Immuno-histologic and histomorphometric evaluation of *Angelica sinensis* adjunctive to β-tricalcium phosphate in critical-sized class II furcation defects in dogs

Periodontal disease leads to destruction of tooth-supporting tissues including alveolar bone, periodontal ligament (PDL), and cementum. Furcation involvement poses a clinical challenge despite maintenance possibilities. True periodontal regeneration is best verified histologically. Regenerative approaches often combine bone grafts and barrier membranes, with various graft materials (autografts, allografts, xenografts, and synthetics such as β-TCP) and resorbable membranes used to guide tissue regeneration. Traditional Chinese medicine, including Angelica sinensis (Dong quai), exhibits anti-inflammatory, immunomodulatory, antioxidative, osteogenic, and angiogenic properties, with evidence of promoting osteoblast proliferation/differentiation and PDL healing in preclinical models. This study investigated whether adding Angelica sinensis to β-TCP would improve regeneration in class II furcation defects in dogs. The null hypothesis stated no significant difference in regeneration with or without As.

Background literature highlights that bone grafts and membranes facilitate periodontal regeneration, with β-TCP widely used and studied in furcation defects. Resorbable membranes provide epithelial exclusion and support bone/PDL repopulation. Angelica sinensis has long-standing medical use and exhibits multiple bioactivities relevant to regeneration: osteogenic (e.g., via VEGF and Wnt/β-catenin signaling), angiogenic, and anti-inflammatory effects. Ferulic acid from As enhances osteoblast proliferation/differentiation in vitro. Topical herbal formulations containing As mitigate periodontal breakdown by reducing collagen degradation and osteoclastic activity. Prior canine and rodent studies of other botanicals and As-derived preparations reported enhanced periodontal healing, PDL fibroblast proliferation, and increased hyaluronic acid production, supporting the rationale for testing As in periodontal regeneration.

Design and sample: Randomized preclinical study on six healthy male mongrel dogs (18–24 months; 10–18 kg), approved by the Alexandria University Animal Ethics Committee (IRBNO:00010556-IORG0008839) and compliant with ARRIVE 2.0 and NIH guidelines. A total of 24 critical-sized class II furcation defects (buccal surfaces of mandibular P3 and P4 bilaterally) were surgically created (3 mm vertical × 4 mm horizontal). Random allocation assigned 12 defects to the experimental group and 12 to control. Materials: Experimental: Angelica sinensis concentrated granules (active ingredient: ferulic acid ≥0.2 mg/g; 3–4 μm granules; 90% extract + 10% maltodextrin) mixed 1:1 by volume with β-TCP (0.1–0.5 mm granules; porosity 80%; pore size 400 μm). Control: β-TCP alone. Both groups covered with a resorbable collagen membrane (Hypo Sorb). Surgical procedure: Under general anesthesia (thiopental 13 mg/kg), sulcular incisions were made and full-thickness mucoperiosteal flaps reflected. Standardized furcation defects were created with a round carbide bur using a stopper and periodontal probe to verify dimensions. Reference notches were placed on mesial and distal root surfaces at the defect base. Root planing and EDTA root conditioning were performed. Experimental defects were filled with As+β-TCP; control with β-TCP only. Collagen membranes were trimmed to extend 2–3 mm beyond defects and adapted without fixation sutures; flaps were coronally repositioned and closed with 3-0 silk interrupted sutures. Postoperative care: Ampicillin and ibuprofen administered; soft diet; plaque control with regular toothbrushing; sutures removed at 10 days. Three dogs were euthanized at 4 weeks and three at 8 weeks using thiopental overdose. Histology and immunohistochemistry: Mandibles were dissected; specimens fixed in 10% neutral buffered formalin, decalcified in 5% trichloroacetic acid, sectioned mesiodistally (5 μm), stained with H&E and Gomori trichrome. Alkaline phosphatase (ALP) immunoreactivity in osteoblasts was assessed semi-quantitatively (intense, strong, moderate, weak) based on cytoplasmic color change. Histomorphometry: Using ImageJ v1.46r, three parameters were quantified from standardized photomicrographs: (1) height of newly formed interradicular bone (mm) from most coronal bone to a line between apical root notches (×40), (2) percentage of newly formed bone surface area (×100), and (3) trabecular thickness (mm; ×100). From each defect, two sections were analyzed (total 48 sections). One calibrated investigator performed duplicate measurements 2 days apart (intra-examiner ICC=0.86; 95% CI 0.81–0.92). Statistics: Normality assessed by Shapiro–Wilk and Q–Q plots. Between-group comparisons used two-tailed unpaired t-tests with α=0.05 (SPSS v28).

Clinical: No postoperative adverse reactions; no infections. Histology: At 4 weeks, experimental defects showed substantial regenerating bone with angiogenesis in PDL, prominent cementoblasts and osteoblasts, and fiber insertion into forming bone; control showed less bone, thin irregular trabeculae, and less organized PDL fibers. At 8 weeks, experimental defects exhibited dense bone nearly filling defects, thicker intercommunicating trabeculae, abundant vessels, remodeling adjacent to Angelica patches, organized cementoblasts/osteoblasts, and thick PDL bundles inserting deeply; control showed moderate bone fill with variable trabeculae, vessels and β-TCP present, PDL fibers less bundled than experimental. Histomorphometry (mean ± SD):

• Height of newly formed interradicular bone: 4 weeks: control 2.51 ± 0.15 mm vs experimental 3.03 ± 0.09 mm (p=0.001). 8 weeks: control 2.95 ± 0.25 mm vs experimental 3.57 ± 0.11 mm (p=0.0001).
• Percentage bone surface area: 4 weeks: control 44.63 ± 7.94% vs experimental 58.88 ± 13.92% (p=0.002). 8 weeks: control 64.70 ± 7.75% vs experimental 78.35 ± 5.76% (p=0.02).
• Trabecular thickness: 4 weeks: control 0.23 ± 0.05 mm vs experimental 0.468 ± 0.08 mm (p=0.0001). 8 weeks: control 0.37 ± 0.05 mm vs experimental 0.57 ± 0.11 mm (p=0.0001). Immunohistochemistry: ALP in osteoblasts—4 weeks: experimental intense vs control moderate; 8 weeks: experimental strong vs control moderate.

Adding Angelica sinensis to β-TCP significantly enhanced bone regeneration and periodontal healing versus β-TCP alone. Improvements in bone height, surface area, and trabecular thickness, along with stronger ALP immunoreactivity, indicate osteoinductive effects and active matrix mineralization. Histologically, As promoted organized cementum and PDL fiber formation and robust angiogenesis, aligning with prior evidence of As’s osteogenic, angiogenic, and anti-inflammatory properties. Compared with other regenerative modalities in canine furcations, the percentage of new bone area with As+β-TCP at 8 weeks (78.35%) was comparable to published outcomes for autogenous bone or MSCs in similar timeframes, underscoring the material’s promise. The findings reject the null hypothesis and support As as a biocompatible adjunct to enhance periodontal regeneration in furcation defects.

Management of class II furcation defects with Angelica sinensis plus β-TCP resulted in superior bone formation—greater rate, quality, and quantity of new bone—and enhanced periodontal regeneration compared with β-TCP alone. Immunohistologic and histomorphometric analyses confirm the high regenerative potential of As. Future work should quantify angiogenesis (e.g., vessel area, VEGF/FGF expression), assess cementum thickness, evaluate alternative alloplasts (e.g., brushite or monetite), and extend observation periods and translational studies.

Preclinical canine model limits direct generalization to humans; the authors note no human studies to date. Short follow-up (4 and 8 weeks) captures early to intermediate healing only. Modality comparisons were limited to β-TCP with/without Angelica sinensis; no additional control arms (e.g., membrane only) were included. Outcome measures were histologic/immunohistochemical and morphometric without clinical periodontal parameters. Sample comprised 24 defects in 6 dogs. Single-blind design (histology assessor blinded) may leave potential for other biases. Quantitative angiogenic markers (e.g., VEGF/FGF) and cementum thickness were not measured.