The specialist in regeneration—the Axolotl—a suitable model to study bone healing?

Bone fractures are common and a significant subset develop delayed healing or non-unions despite current treatments. Mammalian fracture healing proceeds via inflammatory recruitment of progenitors and, depending on stability, intramembranous or endochondral ossification with a cartilaginous callus that remodels to lamellar bone. Critical-sized defects (CSD) model non-unions. Learning from highly regenerative organisms like the axolotl may reveal mechanisms to improve bone regeneration. Although axolotls can regenerate whole limbs, their fracture healing and especially standardized fixation has been poorly characterized. Prior axolotl fracture studies often used zeugopod bones with indirect stabilization by the neighboring bone, resulting in misalignment and variability due to differences in animal size, age, and incomplete diaphyseal ossification. To enable controlled comparisons with mammalian fracture healing and axolotl limb regeneration, this study asks whether a rigid internal plate fixation can be applied to the axolotl femur to create a standardized osteotomy model, and how healing dynamics compare to non-stabilized fractures and to limb amputation.

Previous amphibian studies reported cartilaginous callus formation 20–45 days post-fracture in axolotls, with expression of collagen I/II mRNAs suggesting similarities to mammalian endochondral repair. However, fracture models often cut one of paired zeugopod bones, leaving the other to act as a nonstandardized stabilizer, causing misalignment and periosteal contacts that may accelerate or bias healing. Reported healing times varied widely with animal size/age: ulna cuts repaired within 7 months; fibula osteotomy within 2 months; 10–20% fibula length gaps within 3 months. CSDs also varied (2–4 mm defects in different bones/sizes), sometimes resulting in fusion to the supporting bone. Axolotls can regenerate amputated limbs and some joint defects but not CSDs, a paradox suggesting differing constraints for regeneration vs fracture repair. In mammals, multiple standardized fixation methods (intramedullary pins, external/internal plates) exist, enabling reproducible models including CSDs. An axolotl model using internal plates could better parallel mammalian systems and reduce variability.

Experimental design: Aged axolotls (Ambystoma mexicanum), 5–8 years old and ≥20 cm snout-to-tailtip, and C57BL/6J mice (12–14 weeks; 56–61 weeks) were used under institutional approvals. Groups included axolotl femur osteotomy with plate fixation, non-stabilized osteotomy, and limb amputation; mouse femur osteotomy with plate fixation served for comparison. Axolotl plate osteotomy: Under 0.03% benzocaine anesthesia, upper hind limb femur was exposed with minimal soft-tissue disruption and protected by plastic film. A rigid 7.75 mm MouseFix 4-hole plate (RISystem) was applied with four 2 mm titanium screws (inner screws first to ensure parallel alignment). A saw guide was attached, and a 0.66 mm Gigly wire made a single 0.7 mm mid-diaphyseal osteotomy with constant PBS irrigation. Hardware was removed as appropriate; screw heads were covered with sterile bone wax to prevent skin/muscle irritation. Muscles were repositioned and skin closed (7.0 Optilene). Non-stabilized fractures: identical exposure, mid-shaft cut with iridectomy scissors, closure without fixation. Amputation: mid-femur amputation with trimming to align bone with soft tissue, mimicking the proximal stump of osteotomy groups. Post-op care: artificial pond water with penicillin (50 U/mL) and streptomycin (20 µg/mL) for 3 days; carprofen 5 mg/L for analgesia. Mouse osteotomy: Femur stabilized with 6-hole MouseFix plate (RISystem); 0.66 mm Gigly wire created a 0.7 mm osteotomy under isoflurane anesthesia with standard perioperative analgesia/antibiotic regimen and wound care. Euthanasia at 7, 14, 28 days. Sampling/time points: Axolotl samples at 3 weeks, 3, 6, and 9 months post-surgery; mice at 1, 2, and 4 weeks. Micro-CT (axolotl): Bruker SkyScan 1172; 10.05 µm voxel; 59 kV, 141 µA; 0.5 mm Al filter; 180° orbital, 0.3° steps; reconstruction with NRecon including smoothing, ring artefact reduction, misalignment compensation, beam hardening correction; visualization in CTvox. Histology: Paraffin (axolotl, 5 µm) and cryosections (mouse, 12 µm). Movat’s Pentachrome and Safranin O/Light Green protocols used to distinguish mineralized bone, cartilage, fibrous tissue, muscle. Immunohistology: SOX9 and PCNA staining on 5 µm axolotl paraffin sections and 12 µm mouse cryosections. Blocking with 5% BSA/Tween; antigen retrieval as needed; Alexa-conjugated secondary antibodies; imaging with Pannoramic Slide Scanner and analysis in CaseViewer/Fiji. EdU labeling (axolotl growth plate characterization): 11 cm animals injected with 10 µg EdU/g every other day for 2 weeks; processed for cryosections and Click-iT EdU detection. Comparative anatomy: Movat and Safranin O staining assessed bone structure in axolotls of different sizes (6, 13, 25 cm) vs mice (12 weeks and 2 years).

- Anatomy: Older axolotl femurs have calcified diaphyses with large marrow cavities filled predominantly with adipocyte vacuoles and lack clear secondary ossification centers, resembling aged mouse marrow morphology. Axolotl epiphyses exhibit zones analogous to mammalian growth plates, supporting endochondral longitudinal growth. - Feasibility: Internal plate fixation (7.75 mm MouseFix) can be applied to axolotl femur with careful drilling and screw placement; bone wax helps protect soft tissues in aquatic animals. - Healing dynamics (axolotl): • 3 weeks: No mineralized bridging after 0.7 mm osteotomy in either fixated or non-fixated fractures; gaps filled with cells/ECM. Plate-fixated bones largely aligned (n=4/6) versus severe misalignment in non-fixated (n=7/7). Amputated limbs already formed a cartilaginous stump callus/blastema (n=2/2). • 3 months: Plate-fixated fractures showed mineralized tissue within the gap and cartilaginous soft callus with partial bridging (n=3/4). Non-fixated fractures lacked mineralized bridging, exhibited misalignment/overlap (5/5), with cartilaginous callus weakly mineralized and sealing of medullary cavity at fragment ends. • 6 months: Plate-fixated fractures achieved complete bony bridging with secondary cortex formation (n=4/4). Non-stabilized fractures formed ossifying callus bridging between misaligned segments but with thinner cortices; femoral length not preserved (n=4/4). Amputated limbs showed progressing ossification of regenerated elements. • 9 months: Plate-fixated fractures fully bridged with hard callus and marrow adipocytes (n=3/3). Non-fixated fractures finally completed bony bridging but remained shortened (n=3/3). Amputated limbs restored bone size and structure; amputation plane only evident by denser cortical network. - Comparative mouse healing: 0.7 mm plate-stabilized murine osteotomies typically bridged by 2–3 weeks with woven bone. - Cellular markers: • SOX9: Present in axolotl blastema at 3 weeks and predominant in cartilaginous elements by 3 months; in fractures, SOX9+ cells accumulate by 3 weeks and in callus at 3 months, decreasing by 6–9 months as ossification progresses. In mice, SOX9+ cells present at 1 week in callus prior to bridging; woven bone forms by 2 weeks. • Proliferation (PCNA): High at 3 weeks post-injury in axolotl amputations and fractures; markedly reduced by 3 months and later, confined mostly to soft tissues. In mice, PCNA+ cells peak around 1 week near the fracture and decline by 2 weeks. - Mechanism: Findings support endochondral ossification as the predominant mechanism of axolotl fracture healing, similar to mammals, but with substantially delayed timing in aged axolotls compared to mice. - Alignment/stability effect: Plate fixation improves alignment, reduces callus size, accelerates transition to mineralized bridging, and preserves femoral length compared to non-stabilized fractures, which require larger callus and more time to achieve bridging and fail to restore length.

A standardized, rigid internal fixation model in axolotl enables controlled, mammal-comparable studies of fracture healing and direct comparison to axolotl limb regeneration. Despite axolotls’ robust limb regeneration, their fracture healing in aged, large animals is markedly delayed relative to mice, likely influenced by slower cell cycle kinetics, aged marrow composition dominated by adipocytes, and species-specific biomechanics (aquatic ambulation and loading). Nonetheless, early accumulation of SOX9+ chondrogenic progenitors and PCNA+ proliferating cells, followed by cartilage formation and eventual ossification, mirrors mammalian endochondral repair. Mechanical stability profoundly affects outcomes: rigid fixation preserves alignment and length and expedites bridging, while non-stabilized fractures heal via larger, misalignment-compensating callus with persistent shortening—paralleling displaced fractures in mammals. The paradox that axolotls readily regenerate amputated limbs but struggle to bridge even non-critical fracture gaps within similar time frames suggests that context-dependent cues (wound epidermis, nerve signaling, progenitor source and activation) differ between blastema-mediated regeneration and fracture callus formation. Establishing this model provides a foundation to dissect where these programs overlap or diverge and to test interventions to enhance bone healing.

This study establishes a clinically relevant, internal plate–stabilized femur osteotomy model in axolotl, enabling standardized, controlled fracture healing experiments. Compared with non-stabilized fractures, plate fixation improves alignment, reduces callus size, preserves bone length, and accelerates bridging. Axolotl fracture repair proceeds via endochondral ossification with SOX9+ chondrogenic progenitors and an early proliferative phase, analogous to mammals but substantially delayed in aged axolotls. Limb amputation regenerates bone length and structure more efficiently than fracture bridging within the same time frame. Future work should identify the cellular origins of callus chondro-osteoprogenitors in axolotl fractures (periosteal versus de novo), define the molecular and biomechanical cues distinguishing blastema formation from fracture callus, manipulate signaling pathways and mechanical stability to accelerate bridging, and extend the model to quantitative outcome measures and critical-sized defects.

- Species/age: Experiments used aged, large axolotls (5–8 years, ≥20 cm), which may exhibit slower healing; results may not generalize to younger animals. - Sample size/time points: Small n at each time point (as reported) and long intervals between assessments in axolotl may miss transient events. - Biomechanics: No quantitative assessment or control of in vivo loading; aquatic vs terrestrial biomechanics differ from mammalian conditions. - Model scope: Only a single non-critical gap size (0.7 mm) evaluated; no critical-sized defects in axolotl in this study. - Outcomes: Focus on imaging/histology and marker expression; no functional or mechanical strength testing reported. - Lineage origin unresolved: Whether callus derives from pre-existing SOX9+ periosteal cells versus de novo SOX9 expression remains undetermined. - Marrow differences: Axolotl marrow adiposity and lack of secondary ossification centers complicate direct extrapolation to humans.