Integration failure of regenerated limb tissue is associated with incongruencies in positional information in the Mexican axolotl

Abstract

Introduction: Little is known about how the newly regenerated limb tissues in the Mexican axolotl seamlessly integrate with the remaining stump tissues to form a functional structure, and why this doesn't occur in some regenerative scenarios. In this study, we evaluate the phenomenological and transcriptional characteristics associated with integration failure in ectopic limb structures generated by treating anterior-located ectopic blastemas with Retinoic Acid (RA) and focusing on the "bulbus mass" tissue that forms between the ectopic limb and the host site. We additionally test the hypothesis that the posterior portion of the limb base contains anterior positional identities. Methods: The positional identity of the bulbus mass was evaluated by assaying regenerative competency, the ability to induce new pattern in the Accessory Limb Model (ALM) assay, and by using qRTPCR to quantify the relative expression of patterning genes as the bulbus mass deintegrates from the host site. We additionally use the ALM and qRTPCR to analyze the distribution of anterior and posterior positional identities along the proximal/distal limb axis of uninjured and regenerating limbs. Results: The bulbus mass regenerates limb structures with decreased complexity when amputated and is able to induce complex ectopic limb structure only when grafted into posterior-located ALMs. Expressional analysis shows significant differences in FGF8, BMP2, TBX5, Chrdl1, HoxA9, and HoxA11 expression between the bulbus mass and the host site when deintegration is occuring. Grafts of posterior skin from the distal limb regions into posterior ALMs at the base of the limb induce ectopic limb structures. Proximally-located blastemas express significantly less HoxA13 and Ptch1, and significantly more Alx4 and Grem1 than distally located blastemas. Discussion: These findings show that the bulbus mass has an anterior-limb identity and that the expression of limb patterning genes is mismatched between the bulbus mass and the host limb. Our findings additionally show that anterior positional information is more abundant at the limb base, and that anterior patterning genes are more abundantly expressed in proximally located blastemas compared to blastemas in the more distal regions of the limb. These experiments provide valuable insight into the underlying causes of integration failure and further map the distribution of positional identities in the mature limb.

Keywords: accessory limb model; axolotl; bulbous mass; integration; limb patterning; limb regeneration; retinoic acid.

FIGURE 1

Integration failure phenotypes: (A–D) Cartoons represent previously observed integration phenotypes in regenerating salamanders, where the red arrows indicate the site of injury. (A) Representation of normal integration of the regenerated tissue (distal to injury site) with the stump tissue (proximal to injury site) following a limb amputation. (B) Incomplete joint formation between the regenerated and existing tissue when proximal skeletal elements are removed from the stump prior to amputation (Weiss, 1925), and when an ectopic limb is formed following an ALM surgery (Endo et al., 2004). (C) Fused integration phenotypes are observed when bone processing or maturation are altered (Vieira et al., 2018; Riquelme-Guzmán et al., 2022), when a deep injury is made during ALM Surgery (Makanae et al., 2014a), and occasionally when an amputation blastema is treated with RA (Thoms and Stocum, 1984). (D) Disruption of integration due to growth of ectopic tissue is observed when amputation blastemas are treated with high doses of RA (Thoms and Stocum, 1984), or when an ectopic anterior blastema is treated with RA (McCusker et al., 2014).

FIGURE 2

Characterization of RA-induced ectopic tissue in innervated anterior wound sites: (A) Live image time course over 8 weeks of anterior-located blastemas treated with RA showing bulbous mass formation, and de-integration from the host limb. Red arrows mark the location where the bulbous mass is connected to the host limb at the time points when the bulbous mass is wider than the site of connection. (B) Histogram showing the time point when the RA treated blastemas (shown as % of the total blastemas) exhibited de-integration (N = 13) (C) Cartoon describing experiment design. Anterior located innervated limb wounds will generate blastemas, which upon systemic treatment with RA, will form ectopic structures connected to the host via bulbous mass tissue. Ectopic structures were amputated through the bulbous mass and were monitored for regeneration. (D) Diagram of RA-induced ectopic limb structure and bulbous mass. Red dotted line indicates the position of the amputation plane in the regeneration assay in (E). (E) Live images of RA-induced ectopic limb/bulbus mass (top) and RA-induced bulbous mass alone (lower) before amputation (Pre-amp), and 3 and 21 weeks (W3, W21) post-amputation. Dotted red line delineates the boundary between the remaining bulbous mass and the blastema. (right) The resulting ectopic structures are indicated with red arrows in the whole mount skeletal stained limbs. Table 2 has the full experimental statistics for (E).

FIGURE 3

Characterizing anterior/posterior positional information in the bulbous mass using the ALM assay. (A) Cartoon representing the surgical procedure where bulbous mass tissue form a GFP + animal (green) is grafted to innervated anterior or posterior limb wound sites. Cartoons below explain the possible phenotypes in this assay and how they are interpreted. (B) Live images of the bulbous masses where grafts were obtained form (red dotted line outlines the location where the graft was taken) at 3 and 21 weeks (W3, W21) post-surgery. Host locations are indicated; where “P host” is a posterior wound site, and “A host” is an anterior wound site. (Right) Images of whole mount skeletal preparations of ectopic limb structures (red arrows). Full phenotype data is provided in Table 3.

FIGURE 4

Expression of limb pattern genes in mature limb tissue and developing ectopic limb structures. (A, C, E, G) Histograms representing qRTPCR data for each primer set on mature limb tissue samples from the autopod, zeugopod, stylopod, and flank tissue. Statistical comparisons were made between the stump with each limb position using a two-tailed T-test with equal variance (*p < 0.05, **p < 0.005, ***p < 0.0005, ****p < 0.00005). (B, D, F, H, I, J, K) Histograms representing qRTPCR data for each primer set on limb tissue samples from stylopod (S), 7-day old innervated wound site zero, one, five, nine, and 11 weeks post RA injection (W0 –W11), and flank tissue (F). Statistical comparisons between each time point and the stylopod (S) tissue are marked with black asterisk. In F and H comparisons between bulbous mass and flank tissue are marked with red asterisks. All data represents the average of 4 biological replicates. Error bars are SEM.

FIGURE 5

Distribution of positional information in RA-induced ectopic limbs and normal limbs. (A, B) Cartoon representing the hypothesized distribution of anterior (red), posterior (blue), and the P/D (gradient of yellow, strongest in proximal) positional memory in ALM ectopic limbs (A) and RA-induced ectopic limbs with bulbus mass (B), and host limb tissues. Red dotted lines in bulbus mass indicate anterior positional information. (C) Cartoon representing the hypothesized distribution of A/P positional memories in the (internal) periskeletal tissue (dark red = anterior, dark blue = posterior) and (superficial) dermal tissue (light red and blue, respectively) in the uninjured limb. Squares indicate the hypothesized positional content in posterior ALM wound sites in mid-stylopod (1) and proximal stylopod (2) positions. (D) (left) Cartoon illustrating the localization of the blastemas collected for qRTPCR analysis. (right) Box and whisker plots of expressional data for HoxA13, Shh, Hand2, Ptch1, FGF8, Alx4, Grem1, and Gli3 from autopod (light yellow), stylopod (mid-tone yellow) and zeugopod (dark yellow) blastema samples. Statistical comparisons between the autopod and zeugopod or stylopod were performed using a two-tailed T-test with equal variance (*p < 0.05, **p < 0.005). All data represents the average of 4 biological replicates. (E) Cartoon illustrating the positioning of the nerve deviated wound sites on posterior and anterior limb base locations, flanked by images of whole mount skeletal preparations of the most extreme growth phenotypes in each position. (F) Cartoon illustrating the engraftment of distal anterior (A) and posterior (P) full-thickness skin into nerve-deviated wound sites at the limb base. (G) (Left) Live images of nerve deviated posterior and anterior wound sites at the limb base 3 weeks post-surgery where GFP + distal tissue graft is green. (Right) Whole mount (WM) skeletal preparation of wound sites 15 weeks post-surgery. Phenotype statistics of all surgical manipulations are provided in Table 4.