Zuogui Pill Modulates Osteoblast-Osteoclast Coupling in Ovariectomized Rats: A Three by Three Factorial Study

Summary

This study evaluates the effects of Zuogui Pill on postmenopausal osteoporosis in ovariectomized rats using a three-by-three factorial design. The results support synergistic effects of Yin- and Yang-components on bone metabolism and associated EphB4/ephrinB2 pathway markers, although mechanistic evidence remains associative rather than causal.

Abstract

Postmenopausal osteoporosis is driven by estrogen-deficiency-associated uncoupling of bone formation and resorption. This study quantified the independent and interactive effects of Zuogui Pill Yin-nourishing subformulations in an ovariectomized rat model and examined associations with the EphB4/ephrinB2/RhoA/alkaline phosphatase pathway. In accordance with ARRIVE 2.0, 130 specific pathogen-free female Sprague-Dawley rats were randomized to 13 groups (n=10/group), including nine groups in a three-by-three factorial design. Endpoints included P1NP, CTX-1, femoral bone mineral density measured by dual-energy X-ray absorptiometry (DXA), trabecular histology, and messenger ribonucleic acid and protein expression measured by reverse transcription quantitative polymerase chain reaction and Western blotting. The full formula (I2A2) was associated with higher P1NP, lower CTX-1, and higher BMD than the model group. Factorial analysis identified a Yin-Yang interaction between P1NP and BMD, supporting pharmacological synergy. Molecular changes were associated with increased EphB4/ephrinB2/ALP expression and reduced RhoA expression. These findings support an association between Zuogui Pill treatment, improved bone phenotypes, and modulation of pathway markers; causal pathway dependence requires inhibitor, knockdown, and co-culture validation.

Introduction

Postmenopausal osteoporosis (PMOP) affects a large population of women and is characterized by estrogen-deficiency-associated disruption of balanced bone formation and resorption^1,2. The EphB4/ephrinB2 receptor-ligand system mediates osteoblast-osteoclast communication; EphB4 signaling supports osteoblast differentiation and alkaline phosphatase (ALP) activity, whereas ephrinB2 reverse signaling can restrain osteoclastogenesis through RhoA-related mechanisms^3,4. Reduced coupling in estrogen-deficient bone provides a mechanistic rationale for evaluating interventions that affect both cellular arms of remodeling.

Current osteoporosis therapies, including antiresorptive and anabolic agents, reduce fracture risk but may be limited by adherence, cost, contraindications, rebound effects, or adverse events during long-term use⁵. Therefore, candidate treatments that improve bone formation while limiting excessive resorption remain scientifically relevant.

Zuogui Pill (ZGP; Zuo Gui Wan) is a classical multi-component prescription used in traditional Chinese medicine for kidney-essence deficiency, a theoretical pattern linked to bone weakness. The Yin-nourishing components include Rehmannia glutinosa (prepared root, Shu Di Huang, 24 g), Dioscorea opposita (rhizome, Huai Shan Yao, 12 g), Cornus officinalis (fruit, Shan Zhu Yu, 12 g), Lycium barbarum (fruit, Gou Qi Zi, 12 g), Chinemys reevesii plastron gelatin (plastron-derived gelatin, Gui Ban Jiao, 12 g), and Cyathula officinalis (root, Chuan Niu Xi, 9 g). The Yang-enhancing components include Cuscuta australis (seed, Tu Si Zi, 12 g) and Cervus elaphus antler gelatin (antler-derived gelatin, Lu Jiao Jiao, 12 g). In classical theory, the principle of seeking Yang within Yin predicts that Yang-enhancing herbs may potentiate a Yin-nourishing base rather than act as simple independent additives. Clinical evidence and preclinical studies have reported anti-osteoporotic activity of ZGP or ZGP-containing interventions, including effects on BMD, osteogenesis, and pharmacokinetic behavior^6,7,8,9,10.

The novelty of this study lies in the formal three-by-three factorial separation of Yin-nourishing and Yang-enhancing components in Ovariectomized (OVX) rats, with simultaneous assessment of bone phenotypes and markers of the EphB4/ephrinB2/RhoA/ALP pathway. This design tests formula-compatibility theory more directly than studies that use only the complete prescription or a single comparator, while recognizing that pathway-marker changes remain associative unless mechanistically blocked or silenced¹¹.

Protocol

All procedures in this study were approved by the Institutional Animal Care and Use Committee of Inner Mongolia Medical University (Approval No. YKD202404105) and were conducted in accordance with ARRIVE 2.0¹².

Study design

The study was performed at the animal experimental platform of Inner Mongolia Medical University. A three-by-three full factorial design was applied with two factors: Yin-nourishing base formulation (I; three dose levels: 0, 27, 54 g crude drug equivalent) and Yang-enhancing formulation (A; three dose levels: 0, 8, 16 g crude drug equivalent), yielding nine factorial cells plus four reference groups. The factorial design is shown in Table 1. I2A2 represents the complete ZGP formula at the clinical-equivalent dose. Dose conversion used the body surface area method (rat: human factor = 6.25)¹³.

Animals, randomization, and blinding

One hundred and thirty specific pathogen-free female Sprague-Dawley rats (6 months old, 220–250 g) were allocated by a computer-generated random number table. Allocation codes were placed in sequentially sealed envelopes by an investigator not involved in outcome assessment. Rats were housed 3–4 per ventilated cage under a 12 h light/dark cycle at 22 ± 2 ^oC and 50–60% humidity, with ad libitum water, standard rodent chow with documented calcium and phosphorus content, cage activity, and environmental enrichment. Molecular assays, histological scoring, image quantification, and densitometry were performed by investigators blinded to allocation. Sample size was determined by a priori power analysis, yielding n = 10 per group.

OVX model and drug treatment

Bilateral OVX was performed under inhaled isoflurane delivered by a precision vaporizer. Anesthesia was induced at 3–4% in oxygen and maintained at 1.5–2.5% through a nose cone. Adequate depth was confirmed by loss of pedal withdrawal and palpebral reflexes with stable respiration. OVX success was evaluated by the absence of vaginal estrous cycling at day 5 and BMD reduction at 12 weeks after surgery. After model validation, gavage treatment was administered for 8 weeks (1 mL/100 g/day). Decoctions were prepared by sequential water extraction at a rolling boil (100 °C, reflux), 10x volume for 2 h, followed by 8x volume for 1.5 h, filtered through four layers of medical gauze, and concentrated under reduced pressure to 0.968 g crude drug/mL. The concentrated extract was stored at 4 °C and used within 7 days of preparation. Positive controls were Gusukang granules (1.8 g/kg/day) and diethylstilbestrol (0.04 mg/kg/day). At termination, rats were deeply anesthetized with 5% isoflurane until loss of pedal and corneal reflexes, then exsanguinated and subjected to cervical dislocation. Death was confirmed by cessation of heartbeat and respiration. Carcasses and tissues were discarded as biological waste; chemical residues were disposed of as hazardous waste; and needles or blades were placed in approved sharps containers.

Outcome measures

Serum P1NP and CTX-1 were measured by sandwich ELISA (Jiyinmei)¹⁴. Femoral BMD was quantified by DXA. Trabecular histology was assessed by hematoxylin and eosin staining at 200x. Gene expression was assessed by RT-qPCR using the 2^-ΔΔCt method, normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Protein expression was assessed by Western blot and ImageJ quantification (see Table of Materials). Observed molecular weights and antibody catalog information are provided in Supplementary Table 1. Primer sequences were: EphB4 forward/reverse, 5'-CTAGACTCTTTCCTGCGGCT/GGGAACTTGTGTAGGTGGGA-3'; ephrinB2, 5'-CGGACAAGGCCTGGTACTAT/ACAGTTGAGCAGTGGGGTAT-3'; RhoA, 5'-GTTTATGTGCCCACGGTGTT/ACTATCAGGGCTGTCGATGG-3'; ALP, 5'-CGGACACAAGCATTCCCAGA/TGGTGGAGCTGAACAGGCTT-3'; GAPDH, 5'-ACAGCAACAGGGTGGTGGAC/TTTGAGGGTGCAGCGAACTT-3'.

Statistical analysis

Normality was assessed using the Shapiro-Wilk test, and variance homogeneity was assessed using the Levene test before parametric analysis. Data are expressed as mean ± SD, and all quantitative figure error bars denote SD. Two-way factorial ANOVA was applied to the nine factorial cells to estimate main effects and the I by A interaction, with partial eta-squared (ηp²) as the effect size. Pairwise comparisons across all 13 groups were performed using a one-way ANOVA with Bonferroni correction. Statistical analyses were performed in GraphPad Prism. Significance symbols are defined consistently as follows: *, **, ***, **** indicate p <0.05, p <0.01, p <0.001, p <0.0001 versus model; #, ##, ###, #### indicate the same thresholds versus full-formula ZGP.

Results

Model validation by vaginal cytology

OVX success was confirmed in all ovariectomized animals, as evidenced by loss of estrous cycling within 5 days after surgery (Figure 1A), whereas sham-operated animals retained a 4-day cycle (Figure 1B). At 12 weeks, DXA showed lower femoral BMD in the OVX model group than in sham controls (0.214 ± 0.020 vs. 0.281 ± 0.019 mg/cm³; Bonferroni-adjusted p ...

Discussion

The factorial interaction term supports a non-additive relationship between the Yin-nourishing and Yang-enhancing components of ZGP. This result is consistent with the traditional principle of seeking Yang within Yin, in which Yang-enhancing components are expected to activate or potentiate the Yin-nourishing base¹⁵. The finding should be interpreted as formulation-level pharmacological synergy. It does not identify the individual active constituents responsible for the effect.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
1% ammonia water	N/A	N/A	Bluing solution after hematoxylin differentiation.
1% hydrochloric acid alcohol	N/A	N/A	Differentiation solution for hematoxylin staining.
2x qPCR Mix	N/A	N/A	Quantitative PCR master mix. Source reaction volume: 12.5 uL per reaction.
4% paraformaldehyde solution	N/A	N/A	Fixation of left femur and tibia samples for histological analysis.
5x loading buffer	N/A	N/A	Sample denaturation before electrophoresis.
ALP primary antibody	Affinity Biosciences	N/A	Western blot antibody. Source dilution: 1:2000. RRID and lot number required for JoVE.
ALP primer pair	N/A	N/A	ALP mRNA was measured by qRT-PCR, but the parent primer table does not provide ALP primer sequences. Author confirmation required.
Biotin antibody reagent	Supplied with ELISA kit, Jiyinmei	Kit component	Added to sample wells during ELISA.
Cervus elaphus antler gelatin, Lu Jiao Jiao	Beijing Tongrentang	N/A	Yang-enhancing component. Complete formula dose in source: 12 g, ratio 4. Gelatin was melted into the decoction.
Chinemys reevesii plastron gelatin, Gui Ban Jiao	Beijing Tongrentang	N/A	Yin-nourishing component. Complete formula dose in source: 12 g, ratio 4. Gelatin was melted into the decoction.
Chloroform	N/A	N/A	Phase separation during RNA extraction.
Cooling stage	Leica	1150	Cryo or cooling support for sectioning workflow.
Cornus officinalis, Shan Zhu Yu	Beijing Tongrentang	N/A	Yin-nourishing component. Complete formula dose in source: 12 g, ratio 4.
Cuscuta australis, Tu Si Zi	Beijing Tongrentang	N/A	Yang-enhancing component. Complete formula dose in source: 12 g, ratio 4.
Cyathula officinalis, Chuan Niu Xi	Beijing Tongrentang	N/A	Yin-nourishing component. Complete formula dose in source: 9 g, ratio 3.
DEPC-treated PBS	N/A	N/A	Rinsing of L4 to L5 lumbar vertebrae before liquid-nitrogen freezing.
Diethylstilbestrol	N/A	N/A	Western medicine positive control. Source dose: 0.04 mg/kg/day, 1 mL/100 g/day. Parent document also contains a Chinese nomenclature inconsistency that should be harmonized as diethylstilbestrol.
Dioscorea opposita, Huai Shan Yao	Beijing Tongrentang	N/A	Yin-nourishing component. Complete formula dose in source: 12 g, ratio 4.
dNTP mixture	N/A	N/A	Included in reverse transcription reaction.
ECL chemiluminescence reagent	N/A	N/A	Western blot detection reagent.
ELISA chromogenic reagent A	Supplied with ELISA kit, Jiyinmei	Kit component	Color development reagent for ELISA.
ELISA chromogenic reagent B	Supplied with ELISA kit, Jiyinmei	Kit component	Color development reagent for ELISA.
ELISA stop solution	Supplied with ELISA kit, Jiyinmei	Kit component	Used to terminate ELISA color reaction.
Enzyme-labeled ELISA reagent	Supplied with ELISA kit, Jiyinmei	Kit component	Added to non-blank wells in ELISA.
EphB4 primary antibody	Proteintech	N/A	Western blot antibody. Source dilution: 1:2000. RRID and lot number required for JoVE.
EphB4 primer pair	N/A	PCR product 232 bp	Forward CTAGACTCTTTCCTGCGGCT; Reverse GGGAACTTGTGTAGGTGGGA. Parent table typo Rat RphB4 should read Rat EphB4.
ephrinB2 primary antibody	Affinity Biosciences	N/A	Western blot antibody. Source dilution: 1:2000. RRID and lot number required for JoVE.
ephrinB2 primer pair	N/A	PCR product 172 bp	Forward CGGACAAGGCCTGGTACTAT; Reverse ACAGTTGAGCAGTGGGGTAT.
Ethanol, absolute and graded 95%, 85%, 75%	N/A	N/A	Gradient dehydration and rehydration during hematoxylin and eosin staining.
Female Sprague-Dawley rats	Beijing Vital River Laboratory Animal Technology Co., Ltd.	Animal certificate and license number not supplied in source article	Specific pathogen-free, 6 months old, healthy non-pregnant females. Chinese parent states 230 to 270 g; earlier English draft states 220 to 250 g. Verify one final range before submission.
GAPDH primary antibody	Servicebio	N/A	Loading-control antibody. Source dilution: 1:2000. RRID and lot number required for JoVE.
GAPDH primer pair	N/A	PCR product 253 bp	Forward ACAGCAACAGGGTGGTGGAC; Reverse TTTGAGGGTGCAGCGAACTT.
GraphPad Prism software	GraphPad Software	Version 9.4.1; RRID:SCR_002798	Statistical analysis and graphing. Normality, variance homogeneity, ANOVA, Bonferroni correction, and effect sizes should be retained in output files.
Gusukang granules	Liaoning Wohua Kangchen Pharmaceutical Co., Ltd.	N/A	Traditional Chinese medicine positive control. Source dose: 1.8 g/kg/day or 0.5 mL/100 g/day.
Hematoxylin staining solution	Zhuhai Beisuo Biotechnology Co., Ltd.	N/A	Histological staining. Source procedure used 5 min hematoxylin staining.
High-speed microcentrifuge	N/A	TG16W	Centrifugation for serum and nucleic acid procedures.
Hologic dual-energy X-ray absorptiometer	Hologic	N/A	Bone mineral density analysis. Source reports in vivo and ex vivo femoral scanning.
ImageJ software	National Institutes of Health	RRID:SCR_003070	Densitometry quantification for Western blot bands. Version not supplied in source article.
Induction cooker	Midea	N/A	Listed in Chinese parent instrument table.
Isoflurane	N/A	N/A	Anesthesia for bilateral ovariectomy and DXA positioning. Revised protocol should specify induction, maintenance, route, and anesthetic-depth criteria.
Isopropanol	N/A	N/A	RNA precipitation during extraction.
Liquid nitrogen	N/A	N/A	Snap-freezing of lumbar vertebrae samples.
Lycium barbarum, Gou Qi Zi	Beijing Tongrentang	N/A	Yin-nourishing component. Complete formula dose in source: 12 g, ratio 4.
Microplate reader	N/A	AMR-100	ELISA absorbance detection at 450 nm.
Microplate washer	N/A	APW-200	Washing step for ELISA plates.
MMLV reverse transcriptase	N/A	N/A	Reverse transcription enzyme.
Neutral mounting gum	Sinopharm Group	N/A	Mounting medium for hematoxylin and eosin stained slides.
Normal saline	N/A	N/A	Vehicle gavage for blank control, sham-operated, and model groups.
Oligo(dT)18 primer	N/A	N/A	Used for reverse transcription.
Penicillin	N/A	N/A	Post-operative infection prophylaxis. Source dose: 20,000 U/kg intramuscularly for 3 days.
Pressure cooker	Xinfei	N/A	Listed in Chinese parent instrument table.
Protease inhibitor	N/A	N/A	Added during protein extraction.
Protein lysis buffer	N/A	N/A	Bone tissue protein extraction for Western blotting.
PVDF membrane	N/A	N/A	Western blot transfer membrane activated by methanol.
qRT-PCR instrument	N/A	N/A	Quantitative reverse transcription PCR for EphB4, ephrinB2, RhoA, ALP, and GAPDH.
Rat CTX-1 ELISA kit	Jiyinmei	JYM0026Ra	Sandwich ELISA for serum C-terminal telopeptide of type 1 collagen. Detection at 450 nm.
Rat P1NP ELISA kit	Jiyinmei	JYM0030Ra	Sandwich ELISA for serum procollagen type 1 N-terminal propeptide. Detection at 450 nm.
Rehmannia glutinosa, Shu Di Huang	Beijing Tongrentang	N/A	Yin-nourishing component. Complete formula dose in source: 24 g, ratio 8.
RhoA primary antibody	Proteintech	N/A	Western blot antibody. Source dilution: 1:2000. RRID and lot number required for JoVE.
RhoA primer pair	N/A	PCR product 177 bp	Forward GTTTATGTGCCCACGGTGTT; Reverse ACTATCAGGGCTGTCGATGG. Terminology harmonized from RHoA to RhoA.
RNase inhibitor	N/A	N/A	Included in reverse transcription reaction.
RNase-free water	N/A	N/A	RNA dissolution after drying.
Rotary microtome	Leica	2245	Histological sectioning.
Secondary antibody for Western blotting	N/A	N/A	Source dilution: 1:15000. Host reactivity and RRID required for JoVE.
Skim milk powder, 5%	N/A	N/A	Blocking reagent for Western blotting, room temperature for 2 h.
Slide spreading and baking machine	Jinhua Yidi	N/A	Slide preparation for histology.
Sodium pentobarbital, 3%	N/A	N/A	Terminal anesthesia before abdominal aorta blood collection in the Chinese parent document. Final euthanasia procedure and confirmation method must be verified.
Standard sample diluent	Supplied with ELISA kit, Jiyinmei	Kit component	Used for serial standard dilution in the ELISA procedure.
TBST buffer	N/A	N/A	Wash buffer for Western blotting.
Thermostatic shaking incubator	Shanghai Yiheng Scientific Instrument Co., Ltd.	N/A	Laboratory incubation and shaking.
TRIzol reagent	N/A	N/A	RNA extraction from 100 mg bone tissue.
Upright microscope	Leica	DM1000	Hematoxylin and eosin histology imaging at 200x.
Water-jacket thermostatic incubator	N/A	GNP-9080	Incubation for assay procedures.
Water-soluble eosin staining solution	Zhuhai Beisuo Biotechnology Co., Ltd.	N/A	Histological counterstain. Source procedure used 3 min eosin staining.
Western blot electrophoresis system	N/A	N/A	Protein separation. Source electrophoresis: 80 V initially, then 120 V.
Western blot imaging system	N/A	N/A	Chemiluminescent membrane imaging. Uncropped membranes required for submission.
Western blot transfer system	N/A	N/A	Protein transfer to PVDF membrane. Source transfer condition: 400 mA, 90 min, ice bath.
Xylene	N/A	N/A	Dewaxing and tissue clearing during histology.
Zuogui Pill component herbs	Beijing Tongrentang	N/A	Eight-herb formulation: Rehmannia glutinosa, Dioscorea opposita, Cornus officinalis, Lycium barbarum, Chinemys reevesii plastron gelatin, Cyathula officinalis, Cuscuta australis, and Cervus elaphus antler gelatin.