2.1 CS ameliorated bleomycin-induced IPF in mice

Before investigating the potential of CS in treating IPF, the chemical constituents of the CS utilized in this study were analyzed using a mass spectrometer. The total ion chromatograms (TIC) of the test solution extracted from CS under positive and negative ionization modes are depicted in Supporting Information S1: Figure S1. A total of 58 compounds were identified, comprising 17 nucleosides, 15 amino acids, 5 purine alkaloids, 4 pyrimidine alkaloids, 8 organic acids, 2 nucleotides, 2 sugar alcohols, and 5 other compounds. Detailed mass spectrometry information is provided in Supporting Information S1: Table S1. These findings align well with previous reports concerning the active constituents of CS.²¹

To investigate the potential therapeutic effects of CS on IPF in vivo, we conducted pharmacological experiments on mice as outlined in Figure 1A. Briefly, lung fibrosis was induced in mice via intratracheal instillation of bleomycin. Subsequently, the mice received daily intragastric administration of CS and dexamethasone for a duration of 3 weeks. Upon sacrifice, lung tissues and serum samples were collected for pathological and serum biochemical analyses. Hematoxylin and eosin (H&E) staining results (Figure 1B) revealed that bleomycin instillation led to severe alveolar hemorrhage, edema, thickening of alveolar septa, and infiltration of inflammatory cells. Treatment with dexamethasone (Dexm), a corticosteroid with anti-inflammatory properties, effectively mitigated these observed damages in lung pathology. Dexm is extensively utilized in clinical practice for the management of interstitial pneumonia and pulmonary fibrosis.²⁶ Its protective mechanism involves the suppression of inflammatory mediator production, such as interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α).²⁷ Moreover, previous studies have established dexamethasone as the positive control for pulmonary fibrosis treatment, owing to its notable efficacy, high biological potency, and minimal sodium retention.^{28, 29} Encouragingly, CS treatment demonstrated similar therapeutic effects to Dexm, markedly reducing pulmonary edema and inflammatory cell infiltration. Additionally, Masson's trichrome staining indicated that the administration of CS and Dexm alleviated pulmonary interstitial collagen deposition in the model group (Figure 1B). Serum analysis results revealed a significant elevation in the levels of inflammatory cytokines including IL-1β, IL-6, and TNF-α following bleomycin induction. However, administration of CS and Dexm efficiently suppressed the increase in these cytokine levels in the serum of mice with pulmonary fibrosis (Figure 1C–E). Furthermore, it has been reported that the increase in ROS levels can activate the transforming growth factor β (TGF-β) pathway and promote the onset of pulmonary fibrosis.^{30, 31} Additionally, the transformation of fibroblasts into myofibroblasts is a crucial indicator of IPF progression, with α-smooth muscle actin (α-SMA) and collagen type I alpha 1 (Col1a1) serving as specific marker proteins for myofibroblasts.^{32, 33} Immunohistochemical staining results demonstrated that the expressions of TGF-β, α-SMA, and Col1a1 in the model group were significantly elevated compared to the control group, whereas CS or Dexm treatment markedly suppressed their expression induced by bleomycin (Figure 1G–I). Taken together, these findings demonstrate that CS effectively mitigates bleomycin-induced IPF, similar to dexamethasone.

2.2 CS regulated mitochondrial-mediated oxidative phosphorylation

To elucidate the mechanistic insights into the action of CS against IPF, we conducted a label-free quantitative proteomic analysis of lung tissues, as depicted in Figure 2A. The correlogram illustrating the scaled abundance of proteins across the control (Con), model (Mod), and CS groups revealed a positive correlation coefficient between the CS and Con groups, whereas the Mod group exhibited negative correlation coefficients with both the CS and Con groups (Figure 2B). This suggests that protein expression levels in lung tissues of the Mod group tended to normalize following CS treatment. A total of 4862 proteins were identified across the three groups (Figure 2C,D), with differentially expressed proteins (DEPs) defined using a fold change >2 and p < 0.05, highlighted in scatter plots (Figure 2C). Specifically, 1074 proteins were upregulated and 824 proteins were downregulated when comparing the Mod group to the Con group, whereas in the CS group compared to the Mod group, there were 803 upregulated and 1263 downregulated proteins (Figure 2C). The changes in protein abundance for each group are depicted in a heat map (Figure 2D), wherein cluster analysis revealed a stronger tendency for the CS group to cluster with the Con group rather than the Mod group, consistent with the correlation analysis results (Figure 2B). Furthermore, we analyzed the cellular component of DEPs (Figure 2E). Comparing the cellular localization of differential proteins between the Mod and Con groups, we observed that the majority of differential proteins were localized in the nucleus, cytoplasm, mitochondria, and cell membrane. Notably, the differential proteins in the CS versus Mod group exhibited minimal changes in cellular localization compared to those in the Mod versus Con group.

To further elucidate the changes at the proteomic level, the identified proteins were categorized into five clusters based on their variation patterns (Figure 3A). In cluster 1, proteins initially exhibited downregulation in the model group followed by upregulation in the CS group, whereas proteins in clusters 3 and 4 showed initial upregulation in the model group followed by downregulation in the CS group. Notably, proteins in these clusters displayed a consistent feature across the three groups, wherein CS reversed the alterations in protein abundance induced by bleomycin in IPF mice. Subsequently, proteins from clusters 1, 3, and 4 underwent Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis to elucidate the potential molecular pathways targeted by CS. Pathways including oxidative phosphorylation (OXPHOS), HIF-1 signaling, glycolysis, and ROS were concurrently enriched from the upregulated proteins in the Mod versus Con group and the downregulated proteins in the CS versus Mod group (Figure 3B). Mitochondria serve as pivotal energy metabolism centers wherein a significant array of proteins partake in the biological processes of oxidative phosphorylation and glycolysis.³⁴ Furthermore, in line with the findings from KEGG analysis, bleomycin induction led to an augmentation in the expression of oxidative phosphorylation-related proteins, while CS treatment mitigated this upregulation, as illustrated by the circular heatmap (Figure 3C). Additionally, gene set enrichment analysis (GSEA) was conducted to specifically evaluate the enrichment score of oxidative phosphorylation. Here, CS treatment was observed to restore the gene signatures associated with oxidative phosphorylation in IPF mice towards those observed in healthy mice (Figure 3D,E). Collectively, these proteomic findings underscore mitochondrial-mediated oxidative phosphorylation as a potential molecular pathway through which CS confers protection against IPF.

2.3 CS restored the expression level and activity of mitochondrial complex I/II disturbed by bleomycin

To validate the alterations in the abundance of proteins implicated in mitochondrial oxidative phosphorylation as elucidated by the proteomic analysis, we initially assessed the expression levels of mitochondrial complexes I and II (MC-I and MC-II) in lung tissues. Western blot analysis revealed a significant elevation in the expression levels of mitochondrial complex I components (Ndufs6, Ndufaf1, and Ndufaf3) and complex II component (Sdhc) in the Mod group, which markedly decreased following CS treatment (Figure 4A,B). Consistent with the protein quantification results, the messenger RNA (mRNA) levels of the four genes (Ndufs6, Ndufaf1, Ndufaf3, and Sdhc) assessed via quantitative real-time PCR (qPCR) exhibited similar trends of change across the three groups (Figure 4C–F), collectively indicating that CS partially counteracted the effect of bleomycin on the expression of mitochondrial complexes. Furthermore, the enzyme activity of mitochondrial complexes I and II in lung tissues was hyper-activated following bleomycin induction but returned to normal levels with CS intervention (Figure 4G,H). In summary, these findings suggest that the antifibrotic effect of CS may be closely linked to the expression and activity of mitochondrial complexes I and II.

2.4 CS alleviated mitochondrial dysfunction by reducing mitochondrial ROS production

As crucial contributors to the respiratory electron-transport chain within mitochondria, mitochondrial complexes I and II have the capacity to generate excessive ROS, thereby instigating oxidative stress, which is deemed a pivotal factor in bleomycin-induced IPF.^{35, 36} Therefore, we postulated that a potential mechanism underlying the preventive action of CS against bleomycin-induced IPF may involve the inhibition of ROS release and mitigation of oxidative stress.^{35, 37} To assess the impact of CS on mitochondrial oxidative stress, the levels of total ROS and MitROS in lung tissue were measured. As depicted in Figure 4I,J, compared to the control group, the bleomycin-induced group exhibited elevated levels of both total ROS and MitROS, whereas CS administration mitigated the overproduction of ROS. Excessive ROS can impair mitochondrial function, often accompanied by a reduction in mitochondrial membrane potential (MMP) and ATP production. Consistently, CS effectively restored MMP levels and ATP production in lung tissue, even in the face of significant suppression induced by bleomycin treatment (Figure 4K,L). Similar results were observed in the levels of mtDNA, which were increased in the lung tissue of mice induced by bleomycin and significantly decreased after CS treatment. This indicates that CS can alleviate mitochondrial dysfunction by reducing mitochondrial over-replication caused by bleomycin (Figure 4M).³⁸ In summary, CS modulates the expression and activity of mitochondrial complexes I and II, thereby alleviating oxidative stress and preserving normal mitochondrial functions (Figure 4N).

To further corroborate and extend the in vivo findings, we employed human lung epithelial BEAS-2B cells as an in vitro model. Transmission electron microscopy was utilized to directly visualize mitochondrial morphology in the three groups. The images revealed that bleomycin induced mitochondrial swelling, severe cristae damage, and membrane rupture, while CS treatment significantly alleviated these mitochondrial damages in the model group (Figure 5A). Consistent with the results observed in lung tissues, the elevated levels of total ROS and MitROS induced by bleomycin in BEAS-2B cells were reversed by CS treatment (Figure 5B,C). Furthermore, fluorescence imaging confirmed that CS markedly suppressed the excess ROS in bleomycin-induced BEAS-2B cells (Figure 5D). Moreover, with regard to the levels of ATP (Figure 5E), MMP (Figure 5F), mitochondrial complexes I and II (Figure 5G,H), and mtDNA (Figure 5I), the patterns of change in the in vitro model across the three groups closely mirrored the data observed in lung tissue, wherein CS alleviated the disturbances induced by bleomycin to a certain extent. Additionally, we detected changes in HSP60 expression using immunocytochemistry (ICC). The results confirmed that HSP60 expression was upregulated in the lung tissue of mice induced by bleomycin, while its expression significantly decreased after CS treatment (Figure 5J). As a typical mitochondrial molecular chaperone induced by cell stress, HSP60 causes a strong pro-inflammatory reaction in cells of innate immune system and serves as a danger signal for stressed or damaged cells.^{39, 40} Our findings indicated that CS can target HSP60 to regulate the levels of oxidative damage and inflammation. Furthermore, cordycepin, the principal ingredient of CS,⁴¹ could significantly reduce the activities of MC-I and MC-II as well as cellular ROS level, demonstrating its capacity to regulate oxidative phosphorylation and diminish cellular ROS release (Supporting Information S1: Figure S2). In summary, the congruence between the in vitro and in vivo findings consistently suggests that mitochondrial function, particularly the production of ATP and MitROS, may represent the critical targeted pathway responsible for the ameliorative effect of CS on bleomycin-induced pulmonary fibrosis (IPF).

2.5 CS alleviated oxidative damage in lung tissue of pulmonary fibrosis mice

While the pathogenesis of pulmonary fibrosis has yet to be fully elucidated, oxidative stress is widely acknowledged as a prominent factor.⁴² This oxidative stress arises from a persistent imbalance between the generation of ROS and their clearance by the intracellular antioxidant system.^{43, 44} As mitochondrial dysfunction or impairment is known to trigger oxidative stress, we hypothesize that the attenuation of mitochondrial-mediated oxidative stress by CS may underlie its pharmacological activity in pulmonary fibrosis. The accumulation of ROS during oxidative stress leads to glycoxidation reactions and lipid peroxidation, with malondialdehyde (MDA) being one of the final products of polyunsaturated fatty acid peroxidation.⁴⁵ Additionally, myeloperoxidase (MPO), a heme-containing peroxidase highly expressed in monocytes, neutrophils, and macrophages, catalyzes the production of ROS. Excessive MPO activity is closely associated with tissue damage in inflammation, fibrosis, and other diseases.⁴⁵

As depicted in Figure 6A,B, administration of CS and Dexm significantly reduced the elevated levels of MDA and MPO in lung tissue from bleomycin-induced mice, indicating that CS has the capacity to decrease neutrophil release and alleviate oxidative stress. To further evaluate the impact of CS on the antioxidant system, we assessed the activities of two typical antioxidant enzymes, SOD and glutathione peroxidase (GSH-Px), known for their roles in scavenging free radicals and combating oxidative damage. Administration of CS and Dexm restored the activity of SOD and GSH-Px, which had been compromised by bleomycin in lung tissue, thereby enhancing the antioxidant capacity in the body (Figure 6C,D). Furthermore, it has been reported that lipid peroxidation induced by oxidative stress is closely related to the development of pulmonary fibrosis. Reactive lipid mediators, such as 4-hydroxynonenal (4-HNE) and nitrotyrosine, produced during oxidative stress, serve as the biomarkers of oxidative stress.^{46, 47} Western blot results showed that the expression of 4-hydroxynonenal and nitrotyrosine in the lung tissue of mice induced by bleomycin was upregulated, while their protein expression significantly decreased after CS treatment. This indicates that CS has the potential to regulate oxidative damage in mice with pulmonary fibrosis (Figure 6E,F). Hence, CS not only protects lung tissues against oxidative stress induced by hyper-activated mitochondrial-mediated oxidative phosphorylation but also promotes the normal function of the antioxidant system.

5.1 Reagents and materials

The CS extract in Bailing capsules was supplied by Hangzhou Zhongmei Huadong Pharmaceutical Co., Ltd. Bleomycin hydrochloride (HY-K1053) and dexamethasone acetate (HY-14648A) were procured from MedChem Express. Iodoacetamide (IAA) and dithiothreitol (DTT) were sourced from Sigma. Trypsin (90057) and other LC-MS/MS-related reagents were acquired from Thermo Fisher. Specific primary antibodies against Ndufs6 (ab108208), Sdhc (ab92508), and HSP60 (ab110312) were obtained from Abcam. TNF-α assay kit (ab208348), IL-6 ELISA kit (ab222503), and IL-1β ELISA kit (ab197742) were sourced from Abcam. Ndufaf1 (H00051103) and Ndufaf3 (PAB22870) antibodies were procured from Abnova. The BCA assay (P0012S) was purchased from Beyotime Biotechnology. Activity kits for MDA, MPO, SOD, and GSH-Px were obtained from Shanghai Enzyme-linked Biotechnology Co., Ltd. Kits for assessing the level of total ROS, Mit ROS, ATP, and MMP were purchased from Nanjing Jiancheng Institute of Biological Engineering. Mitochondrial Complex I Activity Assay kit (ab287847) and Mitochondrial Complex II Activity Assay kit (ab287844) were obtained from Abcam. 4-Hydroxynonenal was purchased from Biosis (bs-6313R) and Nitrotyrosine was obtained from Abcam (ab314438). mtDNA fluorescence quantitative PCR kit was purchased from Biolabs (BTN14-141).

5.2 Analysis of the composition of CS by UHPLC-LTQ-Orbitrap mass spectrometry

First, CS powder form six different batches was accurately weighed and dissolved in 50 mL of 70% methanol. The mixture underwent sonication dispersion for 20 min and was subsequently filtered. Then, 25 mL of the filtered solution was vacuum-dried, yielding a dried powder. This powder was re-dissolved in 25 mL of deionized water, which served as the test solution after undergoing filtration once again. The ingredient analysis was conducted using an online linear ion trap (LTQ) Orbitrap Velos Pro (Thermo Fisher Scientific Inc.) coupled with ultrahigh pressure liquid chromatography (UHPLC) via an electrospray ionization (ESI) interface. Liquid chromatogram separation was performed on an ACQUITY UHPLC HSS T3 column (2.1 mm × 100 mm, 1.8 µm) maintained at a temperature of 30°C. The mobile phases consisted of solvent A (methanol) and solvent B (0.1% formic acid in water), with a gradient elution profile as follows: 0–20 min, 1% A; 20–36 min, 1%–15% A; 36–45 min, 15%–60% A; 45–45.1 min, 60%–95% A. The flow rate was set at 0.3 mL/min, with a sample volume of 10 µL for reference and 8 µL for testing. The ESI source parameters were configured for positive and negative ion scanning modes, with a capillary temperature of 350°C and a spray voltage of 3.5 kV in positive ion mode. The primary mass spectra of the samples were scanned in FT mode with a resolution of 30,000 and a m/z scanning range of 50 to 2000. Secondary and tertiary mass spectra were collected in the data-dependent mode. Data acquisition and analysis were performed using Xcalibur, Metworks, and Mass Frontier 7.0 software.

5.3 Animal experiments

Eight-week-old male Kunming mice weighing approximately 20 g were procured from Beijing Vital River Laboratory Animal Technology Co., Ltd. The animal experimental procedures were ethically approved by the Ethics Committee of the Experimental Animal Center of the Chinese Academy of Traditional Chinese Medicine (Approval No. 42020005322). Upon arrival, the mice were acclimatized for 3 days under controlled environmental conditions, including humidity (60%–80%), temperature (21 ± 3°C), and a 12-h light/dark cycle, with ad libitum access to standard diet and water. Following the acclimatization period, the mice were randomly allocated into four groups (n = 5 per group) as follow: control group (Con), bleomycin-treated group (BLM, 5 mg/kg), dexamethasone-treated group (Dexm, 5 mg/kg), Bailing capsule treated group (CS, 5 g/kg). The gavage doses of for CS in mice were determined as 1 and 5 g/kg/d referring to the recommend doses in previous reports.²¹ Dexamethasone was chosen as the positive control for pulmonary fibrosis treatment, aligning with previous studies citing its high efficacy, biological potency, and minimal sodium retention.^{28, 29} Anesthesia was induced in all mice via intraperitoneal injection of pentobarbital sodium before fixation on the operating table for tracheal instillation. A surgical incision was made in the skin near the neck, and 100 μL of bleomycin solution (5 mg/kg) or normal saline (control) was slowly injected into the detached trachea, followed by wound closure. Starting from the following day, the designated treatment drugs were administered to each group via oral gavage once daily for a duration of 3 weeks. On Day 21, all mice were euthanized, and lung tissue and serum samples were collected for subsequent analysis.

5.4 Histopathology and Masson's trichrome staining

The left lung tissue of mice was fixed in a 4% paraformaldehyde solution, followed by staining with hematoxylin-eosin solution (H&E) and Masson's Trichrome staining kits as per the manufacturer's instructions. For immunocytochemistry, tissue slices were incubated overnight at 4°C with primary antibodies against TGF-β, α-SMA, and Col1a1, followed by a 2-h incubation with a secondary antibody at room temperature. Subsequently, the sections were dehydrated, mounted, and imaged using a digital slide scanner (Nanozoomer-SQ; Hamamatsu Photonics). Analysis was performed using an inverted fluorescence microscope (Olympus) equipped with the MShot image analysis system.

5.5 Enzyme-linked immunosorbent assay assay

The proinflammatory cytokines (TNF-α, NO, IL-6, and IL-1β) in mice serum were quantified using commercially available kits following the manufacturer's instructions.

5.6 Evaluation of oxidative stress

We assessed the activity of antioxidant enzymes (GSH-Px, SOD, and MPO) and the level of MDA in lung tissues using kits in accordance with the manufacturer's instructions.

5.7 Cell culture

BEAS-2B cells were procured from the Chinese Academy of Medical Sciences (Beijing) and maintained in high-glucose Dulbecco's modified Eagle's medium (Corning) supplemented with 10% FBS (Corning) and 1% (v/v) penicillin/streptomycin (Thermo Fisher) at 37°C in a 5% CO₂ atmosphere.

5.8 Assays for ATP and MMP measurement

BEAS-2B cells were seeded at a density of 4 × 10³ cells per well in 96-well plates and incubated for 1 day. Following incubation, the cells were treated with bleomycin (40 μg/mL) or CS (250 μg/mL) for 24 h. The ATP and MMP levels were assessed using kits in accordance with the manufacturer's instructions.

5.9 RNA isolation and quantitative real-time reverse-transcription polymerase chain reaction

Total RNA was extracted from mouse lung tissues using TRIzol reagent (Invitrogen) and subsequently reverse-transcribed into cDNA using the TransScript® II Green Two-Step qRT-PCR SuperMix kit (Transgen Biotech). Amplification plots were analyzed, and dissociation curves were examined to ensure data accuracy. Invitrogen synthesized primers for PCR amplification, and their sequences were as follows: for mouse Ndufs6, forward: 5′-GGGGAAAAGATCACGCATACC-3′, reverse: 5′-CAAAACGAACCCTCCTGTAGTC-3′; for mouse Ndufaf1, forward: 5′-CTCTAATCGAGGAAGAATCCG-3′, reverse: 5′-AGAATGGACCATCCACTTT-ATC-3′; for mouse Ndufaf3, forward: 5′-GTGGTCCAGTGGAACGTGG-3′, reverse: 5′-CTCCTGTCACTCGACCTTCG-3′; for mouse Sdhc, forward: 5′-GGGGAATT-CATGGCTTTCTTGCTGAGACATGTCAGC-3′, reverse: 5′-GGGAAGCTTT-CACAGGGCGGCCAGCCC-3′. The relative mRNA levels were normalized to β-actin mRNA.

5.10 Total and mitochondrial ROS measurement

The level of total ROS in mouse lung tissues was evaluated using the ROS detection kit from Beyotime and a fluorescence microplate reader. MitROS levels in the lung tissues were determined after isolating mitochondria with the Beyotime mitochondria isolation kit. The MitROS levels were assessed using the mitochondrial ROS detection kit from Beyotime and were normalized by protein amount. Additionally, the total ROS and MitROS levels in BEAS-2B cells were quantified using the total ROS detection kit and the mitochondrial superoxide kit from Beyotime, respectively.

5.11 Assay for fluorescent staining of ROS

BEAS-2B cells (1 × 10⁵/well) were plated in six-well plates and allowed to culture for 1 day. Subsequently, they were exposed to either bleomycin (40 μg/mL) or CS (250 μg/mL) for a duration of 24 h. After three washes with phosphate-buffered saline (PBS), the cells were sequentially treated with DCFH-DA probe (10 μM, 15 min) and DAPI (5 μg/mL, 15 min) at 37°C in a CO₂ incubator. Following three additional PBS rinses, the cells were visualized under a fluorescence microscope (Olympus).

5.12 Western blot analysis

Mouse lung tissues were lysed using RIPA buffer supplemented with protease inhibitors (Sigma-Aldrich). The protein concentration was determined using the Pierce BCA Protein Assay Kit. Subsequently, 20 μg of proteins were separated by 12% sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred onto polyvinylidene difluoride membranes. After blocking, the membranes were incubated with primary antibodies against β-actin, Ndufs6, Ndufaf1, Ndufaf3, Sdhc, 4-hydroxynonenal and Nitrotyrosine at a dilution of 1:1000, followed by secondary antibody incubation. Protein bands were visualized using electrochemical luminescence (ECL)detection reagents, and Image J software (V1.8.0.112) was utilized for density analysis.

5.13 Transmission electron microscope

BEAS-2B cells were initially fixed in 2.5% glutaraldehyde for 2 h, followed by a 2 h fixation in 1% osmium tetroxide. Subsequently, the samples were sectioned into 50–60 nm slices, stained with 3% uranium acetate and lead citrate, and then analyzed and imaged using a transmission electron microscope (JEM-1011).

5.14 Activity of mitochondrial complex I and II in purified mitochondria from lung tissues and BEAS-2B cells

The level of mitochondrial Complex I and II in purified mitochondria from mice lung tissues were detected using the Mitochondrial Complex I and II Activity Assay kits (Abcam) following the manufacturer's instructions. Mitochondria were isolated using the Beyotime mitochondria isolation kit before the assay. For BEAS-2B cells, the cells were exposed to either bleomycin (40 μg/mL) or CS (250 μg/mL) for 24 h after seeding in six-well plates. Subsequently, the Mitochondrial Complex I and II Activity Assay kits were used according to the manufacturer's instructions.

5.15 Immunofluorescence images of HSP60

BEAS-2B cells cultured in a confocal dish, and treated with either bleomycin (40 μg/mL) or CS (250 μg/mL) for 24 h. Then cells were fixed and the exposed to anti-HSP60 (1:1000) at 4°C overnight, goat anti-rabbit IgG (1:1000) (Alexa Fluor 488) for 2 h, and DAPI (5 μg/mL) for 15 min at 37°C in a CO₂ incubator. Following three additional PBS rinses, the cells were visualized under a fluorescence microscope (Olympus).

5.16 Statistical analysis

All data were presented as mean ± standard deviation (SD) and analyzed using Student's t-test and analysis of variance test. Statistical analysis was conducted using GraphPad Prism 8.0 software (GraphPad Prism). Statistical significance was determined at *p < 0.05, **p < 0.01, ***p < 0.001, or ****p < 0.0001.