Multibatch Fingerprint Integrated with Multi-Indicator Efficacy for Quality Evaluation of Sanhua Decoction and Investigation of Its Mechanistic Basis against Cerebral Ischemia-Reperfusion Injury

Abstract

Building upon the spectrum-effect relationship theory in traditional Chinese medicine, ultraperformance liquid chromatography (UPLC) was utilized to generate fingerprint profiles for 12 batches of Sanhua Decoction (SHD). These profiles were analyzed in relation to pharmacological efficacy, providing a novel strategy to elucidate the material basis of SHD against cerebral ischemia-reperfusion injury (CIRI) and for efficacy evaluation. The UPLC fingerprints identified 33 common peaks, with 28 chemical constituents characterized. Pharmacodynamic assessment using a rat CIRI model demonstrated that SHD significantly improved neurological deficit scores, reduced cerebral infarct volume, and enhanced brain histopathology, with effects correlated to key indicators such as vascular endothelial growth factor. Gray relational analysis showed that 32 of the 33 peaks had a correlation degree greater than 0.706 with efficacy, confirming the synergistic effect of multiple components in SHD’s anti-CIRI action. Orthogonal partial least-squares discriminant analysis identified 11 potential major active compounds significantly associated with anti-CIRI activity. An “efficacy-integrated fingerprint” was constructed by integrating chemical fingerprinting with pharmacodynamic indices, reflecting the intrinsic quality of SHD. This methodology offers a scientific and efficient approach for chemical constituent investigation and pharmacodynamic material basis elucidation of SHD and other traditional Chinese medicines.

1. Introduction

Stroke is a major contributor to global mortality and long-term disability, ranking as the second leading cause of death and the primary cause of disability worldwide. − It results from cerebrovascular rupture or occlusion, causing an abrupt interruption of cerebral blood flow, which leads to physical impairments and multisystem dysfunctions. , Stroke is classified into three subtypes: transient ischemic attack (TIA), hemorrhagic stroke, and ischemic stroke. Among these, ischemic stroke accounts for 80–90% of cases and is primarily caused by thrombosis or embolism-induced blood flow interruption. It is characterized by a high incidence, significant disability burden, elevated mortality, and frequent recurrence. − The severity of ischemic stroke poses a substantial threat to human health, drastically impeding quality of life and burdening patients and families. In Traditional Chinese Medicine (TCM), it is categorized under “apoplexy” (Stroke, Zhongfeng), with pathological mechanisms attributed to qi deficiency and blood stasis. Epidemiological data reveal that the incidence of ischemic stroke in China increased from 129 per 100,000 in 2010 to 145 per 100,000 in 2019. Timely restoration of blood perfusion to the ischemic penumbra remains the most effective therapeutic strategy; however, reperfusion can paradoxically aggravate neuronal death, resulting in secondary brain injury known as cerebral ischemia-reperfusion injury (CIRI). − The development of pharmacological agents capable of effectively mitigating CIRI is thus critical for improving clinical outcomes. Currently available treatments are limited by poor target specificity, narrow mechanisms of action, and adverse gastrointestinal effects. Therefore, identifying novel agents with multimodal neuroprotective properties continues to be a promising direction for preventing and treating CIRI.

(SHD) is listed as the 55th formula in the Catalogue of Ancient Classical Prescriptions (First Batch) promulgated by the Chinese government. Comprising Rheum palmatum (Dahuang), Citrus aurantium (Zhishi), Magnolia officinalis (Houpo), and Notopterygium incisum (Qianghuo), SHD is a traditional herbal formulation widely used in the treatment of stroke. Its therapeutic effects are achieved by eliminating turbidity from the fu-organs, restoring the ascending clarity and descending turbidity, and regulating qi flow to promote physiological harmony. These mechanisms are consistent with the TCM principle of addressing stroke through visceral regulation and metabolic detoxification. Modern investigations have demonstrated that SHD exhibits protective effects on the blood–brain barrier and neurological functions in cerebral ischemia-reperfusion injury, potentially via anti-inflammatory, antioxidative, and endothelial-stabilizing pathways. − Nonetheless, the molecular basis underlying its efficacy against cerebral ischemia-reperfusion injury (CIRI) remains to be fully elucidated, with notable gaps including: (i) unclear relationships between bioactive constituents and therapeutic effects; and (ii) lack of direct quantitative analysis of drug-derived compounds in critical brain regions such as cerebrospinal fluid and parenchyma. Most chemical profiling studies on SHD have focused on its individual herbal components rather than the entire formula. Zhao et al.employed UPLC-FT-ICR MS to tentatively identify 137 compounds in SHD. Luo et al.used UPLC-Q-Orbitrap-MS to analyze Xiao Chengqi Decoctiona formulation structurally similar to SHDidentifying 123 constituents primarily from Zhishi and Dahuang, mainly flavonoids and anthraquinones, but with limited chemical coverage. Liu et al.,using UPLC-Q-Exactive Orbitrap-MS, identified 166 compounds in SHD, including flavonoids, anthraquinones, coumarins, lignans, and alkaloids, with 30 confirmed by reference standards. Wang et al. characterized its chemical profile via UPLC-Q-TOF-MS/MS, focusing on five bioactive anthraquinone aglycones to elucidate the role of Rhei Radix et Rhizoma. However, none of these studies conducted comprehensive pharmacodynamic validation or systemic mechanistic analyses of multicomponent interactions.

Fingerprint profiling effectively characterizes the overall chemical composition, ensures batch consistency, and identifies regional markers of bioactive constituents in traditional Chinese medicine, forming an established approach for authentication and quality control. , When combined with chemometric analysis, this technique enables efficient discrimination of quality variations among different production regions. However, while chromatographic fingerprints confirm chemical uniformity, they lack direct correlation with clinical efficacy. Therefore, establishing spectrum-effect relationships is essential to elucidate the link between chemical profiles and therapeutic outcomes. Spectrum-effect analysis integrates chromatographic fingerprinting with pharmacodynamic data, using linear or nonlinear models to quantitatively define “spectrum-efficacy” relationships and identify active constituents associated with efficacy. As an innovative paradigm beyond conventional fingerprinting, it provides a new approach to elucidate the pharmacodynamic material basis of Chinese medicine formulations. In this study, a spectrum-efficacy model was constructed for Sanhua Decoction (SHD) to clarify its material basis in combating cerebral ischemia-reperfusion injury (CIRI). An efficacy-integrated fingerprint reflecting the intrinsic quality of SHD was developed. This systematic approach aims to (i) elucidate the pharmacodynamic basis underlying SHD’s anti-CIRI effects; (ii) establish a rigorous quality evaluation and efficacy verification framework. Collectively, our methodology offers novel insights, and new analytical strategies, providing a solid scientific foundation for the clinical application of this ancient classical formula.

2. Experimental Section

2.1. Experimental Drugs

All herbal medicines were authenticated by Professor Li Wei from the School of Pharmacy, Guizhou University of Traditional Chinese Medicine. Twelve batches of Sanhua Decoction samples (S1–S12) were obtained by randomly combining these four types of herbal pieces using the random number table method, as shown in Table .

1. Information on 12 Batches of Sanhua Decoction.

	Dahuang		Zhishi		Houpo		Qianghuo
no.	origin	batch number	origin	batch number	origin	batch number	origin	batch number
S1	Sichuan	2104030	Chongqing	210625	Sichuan	210301	Sichuan	210325
S2	Gansu	220901	Jiangxi	220330	Sichuan	220601	Sichuan	220801
S3	Gansu	210701	Jiangxi	220901	Sichuan	221201	Sichuan	221201
S4	Gansu	220101	Sichuan	220801	Sichuan	220101	Sichuan	220301
S5	Sichuan	220617	Hunan	221011	Sichuan	220615	Sichuan	220829
S6	Sichuan	221210	Jiangxi	220401	Sichuan	210801	Sichuan	220829
S7	Sichuan	22040513	Hunan	22120194	Sichuan	220802981	Sichuan	221203491
S8	Sichuan	202212002	Sichuan	20230401	Sichuan	230101	Sichuan	202302001
S9	Sichuan	202204001	Chongqi	221001	Sichuan	220401	Sichuan	221001
S10	Sichuan	C141230101	Sichuan	C648211101	Sichuan	Y220501	Sichuan	Y220801
S11	Gansu	221201	Sichuan	221201	Sichuan	230101	Sichuan	230102
S12	Gansu	220826	Jiangxi	220901	Sichuan	221201	Sichuan	221201

2.2. Instruments

Ultrahigh performance liquid chromatography (Agilent 1290; Agilent Technologies, USA) equipped with ZORBAX Eclipse Plus C18 column (4.6 × 250 mm, 5 μm), electronic analytical balance (XB220A; PRECISA, Switzerland), CNC ultrasonic cleaner (KQ-500DE, 40 kHz/500 W; Kunshan Ultrasonic Instrument Co., Ltd., China), microplate reader (Rayto RT-6100; Rayto Life Sciences, China), electric blast drying oven (101-3AB; Tianjin Qinste Instrument Co., Ltd., China), micro high-speed centrifuge (TG16W, max 15,000 × g; Changsha Xiangzhi Centrifuge Instrument Co., Ltd., China), benchtop refrigerated high-speed centrifuge (TGL16M, max 16,000 × g; Shanghai Luxiangyi Centrifuge Instrument Co., Ltd., China), constant temperature shaker (HT-111B; Shanghai Hetian Scientific Instruments Co., Ltd., China), tissue embedding machine (JB-P5; Wuhan Junjie Electronics Co., Ltd., China), pathology microtome (RM2016; Leica Microsystems, China), upright optical microscope (NIKON ECLIPSE E100; Nikon Corporation, Japan).

2.3. Reagents

Chromatographic-grade solvents were employed, including methanol (batch: 20200911; Sinopharm Chemical Reagent Co., Ltd.), acetonitrile (batch: 106245), and formic acid (batch: A17-50) from Fisher Scientific. Eighteen reference compounds comprised magnolins A (batch: 230502) and B (batch: 230415) from Chengdu Herbpurify Biotechnology Co., Ltd., anthraquinone glycosides (batches: 230319–230425; Chengdu Herbpurify), and flavonoid standards (batches: PS00024–PS011911; Chengdu Pusi Biotechnology Co., Ltd.). Key bioactive constituentsbaicalin, emodin derivatives, magnolol, and honokiolwere procured from Beijing Solarbio Science & Technology Co., Ltd. (batches: 1101P022–1209E023). Three ELISA kits for rat angiogenesis markers included VEGF (cat. no. AF3055-A), ANG-1 (cat. no. AF3045-A), and ANG-2 (cat. no. AF3055-B) supplied by Jiangsu Jingmei Shenwu Biotechnology Co., Ltd. (batch: AF20230731).

2.4. Animals

Specific Male specific pathogen-free (SPF) Sprague-Dawley rats (8 weeks old, body weight 200 ± 20 g) were procured from Tianqin Biotechnology Co., Ltd. (Changsha, China; Animal License No. SCXK­(Xiang) 2022-0011). All experimental procedures were performed in accordance with the protocol approved by the Institutional Animal Ethics Committee (Approval No. 2023116). Animals were maintained under controlled conditions: 12 h light/dark cycle, temperature 25.0 ± 2.0 °C, relative humidity 45.0 ± 15.0%, with ad libitum access to standard laboratory chow and autoclaved water.

2.5. Methods

2.5.1. Establishment of Fingerprint

2.5.1.1. Preparation of Reference Sample

The reference sample of Sanhua Decoction was prepared in accordance with the extraction procedures documented in pharmacopeial literature and Liu Wansu’s Suwen Bingji Qiyi Baoming Ji: Zhongfeng Lun Di Shi (Chapter 10: Essential Mechanisms of Disease and Preservation of Life: On Wind Stroke). Specifically, 30.5 g each of Rhei Radix et Rhizoma (Dahuang), Magnoliae Officinalis Cortex (Houpo), Notopterygii Rhizoma et Radix (Qianghuo), and Aurantii Fructus Immaturus (Zhishi) were decocted with 2,831.4 mL of water. The mixture was soaked for 30 min, then brought to a vigorous boil (high heat), followed by gentle simmering (low heat) until the volume was reduced by 50%. The decoction was filtered while hot and lyophilized using a freeze-dryer to obtain the reference material.

2.5.1.2. Preparation of Test and Reference Solutions

Test Sample Solution: accurately weigh 0.15 g of lyophilized Sanhua Decoction powder and transfer it into a 25 mL stoppered conical flask. Add 15 mL of 50% methanol, sonicate for 15 min, then cool to room temperature, vortex thoroughly, and filter. The filtrate is passed through a 0.22 μm microporous membrane to obtain the final sample solution.

Reference Standard Solution: precisely weigh each reference standard (including Synephrine, Gallic acid, Chlorogenic acid, Syringin, Magnolin B, Magnoflorine, Daidzin, Magnolin A, Ferulic acid, Aloe-emodin-8-O-β-glucoside, Rhein-8-O-β-glucoside, Nodakenin, Rhaponticin, Quercitrin, Rhoifolin, Sennoside A, Hesperidin, Neohesperidin, Emodin-1-O-glucoside, Chrysophanol-8-O-β-glucoside, Aloe-emodin, Rhein, Nobiletin, Tangeretin, Honokiol, Magnolol, Chrysophanol, and Physcion) and dissolve in methanol to prepare a mixed reference standard solution.

2.5.1.3. Chromatographic Conditions

Separation was performed on a Waters ACQUITY UPLC BEH C18 column (100 mm × 2.1 mm, 1.7 μm) with mobile phase A (aqueous 0.1% formic acid) and C (acetonitrile) under gradient elution: 0–3 min (5 → 12% C), 3–25 min (12 → 15% C), 25–32 min (15 → 18% C), 32–35 min (18 → 21% C), 35–40 min (21 → 30% C), 40–45 min (30 → 45% C), 45–52 min (45 → 75% C), 52–55 min (75 → 95% C). The flow rate was 0.2 mL/min with the column temperature maintained at 35 °C. Detection wavelengths were programmed as 0–17 min at 254 nm, 17–38 min at 310 nm, and 38–55 min at 254 nm. Injection volume: 5 μL.

2.5.1.4. Precision Test

Test solutions were prepared from the lyophilized powder of the same Sanhua Decoction batch (S1) in accordance with the procedure described in Section 1.5.1.3. Six consecutive injections were conducted under the chromatographic conditions outlined in the same section.

2.5.1.5. Repeatability Test

Six parallel test solutions were prepared from the same batch of Sanhua Decoction lyophilized powder (S1) in accordance with the procedure outlined in Section 1.5.1.3. Each solution was subsequently injected under the chromatographic conditions specified in the same section.

2.5.1.6. Stability Test

A test solution was prepared from Sanhua Decoction lyophilized powder (Batch S1) in accordance with Section 1.5.1.3 and injected at 0, 2, 4, 8, 12, and 24 h under identical chromatographic conditions to assess stability at room temperature.

2.5.1.7. Establishment of Fingerprint

Twelve batches of Sanhua Decoction lyophilized powders were processed into test solutions according to Section 1.5.1.2 and analyzed under the chromatographic conditions specified in Section 1.5.1.3. The UPLC fingerprint data were recorded in CDF format and imported into the Traditional Chinese Medicine Chromatographic Fingerprint Similarity Evaluation System (2012 Edition). Using Batch S1 as the reference chromatogram, peak alignment was performed with the following parameters: mean vector method, a time window width of 0.1 min, and multipoint calibration. This procedure generated 12 batch-specific characteristic chromatograms and a consolidated reference fingerprint for Sanhua Decoction.

2.5.2. Pharmacodynamic Study of Sanhua Decoction on Cerebral Ischemia-Reperfusion Injury in Rats

2.5.2.1. Animal Grouping and Administration

Following a 1 week acclimatization period, SD rats were randomly assigned to four groups (n = 10 per group):

Sham group: no ischemia induction;

Model group: cerebral ischemia-reperfusion injury (CIRI) without treatment;

Positive control group: nimodipine (10.8 mg/kg, oral gavage);

Sanhua Decoction groups: 12 batches (S1–S12, 10.98 g/kg, oral gavage; dose converted from clinical equivalence).

Treatment began 1 week prior to modeling. The positive control and Sanhua Decoction groups received daily administration, whereas the sham and model groups were given equal volumes of saline. The final dose was administered 24 h after modeling, and tissue collection was conducted 2 h postdose.

2.5.2.2. Establishment of Cerebral Ischemia-Reperfusion Injury Model

Following 12 h of fasting (with water provided ad libitum), rats were anesthetized with 2% pentobarbital sodium (40 mg/kg, i.p.) and secured in a supine position. The middle cerebral artery occlusion (MCAO) model was induced using the intraluminal suture technique:

Surgical exposure: a midline neck incision was made to expose the right common carotid artery (CCA), external carotid artery (ECA), and internal carotid artery (ICA).

Vascular occlusion: the proximal CCA and ECA were ligated, and a microvascular clip was used to temporarily occlude the ICA.

Suture insertion: a nylon monofilament (0.26 mm in diameter) coated with silicone (0.36 ± 0.02 mm tip diameter) was inserted through an arteriotomy in the CCA into the ICA until resistance was encountered (≈18–22 mm insertion depth), with final positioning confirmed when the suture’s black marker reached the carotid bifurcation.

Reperfusion: after 2 h of ischemia, the filament was withdrawn by 2 cm to restore blood flow.

2.5.2.3. Neurological Assessment

Postawakening, rats were subjected to neurological evaluation using the Zea Longa scoring system (0 = no deficit; 4 = severe impairment) as detailed in Table . Rats scoring 1–4 points were included for subsequent analysis, with higher scores indicating more severe neurological deficits.

2. Neurological Function Scoring Criteria for Rats with Cerebral Ischemia-Reperfusion Injury.

score	behavioral manifestation
0	normal neurological function; no deficits
1	mild neurological deficit (inability to extend the contralateral forepaw)
2	moderate neurological deficit (circling toward the paretic side when walking)
3	moderate neurological deficit (tilting toward the paretic side)
4	inability to walk spontaneously; loss of consciousness
5	death

2.5.2.4. Determination of Serum Ang1, Ang2, and VEGF Levels

Following anesthesia, blood was collected via abdominal aorta puncture and centrifuged at 4 °C (4,000 rpm, 10 min). Serum aliquots were analyzed using ELISA kits (AF3045-A for Ang1, AF3055-A for Ang2/VEGF; Jiangsu Jingmei Shenwu Technology Co., Ltd.) strictly following the manufacturer’s protocols. Angiogenic factor concentrations (angiopoietin-1, angiopoietin-2, and vascular endothelial growth factor) were quantified spectrophotometrically at 450 nm.

2.5.2.5. TTC Staining

Following anesthesia, rats were euthanized, and the whole brain (excluding the cerebellum, olfactory bulb, and lower brainstem) was excised. The tissues were rinsed in ice-cold PBS (0–4 °C) and frozen at −20 °C) for 20 min. Coronal sections (2 mm thickness) were prepared using a rodent brain matrix. The slices were immersed in a 2% TTC solution (37 °C), in the dark) for 30 min with gentle agitation every 5 min. Stained sections were fixed overnight in 4% paraformaldehyde, air-dried, and photographed for infarction area quantification. White, unstained regions indicated ischemic lesions. The percentage of cerebral infarction area was calculated using ImageJ 1.41 software.

2.5.2.6. Observation of Pathological Sections

The brain of the rat was excised, and surface bloodstains were rinsed off with normal saline. After fixation in 4% paraformaldehyde for 24 h, the tissue underwent dehydration and clearing. The specific steps were as follows: the tissue was removed from the fixative and soaked in pure water for 2 h, with frequent water changes (approximately 10 times). Gradient alcohol dehydration was then performed: 60, 70, 80, 90, and 100% ethanol (I and II) for 1 h each. The tissue was then immersed in a xylene:methanol (1:1) mixture for 30 min, followed by pure xylene for 45 min and paraffin for 1 h. After cooling, the tissue was embedded in paraffin and sectioned. The paraffin sections were successively immersed in xylene I and xylene II for 7 min each, followed by anhydrous ethanol, 95% ethanol, 85% ethanol, and 75% ethanol for 3 min each, and rinsed with distilled water for 2 min. The dewaxed sections were stained with hematoxylin solution for 5 min and rinsed twice with tap water for 2 min each to blueify. They were then stained with 1% water-soluble eosin solution for 2 min and rinsed with distilled water for 2–3 s. The sections were successively dehydrated in 75, 85, and 95% ethanol and anhydrous ethanol I for 2–3 s each, followed by anhydrous ethanol II for 1 min and xylene I and xylene II for 1 min each. After air-drying, the sections were mounted with neutral gum and photographed under an optical microscope for the analysis of pathological changes in each group of rats.

2.5.3. Statistical Analysis

Data were analyzed using SPSS 26.0 statistical software. Measurement data were expressed as mean ± standard deviation (SD). One-way analysis of variance (ANOVA) was used for comparisons among multiple groups. The least significant difference (LSD) method was applied for pairwise comparisons when homogeneity of variance was met. A value of P < 0.05 was considered statistically significant. All analyses were conducted using SPSS software (version 26.0; IBM Corp., Armonk, NY, USA).

3. Results

3.1. Precision Test Results

Using peak 12 as the reference peak, the relative retention times (RRT) and relative peak areas (RPA) of all characteristic peaks were calculated. As shown in Figure , The results demonstrated that both RRT and RPA of common peaks exhibited RSD values below 3% (n = 6), confirming excellent instrumental precision.

1.

Precision test results.

3.2. Repeatability Test Results

Using Peak 12 as the reference, relative retention times (RRT) and relative peak areas (RPA) were calculated for all characteristic peaks. As shown in Figure , both RRT and RPA demonstrated relative standard deviations (RSD) below 3.0% (RRT RSD: 0.11–1.37%; RPA RSD: 0.19–2.94%), confirming good method repeatability (ICH Q2­(R1) acceptance criteria: RSD ≤ 3.0% for method validation).

2.

Repeatability test results.

3.3. Stability Test Results

Using peak 12 as the reference peak, relative retention times (RRT) and relative peak areas (RPA) were calculated for all characteristic peaks. As shown in Figure , The RSD values for both RRT and RPA of common peaks remained below 3.0% (n = 6) over 24 h, confirming excellent solution stability of the test samples.

3.

Stability test results.

3.4. Results of Fingerprint Establishment

The chromatographic data were imported into the Traditional Chinese Medicine Chromatographic Fingerprint Similarity Evaluation System (2012 version). Using the S1 sample chromatogram as the reference spectrum, the average method was adopted with a time window width of 0.1. After multipoint calibration and peak matching, the UPLC characteristic fingerprints of 12 batches of Sanhua Decoction were generated, along with the UPLC reference characteristic fingerprint (Figures and ). A total of 33 common peaks were identified in the fingerprints of the 12 batches. Using peak 12 as the reference peak, the relative retention times and relative peak areas of the common peaks were calculated. The relative standard deviation (RSD%) values of the relative retention times for all common peaks were less than 0.947%, further confirming the stability and reliability of the established analytical method. However, the relatively high RSD% values of the relative peak areas indicated batch-to-batch variations in the chemical composition of Sanhua Decoction.

4.

UPLC characteristic fingerprints of 12 batches of Sanhua Decoction.

5.

UPLC reference characteristic fingerprint of Sanhua Decoction.

3.5. Identification of Common Peaks

The mixed reference standard solution prepared under Section was analyzed using the chromatographic conditions specified in Section , and the chromatogram was recorded. By comparing the retention times of peaks in the chromatogram of the mixed reference standards with those in the reference characteristic fingerprint, a total of 28 common peaks were identified, as shown in Figure .

6.

Chromatogram of mixed reference standards. (1) Synephrine; (2) gallic acid; (5) chlorogenic acid; (6) syringin; (7) magnolin B; (8) magnoflorine; (9) daidzin; (10) magnolin A; (11) ferulic acid; (12) aloe-emodin-8-O-β-glucoside; (13) rhein-8-O-β-glucoside; (14) decursin; (15) rhaponticin; (16) quercitrin; (18) rhoifolin; (19) sennoside A; (20) hesperidin; (21) neohesperidin; (22) emodin-1-O-glucoside; (24) chrysophanol-8-O-β-glucoside; (26) aloe-emodin; (27) rhein; (28) nobiletin; (29) tangeretin; (30) honokiol; (31) magnolol; (32) chrysophanol; (33) physcion.

3.6. Similarity Evaluation

Similarity evaluation was performed using the Traditional Chinese Medicine Chromatographic Fingerprint Similarity Evaluation System (2012 version). The results indicated that batches S10 and S11 exhibited significantly lower similarity compared to other batches, suggesting that chemical composition variations exist among different batches of Sanhua Decoctionas detailed in Table .

3. Fingerprint Similarity of 12 Batches of Sanhua Decoction.

batches	S1	S2	S3	S4	S5	S6	S7	S8	S9	S10	S11	S12
S1	1	0.974	0.913	0.927	0.935	0.961	0.935	0.907	0.903	0.769	0.768	0.919
S2	0.974	1	0.952	0.962	0.939	0.962	0.934	0.93	0.924	0.767	0.759	0.905
S3	0.913	0.952	1	0.964	0.934	0.937	0.831	0.939	0.943	0.726	0.698	0.825
S4	0.927	0.962	0.964	1	0.972	0.966	0.904	0.959	0.953	0.802	0.766	0.896
S5	0.935	0.939	0.934	0.972	1	0.978	0.915	0.95	0.931	0.763	0.721	0.938
S6	0.961	0.962	0.937	0.966	0.978	1	0.926	0.932	0.909	0.782	0.758	0.956
S7	0.935	0.934	0.831	0.904	0.915	0.926	1	0.867	0.864	0.804	0.798	0.932
S8	0.907	0.93	0.939	0.959	0.95	0.932	0.867	1	0.98	0.748	0.692	0.871
S9	0.903	0.924	0.943	0.953	0.931	0.909	0.864	0.98	1	0.792	0.742	0.813
S10	0.769	0.767	0.726	0.802	0.763	0.782	0.804	0.748	0.792	1	0.986	0.708
S11	0.768	0.759	0.698	0.766	0.721	0.758	0.798	0.692	0.742	0.986	1	0.688
S12	0.919	0.905	0.825	0.896	0.938	0.956	0.932	0.871	0.813	0.708	0.688	1

3.7. Amelioration of Neurological Impairment by Sanhua Decoction in Rats with Cerebral Ischemia-Reperfusion Injury

Neurological function scoring is commonly used to assess the severity of cerebral ischemia-reperfusion injury and the efficacy of pharmacological treatments. Similar to humans, rats exhibit contralateral motor deficits due to crossed corticospinal innervation from the middle cerebral artery. For example, right-sided cerebral ischemia-reperfusion injury typically results in left hemiparesis, contralateral (leftward) circling, and ipsiversive (leftward) tilting, whereas left-sided injury leads to reciprocal deficits.

In the experimental protocol, neurological impairment was evaluated using Longa’s 5-point scale. Comparative results between experimental groups are shown in Figure . Sham-operated rats exhibited no neurological deficits (score = 0), displaying normal ambulation and intact neurological function. In contrast to the sham group, the model group demonstrated significantly higher neurological scores (P < 0.05), characterized by gait abnormalities, inability to fully extend the contralateral forepaw, and contralateral circling or body tilting during locomotion. Treatment with Sanhua Decoction significantly reduced neurological scores compared to the model group (P < 0.05), suggesting amelioration of ischemia-reperfusion-induced neurological dysfunction.

7.

Comparison of neurological function scores among groups in rats. Note: data are presented as mean ± SD (n = 8); #P < 0.05 vs Sham group; *P < 0.05 vs model group.

3.8. Modulation of Angiogenesis by Sanhua Decoction in Cerebral Ischemia-Reperfusion Injury Rats

Angiogenesis represents a critical multitarget intervention following ischemic stroke, inducible by inflammation and hypoxia. Both nascent vasculature and the angiogenic process itself mitigate postischemic brain injury. Multiple angiogenic growth factors (e.g., VEGF, Ang1, and Ang2) promote angiogenesis.

Following anesthesia, rat blood samples were collected from the abdominal aorta. Serum was separated by centrifugation at 4 °C and 4000 × g for 10 min. Serum concentrations of Ang1, Ang2, and VEGF were quantified using ELISA kits according to the manufacturer’s protocols. Results are presented in Figure .

8.

Comparison of serum Ang1 (A), Ang2 (B), and VEGF (C) levels among groups in rats. Note: data are presented as mean ± SD (n = 8); #P < 0.05 vs Sham group; *P < 0.05 vs model group.

Compared with the sham-operated group, serum levels of Ang1 (Figure A), Ang2 (Figure B), and VEGF (Figure C) were significantly elevated in the model group (P < 0.05). Administration of different Sanhua Decoction batches (S1–S12) progressively restored these levels toward sham-operated values relative to the model group (P < 0.05). These findings suggest that CIRI-induced rats spontaneously augment angiogenesis to ameliorate cerebral ischemia-hypoxia. Conversely, reduced tissue damage in Sanhua Decoction-treated groups corresponded to diminished angiogenic responses.

3.9. Attenuation of Cerebral Infarct Volume by Sanhua Decoction in Rats with Cerebral Ischemia-Reperfusion Injury

TTC Staining Methodology and Results: TTC (2,3,5-triphenyltetrazolium chloride) staining serves as a validated method for assessing cerebral infarction in ischemia-reperfusion injury. Viable brain tissue metabolizes TTC into formazan via dehydrogenase activity, exhibiting a brick-red color, whereas infarcted areas remain pale due to enzymatic failure. Representative TTC-stained brain sections from each experimental group are shown in Figure . The sham-operated group exhibited uniform brick-red coloration throughout the cerebral tissue with no detectable infarct zones. In contrast, the model group displayed extensive pale infarct regions in the ipsilateral hemisphere. After drug administration, the infarct volume was reduced, as shown in Figure . Quantitative analysis revealed a significant increase in infarct volume compared to the sham group (P < 0.05). Both the positive control group (nimodipine-treated) and Sanhua Decoction-treated groups demonstrated significantly reduced infarct areas relative to the model group (P < 0.05). These findings collectively indicate that Sanhua Decoction effectively attenuates cerebral infarction following ischemia-reperfusion injury.

9.

Comparison of cerebral infarct areas among groups in rats (TTC staining, n = 3). Note: For each group (n = 3 independent rats), four consecutive coronal sections from one representative rat are shown. The photo was taken by Xue-li Song.

10.

Ratio of cerebral infarct area to total brain tissue in each experimental group. Note: data are presented as mean ± SD (n = 8); #P < 0.05 vs Sham group; *P < 0.05 vs model group.

3.10. Attenuation of Brain Tissue Damage by Sanhua Decoction in Rats with Cerebral Ischemia-Reperfusion Injury

Histopathological assessment: histopathological examination remains the gold standard for evaluating cerebral ischemia-reperfusion injury (CIRI). Representative histological sections are shown in Figure . Sham group: hippocampal neurons maintained their density and laminar organization. Neuronal morphology was preserved, with distinct nucleocytoplasmic boundaries and prominent nucleoli. Model group: ipsilateral cortical regions exhibited extensive neuronal necrosis, characterized by pyknotic hyperchromatic nuclei and surrounding perineuronal vacuolization. Treatment groups: both nimodipine and Sanhua Decoction treatments demonstrated reduced pathology, with only scattered neurons showing nuclear pyknosis and minimal vacuolization. These findings suggest that Sanhua Decoction significantly mitigates brain tissue damage induced by CIRI.

11.

Histopathological examination of the ischemic penumbra in rat brains among groups (HE staining, ×20 magnification, n = 3). Note: black arrows indicate pyknotic nuclei with hyperchromasia and vacuolation (cystic cavities).

4. Spectrum-Effect Relationship Analysis

4.1. Calculation of Indicator Weights and Comprehensive Scores Using Entropy Weight Method

The entropy weight method (EWM) objectively determines the weights of indicators based on data variability, effectively addressing the limitations of single-parameter efficacy assessments. This method provides a computational foundation for comprehensive efficacy scoring. EWM-derived weights were assigned to each pharmacodynamic indicator to calculate composite efficacy scores. Spectrum-effect relationships were modeled by correlating the composite efficacy scores of 12 Sanhua Decoction batches (S1–S12) with the peak areas of their shared chromatographic peaks. Batch efficacy rankings were then established based on the model outputs.

Key results (Tables and ): composite efficacy scores were ranked in descending order as follows: S9 > S12 > S11 > S10 > S8 > S7 > S5 > S1 > S4 > S6 > S2 > S3. This ranking indicates that batch S9 exhibits the optimal anti-CIRI efficacy.

4. Weight Coefficients of Indicators for Sanhua Decoction against Cerebral Ischemia-Reperfusion Injury.

indicator	weight coefficient (%)
Ang-1	31.574
Ang-2	32.271
VEGF	36.154

5. Comprehensive Scores of Indicators for Sanhua Decoction against Cerebral Ischemia-Reperfusion Injury.

no	Ang-1	Ang-2	VEGF	comprehensive score (M)
S1	2363.7209	8.1121	101.9223	0.4554
S2	3593.8611	14.8187	192.6785	0.4050
S3	2542.0277	9.6859	101.0064	0.3880
S4	3114.8828	11.8988	144.0202	0.4287
S5	3083.2197	11.8589	159.3687	0.5188
S6	2979.2338	13.2893	152.9943	0.4081
S7	2837.5572	12.3508	163.3249	0.5429
S8	2850.6698	12.1171	142.4853	0.5908
S9	3228.5885	10.6408	163.9773	0.8307
S10	2918.7158	11.7273	136.7048	0.6232
S11	2739.9165	10.2886	153.7371	0.7150
S12	2422.9105	10.7409	108.0247	0.7386

4.2. Gray Relational Analysis (GRA)

Gray relational analysis (GRA) was employed to quantify the contribution of common chromatographic peaks to pharmacological efficacy by calculating relational grades. The comprehensive efficacy scores of Sanhua Decoction against cerebral ischemia-reperfusion injury were used as the reference sequence, while the peak areas of 33 common peaks from 12 fingerprint batches served as the comparative sequences. Data normalization was performed using the mean-value method to eliminate dimensional differences, followed by the calculation of relational grades. The results, summarized in Table , showed that all relational grades exceeded 0.669, indicating that the neuroprotective effects of Sanhua Decoction may result from the synergistic action of multiple bioactive constituents.

6. Grey Relational Grades for Sanhua Decoction against Cerebral Ischemia-Reperfusion Injury.

peak no.	peak name	gray relational grade	rank	peak no.	peak no.	peak no.	rank
5	chlorogenic acid	0.884	1	9	daidzin	0.826	18
23		0.88	2	26	aloe-emodin	0.825	19
33	physcion	0.875	3	31	magnolol	0.812	20
12	aloe-emodin-8-O-β-glucoside	0.873	4	15	rhaponticin	0.805	21
8	magnoflorine	0.873	5	2	gallic acid	0.804	22
24	chrysophanol-8-O-β-glucoside	0.858	6	10	magnoside A	0.804	23
22	emodin-1-O-glucoside	0.857	7	3		0.798	24
11	ferulic acid	0.852	8	6	syringin	0.795	25
14	decursin	0.851	9	32	chrysophanol	0.794	26
29	tangeretin	0.845	10	18	rhoifolin	0.792	27
21	neohesperidin	0.844	11	4		0.774	28
25		0.844	12	27	rhein	0.725	29
17		0.841	13	7	magnolin B	0.718	30
20	hesperidin	0.836	14	19	sennoside A	0.713	31
1	synephrine	0.828	15	30	honokiol	0.706	32
16	quercitrin	0.828	16	13	rhein-8-O-β-glucoside	0.669	33
28	nobiletin	0.826	17

4.3. Orthogonal Partial Least Squares-Discriminant Analysis (OPLS-DA)

Gray Gray relational analysis (GRA) is limited in its ability to indicate the direction (positive or negative) of the correlation between each common chromatographic peak and pharmacological efficacy. This limitation is effectively addressed by orthogonal partial least-squares-discriminant analysis (OPLS-DA). In OPLS-DA models, the sign of the regression coefficient indicates whether a given peak promotes or inhibits the therapeutic effect. Therefore, both GRA and OPLS-DA were employed in this study to enable a more comprehensive evaluation.

OPLS-DA was performed using the peak areas of 33 common chromatographic peaks in the Sanhua Decoction fingerprint as independent variables (X ₁–X _{3 3}), and the comprehensive scores of therapeutic efficacy indicators for cerebral ischemia-reperfusion injury were used as the dependent variable (Y). The analysis was conducted using SIMCA 14.1 software. The score plot and variable importance in projection (VIP) values are shown in Figures and , respectively. The regression equation was as follows:

$$Y=-{0.0564084}_{X1}-{0.0286857}_{X2}-{0.0188484}_{X3}-{0.070271}_{X4}+{0.06519}_{X5}+{0.0112551}_{X6}+{0.0123921}_{X7}+{0.0655598}_{X8}+{0.0313794}_{X9}-{0.0288849}_{X10}+{0.0911782}_{X11}+{0.0697847}_{X12}+{0.0442154}_{X13}+{0.0339244}_{X14}-{0.0105337}_{X15}+{0.0488775}_{X16}+{0.0434022}_{X17}-{0.0408714}_{X18}+{0.041368}_{X19}-{0.0577619}_{X20}+{0.00296451}_{X21}+{0.0654654}_{X22}+{0.0716996}_{X23}+{0.0557678}_{X24}+{0.0527069}_{X25}-{0.0580861}_{X26}+{0.0257547}_{X27}-{0.0558143}_{X28}+{0.00605808}_{X29}-{0.0617926}_{X30}-{0.0357652}_{X31}-{0.0290949}_{X32}+{0.00658657}_{X33}$$

12.

OPLS-DA score plot.

13.

VIP values plot.

4.3.1. Key Findings

Positive regression coefficients (β > 0), indicating a potential therapeutic contribution, were identified for the following peaks: Peak 5 (Chlorogenic Acid), Peak 6 (Syringin), Peak 7 (Magnolin B), Peak 8 (Magnoflorine), Peak 9 (Daidzin), Peak 11 (Ferulic Acid), Peak 12 (Aloe-emodin-8-O-β-glucoside), Peak 13 (Rhein-8-O-β-glucoside), Peak 14 (Decursin), Peak 16 (Quercitrin), Peak 17 (Unknown), Peak 19 (Sennoside A), Peak 21 (Neohesperidin), Peak 22 (Emodin-1-O-glucoside), Peak 23 (Unknown), Peak 24 (Chrysophanol-8-O-β-glucoside), Peak 26 (Aloe-emodin), Peak 27 (Rhein), Peak 29 (Tangeretin), and Peak 33 (Physcion). Negative regression coefficients (β < 0) were observed for the remaining peaks.

VIP analysis (VIP > 1) identified the following peaks as key contributors to therapeutic efficacy: Peaks 1, 4, 5, 8, 11, 12, 13, 16, 17, 19, 22, 23, and 25.

4.3.2. Conclusions

Peaks with both positive regression coefficients (β > 0) and VIP values >1namely Peaks 5, 8, 11, 12, 13, 16, 19, 22, 23, 25, and 33were strongly associated with therapeutic efficacy. These 11 peaks were thus identified as the principal bioactive components in Sanhua Decoction responsible for its protective effects against cerebral ischemia-reperfusion injury. The constructed spectrum-effect relationship model was rigorously validated. The cross-validation results demonstrated that the model exhibited an R ² of 99.3% and a Q ² of 53.50%, indicating good fitting quality and predictive ability. Furthermore, the permutation test results (with 200 permutations) showed that the original R ² and Q ² values (located on the far right of the plot) were significantly higher than those of the permuted models, and the intercept of the Q ² regression line was negative (−0.656), confirming that the model is valid without overfitting and possesses good predictive capability, as illustrated in Figure A,B.

14.

(A) Cross-validation plot; (B) permutation test plot.

5. Discussion

Sanhua Decoction exerts its protective effects against cerebral ischemia-reperfusion injury through a synergistic, multicomponent, multitarget mechanism. Its efficacy is intrinsically linked to its chemical composition, making the study of the relationship between its chemical constituents and pharmacological effects critically important. Chromatographic efficacy relationship analysis is a method that integrates chemical components from chromatographic fingerprints with pharmacological effects using mathematical models. This approach facilitates the screening of bioactive substances in traditional Chinese medicine (TCM) and provides a scientific assessment of TCM quality. It preserves the holistic perspective of TCM and promotes its modernization.

Previous studies have demonstrated that Sanhua Decoction ameliorates pathological brain morphology in rats with perfusion injury; reduces cerebral edema and decreases infarction volume; improves hemodynamics in aged ischemic rats, and enhances neurological recovery following ischemia-reperfusion injury. The present study aimed to establish associations between the HPLC fingerprint of Sanhua Decoction and therapeutic end points in MCAO (middle cerebral artery occlusion) rats. Specifically, we investigated correlations with cerebral infarct size, angiogenesis-related biomarkers (Ang1, Ang2, VEGF), histopathological brain alterations, and neurological function scores. Several chemical constituents exhibiting strong correlations with these efficacy indicators were identified through chromatographic-efficacy analysis.

This study is the first to systematically associate the material basis of Sanhua Decoction (SHD) with its therapeutic effects against cerebral ischemia-reperfusion injury (CIRI) through chromatographic-efficacy relationships. By integrating gray relational analysis and orthogonal partial least-squares-discriminant analysis (OPLS-DA), 11 key constituentsincluding ferulic acid, aloe-emodin-8-O-β-glucoside, and chlorogenic acidwere identified as highly correlated with pharmacological efficacy based on their chromatographic profiles. This approach provides a systematic elucidation of the bioactive material basis underlying SHD’s protective effects against CIRI. Correlation analysis further confirmed that SHD significantly improved neurological deficits, reduced cerebral infarct volume, and modulated angiogenesis-related biomarkers, including Angiopoietin-1 (Ang1), Angiopoietin-2 (Ang2), and Vascular Endothelial Growth Factor (VEGF). These outcomes are consistent with prior findings. Collectively, this integrated chromatographic-efficacy strategy offers a novel and reliable framework for identifying active compounds and elucidating the therapeutic mechanisms of traditional Chinese medicine (TCM) formulas.

The anthraquinone aglyconesincluding aloe-emodin-8-O-β-glucoside, emodin-1-O-glucoside, rhein-8-O-β-glucoside, and sennoside Aidentified through HPLC combined with gray relational analysis and orthogonal partial least-squares-discriminant analysis (OPLS-DA), constitute the principal active components of Rhei Radix et Rhizoma (rhubarb), the monarch herb in Sanhua Decoction (SHD). These compounds have been recognized as effective agents for the treatment of stroke. Given that cerebral ischemia-reperfusion injury (CIRI) falls within the pathological domain of stroke, and considering Professor Yan Zhenghua’s assertion that stroke pathogenesis fundamentally involves a combination of deficiency (primarily qi deficiency) and excess (primarily blood stasis), the therapeutic role of rhubarb in SHD merits re-evaluation. Its inclusion without postdecoction addition suggests pharmacological activities beyond intestinal purgation, notably its potent blood stasis-resolving effects. This aligns with traditional Chinese medical principles such as “eliminating stasis to promote circulation and regenerate tissues” (yuxue qu er xuexing changtong, xinxue sheng), reflecting rhubarb’s dual function of “expelling the stale and generating the new” (tuichen zhixin) as fully realized within the SHD formulation.

Modern pharmacological studies indicate that angiogenesis, triggered by inflammation and hypoxia following ischemic stroke, represents a critical multitarget therapeutic pathway. Both the neovascular structures and the angiogenic process itself play essential roles in mitigating postischemic brain injury, facilitated by factors such as VEGF, Ang1, and Ang2. Our results demonstrated that serum levels of Ang1, Ang2, and VEGF were significantly elevated in the model group compared to the sham group, indicating a compensatory pro-angiogenic response. In contrast, SHD treatment normalized these levels, suggesting reduced ischemic injury and inflammation. Histopathological analysis further supported these findings. In the sham group, hippocampal neurons exhibited normal morphology, clear laminar organization, distinct nuclear-cytoplasmic boundaries, and prominent nucleoli. The model group displayed extensive cortical damage, including widespread neuronal necrosis, nuclear pyknosis, cavitation, and inflammatory infiltration. In the SHD-treated group, only mild pathological changesisolated necrosis, nuclear pyknosis, and minor cavitationwere observed. Collectively, these results indicate that the neuroprotective effects of SHD against CIRI are likely mediated by anthraquinone aglycones derived from rhubarb.

Ferulic acid (primarily derived from Notopterygii Rhizoma et Radix [Qianghuo]), chlorogenic acid (primarily from Notopterygii Rhizoma et Radix and Magnoliae Officinalis Cortex [Houpo]), magnoflorine (mainly from Magnoliae Officinalis Cortex), and quercetin (derived from Aurantii Fructus Immaturus [Zhishi] and Magnoliae Officinalis Cortex) were identified as key constituents. In this formula, Aurantii Fructus Immaturus (Zhishi) alleviates epigastric distension, eliminates phlegm-fluid accumulation, and relieves stagnation. Magnoliae Officinalis Cortex (Houpo) regulates qi movement by promoting downward flow and exerting a mild exterior-releasing effect that facilitates the upward and outward dispersion of qi, exemplifying the principle of “descending action containing ascending properties” (jiang zhong yu sheng). When combined with the wind-expelling herb Notopterygii Rhizoma et Radix (Qianghuo), these effects extend to qi regulation, pathogenic dispersion, elevation of clear yang, and guidance of herbal actions toward the vertex. Modern pharmacological studies demonstrate that ferulic acid, a potent antioxidant, mitigates neuronal apoptosis by scavenging reactive oxygen species (ROS) and enhancing intrinsic antioxidant defenses. Consistent with these findings, neurological scoring revealed that CIRI model rats showed marked locomotor deficits and ipsilateral rotation, with some prone to lateral falls upon provocation, validating successful model induction. In contrast, SHD-treated rats displayed only mild forelimb flexion with substantially improved gait performance, indicating therapeutic efficacy in ameliorating CIRI-induced neurological dysfunction. Chlorogenic acid exerts neuroprotective effects on ischemic brain tissue by activating the Nrf2 signaling pathway, reducing lipid peroxidation, and upregulating HO-1 expression. Of the 11 bioactive constituents identified, three remain uncharacterized and require further validation through preclinical and clinical studies. The characterized components act synergistically against CIRI through multiple mechanisms: reducing cerebral infarct size, improving histopathological and functional neurological outcomes, and modulating the VEGF and Ang1/Ang2 signaling pathways to exert both anti-inflammatory and pro-angiogenic effects. This multicomponent integration underscores SHD’s multitarget therapeutic profile. The comprehensive chromatographic-efficacy correlation analysis further supports SHD as a representative formula embodying the principle of “unblocking, dispersing, and restoring” (kaitong xuanfu).

Although this study has confirmed through multi-indicator comprehensive evaluation and spectrum–effect relationship analysis that SHD exerts protective effects against CIRIby improving neurological function and reducing cerebral infarct volume via regulation of the VEGF/Ang-mediated angiogenesis pathwaycertain limitations are acknowledged. As appropriately noted by the reviewers, inflammatory response (e.g., TNF-α levels) and oxidative stress (e.g., T-SOD activity) represent core pathological processes in CIRI. Although direct measurements of TNF-α and T-SOD in brain tissue or serum were not performed in this study, the anti-inflammatory and antioxidant properties of the key active components identified in our spectrum–effect analysis are well-documented in the literature. For instance, ferulic acid and chlorogenic acid, which were identified as key constituents in this study, have been shown to significantly suppress the expression of pro-inflammatory factors such as TNF-α and IL-1β and alleviate oxidative stress damage. , Similarly, anthraquinone components derived from Rhei Radix et Rhizoma (e.g., aloe-emodin) have been demonstrated to effectively activate the Nrf2/ARE pathway and enhance the activity of antioxidant enzymes including SOD and GSH-Px. Therefore, it can be reasonably hypothesized that the anti-CIRI effects of SHD are attributable to the synergistic actions of its multiple constituents, which likely involve suppression of inflammatory and oxidative stress pathways. Direct assessment of key markers within these pathways will be prioritized in future studies to further elucidate the comprehensive mechanism of action of SHD.

Through a spectrum-effect relationship model, 11 potential active components closely associated with the efficacy of SHD against CIRI were successfully identified in this study. However, it must be emphasized that the in vitro activity of chemical components should be integrated with their in vivo pharmacokinetic behavior to ultimately confirm the true pharmacodynamic material basis. To address this, an independent pharmacokinetic study was conducted (results to be published separately), in which the blood-transiting components of SHD in CIRI model rats were systematically analyzed. Preliminary results revealed that several key components identified in the current studysuch as ferulic acid, chlorogenic acid, and aloe-emodin-8-O-β-glucosidecould be detected in plasma after administration. This provides important evidence supporting their identification as direct in vivo efficacy-related substances of SHD. Subsequent research will be built upon these findings to further explore the distribution of these blood-absorbed components in brain tissue and their synergistic pharmacological mechanisms.

A notable finding from the similarity assessment was that although batches S10 and S11 exhibited significantly lower fingerprint similarity compared to the other batches (Table ), their pharmacological efficacyreflected by improvements in neurological deficits, reduction in cerebral infarct volume, and modulation of serum VEGF, Ang-1, and Ang-2 levels (Figures , , )was not correspondingly reduced. This apparent discrepancy may be explained by the following factors. First and most importantly, therapeutic efficacy is determined by key active constituents (the “efficacy foundation”) rather than by the entire chemical profile. Through OPLS-DA analysis, 11 potential major active components were identified from the 33 common peaks, which were strongly correlated with anti-CIRI efficacy, including ferulic acid, chlorogenic acid, and aloe-emodin-8-O-β-glucoside. It is speculated that although the overall chemical profiles of S10 and S11 differed (as indicated by low similarity), these core efficacy components may have been maintained at sufficient concentrations to preserve the fundamental pharmacological activity. Batch-to-batch variations are thus likely to have occurred primarily among components with minimal contribution to the core efficacy. Second, the multicomponent, multitarget synergistic nature of traditional Chinese medicine may provide a compensatory mechanism. The efficacy of SHD is not dependent on a single component but is attributed to the synergistic actions of multiple active constituents. A phenomenon of “efficacy equivalence” may exist, whereby decreases or increases in certain noncritical components across batches could be compensated by other compounds with similar pharmacological activities, thereby maintaining overall therapeutic stability. Finally, this observation highlights the limitations of conventional fingerprint similarity assessment. Although it reflects overall chemical profile similarity, it cannot distinguish the relative contributions of individual chromatographic peaks to efficacy. To overcome this limitation, an “efficacy-integrated fingerprint” and a spectrum-effect relationship model were established in this study, enabling chemical analysis to be directly linked with pharmacological outcomes and thereby more accurately reflecting the intrinsic quality of the herbal medicine. This finding underscores the importance of shifting from “component monitoring” to “efficacy component monitoring” in the quality control of traditional Chinese medicines. Future quality standards for SHD should be established based on the key active components identified in this study to better reflect its clinical efficacy.

In the application of OPLS-DA to explore the spectrum-effect relationships, a statistical challenge was encountered due to the number of fingerprint variables exceeding the number of sample batches. Although the validity of the current model was confirmed by permutation testing and cross-validation, it must be acknowledged that this limitation may affect the generalizability of the model. In future investigations, this issue will be addressed by increasing the number of sample batches and introducing in vitro activity screening to reduce variable dimensionality prior to modeling, thereby enabling the construction of more universally applicable predictive models. Nevertheless, the conclusions derived from the OPLS-DA are supported by the results of the gray relational analysis, with both methods mutually reinforcing the inferences regarding the pharmacodynamic material basis of Sanhua Decoction against cerebral ischemia-reperfusion injury.

This study is the first to establish chromatographic-efficacy relationships between the HPLC fingerprint of Sanhua Decoction (SHD) and multiple pharmacodynamic indicators in MCAO (middle cerebral artery occlusion) rats modeling cerebral ischemia-reperfusion injury (CIRI). Our findings clearly identified the primary active constituents responsible for SHD’s therapeutic effects against CIRI, providing a solid foundation for future investigations into the pharmacokinetic behavior of these components in the plasma of CIRI rats. A notable limitation of this study is the absence of metabolomic analysis. Future research will address this gap by analyzing serum, urine, feces, and brain tissue samples from MCAO rats. Based on the chromatographic peaks exhibiting strong correlations with pharmacological efficacy, potential biomarkers of SHD’s anti-CIRI effects will be screened, followed by pathway enrichment analysis to elucidate its underlying mechanisms of action. Interestingly, although batches S10 and S11 showed significant deviations in chemical similarity compared to other batches, no substantial differences were observed in their pharmacodynamic outcomes. This finding highlights the critical importance of accounting for geographical origin in the quality control of traditional Chinese medicines.

6. Conclusions

This This study established UPLC fingerprints for 12 batches of Sanhua Decoction (SHD), identifying 33 common peaks and characterizing 28 of these components. Using gray relational analysis and orthogonal partial least-squares-discriminant analysis (OPLS-DA), the following core bioactive constituents were identified: ferulic acid, aloe-emodin-8-O-β-glucoside, emodin-1-O-glucoside, magnoflorine, chlorogenic acid, rhein-8-O-β-glucoside, sennoside A, and quercitrin. These constituents exert protective effects against cerebral ischemia-reperfusion injury (CIRI) through synergistic, multitarget mechanisms, including antioxidant activity (via activation of the Nrf2 signaling pathway); anti-inflammatory effects (through modulation of cytokine expression); promotion of angiogenesis (by regulating VEGF/Ang1 signaling); reduction of cerebral infarct volume; improvement of brain tissue histopathology; and amelioration of neurological deficits associated with CIRI. These findings provide a scientific foundation for optimizing SHD quality control using these marker components and offer guidance for its clinical application in the treatment of CIRI.

L.Z.: conceptualization, data curation, formal analysis, writing-original draft, methodology. X.-l.S.: project research, data organization, and analysis. G.F.: data curation, funding acquisition, project administration, resources, writing review and editing. T.L.: literature collection. X.-p.W.: scientific editing. Z.-m.L.: data amendment. S.-q.Q.: data amendment. Y.-l.W.: data amendment. G.L.: visualization. K.M.: investigation. J.H.: investigation. W.L.: resources, validation. Y.S.: visualization.

This work was supported by the National and Provincial Science and Technology Innovation Talent Team Cultivation Project of Guizhou University of Traditional Chinese Medicine (No. Guizhong Medicine TD He Zi [2024] 001). the “Thousand” level Innovative Talents Project in Guizhou Province (grant no. Qrlf [2020]­4), the Guizhou University of Traditional Chinese Medicine (“Thousand level” Talent) Research Project (No. ZQ2018001), the Guizhou Traditional Chinese Medicine Processing Technology Inheritance Base Construction Project (Project No. Qian Traditional Chinese Medicine Letter ([2024] No. 22), the Central-level major increase or decrease expenditure projects (No. 222060302), and the 2024 Guizhou Provincial Enterprise ‘Science and Technology Deputy Positions’ and ‘Science and Technology Commissioners’ Project (No. Qian Ke He Ren Cai KJZY [2025] 060).

The authors declare no competing financial interest.