Abstract
Background
High-altitude pulmonary edema (HAPE), a severe manifestation of
hypoxia-induced pulmonary hypertension, continues to present a major
health concern in high-altitude environments due to the absence of
efficient preventive measures. This investigation explores the
protective influence of ginsenoside Rg3 (G-Rg3), an active substance
derived from the botanical drug Panax ginseng C.A.Mey., on the
prevention of HAPE progression.
Methods
A mouse model mimicking exposure to 6000-m altitude (n = 63 C57BL/6
mice) was employed to evaluate the impact of G-Rg3 (15/30 mg/kg) using
histopathological, biochemical, and multi-dimensional molecular
assessments. Western blotting, network pharmacology and computational
simulations were utilized to identify molecular targets of G-Rg3. The
role of the PI3K/AKT signaling pathway was further validated through
experiments using the PI3K/AKT inhibitor LY294002.
Results
Pre-treatment with G-Rg3 effectively alleviated HAPE, maintained the
stability of lung ultrastructure, and inhibited inflammatory mediators
and oxidative stress indicators. Mechanistically, G-Rg3 prevented
ferroptosis by stimulating the PI3K/AKT signaling pathway, as evidenced
by the upregulation of protective proteins (GPX4, Nrf2, HO-1, SLC7A11,
FTH1, FLC) and the downregulation of iron metabolism regulatory factors
(TFRC, COX2). Network pharmacology and molecular docking analysis
confirmed that PI3K/AKT is the core target of G-Rg3, and the protective
effect disappeared when this pathway was inhibited. G-Rg3 uniquely
regulated oxidative stress and inflammation by inhibiting ferroptosis,
demonstrating adaptability to high-altitude environments.
Conclusion
This research examined the pharmacological impacts and molecular
pathways of ginseng active monomers on HAPE, suggesting the potential
of G-Rg3 as a promising treatment option for this condition.
Keywords: high-altitude pulmonary edema (HAPE), ginsenoside Rg3
(G-Rg3), PI3K/Akt pathway, ferroptosis, hypoxic pulmonary hypertension
1 Introduction
High-altitude pulmonary edema (HAPE) represents a critical form of
hypoxic pulmonary hypertension, typically emerging in settings with
diminished atmospheric pressure and oxygen levels ([42]Bärtsch and
Swenson, 2013; [43]Persson and Bondke Persson, 2017). Extended exposure
to high-altitude hypoxia is the primary driver of HAPE, which, without
intervention, can evolve into chronic hypoxic pulmonary hypertension.
Clinically, HAPE is marked by non-cardiogenic pulmonary edema and
worsening hypoxemia, with an untreated fatality rate estimated at
nearly 50% ([44]Bouzat et al., 2013; [45]Li et al., 2024; [46]Menon,
1965). Individual susceptibility to hypoxic stress and the development
of HAPE may be influenced by genetic variations, particularly in
mitochondrial DNA and components of the renin-angiotensin-aldosterone
system (RAAS) ([47]Bhagi et al., 2015; [48]Luo et al., 2012; [49]Sharma
et al., 2019; [50]Srivastava et al., 2012). Furthermore, hypoxic
conditions can induce oxidative stress via mitochondrial dysfunction
and disruptions in metabolic pathways, thereby amplifying abnormalities
in the pulmonary vasculature ([51]Ali et al., 2012; [52]Sharma et al.,
2021). Importantly, in high-altitude environments, cold temperatures
may interact synergistically with hypoxia to exacerbate cardiopulmonary
dysfunction, thus elevating the risk of HAPE ([53]Eichstaedt et al.,
2023). These underlying pathophysiological mechanisms ultimately result
in abnormal pulmonary vascular reactions and sustained pulmonary
hypertension, leading to the formation of edema.
In recent years, there has been a growing emphasis on the use of
natural medicines for treating HAPE, particularly those derived from
botanical drugs with lung-protective active metabolites. Recent
clinical studies have highlighted the significant potential of
traditional botanical drugs, such as ginseng, in preventing and
managing HAPE through multi-target mechanisms ([54]Huang et al., 2024;
[55]Shen et al., 2023). Panax ginseng C.A.Mey. (known as Renshen in
Chinese), a classic botanical drug historically used for respiratory
enhancement, exemplifies this potential through its anti-inflammatory,
antioxidant, and cellular protective properties ([56]Ratan et al.,
2021). One notable metabolite, ginsenoside Rg3 (G-Rg3), demonstrates
potential efficacy in mitigating hypoxic lung injury ([57]Bae et al.,
2014; [58]Chen et al., 2010; [59]Wang et al., 2016). Its mechanism of
action involves promoting cell survival by activating the PI3K/AKT
signaling pathway ([60]Yang et al., 2018). Our previous study
demonstrated that intraperitoneal administration of G-Rg3 (15/30 mg/kg)
significantly mitigated acute mountain sickness (AMS) in C57BL/6 mice
through ferroptosis regulation ([61]Liu et al., 2025a). These
advancements not only underscore the scientific significance of
traditional medicinal resources but also offer a theoretical foundation
for developing plant-based therapies against high-altitude hypoxia.
Oxidative stress plays a central role in the pathogenesis of HAPE.
Under normoxic conditions, endogenous antioxidant systems, including
catalase, glutathione (GSH), and superoxide dismutase (SOD), maintain
redox balance by neutralizing reactive oxygen species (ROS)
([62]Diaz-Vivancos et al., 2015; [63]Sies, 2017). Hypoxia disrupts this
balance through two main mechanisms: impaired electron transport chain
function due to oxygen deficiency increases ROS production ([64]Gaur et
al., 2021), while simultaneous depletion of antioxidant reserves
exacerbates oxidative damage to lipids, proteins, and DNA ([65]Wang et
al., 2022b). This dual disruption creates a self-perpetuating cycle of
oxidative injury, eventually overwhelming cellular repair mechanisms
and triggering redox imbalance ([66]Yu et al., 2021).
The resulting oxidative stress induces ferroptosis, an iron-dependent
form of programmed cell death marked by three key features: (1)
membrane rupture via lipid peroxidation cascades, (2) intracellular
iron overload, and (3) failure of the glutathione peroxidase 4 (GPX4)
system ([67]Dixon et al., 2012; [68]Yang and Stockwell, 2016).
Mechanistically, hypoxia-induced ROS overproduction initiates lipid
peroxidation through Fenton reactions, while iron accumulation
amplifies oxidative damage by catalyzing hydroxyl radical formation
([69]Djulbegovic and Uversky, 2019). These processes compromise
membrane integrity and inactivate iron-regulatory proteins,
establishing a pathological feedback loop. Recent evidence links
ferroptosis to the activation of Phosphoinositide 3-kinase/Protein
Kinase B (PI3K/AKT) and Mitogen-Activated Protein Kinase (MAPK)
signaling pathways, which are involved in both cellular survival
decisions and hypoxic adaptation ([70]Gao et al., 2025; [71]Guo et al.,
2022). Notably, preclinical models demonstrate ferroptosis involvement
in hypoxia-induced organ damage, including neurological impairment
following acute altitude exposure ([72]Han et al., 2025), suggesting
its potential as a therapeutic target in HAPE pathophysiology.
In view of the aforementioned research context, our study puts forward
the hypothesis that G-Rg3 may prevent HAPE by suppressing ferroptosis.
To verify this hypothesis, we developed a HAPE model induced by
hypobaric hypoxia in C57BL/6 mice. The therapeutic efficacy of G-Rg3 on
HAPE was assessed using multiple approaches, including quantitative
analysis of pulmonary edema, histopathological scoring, profiling of
inflammatory cytokines, and evaluation of oxidative stress biomarkers.
To investigate the mechanistic regulation of ferroptosis by G-Rg3, we
employed an integrated strategy combining computational techniques
(network pharmacology, molecular docking, and molecular dynamics
simulations) with experimental validation methods: immunofluorescence
detection of ferroptosis markers, transmission electron microscopy for
mitochondrial ultrastructure analysis, and Western blot examination of
PI3K/AKT signaling pathway components.
This holistic approach not only facilitates the identification of
therapeutic targets involved in ferroptosis-driven progression of HAPE
but also establishes a theoretical foundation for G-Rg3 as a promising
candidate for treating this condition.
2 Materials and methods
2.1 Materials
Ginsenoside Rg3 (G-Rg3, ≥99.15% purity, Cat.14197-60-5) was procured
from Must Biotechnology (Chengdu, China). The PI3K/AKT pathway
inhibitor LY294002 (Cat.S1105) and all analytical-grade chemicals were
obtained from Selleck Chemicals (United States) and commercial
suppliers, respectively. Oxidative stress parameters were measured
using Nanjing Jiancheng kits: Malondialdehyde (MDA, A003-1), GSH
(A006-2-1), SOD (A001-3), and tissue iron quantification (A039-2-1).
Proinflammatory cytokines Interleukin-1 beta (IL-1β, ZC-37974W),
Interleukin 6 (IL-6, ZC-37988W), and Tumor necrosis factor-alpha
(TNF-α, ZC-39024W) were analyzed with Zhucai Biotechnology ELISA kits
(Shanghai).
Antibody Specifications Immunoblotting employed the following primary
antibodies: Abcam (UK): Vascular Endothelial Growth Factor (VEGF,
ab32152), Hypoxia-inducible factor 1 alpha (HIF-1α, ab179483), Nuclear
Factor Erythroid 2-Related Factor 2 (Nrf2, ab92946), Heme Oxygenase-1
(HO-1, ab68477), GPX4 (ab125066), Ferritin light chain (FLC, ab75973),
Transferrin receptor gene (TFRC, ab269513), PI3K/AKT pathway components
(PI3K ab191606, AKT ab185633, p-PI3K ab182651, p-AKT ab192623), β-actin
(ab227387). Cell Signaling Technology (United States): Ferritin Heavy
Chain 1 (FTH1, #4393S), Cyclooxygenase-2 (COX2, #12282S). ABclonal
(Wuhan): Solute Carrier Family 7 Member 11 (SLC7A11, A2413) Secondary
detection used HRP-conjugated goat anti-rabbit IgG (BF03008, Biodragon
Biotech).
All supplementary chemicals meeting analytical-grade specifications.
2.2 Animals
Sixty-three male C57BL/6 mice (6–8 weeks, 18–22 g, specific
pathogen-free) were sourced from Chengdu Dasuo Biological Technology
(Certification SCXK 2022–0345). Following 7-day acclimation under
controlled conditions (22°C ± 1°C, 55% ± 5% humidity; 12-hr light/dark
cycle), animals received standardized feeding with autoclaved water and
chow ([73]Zhang et al., 2025a). The study received ethical approval
(2022-18) from Chengdu University of Traditional Chinese Medicine’s
Animal Welfare Experimental Center, with all procedures complying with
institutional ethics guidelines and national welfare legislation.
2.3 Experimental design
Animals were randomly divided into five experimental subgroups (n = 7
per group): (1) Sham control, (2) G-Rg3 monotherapy (30 mg/kg), (3)
HAPE model, (4) HAPE + G-Rg3-L (15 mg/kg), and (5) HAPE + G-Rg3-H
(30 mg/kg), with G-Rg3 dosages validated by prior pharmacological
studies ([74]Cheng and Li, 2016; [75]Heinrich et al., 2020; [76]Liu et
al., 2025a). While Sham and HAPE groups received intraperitoneal (i.p.)
phosphate-buffered saline (PBS), other cohorts were administered G-Rg3
via i. p. injection for 72 consecutive days. To investigate
PI3K/AKT-ferroptosis regulatory mechanisms, an additional cohort of 28
mice underwent randomization into four groups: (1) HAPE baseline, (2)
HAPE + G-Rg3-H (30 mg/kg), (3) HAPE + LY294002 (5 mg/kg PI3K/AKT
inhibitor), and (4) HAPE + G-Rg3-H/LY294002 combinatorial therapy.
LY294002 formulations utilized a solvent system containing 50%
distilled water, 40% PEG300, 5% Tween 80, and 5% DMSO. All
interventions employed standardized i. p. administration protocols
(5 mL/kg injection volume, daily dosing over 3 days).
2.4 HAPE modelling
To replicate the HAPE condition in mice, we employed a hypobaric
hypoxic chamber model ProOx-830 from Tawang Intelligent Technology
(Shanghai, China), following the protocols outlined in our previous
research and relevant publications ([77]Ma et al., 2020; [78]Tan et
al., 2020; [79]Wang et al., 2022b). After a 3-day administration
period, the mice were transferred to a hypobaric hypoxia chamber set at
an altitude of 6000 m, with an oxygen partial pressure of 9.6 kPa,
humidity of 60%, and temperature of 20°C. The animals were rapidly
elevated to this altitude within 5 min at a rate of 20 m per second and
maintained there for 48 h.
Following the 48-h modeling period, the altitude was gradually adjusted
to normal levels, and then the mice were taken out of the chamber and
euthanized via an i. p. injection of sodium pentobarbital. Following
euthanasia, blood was collected from the abdominal aorta for serum
extraction, bronchoalveolar lavage fluid (BALF) was obtained, and lung
tissues were sectioned for additional analyses. The experimental
protocol for the HAPE animal model is depicted in [80]Figure 1.
FIGURE 1.
[81]Diagram illustrating a research process involving mice. Mice
undergo G-Rg3 pretreatment for three days, subjected to hypobaric
hypoxia, followed by sample analysis. Methods include H&E/TEM, Western
blot for oxidative stress and cytokines, and IF for tissue iron.
Samples are collected from lungs and analyzed post-euthanasia.
[82]Open in a new tab
The flowchart of HAPE animal experiments.
2.5 Histology of lung tissue
Post-euthanasia pulmonary specimens underwent standardized
histoprocessing: The right middle lung lobe was immediately harvested,
cleared of extraneous tissue, and immersion-fixed in ice-cold 4%
paraformaldehyde (12 h, 4°C). Sequential ethanol dehydration preceded
paraffin embedding, with resultant blocks sectioned at 4 μm thickness.
Following hematoxylin-eosin (H&E) staining, slides were imaged using an
OLYMPUS BX41 microscope. Two professional pathologists blindly scored
the degree of lung tissue damage using the McGuigan pathology scoring
method ([83]Faller et al., 2012; [84]McGuigan et al., 2003).
2.6 Lung wet/dry (W/D) weight ratio
The right upper lung lobe was surgically harvested post-euthanasia for
hydrostatic evaluation. Fresh tissue mass (W) was immediately recorded
before dehydration at 60°C until mass stabilization (D), enabling
calculation of the W/D ratio - a validated indicator of alveolar fluid
accumulation severity in hypobaric pulmonary edema models.
2.7 Cytokine content in BALF
Following right lung ligation, the left lung underwent dual
bronchoalveolar lavage procedures utilizing 0.2 mL ice-cold PBS
administered via tracheal cannulation. Pooled lavage fluid was
centrifuged (12,000×g, 10 min, 4°C) to separate cellular components,
with the resultant supernatant aliquoted for cryopreservation at −80°C
pending cytokine analysis. Quantitative assessment of IL-1β, IL-6, and
TNF-α concentrations in BALF was conducted through enzyme-linked
immunosorbent assay (ELISA) following manufacturer-specified protocols
(Thermo Fisher Scientific).
2.8 Oxidative stress and iron content in lung tissues
The right posterior lobe of the lung was surgically removed and
perfused with sterile saline prior to mechanical homogenization in PBS.
Cellular debris was eliminated by centrifugation at 12,000×g for 15 min
at 4°C, allowing us to collect clear supernatants for further
biochemical evaluation. Key markers of oxidative stress, including MDA
levels, SOD activity, and GSH concentrations, as well as parameters
related to iron metabolism, were quantified using commercially
available assay kits. Absorbance values were measured using a Thermo
Scientific Varioskan LUX microplate reader. To ensure accurate
quantitative comparisons across samples, total protein content was
normalized using the bicinchoninic acid (BCA) protein assay method.
2.9 Western blotting
The right inferior pulmonary lobe was immediately excised from
euthanized mice and flash-frozen in liquid nitrogen prior to
cryostorage at −80°C for subsequent molecular characterization. Tissue
homogenates prepared in RIPA buffer containing protease/phosphatase
inhibitors (1 mM PMSF) underwent centrifugation (12,000×g, 10 min,
4°C), with supernatants subjected to BCA protein quantification (Pierce
Biotechnology) per manufacturer specifications. Aliquots (30 μg)
underwent electrophoretic separation on 8%–15% gradient SDS-PAGE gels
(Bio-Rad Laboratories) followed by semi-dry transfer to PVDF membranes.
Post-blocking with 5% BSA/TBST (1 h, 25°C), membranes were probed with
primary antibodies at 4°C (16 h) targeting: VEGF (1:2000), HIF-1α
(1:1000), Nrf2 (1:1000), GPX4 (1:5000), HO-1 (1:10,000), FLC (1:1000),
TFRC (1:5000), PI3K (1:1000), AKT (1:2000), p-PI3K (1:800), p-AKT
(1:1000), FTH1 (1:1000), COX2 (1:1000), SLC7A11 (1:2000) and β-actin
(1:10,000)as normalization control. Post-TBST washes, HRP-conjugated
secondary antibodies (1:5000, 2 h, 25°C) enabled chemiluminescent
detection via ECL substrate, with band visualization (GelView 6000Plus)
and densitometric quantification (Image-Pro Plus 6.0) relative to
β-actin expression.
2.10 Transmission electron microscopy (TEM)
The excised apex of the right pulmonary lobe underwent dual fixation
protocol involving primary stabilization in 3% glutaraldehyde (24 h,
4°C) followed by secondary osmication with 1% osmium tetroxide (2 h,
25°C). Processed through acetone-gradient dehydration and epoxy resin
(Epon 812) polymerization, ultrastructural specimens were microtomed at
70 nm thickness. Transmission electron microscopy was conducted using a
JEOL JEM-1400-FLASH system operating at 120 kV, with contrast
enhancement achieved through sequential uranyl acetate and lead citrate
staining regimens.
2.11 Immunofluorescence staining
Paraffin-embedded lung tissue sections (4 μm) underwent sequential
processing through xylene deparaffinization and graded alcohol
rehydration. For antigen retrieval, heat-mediated treatment was
performed in 0.01 M citrate buffer (pH 6.0) for 10 min. Subsequent
pre-treatment included 15-min endogenous peroxidase blockade with 0.3%
H[2]O[2] and 20-min non-specific binding inhibition using 10% normal
goat serum. Primary antibody incubation proceeded overnight at 4°C with
anti-GPX4 (1:100) and anti-p-PI3K (1:100), followed by 30-min room
temperature exposure to species-matched secondary antibodies. Nuclear
counterstaining was achieved through 10-min DAPI application (1 μg/mL)
at 25°C, with residual dye removal via triple PBS washing (5 min each).
Fluorescent signal acquisition utilized a Nikon C2 confocal system
(Tokyo, Japan), with subsequent quantitative analysis performed using
ImageJ (NIH, United States) for fluorescence intensity measurements.
2.12 Network pharmacology
2.12.1 Target genes acquisition
The identification of G-Rg3 target genes incorporated a multi-platform
strategy utilizing both computational prediction tools and
pharmacological databases. Initial screening was performed through the
Encyclopedia of Traditional Chinese Medicine (ETCM,
[85]http://www.tcmip.cn/ETCM), TargetNet
([86]http://targetnet.scbdd.com), Swiss Target Prediction
([87]http://www.swisstargetprediction.ch), and PharmMapper
([88]http://www.lilab-ecust.cn/pharmmapper). To compensate for
potential database limitations, literature-curated targets from prior
ligand-receptor interaction studies were integrated ([89]Wang et al.,
2022a). Concurrently, HAPE-associated genes were systematically
collated from four disease-specific repositories: GeneCards
([90]https://www.genecards.org), OMIM ([91]https://omim.org), CTD
([92]http://ctdbase.org), and DisGeNET ([93]http://www.disgenet.org).
All identified genes underwent nomenclature standardization via a
bioinformatics pipeline combining UniProt ([94]https://www.uniprot.org)
for sequence annotation and STRING ([95]https://string-db.org) for
orthology mapping, ensuring compliance with HUGO Gene Nomenclature
Committee guidelines ([96]Yin et al., 2024).
2.12.2 Development of the protein-protein interaction (PPI) network
Following deduplication, gene set intersections between pharmacological
targets and pathological associations were graphically represented
through Venn diagram construction. Hub gene identification employed PPI
network modeling via the STRING platform (organism: Homo sapiens;
confidence threshold ≥0.7). The resultant interactome data underwent
advanced topological analysis using Cytoscape v3.6.2, facilitating
three-dimensional visualization of the metabolite-target-disease triad
and systematic interrogation of network architecture through integrated
bioinformatics tools.
2.12.3 Gene pathway analysis
Following the identification of key genes in the network through
topological analysis, a multi-level functional assessment was carried
out using an integrated bioinformatics approach (with the
clusterProfiler package). This framework executed Gene Ontology (GO)
enrichment analysis for cellular process annotation and Kyoto
Encyclopedia of Genes and Genomes (KEGG) pathway mapping to
contextualize biological systems simultaneously. As a result, it
enabled a comprehensive functional evaluation connecting molecular
functions, biological processes, and pathways.
2.13 Molecular docking
2.13.1 Molecular acquisition
The PI3Kγ structural coordinates for molecular docking were acquired
from the RCSB Protein Data Bank (accession code 3ML9) ([97]Burley et
al., 2017), while the bioactive conformation of Ginsenoside Rg3 was
obtained from PubChem (CID: 12855989) and energetically optimized
through MMFF94 molecular mechanics refinement ([98]Sunghwan et al.,
2016).
Molecular docking utilized AutoDock Vina version 1.2, with receptor
proteins preprocessed in PyMOL 2.5 to remove water, salt ions, and
other small molecules beforehand ([99]Eberhardt et al., 2021). The
central coordinates for the docking box were determined by the
center_of_mass.py plugin, using the active site as a reference, with
each side of the box being 22.5 Å long. Moreover, all preprocessed
small molecules and receptor proteins were altered into the PDBQT
format necessary for docking with AutoDock Vina 1.2, utilizing ADFR
Suite 1.0 ([100]Pradeep et al., 2015). The global docking search
exhaustiveness was configured to 32, whereas the remaining parameters
were kept at their default values. The highest-ranked docking
conformation was selected as the bound structure for follow-up
molecular dynamics simulations in this investigation.
2.13.2 Molecule dynamics
Informed by the docking results, complexes derived from small molecules
and proteins served as the initial configurations for extensive
molecular dynamics simulations, which were performed using the AMBER 22
software ([101]Salomon-Ferrer et al., 2013). Prior to initiating the
simulations, the partial charges of the small molecules were computed
using the Antechamber module and the Hartree-Fock (HF) SCF/6-31G*
method via the Gaussian 09 software ([102]Frisch et al., 2009;
[103]Wang et al., 2005). For the small molecules, the GAFF2 force field
was utilized, while the ff14SB force field was applied for the proteins
([104]Maier et al., 2015; [105]Wang J. et al., 2024; [106]Wang X. et
al., 2004). For each system configuration, the LEaP module was employed
to add hydrogen atoms. A truncated octahedral TIP3P water box, with a
10 Å buffer around the system, was then positioned. Additionally,
Na^+/Cl^− ions were introduced to neutralize any charge imbalances
([107]Mark and Nilsson, 2001). Ultimately, the topology and parameter
files needed for the simulations were produced.
Using the AMBER 22 software package, molecular dynamics simulations
were executed ([108]Salomon-Ferrer et al., 2013). An energy
minimization process was applied to the system before initiating the
simulation that consisted of 2500 iterations of steepest descent,
succeeded by a further 2500 iterations employing the conjugate gradient
technique. Upon finishing the energy minimization, the system underwent
a controlled temperature increase from 0 K to 298.15 K over 200
picoseconds at a uniform rate, maintaining a constant volume throughout
this process. To ensure a homogeneous dispersion of solvent molecules
within the solvent box and maintain a temperature of 298.15 K, an NVT
ensemble simulation was performed for a duration of 500 picoseconds.
Subsequently, to stabilize the system, an equilibrium simulation under
NPT conditions was carried out for an additional 500 picoseconds
throughout the entire system. The composite system underwent an
extended simulation lasting 100 nanoseconds within an NPT ensemble,
with periodic boundary conditions applied consistently. During this
period, non-bonded interactions were truncated at a cutoff distance of
10 Å. Long-range electrostatic interactions were evaluated using the
Particle Mesh Ewald (PME) method ([109]Sagui and Darden, 1999). The
SHAKE algorithm was utilized to fix the lengths of hydrogen bonds
([110]Kräutler et al., 2015), while Langevin dynamics with with a
collision frequency of γ = 2 ps^-1 was used for temperature control
([111]Larini et al., 2007). The pressure was maintained at 1 atm,
integration time steps were established at 2 femtoseconds, and
trajectory data were captured every 10 picoseconds for further
analysis.
2.13.3 MM/GBSA binding free energy calculation
The MM/GBSA method was applied to evaluate the binding free energy
between the protein and ligand across all systems ([112]Chen et al.,
2020; [113]Genheden and Ryde, 2015; [114]Hou et al., 2011;
[115]Rastelli et al., 2010). Acknowledging that extended MD simulations
can significantly affect the accuracy of MM/GBSA outcomes is crucial
([116]Hou et al., 2011). Hence, this investigation used an MD
trajectory covering 90–100 nanoseconds for the calculations, based on
the formula described below:
[MATH:
ΔGbi
nd=ΔGcomplex
– ΔGreceptor+ΔGligand :MATH]
[MATH: =ΔEinternal+ΔEVDW+ΔEelec+ΔGGB+ΔGSA
:MATH]
In the formula, internal energy is denoted by ΔE[internal], van der
Waals interaction by ΔE[VDW], and electrostatic interaction by
ΔE[elec]. The internal energies, which include bond energy (E[bond]),
angle energy (E[angle]), and torsion energy (E[torsion]), are
collectively known as solvation free energy. Here, ΔG[GB] represents
polar solvation free energy, while ΔG[SA] denotes non-polar solvation
free energy. In this study, the ΔG[GB] model developed by Nguyen et al.
(igb = 2) was utilized for calculations ([117]Nguyen et al., 2013). To
calculate the non-polar solvation free energy (ΔG[SA]), the surface
tension (γ) was multiplied by the solvent-accessible surface area
(ΔSASA), given by the formula ΔG[SA] = 0.0072 × ΔSASA ([118]Weiser et
al., 1999). Entropy change calculations were not included in this study
because they require significant computational resources and have low
accuracy ([119]Chen et al., 2020; [120]Hou et al., 2011).
2.14 Statistical analyses
This research employed GraphPad Prism 8 (San Diego, California, United
States) for data analysis and figure generation. For normally
distributed datasets, one-way analysis of variance was conducted. The
findings are expressed as mean ± SEM. A threshold of P < 0.05 was used
to establish statistical significance.
3 Results
3.1 Effects of G-Rg3 pre-treatment on the prevention of HAPE
In instances of acute high-altitude pulmonary edema, patients typically
exhibit clinical symptoms such as breathing difficulties, cyanosis of
the mucous membranes, delayed reactions, and reduced activity levels
([121]Gatterer et al., 2024). Consequently, we monitored the overall
condition of mice to evaluate the pathological impacts of low-pressure
hypoxia on the body, as well as the effectiveness of drug
interventions. In their baseline state, all mice displayed normal
physiological signs, characterized by glossy fur, steady respiration,
and a heightened stress response. Following exposure to low-pressure
hypoxia, distinct variations emerged among the groups: both the NC
group and the G-Rg3 groups maintained normal activity levels, with
regular breathing and feeding patterns; in contrast, the HAPE group
demonstrated classic pathological features, including mucosal cyanosis,
rapid breathing, and lethargic responses. The two dosage levels of
G-Rg3 exhibited dose-related improvements, particularly the 30 mg/kg
group, which showed markedly enhanced recovery in terms of activity,
mucosal coloration, and respiratory rhythm.
Histopathological analysis demonstrated that lung tissues in both the
sham operation group and the G-Rg3 pre-treatment group remained
structurally intact. Conversely, lung tissues in the HAPE group
exhibited significant damage, including thickened alveolar walls,
widened alveolar septa, alveolar cavity congestion, and prominent
inflammatory cell infiltration ([122]Figure 2A). These observations
confirm the successful establishment of the rat model for high-altitude
pulmonary edema. Following G-Rg3 pre-treatment, the extent of
inflammatory cell infiltration, hemorrhage, and alveolar wall
thickening was dose-dependently attenuated. Semi-quantitative
assessment of lung tissue injury revealed that G-Rg3 pre-treatment
markedly decreased the lung injury score associated with HAPE
([123]Figure 2B) ([124]McGuigan et al., 2003). Additionally, the
wet/dry weight ratio of lung tissue was substantially elevated in the
HAPE group compared to the sham operation group. However, this ratio
was significantly lowered following G-Rg3 pre-treatment ([125]Figure
2C).
FIGURE 2.
[126]Panel A shows lung tissue histological sections under different
conditions: Sham, G-Rg3, HAPE, G-Rg3-L, and G-Rg3-H, at magnifications
of 200x and 400x. Panels B and C present bar graphs illustrating the
lung injury score and W/D ratio for these conditions. Panel D includes
Western blot images for VEGF and HIF-1α with bar graphs indicating
their relative expression levels.
[127]Open in a new tab
Effects of G-Rg3 pre-treatment on the prevention of HAPE. (A)
H&E-stained lung sections (upper: ×200, 50 μm scale; lower: ×400, 20 μm
scale). (B) Semi-quantified lung injury scores. (C) Hydrostatic
imbalance via W/D weight ratio. (D) Hypoxic response markers VEGF and
HIF-1α by immunoblotting. Data represent mean ± SEM (n = 6 biological
replicates). One-way ANOVA with Tukey’s post hoc analysis: ^##/### P <
0.01/0.001 vs. Sham; ^*/**/*** P < 0.05/0.01/0.001 vs. HAPE controls.
As a pivotal transcription factor, HIF-1α modulates the expression of
genes involved in hypoxic responses and is crucial for the
physiological adaptation of organisms to hypoxic settings
([128]Diaz-Vivancos et al., 2015). Elevated levels of HIF-1α have been
conspicuously observed in individuals prone to HAPE under normoxic
conditions. Moreover, empirical evidence indicates that HIF-1α
contributes to the progression of HAPE through the regulation of its
target gene, VEGF. Western blotting showed increased expression of VEGF
and HIF-1α in the HAPE group, which was significantly attenuated
following G-Rg3 pre-treatment ([129]Figure 2D). The results showed that
G-Rg3 pre-treatment could effectively reduce the inflammation and
oxidative stress in the HAPE model.
3.2 Effects of G-Rg3 pre-treatment on oxidative stress and inflammatory
cytokines by HAPE
HAPE progression is mechanistically linked to acute immune-inflammatory
cascade activation, characterized by neutrophilic infiltration,
macrophage aggregation, and mediator-induced alveolar-capillary
hyperpermeability ([130]Lee et al., 2018; [131]Wang et al., 2022b). Our
experimental analyses revealed significant upregulation of
pro-inflammatory IL-1β, IL-6, and TNF-α in BALF during hypobaric
hypoxia exposure, with prophylactic G-Rg3 administration demonstrating
potent cytokine suppression efficacy ([132]Figure 3A). Concomitant
evaluation of pulmonary redox homeostasis identified characteristic
oxidative imbalance patterns: HAPE-induced samples exhibited heightened
lipid peroxidation (elevated MDA), compromised antioxidant defenses
(reduced SOD and GSH), and tissue iron overload - all metabolic
disturbances effectively normalized through G-Rg3 pre-treatment, as
evidenced by complete restoration of baseline oxidative stress markers
and iron homeostasis parameters ([133]Figure 3B).
FIGURE 3.
[134]Bar graphs labeled A and B show various biochemical parameters. A
displays IL-1β, IL-6, and TNF-α levels across five groups: Sham, G-Rg3,
HAPE, G-Rg3-L, and G-Rg3-H. B shows GSH, MDA, SOD, and tissue iron
levels in the same groups. Significant differences are indicated by
asterisks and hash marks. Each graph uses different y-axes units
relevant to the parameter measured.
[135]Open in a new tab
Effects of G-Rg3 pre-treatment on oxidative stress and inflammatory
cytokines by HAPE. (A) G-Rg3-mediated suppression of proinflammatory
cytokines (IL-1β/IL-6/TNF-α). (B) Antioxidant regulation through
SOD/GSH/MDA biomarkers. Data reflect mean ± SEM (n = 6 biological
replicates). One-way ANOVA with Tukey’s post hoc analysis: ^##/### P <
0.01/0.001 vs. Sham controls; ^*/** P < 0.05/0.01 vs. HAPE group.
3.3 Effects of G-Rg3 pre-treatment on ferroptosis in HAPE
Given that excessive oxidative stress responses can trigger ferroptosis
([136]Djulbegovic and Uversky, 2019), we next investigated the
involvement of ferroptosis in HAPE pathogenesis. Western blot analysis
indicated that the expression of anti-ferroptosis proteins, including
GPX4, Nrf2, HO-1, SLC7A11, FTH1, and FLC, was markedly reduced in model
tissues, whereas pro-ferroptosis proteins COX2 and TFRC exhibited
elevated expression ([137]Figure 4A). Notably, pre-treatment with G-Rg3
reversed these trends in a dose-dependent manner. To further elucidate
the cellular phenotypes linked to ferroptosis, we conducted TEM
analysis on lung tissue samples ([138]Figure 4B). Electron microscopic
observations revealed characteristic ultrastructural alterations
indicative of ferroptosis in type II alveolar epithelial cells from the
HAPE group, such as mitochondrial shrinkage, reduced cristae density,
and outer mitochondrial membrane damage. These changes were mitigated
following G-Rg3 pre-treatment. Based on these findings, we conclude
that G-Rg3 pre-administration alleviates HAPE by suppressing
ferroptosis.
FIGURE 4.
[139]The image contains two panels labeled A and B. Panel A shows
Western blot results and corresponding bar graphs for several proteins
(e.g., GPX4, Nrf2, HO-1) across different treatments (Sham, G-Rg3,
HAPE). The graphs display protein expression as a fold of Sham, with
significant differences indicated by asterisks. Panel B presents
electron microscopy images of tissue samples under various conditions:
Sham, G-Rg3, HAPE, G-Rg3-L, and G-Rg3-H, at magnifications of x6000 and
x25000. Red arrows in the images highlight specific structures or
changes.
[140]Open in a new tab
Effects of G-Rg3 pre-treatment on ferroptosis in HAPE. (A) Immunoblot
quantification of key ferroptotic regulators (GPX4/SLC7A11/FTH1). (B)
Ultrastructural evidence of mitochondrial pathology in AT2 cells (Red
arrows: cristae disruption; scale bars: 2 μm [×6,000], 500 nm
[×25,000]). Data represent mean ± SEM (n = 6 biological replicates).
One-way ANOVA with Tukey’s post hoc analysis: ^#/##/### P <
0.05/0.01/0.001 vs. Sham; ^*/**/*** P < 0.05/0.01/0.001 vs. HAPE group.
3.4 Network pharmacology prediction and molecular docking analysis of G-Rg3
pre-treatment in HAPE
Our multi-modal investigation combining network pharmacology and
computational modeling revealed G-Rg3’s mechanistic actions against
HAPE. Bioinformatic interrogation identified 1,285 HAPE-associated
targets and 248 G-Rg3-related genes, with 52 shared candidates forming
the core interaction network ([141]Figures 5A–C). Protein interactome
mapping prioritized five hub genes (TNF, IL6, AKT1, IL1B, ESR1) through
topological centrality analysis ([142]Figure 5D). Functional enrichment
demonstrated these mediators coordinate apoptotic regulation
(biological process), granular secretory mechanisms (cellular
component), and nuclear receptor activation (molecular function), with
pathway analysis implicating PI3K/AKT signaling and endocrine
resistance as primary therapeutic targets ( [143]Figures 5E,F).
FIGURE 5.
[144]Panel A shows a Venn diagram comparing Ginsenoside Rg3 and HAPE,
with an overlap of 54 entities. Panel B features a complex network
diagram of interacting proteins. Panel C displays a network diagram
centered around Ginsenoside Rg3 with connected genes. Panel D is a
dense gene interaction network. Panel E presents a bar graph detailing
various biological processes and their significance, while Panel F is a
dot plot illustrating KEGG pathway enrichment, with color and size
indicating statistical significance and gene count.
[145]Open in a new tab
Network pharmacology prediction of G-Rg3 pre-treatment in HAPE. (A)
Pharmacological target convergence between G-Rg3 and HAPE-associated
genes. (B) Protein-protein interaction (PPI) network of hub targets,
with node size/color intensity reflecting interaction centrality. (C)
Tripartite network mapping G-Rg3-target-disease-pathway
interrelationships. (D) High-fidelity subnetwork of critical protein
interactions, highlighting topological significance through node degree
gradation. (E) GO functional annotation categorizing targets into
biological processes, molecular functions, and cellular components. (F)
KEGG pathway enrichment analysis of core therapeutic targets.
Further molecular docking and dynamics simulations revealed the
interaction between Ginsenoside_Rg3 and PI3K. [146]Figure 6A
illustrates the binding mode of the PI3K_Ginsenoside_Rg3 complex,
demonstrating eight hydrogen bonds (including VAL-803 and LYS-807) and
hydrophobic interactions within the active pocket, with a docking
energy of −8.48 kcal/mol ([147]Table 1). The 100 ns molecular dynamics
simulations confirmed binding stability, showing low RMSD (<2 Å) and
RMSF (<2 Å) values ([148]Figures 6B,C) ([149]Chen et al., 2024).
MM-GBSA calculations yielded a binding energy of −34.10 ± 3.10 kcal/mol
([150]Table 2), primarily driven by van der Waals and electrostatic
interactions. Ten key residues (including TRP-812 and ILE-963)
contributed significantly to binding ([151]Figure 6D). Sustained
hydrogen bonding (4 bonds/frame on average, [152]Figure 6E) validated
the high-affinity binding potential of this interaction.
FIGURE 6.
[153]Five panels show molecular dynamics results. Panel A depicts a
protein-ligand interaction with highlighted amino acids. Panel B is a
graph of ligand RMSD over time, measuring conformational stability.
Panel C illustrates protein RMSF across residues, indicating
flexibility. Panel D is a bar chart showing binding free energy changes
by residue. Panel E displays the number of hydrogen bonds over time,
examining stability fluctuations.
[154]Open in a new tab
Molecular docking analysis of G-Rg3 pre-treatment in HAPE. (A)
Molecular docking pose visualization: Gold stick = G-Rg3; Cyan cartoon
= PI3K backbone; Blue dashes = hydrogen bonds; Yellow surfaces =
hydrophobic interfaces. (B) Conformational stability assessment via
RMSD. (C) Per-residue flexibility mapping (RMSF) with mobile loop
regions. (D) Binding energy hotspot identification through MM-PBSA
decomposition (Top 10). (E) Time-dependent hydrogen bond population
analysis.
TABLE 1.
Scores for the binding affinity of the complexes.
Target_name Ligand_name Docking_score (kcal/mol)
3ml9-PI3K Ginsenoside_Rg3 −8.484
[155]Open in a new tab
TABLE 2.
MM/GBSA predictions for binding free energies and energy components
(kcal/mol).
System name PI3K/Ginsenoside_Rg3
ΔE [vdw] −58.96 ± 2.20
ΔE [elec] −38.11 ± 2.68
ΔG[GB] 71.67 ± 5.68
ΔG[SA] −8.70 ± 0.68
ΔG[bind] −34.10 ± 3.10
[156]Open in a new tab
ΔE[vdW]: van der Waals energy.
ΔE[elec]: electrostatic energy.
ΔG[GB]: electrostatic contribution to solvation.
ΔG[SA]: non-polar contribution to solvation.
ΔG[bind]: binding free energy.
According to the findings from network pharmacology and docking
analysis, G-Rg3 might mitigate HAPE and ferroptosis by stimulating the
PI3K/AKT signaling pathway via strong affinity binding with PI3K, which
was needed additional experimental validation.
3.5 G-Rg3 ameliorates HAPE by inhibiting ferroptosis through activation of
the PI3K/AKT pathway
To confirm the predictions of network pharmacology through biochemical
approaches, we conducted Western blot and immunofluorescence analyses.
The Western blot results showed that in mice exposed to high-altitude
hypobaric hypoxia-induced HAPE, the phosphorylation levels of PI3K and
AKT were substantially decreased. Nevertheless, G-Rg3 pre-treatment
markedly inhibited the reduction in phosphorylation levels of PI3K and
AKT ([157]Figure 7A). Immunofluorescence analysis further indicated
that in the HAPE group, the co-localization between PI3K activation and
GPX4 antioxidant enzyme expression was inhibited. In contrast,
prophylactic treatment with G-Rg3 successfully restored the expression
of these two biomarkers ([158]Figure 7B). Taken together, these
findings suggest that G-Rg3 can effectively activate the PI3K/AKT
signaling pathway.
FIGURE 7.
[159]Panel A shows Western blot results for proteins p-PI3K, PI3K,
p-AKT, AKT, and β-actin with bar graphs comparing their expression.
Panel B presents immunofluorescence images stained with DAPI, p-PI3K,
and GPX4 across different conditions: Sham, G-Rg3, HAPE, G-Rg3-L, and
G-Rg3-H. Each row represents a different stain, culminating in a merged
image.
[160]Open in a new tab
G-Rg3 ameliorates HAPE by inhibiting ferroptosis through activation of
the PI3K/AKT pathway. (A) Immunoblot quantification of PI3K, p-PI3K,
AKT, and p-AKT. (B) Confocal microscopy of merged p-PI3K (green)/GPX4
(red) fluorescence signals (scale bar: 20 μm, ×400). Data represent
mean ± SEM (n = 6 biological replicates). One-way ANOVA with Tukey’s
post hoc analysis: ^##/### P < 0.01/0.001 vs. Sham; ^*/*** P <
0.05/0.001 vs. HAPE.
3.6 G-Rg3 pre-treatment inhibition of ferroptosis was dependent on the
PI3K/AKT pathway
To determine whether the inhibitory action of G-Rg3 on ferroptosis in a
HAPE mouse model is dependent on the activation of the PI3K/AKT
signaling pathway, this study utilized the specific PI3K inhibitor
LY294002. Western blotting analysis demonstrated G-Rg3’s capacity to
enhance PI3K/AKT phosphorylation, an effect abrogated by concurrent
LY294002 administration ([161]Figure 8A). Subsequent ferroptotic
biomarker profiling revealed G-Rg3’s dual regulatory function:
coordinated upregulation of GPX4, Nrf2, HO-1, SLC7A11, and ferritin
complexes (FTH1/FLC) concurrent with suppression of TFRC/COX2
expression ([162]Figure 8B). Crucially, the inhibitor LY294002 negated
these modulatory effects. Complementary validation through
immunofluorescence and transmission electron microscopy corroborated
the molecular analyses, demonstrating restored subcellular architecture
and antioxidant enzyme localization ([163]Figures 8C,D). These findings
mechanistically establish PI3K/AKT signaling as the principal conduit
mediating G-Rg3’s anti-ferroptotic efficacy in HAPE pathophysiology.
FIGURE 8.
[164]The figure consists of four panels (A, B, C, and D). Panel A shows
Western blot results for proteins like p-PI3K, PI3K, p-AKT, and AKT,
and corresponding bar graphs comparing expression levels under
different treatments involving G-Rg3 and LY294002 in a HAPE context.
Panel B displays Western blots for proteins such as GPX4, Nrf2, HO-1,
and others, with bar graphs illustrating their expression levels under
various conditions. Panel C contains fluorescent microscopy images for
DAPI, p-PI3K, GPX4, and a merged view under different treatment
conditions. Panel D features electron microscopy images at
magnifications ×6000 and ×25000 comparing cellular ultrastructure
across treatments.
[165]Open in a new tab
G-Rg3 pre-treatment inhibition of ferroptosis was dependent on the
PI3K/AKT pathway. (A) Immunoblot quantification of PI3K, p-PI3K, AKT,
and p-AKT. (B) Ferroptosis biomarker analysis. (C) Confocal microscopy
of p-PI3K (green, Alexa Fluor 488) and GPX4 (red, Cy3) co-localization
in alveolar epithelia (scale bar: 20 μm; ×400). (D) Ultrastructural
ferroptosis markers: mitochondrial cristae dissolution (red arrows) in
type II alveolar epithelial cells (scale bars: 2 μm [×6,000]; 500 nm
[×25,000]). Data represent mean ± SEM (n = 6 biological replicates).
One-way ANOVA with Tukey’s post hoc analysis: ^*/**/*** P < 0.05/0.001
vs. HAPE, ^$/$$ P < 0.05/0.01 vs. G-Rg3-H group.
3.7 The protective effect of G-Rg3 pre-treatment on HAPE was mediated through
the PI3K/AKT signaling pathway
Histopathological evaluation of LY294002-treated lung tissues revealed
pronounced structural damage characterized by inflammatory
infiltration, alveolar wall thickening, and intra-alveolar congestion
([166]Figure 9A). These pathological changes were corroborated through
semi-quantitative scoring and edema quantification ([167]Figures 9B,C).
G-Rg3 pre-treatment notably mitigated these alterations, restoring
near-normal pulmonary architecture. Concurrent analysis demonstrated
LY294002-induced upregulation of inflammatory mediators and oxidative
stress markers, effects substantially attenuated by G-Rg3
administration ([168]Figures 9D,E). This integrated analysis confirms
PI3K/AKT signaling modulation as the central mechanism underlying
G-Rg3’s therapeutic efficacy against HAPE-associated pulmonary injury.
FIGURE 9.
[169]Histological images and bar graphs illustrate the effects of
treatments on lung tissue in different conditions. Panel A shows
stained lung tissue sections at two magnifications (200x and 400x)
under four conditions: HAPE, G-Rg3-H, LY294002, and G-Rg3-H+LY294002.
Panels B to E present bar graphs comparing lung injury scores,
wet-to-dry ratios, cytokine levels (IL-1, IL-6, TNF-α), and oxidative
stress markers (GSH, MDA, SOD, tissue iron) across the conditions, with
statistical annotations indicating significance levels.
[170]Open in a new tab
The protective effect of G-Rg3 pre-treatment on HAPE was mediated
through the PI3K/AKT signaling pathway. (A) H&E-stained lung sections
(upper: ×200, 50 μm scale; lower: ×400, 20 μm scale). (B)
Semi-quantitative histopathological assessment of HAPE. (C) Pulmonary
edema quantification (W/D ratio). (D) Analysis of the effect of
LY294002 on inflammatory markers (IL-1β, IL-6, and TNF-α) in BALF. (E)
Evaluation of LY294002’s influence on antioxidant enzymes (GSH, SOD)
and oxidative stress markers (MDA) as well as tissue iron levels in
lung tissues. Data represent mean ± SEM (n = 6 biological replicates).
One-way ANOVA with Tukey’s post hoc analysis: ^*/**/*** P < 0.05/0.001
vs. HAPE, ^$/$$/$$$ P < 0.05/0.01 < 0.001 vs. G-Rg3-H group.
4 Discussion
Recent years have witnessed increased human activity in high-altitude
regions, amplifying clinical demands for managing hypoxia-related
cardiopulmonary disorders. Particularly concerning is the prevalence of
HAPE. Our study employed an HAPE mouse model to systematically evaluate
the therapeutic potential of G-Rg3. Pretreated mice exhibited markedly
attenuated HAPE-associated pathological features, concurrent with
normalized HIF-1α overexpression and robust
anti-inflammatory/antioxidant responses. Further mechanistic
investigations identified ferroptosis suppression as a critical
component of G-Rg3’s action. Central to this protective effect is the
metabolite’s activation of PI3K/AKT signaling, as evidenced by
integrated network pharmacology analyses, computational modeling, and
targeted pathway inhibition experiments. Investigations reveal that the
traditional botanical metabolite G-Rg3 holds promise for high-altitude
HAPE treatment through its multi-action regulatory effects. By
targeting acute pathological changes and alleviating hypoxia-induced
secondary injuries, it offers significant pharmacological insights
supporting the clinical translation of plant-based therapies.
The complex physiological demands imposed by high-altitude hypoxia
present significant challenges. A key mechanism driving the development
of HAPE involves the deregulation of HIF-1α and VEGF ([171]Wang et al.,
2022b). Under conditions of reduced pressure and oxygen availability,
cells initiate an adaptive response by stabilizing HIF-1α, which
subsequently triggers elevated VEGF expression. Both factors are
crucial in the progression of HAPE. Studies have shown that hypoxic
conditions amplify the activity of HIF-1α, positioning it as the
principal regulator of genes responsive to hypoxia—genes that are vital
for cellular survival. In pathological scenarios, the heightened
induction of VEGF, mediated by HIF-1α, fosters irregular angiogenesis
and accelerates disease progression. Our experimental data corroborate
these findings, indicating that increased levels of HIF-1α and VEGF
intensify the pathological processes associated with pulmonary edema.
Importantly, this investigation expands on earlier studies ([172]Ahmmed
et al., 2019; [173]Lv et al., 2023) by illustrating the suppressive
impact of G-Rg3 on the expression levels of HIF-1α and VEGF in HAPE
models.
Simultaneously, pro-inflammatory cytokines such as TNF-α, IL-1β, and
IL-6 have been identified as pivotal factors influencing the dynamic
changes in white blood cells during HAPE, which play critical roles in
regulating the proliferation, migration, and differentiation of immune
cells ([174]Tanaka et al., 2014) In mouse models, the systemic levels
of these cytokines were significantly elevated following HAPE
induction. These observations align with previous reports associating
cytokine storms with inflammatory lung injury ([175]Arya et al., 2013).
The ability of G-Rg3 to modulate these pathways underscored its dual
therapeutic potential, illustrating its influence on the inflammatory
cascade within the pathophysiology of HAPE.
Hypobaric hypoxia in HAPE induces mitochondrial dysfunction and
excessive ROS generation (superoxide, hydrogen peroxide), driving
oxidative damage and depleting antioxidant reserves ([176]Irarrázaval
et al., 2017). his oxidative cascade triggers membrane lipid
peroxidation, quantified by elevated MDA levels ([177]Weismann et al.,
2011), while concurrently suppressing SOD activity which is a critical
antioxidant enzyme ([178]Wang et al., 2018)). The hypoxia-induced
inhibition of system Xc^− (SLC7A11/SLC3A2) further diminishes GSH
synthesis through cysteine deprivation, impairing both ROS
neutralization and GPX4-mediated ferroptosis prevention ([179]Endale et
al., 2023; [180]Mailloux et al., 2013). Our findings reveal a
pathological feedback loop in HAPE: diminished SOD activity facilitates
superoxide accumulation, heightened MDA concentrations signify
accelerated lipid peroxidation, and GSH depletion undermines
antioxidant mechanisms, all of which contribute to pulmonary edema.
Importantly, pre-treatment with G-Rg3 can counteract these disruptions,
restore SOD function and GSH levels in lung tissue, and lower MDA,
effectively interrupting this disease progression pathway.
Emerging evidence identifies oxidative stress-mediated iron metabolism
dysregulation and ferroptosis as pivotal mechanisms in HAPE. The
pathogenic triad of ROS accumulation, lipid peroxidation, and iron
overload which established contributors to pulmonary disorders, is
exacerbated under hypoxic conditions ([181]Wang et al., 2024).
Mechanistically, hypoxia inhibits Nrf2, suppressing key iron regulators
(SLC7A11, HO-1, FTH1) and weakening antioxidant defenses ([182]Loboda
et al., 2016; [183]Su et al., 2025; [184]Zhang S. et al., 2025). This
suppression triggers two detrimental pathways: 1) impaired cystine
transport through System Xc^− disrupts glutathione synthesis and GPX4
antioxidant activity; 2) reduced HO-1/FTH1 expression enhances free
iron toxicity via Fenton reactions ([185]Basiouny et al., 2025; [186]Fu
et al., 2022; [187]Qian et al., 2024). Concurrent elevation of
ferroptosis markers TFRC and COX2 further drives pathological iron
uptake and inflammatory lipid damage ([188]Liu M. J. et al., 2025;
[189]Zhuang et al., 2024). Our experimental data validate this
mechanistic model in HAPE progression. Diseased mice exhibited
characteristic ferroptosis signatures: depressed anti-ferroptosis
proteins (GPX4, Nrf2, HO-1, SLC7A11, FTH1/FLC) alongside elevated TFRC,
COX2, and tissue iron. Crucially, G-Rg3 treatment ameliorated these
imbalances, reactivating ferroptosis-suppressing pathways and restoring
iron homeostasis. This dual antioxidant-iron regulatory capability
positions G-Rg3 as a promising therapeutic agent targeting the
oxidative-ferroptosis axis in HAPE.
The PI3K/AKT signaling axis ([190]Liu et al., 2025c) functions as a
master regulator of cellular homeostasis, coordinating processes from
metabolic regulation to survival pathways ([191]Huang et al., 2022).
Mechanistically, PI3K initiates signaling by phosphorylating membrane
phosphatidylinositol to generate PIP3, which facilitates AKT membrane
recruitment and subsequent activation through phosphorylation events.
Activated AKT then orchestrates downstream biological responses through
its effector network ([192]Vanhaesebroeck et al., 2010). This pathway
exhibits dual ferroptosis-inhibitory capacities: (1) SREBP1-mediated
lipogenesis enhancement reduces membrane phospholipid peroxidation
susceptibility ([193]Zheng et al., 2024); (2) AKT-driven Nrf2
phosphorylation (Ser40) promotes KEAP1 dissociation, enabling nuclear
translocation to boost antioxidant defenses (glutathione synthesis) and
iron regulatory capacity ([194]Cheng et al., 2023). Our integrated
pharmacological investigation reveals G-Rg3’s therapeutic mechanism
through PI3K/AKT modulation. Network pharmacology prioritized this
pathway as the prime HAPE intervention target, corroborated by
computational modeling showing stable G-Rg3-PI3K binding through
hydrogen bonds and hydrophobic interactions. Immunoblot analyses
demonstrated G-Rg3’s capacity to restore depressed PI3K/AKT
phosphorylation in HAPE lungs. Crucially, co-treatment with LY294002
(PI3K inhibitor) abolished G-Rg3’s protective effects against both
inflammatory cytokines and ferroptosis markers (GPX4/HO-1
downregulation), conclusively validating pathway centrality.
This study identified G-Rg3 as a selective PI3K/AKT activator that can
alleviate ferroptosis in high-altitude pulmonary edema (HAPE) by
coordinating the regulation of lipid metabolism, iron homeostasis, and
oxidative defense. Four limitations need to be addressed when
establishing the therapeutic framework of G-Rg3: (1) The limitations of
the acute hypobaric model require gradient altitude studies that
simulate the progression of human HAPE; (2) Further in vitro studies
are needed to clarify the mechanism by which G-Rg3 prevents HAPE; (3)
Orthogonal validation through gene editing, co-immunoprecipitation
target identification, and surface plasmon resonance is necessary for
the confirmation of the PI3K-based pathway; (4) GLP-compliant
toxicological analyses (chronic toxicity, tissue-specific
pharmacokinetics) are crucial for clinical translation. These further
studies will provide a more solid theoretical basis for G-Rg3-based
intervention strategies for HAPE.
5 Conclusion
This research highlights that the active constituent of ginseng, G-Rg3,
can successfully mitigate acute high-altitude pulmonary edema triggered
by hypobaric hypoxia. This effect is achieved through the activation of
the PI3K/AKT signaling pathway and the cooperative modulation of
inflammatory responses, oxidative stress, and ferroptosis ([195]Figure
10). Importantly, the therapeutic timing for G-Rg3 includes preventive
administration up to 72 h prior to ascent, with optimal
efficacy-to-dose performance observed at a dosage of 30 mg/kg. Beyond
counteracting the HIF-driven inflammatory-oxidative cascade, focusing
on the PI3K/AKT regulatory axis offers a molecular foundation for
designing combination therapies. These underlying mechanisms indicate
that G-Rg3 could serve both as a standalone treatment in high-altitude
emergency care and leverage its distinctive iron homeostasis regulation
capabilities for managing chronic high-altitude illnesses or systemic
inflammation-related hypoxic conditions. This provides a comprehensive
approach for advancing traditional medicinal plants into modern
applications.
FIGURE 10.
[196]Illustration of cellular mechanisms influenced by hypoxia, showing
the role of G-Rg3 and pulmonary edema. Arrows indicate pathways from
hypoxia through cytokines, oxidative stress, and ferroptosis. Molecular
interactions involve PI3K/Akt, COX2, ROS, and various proteins like
FTH1 and SLC7A11. The diagram includes mitochondria-related processes
like the TCA cycle and Fenton reaction, highlighting oxidative stress
and cellular responses.
[197]Open in a new tab
Mechanistic Framework of G-Rg3-Mediated Pulmonary Protection in Murine
HAPE Models. Pharmacological preconditioning with G-Rg3 (30 mg/kg, i.
p.) ameliorates hypoxia-induced pulmonary edema through coordinated
ferroptosis-PI3K/AKT axis modulation.
Funding Statement
The author(s) declare that financial support was received for the
research and/or publication of this article. This study was financially
supported by National Natural Science Foundation of China (82205043),
China Association of Chinese Medicine Qiushi Project (2023-QNQS-11),
Natural Science Foundation of Sichuan Province (2024NSFSC1865,
2025ZNSFSC1851), Yunnan Provincial Science and Technology Department
(202301AZ070001-119, C012018005, Y0120180018), Sichuan Cadre Health
Care Research Project (2024-503), Research Project of Sichuan
Provincial Administration of Traditional Chinese Medicine (2023MS555),
Scientific Research Project of Health Commission of Chengdu
(WXLH202403114, WXLH202402022) and “Xinglin Scholars” Discipline Talent
Research Enhancement Program Postdoctoral Special Project
(QJRC2023011).
Data availability statement
The original contributions presented in the study are included in the
article/supplementary material, further inquiries can be directed to
the corresponding authors.
Ethics statement
The animal study was approved by the Medical Ethics Committee at
Chengdu University of Traditional Chinese Medicine (Approval number:
2022-18). The animal experiments followed 3R (reduction, replacement,
and refinement) principles, where all measures are taken to ensure
minimal discomfort and minimal suffering. The study was conducted in
accordance with the local legislation and institutional requirements.
Author contributions
YH: Writing – original draft. YW: Writing – original draft. HD: Writing
– original draft. DH: Writing – original draft. NJ: Writing – original
draft. ZS: Writing – original draft. ZW: Writing – review and editing.
MW: Writing – review and editing. TZ: Writing – review and editing.
Conflict of interest
The authors declare that the research was conducted in the absence of
any commercial or financial relationships that could be construed as a
potential conflict of interest.
Generative AI statement
The author(s) declare that no Generative AI was used in the creation of
this manuscript.
Publisher’s note
All claims expressed in this article are solely those of the authors
and do not necessarily represent those of their affiliated
organizations, or those of the publisher, the editors and the
reviewers. Any product that may be evaluated in this article, or claim
that may be made by its manufacturer, is not guaranteed or endorsed by
the publisher.
References