in A-P and P-D axis patterning. We observed strong and prolonged upregulation of Fgf8, Shh and Gremlin but a loss of Bmp4 and never observed ectopic digits, nor did we see a rescue of limb pattern. We know that Gremlin operates in a feedback loop that regulates Fgf8 and Shh expression (Zeller et al. 2009; Zeller, 2010; Probst et al. 2011). Removal of medial tissue at HH St21 could result in a transient deregulation of this feedback loop, leading to an upregulation of Shh and Fgf8 and an increase in localised cell proliferation (and dispersal) in the posterior margin. After the Gremlin expression and Bmp4 expression patterns are normalised, Shh and Fgf8 expression return to normal, potentially explaining why expression of these signals is prolonged. We cannot rule out that the transient gene misexpression is due to a wound-healing response. The increased Fgf8 and Shh expression could also indicate that the remaining tissue is distalised; this has been proposed in recombinant and limb grafting experiments looking at regulative events (Cooper et al. 2011; Roselló-Díez & Torres, 2011; Roselló-Díez et al. 2011). This assay makes for an interesting assay to study the Fgf-Shh-Bmp4-Grm feedback loop further in the chick limb bud.

Limb dysmelia as well as phocomelia occurs after St21, not before

The regulative capacity of the HH St19 chick limb does not rely upon changes in cell lineage, proliferation or gene expression domain changes in the proximo-distal or antero-posterior axis. In fact, following large tissue loss at HH St19 the limb does not appear to attempt to replace the missing tissue, or change molecularly, but pattern is maintained, with the limb being shorter than normal. Indeed, all the limb elements are shorter in length, but always present (Fig.7). We conclude that following tissue loss at HH St19, the remaining tissue heals and re-establishes pattern and, as development proceeds, distal genes are normally activated, maintaining a normal limb segmental pattern. In contrast, large tissue loss at HH St21 always resulted in limb dysmelia, most commonly phocomelia-like malformations.

Strikingly, small tissue loss at HH St21 appears to behave exactly as large tissue loss at HH St19, where normal pattern is obtained but through no observable changes in cell proliferation, cell dispersal or gene expression (Fig.7A). We propose this ability is due to the majority of the signalling environment remaining, allowing cells to maintain identity and pattern accordingly. Why distal structures are smaller than normal following small tissue loss in proximal regions is unclear, particularly as we observe no changes in proximal or distal patterning gene expression patterns (Fig.7E-J).

Conclusions

We have identified molecular differences in intact and in vivo limb buds, following mesenchyme loss, between HH St19 and HH St21 limbs. Our data suggest it is the early signalling environment that is the key to allowing regulation, supporting recent findings using recombinant limbs and grafted limb tips (Cooper et al. 2011; Roselló-Díez & Torres, 2011; Roselló-Díez et al. 2011).

Many models attempt to explain proximo-distal outgrowth of the limb bud and our data best fit the two-signal model (Cooper et al. 2011; Roselló-Díez et al. 2011). In the HH St21 experiments we removed tissue as the limb was defining the medial and distal segments. In the HH St19 experiments the distal signals had not yet defined the distal Hox genes, or the distal segment of the limb. Thus, following tissue removal activation of these distal genes can still occur normally, allowing a normal limb pattern (Fig.8C). At HH St21, as such distal signals have been activated (Fig.8A,B) they can not be replaced or downregulated to allow re-establishment of the medial segment.

Finally, this work confirms that loss of large amounts of central limb mesenchyme at HH St21 results in phocomelia, which has also been demonstrated in an x-irradiation-induced model of phocomelia (Wolpert et al. 1979; Galloway et al. 2009). We show this is because of a permanent loss of this tissue and coincides with a failure to re-establish medial limb identity (through loss of Hoxa11 expression) (Fig.8B). We have never seen dysmelic or phocomelic limbs when similar experiments were carried out at HH St19, at which stage we hypothesise that these limbs can re-establish the molecular signalling environment to maintain pattern before onset of activation of distal Hox genes (Fig.8C). Our data further suggest that the onset of phocomelia itself is a relatively late event in limb development. It will be interesting to establish whether other causes of phocomelia, for example thalidomide teratogenesis (for review see Vargesson, 2009, 2011), occur through a similar manner.

Acknowledgments

Parts of this work were started by N.V. in Profs Cheryll Tickle and Lewis Wolpert's labs. N.V. is immensely grateful to Cheryll Tickle and Lewis Wolpert for their continued support, advice and enthusiasm. The authors thank Prof. Lynda Erskine for help with statistics and analyses. The authors also thank Dr Martin Collinson, Prof. Lynda Erskine, Scott McMenemy, Prof. Cheryll Tickle and Prof. Lewis Wolpert for comments on the manuscript. C.M. is funded by a PhD studentship from the University of Aberdeen and we also thank Prof. Ruth Ross for her support. The authors declare no conflicts of interest.

Supporting Information

Additional Supporting Information may be found in the online version of this article:

Fig. S1. Expression chondrogenic precursors at the time of operation. Whole-mount RNA in situ hybridisation was completed for Sox9 to identify whether chondrogenic precursors were present in the limb at the time of the experiment. (A) A control, unoperated limb at HH St21 shows no Sox9 expression in the embryo (n = 3/3). (B) A control, unoperated limb at HH St27 (48 h later) showing normal Sox9 expression (n = 3/3). Images in (A) and (B) were carried out at the same time and confirm Sox9 expression is not seen in HH St21 limbs. Scale bar: 500 μm.

joa0223-0001-sd1.tif (737.7KB, tif)

References

  1. Amprino R. Aspects of Limb morphogenesis in the chicken. In: DeHann RL, Ursprung H, editors. Organogenesis. New York, USA: Holt, Rinehart and Winston; 1965. pp. 255–281. [Google Scholar]
  2. Amprino R, Camosso M. Experimental observations on influences exerted by the proximal over the distal territories of the extremities. Experientia. 1958;14:241–243. doi: 10.1007/BF02159179. [DOI] [PubMed] [Google Scholar]
  3. Barasa A. On the regulative capacity of the chick embryo limb bud. Experientia. 1964;20:444. doi: 10.1007/BF02152140. [DOI] [PubMed] [Google Scholar]
  4. Cooper KL, Hu Berge JK, Fernandez-Teran M, et al. Initiation of proximal-distal patterning in the vertebrate limb by signals and growth. Science. 2011;332:1083–1086. doi: 10.1126/science.1199499. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Davis A, Witte D, Hsieh-Li H, et al. Absence of radius and ulna in mice lacking hoxa11 and hoxd11. Nature. 1995;375:791–795. doi: 10.1038/375791a0. [DOI] [PubMed] [Google Scholar]
  6. Duboule D, Morata G. Colinearity and functional hierarchy among genes of the homeotic complexes. Trends Genet. 1994;10:358–364. doi: 10.1016/0168-9525(94)90132-5. [DOI] [PubMed] [Google Scholar]
  7. Galloway JL, Delgado I, Ros MA, et al. A revaluation of x-irradiation induced proximo-distal limb patterning. Nature. 2009;460:400–404. doi: 10.1038/nature08117. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Hamburger V, Hamilton HL. A series of normal stages in the development of the chick embryo. J Morphol. 1951;88:49–92. [PubMed] [Google Scholar]
  9. Hayamizu T, Wanek N, Taylor G, et al. Regeneration of the HoxD expression domains during pattern regulation in chick wing buds. Dev Biol. 1994;161:504–512. doi: 10.1006/dbio.1994.1048. [DOI] [PubMed] [Google Scholar]
  10. Healy C, Uwanogho D, Sharpe PT. Regulation and role of Sox9 in cartilage formation. Dev Dynamics. 1999;215:69–78. doi: 10.1002/(SICI)1097-0177(199905)215:1<69::AID-DVDY8>3.0.CO;2-N. [DOI] [PubMed] [Google Scholar]
  11. Hornbruch A. Abnormalities along the proximo-distal axis of the chick wing bud: the effect of surgical intervention. In: Merker HJ, Nou H, Neubert D, editors. Teratology of the limbs. Berlin: Walter de Gruyter and Co; 1980. pp. 191–197. [Google Scholar]
  12. Kostakopoulou K, Vogel A, Brickell P, et al. ‘Regeneration’ of wing bud stumps of chick embryos and reactivation of Msx1 and Shh expression in response to FGF4 and ridge signals. Mech Dev. 1996;55:119–131. doi: 10.1016/0925-4773(95)00492-0. [DOI] [PubMed] [Google Scholar]
  13. Kostakopoulou K, Vargesson N, Clarke JD, et al. Local origin of cells in FGF4-induced outgrowth of amputated chick wing bud stumps. Int J Dev Biol. 1997;41:747–750. [PubMed] [Google Scholar]
  14. Krabbenhoft KM, Fallon JF. The formation of leg or wing specific structures by leg bud cells grafted to the wing bud is influenced by proximity to the apical ridge. Dev Biol. 1989;131:373–382. doi: 10.1016/s0012-1606(89)80011-9. [DOI] [PubMed] [Google Scholar]
  15. Ozpolat B, Zapata M, Fruge J, et al. Regeneration of the elbow joint in the developing chick embryo recapitulates development. Dev Biol. 2012;372:229–238. doi: 10.1016/j.ydbio.2012.09.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Pautou M. Proximo-distal pattern regulation in deficient avian limb buds. Roux's Arch Develop Biol : 1977. pp. 183,177–191. [DOI] [PubMed] [Google Scholar]
  17. Post L, Innis J. Altered Hox expression and increased cell death distinguish Hypodactyly from Hoxa13 null mice. Int J Dev Biol. 1999;43:287–294. [PubMed] [Google Scholar]
  18. Probst S, Kraemer C, Demougin P, et al. SHH propagates distal limb bud development by enhancing CYP26B1-mediated retinoic acid clearance via AER-FGF signalling. Development. 2011;138:1913–1923. doi: 10.1242/dev.063966. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Roselló-Díez A, Torres M. Regulative patterning in limb bud transplants is induced by distalizing activity of apical ectodermal ridge signals on host limb cells. Dev Dyn. 2011;240:1203–1211. doi: 10.1002/dvdy.22635. [DOI] [PubMed] [Google Scholar]
  20. Roselló-Díez A, Ros MA, Torres M. Diffusible signals, not autonomous mechanisms, determine the main proximodistal limb subdivision. Science. 2011;332:1086–1088. doi: 10.1126/science.1199489. [DOI] [PubMed] [Google Scholar]
  21. Sato K, Koizumi Y, Takahashi M, et al. Specification of cell fate along the proximal-distal axis in the developing chick limb bud. Development. 2007;134:1397–1406. doi: 10.1242/dev.02822. [DOI] [PubMed] [Google Scholar]
  22. Saunders JW., Jr The proximo-distal sequence of origin of the parts of the chick wing and the role of the ectoderm. J Exp Zool. 1948;108:363–403. doi: 10.1002/jez.1401080304. [DOI] [PubMed] [Google Scholar]
  23. Stark RJ, Searls RL. The establishment of the cartilage pattern in the embryonic chick wing and evidence for a role of the dorsal and ventral ectoderm in normal wing development. Dev Biol. 1974;38:51–63. doi: 10.1016/0012-1606(74)90258-9. [DOI] [PubMed] [Google Scholar]
  24. Summerbell D. Regulation of the deficiencies along the proximal distal axis of the chick wing-bud: a quantitative analysis. J Embryol Exp Morphol. 1977;41:137–159. [PubMed] [Google Scholar]
  25. Summerbell D, Lewis JH. Time, place and positional value in the chick limb-bud. J Embryol Exp Morphol. 1975;33:621–643. [PubMed] [Google Scholar]
  26. Tabin C, Wolpert L. Rethinking the proximodistal axis of the vertebrate limb in the molecular era. Genes Dev. 2007;21:1433–1442. doi: 10.1101/gad.1547407. [DOI] [PubMed] [Google Scholar]
  27. Tamura K, Nomura N, Seki R, et al. Embryological evidence identifies wing digits in birds as digits 1, 2, and 3. Science. 2011;331:753–757. doi: 10.1126/science.1198229. [DOI] [PubMed] [Google Scholar]
  28. Taylor G, Anderson R, Reginelli A, et al. FGF-2 induces regeneration of the chick limb bud. Dev Biol. 1994;163:282–284. doi: 10.1006/dbio.1994.1144. [DOI] [PubMed] [Google Scholar]
  29. Tickle C. Morphogen gradients in vertebrate limb development. Semin Cell Dev Biol. 1999;10:345–351. doi: 10.1006/scdb.1999.0294. [DOI] [PubMed] [Google Scholar]
  30. Towers M, Tickle C. Generation of pattern and form in the developing limb. Int J Dev Biol. 2009;53:805–812. doi: 10.1387/ijdb.072499mt. [DOI] [PubMed] [Google Scholar]
  31. Towers M, Signolet J, Sherman A, et al. Insights into bird wing evolution and digit specification from polarizing region fate maps. Nat Commun. 2011;2:426. doi: 10.1038/ncomms1437. [DOI] [PubMed] [Google Scholar]
  32. Vargesson N. Thalidomide-induced limb defects: resolving a 50 years old puzzle. BioEssays. 2009;31:1327–1336. doi: 10.1002/bies.200900103. [DOI] [PubMed] [Google Scholar]
  33. Vargesson N. Thalidomide. In: Gupta R, editor. Reproductive and Developmental Toxicology. Amsterdam: Academic Press; 2011. pp. 395–403. [Google Scholar]
  34. Vargesson N, Laufer E. Negative Smad expression and regulation in the developing chick limb. PLoS ONE. 2009;4:e5173. doi: 10.1371/journal.pone.0005173. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Vargesson N, Clarke JD, Vincent K, et al. Cell fate in the chick limb bud and relationship to gene expression. Development. 1997;124:1909–1918. doi: 10.1242/dev.124.10.1909. [DOI] [PubMed] [Google Scholar]
  36. Vargesson N, Luria V, Messina I, et al. Expression patterns of Slit and Robo family members during vertebrate development. Mech Dev. 2001;106:175–180. doi: 10.1016/s0925-4773(01)00430-0. [DOI] [PubMed] [Google Scholar]
  37. Wang Z, Young RL, Xue H, et al. Transcriptomic analysis of avian digits reveals conserved and derived digit identities in birds. Nature. 2011;477:583–586. doi: 10.1038/nature10391. [DOI] [PubMed] [Google Scholar]
  38. Wolpert L, Tickle C, Sampford M. The effect of cell killing by X-irradiation on pattern formation in the chick limb. J Embryol Exp Morphol. 1979;50:175–198. [PubMed] [Google Scholar]
  39. Wyngaarden LA, Hopyan S. Plasticity of proximal-distal cell fate in the mammalian limb bud. Dev Biol. 2008;313:225–233. doi: 10.1016/j.ydbio.2007.10.039. [DOI] [PubMed] [Google Scholar]
  40. Yallup B, Hinchliffe JR. Regulation along the antero-posterior axis of the chick wing bud. In: Fallon J, Caplan A, editors. Limb Development and Regeneration. New York: A. R. Liss Inc; 1983. pp. 131–140. [PubMed] [Google Scholar]
  41. Zeller R. The temporal dynamics of vertebrate limb development, teratogenesis and evolution. Curr Opin Genet Dev. 2010;20:384–390. doi: 10.1016/j.gde.2010.04.014. [DOI] [PubMed] [Google Scholar]
  42. Zeller R, López-ríos J, Zuniga A. Vertebrate limb bud development: moving towards integrative analysis of organogenesis. Nat Rev Genet. 2009;10:845–858. doi: 10.1038/nrg2681. [DOI] [PubMed] [Google Scholar]
  43. Zwilling E. Interaction between limb bud ectoderm and mesoderm in the chick embryo: II Experimental limb duplication. J Exp Zool. 1956;132:173–187. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Fig. S1. Expression chondrogenic precursors at the time of operation. Whole-mount RNA in situ hybridisation was completed for Sox9 to identify whether chondrogenic precursors were present in the limb at the time of the experiment. (A) A control, unoperated limb at HH St21 shows no Sox9 expression in the embryo (n = 3/3). (B) A control, unoperated limb at HH St27 (48 h later) showing normal Sox9 expression (n = 3/3). Images in (A) and (B) were carried out at the same time and confirm Sox9 expression is not seen in HH St21 limbs. Scale bar: 500 μm.

joa0223-0001-sd1.tif (737.7KB, tif)

Articles from Journal of Anatomy are provided here courtesy of Anatomical Society of Great Britain and Ireland

RESOURCES