Post-amputation reactive oxygen species production is necessary for axolotls limb regeneration

Abstract

Introduction: Reactive oxygen species (ROS) represent molecules of great interest in the field of regenerative biology since several animal models require their production to promote and favor tissue, organ, and appendage regeneration. Recently, it has been shown that the production of ROS such as hydrogen peroxide (H₂O₂) is required for tail regeneration in Ambystoma mexicanum. However, to date, it is unknown whether ROS production is necessary for limb regeneration in this animal model. Methods: forelimbs of juvenile animals were amputated proximally and the dynamics of ROS production was determined using 2'7- dichlorofluorescein diacetate (DCFDA) during the regeneration process. Inhibition of ROS production was performed using the NADPH oxidase inhibitor apocynin. Subsequently, a rescue assay was performed using exogenous hydrogen peroxide (H₂O₂). The effect of these treatments on the size and skeletal structures of the regenerated limb was evaluated by staining with alcian blue and alizarin red, as well as the effect on blastema formation, cell proliferation, immune cell recruitment, and expression of genes related to proximal-distal identity. Results: our results show that inhibition of post-amputation limb ROS production in the A. mexicanum salamander model results in the regeneration of a miniature limb with a significant reduction in the size of skeletal elements such as the ulna, radius, and overall autopod. Additionally, other effects such as decrease in the number of carpals, defective joint morphology, and failure of integrity between the regenerated structure and the remaining tissue were identified. In addition, this treatment affected blastema formation and induced a reduction in the levels of cell proliferation in this structure, as well as a reduction in the number of CD45⁺ and CD11b + immune system cells. On the other hand, blocking ROS production affected the expression of proximo-distal identity genes such as Aldha1a1, Rarβ, Prod1, Meis1, Hoxa13, and other genes such as Agr2 and Yap1 in early/mid blastema. Of great interest, the failure in blastema formation, skeletal alterations, as well as the expression of the genes evaluated were rescued by the application of exogenous H₂O₂, suggesting that ROS/H₂O₂ production is necessary from the early stages for proper regeneration and patterning of the limb.

Keywords: Ambystoma mexicanum; axolotl; blastema; hydrogen peroxide; immune cells; limb regeneration; macrophages; reactive oxygen species.

FIGURE 1

ROS production during limb regeneration in A. mexicanum. (A) representative images of ROS production acquired by confocal microscopy. The pre-amputation image represents the medial area of the non-amputated humerus (basal level of ROS production). ROS production was detected at the amputation plane at 1, 2, 3, and 5 dpa. At 7 dpa and 9 dpa ROS was detected in the apical epithelial cap (AEC) and at 11 dpa in the AEC and developing blastema region. White arrowheads represent the amputation plane. A white box with a dashed line is shown at higher magnification in lower panel images acquired with differential interference contrast (DIC) in conjunction with fluorescence images for anatomical details. Yellow arrows indicate epithelial localization of ROS and red arrows localization in blastema cells. (B), Semi-quantification of relative fluorescence intensity. Fluorescence levels show two apparent waves of ROS production between 1dpa and 3 dpa for the first wave, and between 7 and 15 dpa, for the second wave of ROS production. One-way ANOVA with a Tukey post-hoc was performed to compare each point with basal levels of pre-amputation ROS production (n = 10 for each point, 5 animals per point). Data are expressed as mean ± SEM (mean standard error). ***p < 0.001, **p < 0.01, *p < 0.05.

FIGURE 2

Blocking ROS production by apocynin affects limb regeneration in A. mexicanum. (A), Representative images of animals exposed to different treatments compared to the control group in 0.1% DMSO. Defects generated by inhibition of NOX activity and rescue treatments (apocynin 400 µM + H₂O₂) are shown. Solid red lines represent the amputation plane. (B), Quantification of regenerated limb size at 72 dpa. Apocynin-exposed group evaluated at 124 dpa was included. One-way ANOVA and post-hoc Tukey were performed to compare the evaluated groups (n = 10 per group). (C), Semi-quantification of ROS levels post-exposure to apocynin inhibitor (n = 7). ROS levels decrease significantly under inhibitory treatment. Student’s t-test was performed to compare the two groups. Data are expressed as mean ± SEM. ***p < 0.001, **p < 0.01, *p < 0.05.

FIGURE 3

Exogenous H₂O₂ induces a rescue effect on the regenerating structure. (A) Illustrative scheme of the rescue assay performed. (B) Representative images of regenerating tissues at 11 and 31 dpa of controls in 0.1% DMSO, inhibitory treatments with apocynin and without rescue, and inhibitory treatments rescued with exogenous H₂O₂. (C) Quantification of regenerate size at 31 dpa. White arrowheads represent the amputation plane. One-way ANOVA and post-hoc Tukey were performed to compare the evaluated groups (n = 10 per group). Data are expressed as mean ± SEM. ***p < 0.001, **p < 0.01, *p < 0.05.

FIGURE 4

Inhibition of NOXs activity reduces the size of the regenerated skeleton and induces a decrease in the number of carpals in A. mexicanum at 72 dpa. (A), Representative images of the treatments evaluated. The skeleton of apocynin (APO)-treated limbs confirms a miniature limb phenotype. Treatments with exogenous H₂O₂ partially rescue the phenotype generated by NOXs inhibition. (B–E), Quantification of the skeletal elements of the stylopod (humerus), zeugopod (radius and ulna), and autopod (carpals, metacarpals, and phalanges). One-way and post-hoc Tukey ANOVA was performed to compare the evaluated groups (n = 10 per group). (F), inhibition of ROS production by APO generated a decrease in the number of carpals (5 animals had only 5 carpals and 2 animals had six carpals out of 10 animals evaluated). Red arrow, black arrow and dotted circle indicate apparent ectopic fusions in anterior (ant), posterior (post), and intermediate carpals, respectively. (G), Porcetage of animals with reduction in the number of carpals observed in the different experimental groups. h, humerus; r, radius; u, ulna. Data are expressed as mean ± SEM. ***p < 0.001, **p < 0.01, *p < 0.05.

FIGURE 5

Inhibitory treatment of NOXs-dependent ROS production affects the integration of the regenerated skeleton and joint morphology in A. mexicanum at 72 dpa. (A), Illustration of the areas of interest evaluated: ai (integration area) delimited with red brackets. The red circle indicates the joint area studied. (B–E), representative images of the experimental groups evaluated by alcian blue and alizarin red staining: Controls in 0.1% DMSO, apocynin (APO) treatment, rescue assay (APO + H₂O₂) and control exposed to 0.1% DMSO +50 µM H₂O₂. Yellow dotted lines indicate remnant bone tissue. (F–I), Histological sections with Masson’s trichrome staining at 5 µm. (F), shows continuity between the regenerated humerus and remnant skeletal tissue in the “ai”. (G), the integration area shows discontinuity between the regenerated and remnant humerus. (H), application of exogenous H₂O₂ rescues the phenotype affected by ROS inhibition. (I), Control animals exposed to 0.1% DMSO + H₂O₂ show integration between regenerated skeletal tissue and remnant skeletal tissue. (J), Percentage of animals that presented integration failure in the different experimental groups. (K,L) Representative images of joint morphology. (K), typical synovial joint morphology. (L), APO treatment affects the joint capsule and generates constriction of synovial spaces. (M), treatment with exogenous H₂O₂ rescues the articular morphology. (N), DMSO + H₂O₂ treatments do not affect typical joint morphology. (O–R), magnification of area delimited with red dotted lines in K, L, M, and N. (O) A well-defined synovial space can be observed. (P) Animals treated with apocynin show a fairly constricted synovial space with apparent continuity between the connective tissue of the articular capsule and the articular surface of the regenerated humerus (orange arrows). (Q) Similar to controls, a more defined synovial space is observed. (R) An articular morphology as seen in the DMSO controls can be observed. (S), Percentage of animals that presented joint morphological abnormality in the evaluated groups. Yellow arrows indicate connective tissue joint capsule. Black arrows indicate synovial spaces. ai, integration area; h, regenerated humerus; r, radius; u, ulna.

FIGURE 6

ROS is required for cell cycle re-entry, blastema formation, and blastema growth during limbregeneration in A. mexicanum. (A–C), representative images of blastema formation at 11 dpa of controls (DMSO 0.1%), APO and rescue treatment (APO + H₂O₂ 50 µM). White arrowheads indicate the amputation plane. (D–F), Hematoxylin-Eosin histologic sections of blastemas at 11 dpa of evaluated groups. (G), Quantification of blastema size. (H–J), Imnunofluorescence anti-BrdU at 11 dpa. Hydrogen peroxide rescues proliferation (re-entry into the cell cycle) of blastema cells affected by inhibition of ROS production. White arrows indicate the amputation plane. (K), Quantification of cell proliferation expressed as percentage of BrdU-positive cells. (L–Q), representative images of blastema growth at 14 dpa and 21 dpa in the different groups evaluated. (R), quantification of blastema size at 14 and 21 dpa. b, blastema; c, cartilage; ep, epithelium.*, indicates area of fibrous tissue accumulation between epithelium and remnant cartilage. Black and white arrowheads indicate the amputation plane. A one-way ANOVA with a Tukey post hoc was performed to compare the groups evaluated (n = 10 for blastema sizes and n = 5 for immunofluorescence analysis). Data are expressed as mean ± SEM (mean standard error). ***p < 0.001, **p < 0.01, *p < 0.05.

FIGURE 7

NOXs-dependent production ROS inhibition and exogenous H₂O₂ induces expression changes of genes related to blastema formation and positional identity during limb regeneration in A. mexicanum at 11 dpa. (A–G), RT-qPCR of Prod1, Meis1, Hoxa13, Aldh1a1, Rarβ, Agr2 and Yap1 genes. The results show that blocking ROS production significantly affects the expression of the genes described here and their expression levels can be rescued by the application of exogenous H₂O₂. Gene expression levels were normalized to the expression of the endogenous reference gene 18S. Data are expressed as mean ± SEM. One-way ANOVA followed by Tukey’s post hoc test was performed for comparisons between groups treated with apocynin, exogenous H₂O₂ and controls in 0.1% DMSO. ***p < 0.001, **p < 0.01, *p < 0.05.

FIGURE 8

NOXs-dependent production ROS inhibition and exogenous H₂O₂ from 12 dpa to 13 dpa regulate the expression of positional identity and blastema formation-related genes during limb regeneration in A. mexicanum. (A), scheme of treatments performed. Treatments used are symbolized with colored bars. (B–I) RT-qPCR of Prod1, Meis1, Meis2 Hoxa13, Aldh1a1, Rarβ, Agr2 and Yap1 genes. The results show that blocking ROS production significantly regulate the expression of these genes and their expression levels can be rescued by the application of exogenous H₂O₂. As in the previous trials, the expression levels were normalized to the expression of the endogenous reference gene 18S. Data are expressed as mean ± SEM. One-way ANOVA followed by Tukey’s post hoc test was performed for comparisons between groups treated with apocynin, exogenous H₂O₂ and controls in 0.1% DMSO. ***p < 0.001, **p < 0.01, *p < 0.05.

FIGURE 9

The production of ROS generated post-amputation of the limb is necessary for the recruitment and phagocytic activity of inflammatory cells. (A–C), Representative immunofluorescence images against the leukocyte pan marker CD45 at 11 dpa. Boxes in white dotted lines are shown at higher magnification in solid line boxes for each image. (A) CD45 ⁺ cells are predominantly located in the blastema region adjacent to the amputation plane. (B) A reduced number of CD45 ⁺ cells are observed near the amputation plane. (C) a notable increase of CD45 ⁺ cells are observed in the region of the blastema and the amputation plane. (D–F), Representative immunofluorescence images against CD11b. A higher presence of CD11b + cells can be seen in the control group compared with the apocynin-treated group. Animals exposed to rescue treatment show a prominent accumulation of CD11b + cells in the blastema and amputation plane. (G,H), quantification of CD45 and CD11b + positive cells, respectively. (I–K), vital staining with neutral red at 11 dpa. Representative images of control animals in 0.1% DMSO, exposed to apocynin inhibitor (APO) and rescue assays. Boxes in dotted white lines are shown at higher magnification in (I′,j′,K′) for each experimental group. (L), quantification of cells positive for neutral red staining. Data are expressed as mean ± SEM. One-way ANOVA followed by Tukey’s post hoc test was performed for comparisons between groups treated with apocynin, exogenous H₂O₂ and controls in 0.1% DMSO. ***p < 0.001, **p < 0.01, *p < 0.05.

FIGURE 10

Graphical abstract of the results obtained and putative mechanistic insight of ROS in regeneration. Following limb amputation, ROS are produced. The production of ROS is necessary for the re-entry into the cell cycle of the remaining tissues and consequently for the formation and growth of the blastema. ROS also regulates the expression of genes previously involved in blastema formation such as Yap1 and Agr2. This suggests that ROS production-dependent blastema size influences the final size of the limb and its regenerating skeleton. Moreover, ROS regulates the expression of proximal-distal identity genes, suggesting that the obtained phenotype of integration failure between the regenerated and remnant skeleton as well as alterations in joint morphogenesis could be regulated by ROS production. In addition, ROS are necessary to promote the recruitment of inflammatory cells such as leukocytes including monocytes/macrophages during blastema formation, as well as their phagocytic activity. Therefore, we hypothesize that the regulation of the inflammatory response represents a potential mechanism by which ROS mediate its function during axolotl limb regenearation. The solid black lines represent the results obtained in this work and the dotted blue lines represent the proposed relationships according to previous studies and those obtained in this work.