Intrinsic properties of limb bud cells can be differentially reset

Abstract

An intrinsic timing mechanism specifies the positional values of the zeugopod (i.e. radius/ulna) and then autopod (i.e. wrist/digits) segments during limb development. Here, we have addressed whether this timing mechanism ensures that patterning events occur only once by grafting GFP-expressing autopod progenitor cells to the earlier host signalling environment of zeugopod progenitor cells. We show by detecting Hoxa13 expression that early and late autopod progenitors fated for the wrist and phalanges, respectively, both contribute to the entire host autopod, indicating that the autopod positional value is irreversibly determined. We provide evidence that Hoxa13 provides an autopod-specific positional value that correctly allocates cells into the autopod, most likely through the control of cell-surface properties as shown by cell-cell sorting analyses. However, we demonstrate that only the earlier autopod cells can adopt the host proliferation rate to permit normal morphogenesis. Therefore, our findings reveal that the ability of embryonic cells to differentially reset their intrinsic behaviours confers robustness to limb morphogenesis. We speculate that this plasticity could be maintained beyond embryogenesis in limbs with regenerative capacity.

Keywords: Chick; Hoxa13; Limb development; Proliferation; Proximo-distal.

Fig. 1.

Proximal and distal autopod progenitors display similar positional values when grafted to a young environment. (A-C) GFP-expressing HH24 distal cells (150 µm blocks) grafted under the AER of earlier wild-type HH20 wings give rise to structures distal to the wrist without perturbing the morphogenesis of the host. (D-F) GFP-expressing HH27 distal cells (150 µm blocks) display similar positional values but the development of the host autopod is disrupted. (G-I) GFP-expressing HH24 distal mesenchyme tissue subject to two consecutive grafts spaced 24 h apart also give rise to structures distal to the wrist without altering the morphogenesis of the host. For each experiment, the schematic is depicted on the left followed by the picture of the 11-day-old specimen under UV light to visualize the graft and the skeletal preparation. (J,K) Consecutive sections of a HH27 graft 7 days after implantation hybridized for Gfp and Sox9 as indicated. Note that the proximal boundary of the grafted tissue lies at the wrist and that the Gfp-expressing cells have integrated in all type of tissues. (L) Schematic summary representation of the developmental potential of the grafts. Each asterisk represents the proximal boundary of the grafted tissue for each experiment. The fate of homochronic HH20 to HH20 (blue) and HH24 to HH24 (black) from Saiz-Lopez et al. (2015) is also depicted. (M) Histogram showing the length distribution of the host limb segments in each of the grafting experiments performed. Only the autopod of HH20 host with HH27 grafts shows a statistically significant reduction in length. The GFP-expressing tissue is depicted as green in this and all schematic representations. Data are mean±s.e.m. of n=34 for normal, 8 for HH24-20, 13 for HH24-20x2 and 15 for HH27-20. Scale bars: 1 mm (B-I) and 600 µm (J,K).

Fig. 2.

Hoxa13 expression is robustly maintained in the grafted tissue. Frontal (flat) sections of host limbs showing stable expression of Hoxa13 (also hybridized for Fgf8) in the graft irrespective of the earlier host environment (A,D,G,J,M,P). The position of the graft is assessed by in situ hybridization for Gfp in consecutive, 7 µm apart, sections (B,E,H,K,N,Q). The type of graft is indicated on the left and the schematics, including the expression patterns of Hoxa13 (dark blue) and Gfp (bright green) on the left (C,F,I,L,O,R). The age of the host (brown) and grafts (green) at the time of the analysis is also indicated in the schematics. Note that 24 h after implantation, the graft is only partially immersed into the host Hoxa13 domain of expression (G-O) but that by 48 h after grafting the entire graft is embedded in the host Hoxa13 domain (P-R). Note also that the graft does not interfere with the dynamics of the host Hoxa13 expression. The red asterisks mark the proximal limit of the graft. At least three examples for each experimental condition were analysed. Scale bars: 175 µm.

Fig. 3.

Sorting out of Hoxa13-expressing and Hoxa13-non-expressing distal progenitor cells in vivo. (A-G) HH20 GFP-expressing and wild-type progenitor cells (A) randomly distribute in reaggregated grafts after 24 h as indicated by the expression of Gfp (E,F) as Hoxa13 (B,C) has not been activated yet. (H-N) GFP-expressing HH20 and wild-type HH24 progenitor cells (H) sort out in reaggregated grafts within 24 h with the labelled cells preferentially located in the periphery of the graft in contact with the age-matched host cells, as indicated by the expression of Hoxa13 (I,J) and Gfp (L,M). The distribution of cells in the re-aggregated grafts according to the expression patterns is schematically represented in D,G,K,N. Hoxa13-ve: Hoxa13 negative; Hoxa13+ve: Hoxa13 positive. Scale bars: 250 µm (B,E,I,L) and 100 µm (C,F,J,M).

Fig. 4.

Reduced sorting out of different stage Hoxa13-expressing cells in vitro. Mixtures of HH27 GFP-expressing and wild-type cells do not segregate in micromass cultures (A,B). Mixtures of GFP-expressing HH24 and wild-type HH27 distribute much more evenly (C,D) than mixtures of GFP-expressing HH20 and wild-type HH27 cells which readily sort out (E,F). Three examples for each experimental condition were analysed. (G) Granulometry graph showing the proportion of the total area occupied by different size aggregates, in each type of mixed micromass culture. The size of the minor diameter of the aggregate is indicated on the right. Scale bars: 200 µm.

Fig. 5.

Normal Fgf8 expression in the AER over old grafts. Strong expression of Fgf8 is observed in the AER over the graft of GFP-expressing HH27 to HH20 host wing buds 24 h after grafting (A-C). Fgf8 expression is also normal over grafts of GFP-expressing HH24 serially transferred two times to HH20 host wing buds (D-F). The expression of Fgf8 is also normal over serial HH27 grafts to HH20 host 48 h after grafting (G-I) even if the serial grafting is spaced by 48 h. (A,D,G) Fgf8 is also hybridized with Hoxa13. Fgf8 expression is shown in frontal (A,G) and longitudinal (D) sections also hybridized for Hoxa13. Note area of grafted tissue shown by Gfp expression (B,E,H) in consecutive sections to those hybridized for Fgf8 and Hoxa13 (A,D,G). On the right, the schematics for each experiment are shown in which the final age of the host and grafted tissue is indicated with the corresponding colour as well as the limb axes (C,F,I). Frontal sections are shown except for D-F, which are longitudinal sections. At least three examples for each experimental condition were analysed. Scale bars: 150 µm.

Fig. 6.

Cell cycle parameters can be reset at HH24 but not at HH27. Distal HH27 cells maintain their cell cycle parameters even in a host earlier environment (A) while HH24 distal cells reset their cell cycle parameters to that of host equivalent cells within 24 h after grafting (B) and this resetting is maintained in serial grafts (C). Note that in A, there is a significant difference in G1-phase numbers between left and right wing buds (Pearson's χ² test, P<0.05) consistent with cell cycle parameters being maintained in donor tissue, and in B,C, there is a significant difference in G1-phase numbers between host values and expected values (Saiz-Lopez et al., 2015) for the stage of the donor tissue (Pearson's χ² test, P<0.05) consistent with cell cycle parameters of the graft being reset close to host levels. Note in all cases, 10-12 blocks of tissue were pooled from replicate experiments and run once. There is no abnormal cell death in heterochronic distal grafts (D-G). The TUNEL assay fails to detect abnormal cell death at any stage analysed during the development of the grafted limbs. One example for a HH24 to HH20 graft (D,E, frontal sections) and another for a HH27 to HH20 graft (F,G, longitudinal sections) are shown. In each case, the graft is detected in consecutive sections hybridized for Gfp (E,G). Note that the graft in D still maintains weak Gfp expression. Scale bars: 200 µm (D,F) and 275 µm (E,G).