A randomized, open-label, single-dose, two-period, two-sequence, crossover study was performed with healthy Korean male and female subjects. This study was performed in accordance with the principles of the Declaration of Helsinki and Korean Good Clinical Practice guidelines. Informed written consent was obtained from each subject in advance. The study was approved by the Institutional Review Board of Kyung Hee University Hospital (KHGIRB-21-419). Eleven healthy male and female Koreans were enrolled in this study, and their age details are shown in Supplementary Table S1. We expected to find large individual variability in the pharmacokinetic profile of CK. To reduce the individual variability in pharmacokinetics caused by sex, we combined and calculated the means of the pharmacokinetic data from the male and female groups separately. The exclusion criteria were any significant clinical illness within 2 weeks before the study, i.e., history of high blood pressure, diabetes, and cardiovascular, hepatic, renal, hematological, gastrointestinal, neurologic, or psychiatric disease; blood donation within 8 weeks before the study; and use of any medications, including prescription and over-the-counter drugs, within 2 weeks before the study. In addition, subjects who previously experienced adverse reactions to ginseng were excluded.

The enrolled subjects were assigned to receive a single oral dose of one of two extracts, DDK-401 (100 mL spout pouch, combination of well-known representative ginsenosides Rg1, Rb1, and Rg3 at 21.51 mg and ginsenoside CK at 31.19 mg) or DDK-204 (100 mL spout pouch, ginsenosides Rg1, Rb1, and Rg3 at 11.29 mg and ginsenoside CK at 0 mg) during the first period. After a 7-day washout, each subject received the other extract. The dose for oral administration was chosen based on the recommended total daily intake of each investigational product.

CK-enriched fermented ginseng extract (DDK-401) and common red ginseng extract (DDK-204) were supplied by Deadong Korea Ginseng Co., Ltd. (Geumsan, Korea). First, the red ginseng powder was dissolved in a mixture of water and food-grade alcohol and extracted at 75 ± 5 °C. This extraction procedure was repeated 4 times. Then, the supernatant was collected and evaporated at 60 ± 5 °C with 500–760 mmHg vacuum until the sugar content was 65 brix and solid content was ≥60%. Finally, the sample was sterilized at 80–85 °C for 30–40 min and aged at 60 ± 5 °C for 24–75 h; this product was named DDK-204 and used as the control.

Second, the red ginseng concentrate (60 brix) was diluted in water until the solid content was 5%. Then, an enzyme mixture (pectinase and β-glucosidase) was added to the diluted red ginseng concentrate at a concentration of 3% and reacted at 60 ± 2 °C for 114–168 h at 3000 rpm. Thereafter, the enzyme was inactivated at 90 °C for 30 min. Next, the sample was evaporated at 55 ± 5 °C and 500–760 mmHg vacuum until the solid content reached 40%. Then the sample was dissolved in 80% food-grade alcohol and incubated for 1–2 h. Following this, centrifugation was performed at 0.5 m³/h for 5–6 h, and the supernatant was collected. A second evaporation was then performed at 55 ± 5 °C and 500–760 mmHg vacuum until the sugar content was 65 brix and solid content was ≥60%. The concentrate was then fermented with a mixture of Lactobacillus species at 1% concentration and 37 °C for one day. The resulting CK-enriched red ginseng concentrate was named DDK-401 and stored in the refrigerator until it was used for the analysis and bioassays.

One g each of DDK-401 and DDK-204 were dissolved in 50 mL of 70% methanol and filtered with a 0.45 μm membrane filter. The samples were then injected into an Ultimate 3000 HPLC system with a PRONTOSIL 120-5-C18 ACE-EPS (250 × 4.6 mm i.d., 5 µm particle size) (Bischoff Chromatography, Leonberg, Germany). The mobile phase consisted of water (solvent A) and acetonitrile (solvent B) in the following gradients: 0–10 min, 20% B; 10–42 min, 29% B; 42–67 min, 41% B; 67–70 min, 47% B; 70–90 min, 71% B; 90–95 min, 71% B. The flow rate of the mobile phase was 1.0 mL/min, and an injection volume of 10 µL was used in the quantitative analysis. The column temperature was maintained constant at 40 °C. The ginsenoside profiles were determined at 203 nm.

The quantitative determination of CK concentration in the plasma was achieved using 2 mL of intravenous blood collected from each volunteer before administration and at 2, 4, 8, and 12 h after dosing during each period. The blood samples were centrifuged at 3000 rpm for 10 min, and the supernatant was separated and frozen at -80 °C until analysis. The plasma concentrations of ginsenoside CK were determined by PCAM KOREA Co., Ltd. (Daejeon, Korea) using a HPLC–tandem mass spectrometry system. The chromatographic analysis was performed using a Waters I-class (Waters, USA), with Berberine (Dr. Ehrenstorfer GmbH, Germany) as the internal standard. Chromatographic separation was achieved with an Acquity UPLC BEH C18 column (100 mm × 2.1 mm, 1.7 μm; Waters, USA) maintained at 45 °C. The mobile phase was a gradient of 0.1% formic acid in water and 100% acetonitrile. Mass spectrometry was performed in the positive mode on an API Xevo TQ-XS instrument (Waters, USA) equipped with an electrospray ionization probe. The temperature of the ion source was set to 150 °C, and the voltage of the ion spray was 3 kV. The quantifications were performed by multiple reaction monitoring of the transitions at 645.2–203 nm for ions of ginsenoside CK, with a dwell time of 11.28 min. To validate the quantitative data in terms of linearity, the limit of detection (LOD) and limit of quantification metrics were calculated (Supplementary Table S2).

The total mRNA was extracted from each blood plasma sample to build the mRNA-seq libraries that were generated using a TruSeq stranded mRNA LT sample prep kit (Illumina, San Diego, CA, USA) following manufacturer protocols and sequenced using a Novaseq 6000 sequencing system (Illumina). The reads were trimmed with Trimmomatic (Bolger et al., 2014) to remove any adapters and low-quality reads, resulting in clean reads for improved paired-end mapping. The trimmed reads were mapped to the Homo sapiens reference genome (GRCm38) transcriptome using Salmon software version 1.3.0 (Patro et al., 2017). Differential gene expressions among the three experimental groups were evaluated using edgeR (version 3.30.3) software (McCarthy et al., 2012). The differentially expressed genes were identified based on a cutoff threshold of p < 0.05 and log-fold change >1 before being subjected to further analyses.

Functional annotations for each gene were made using the drug discovery protocol. The seven datasets used are included in Supplementary Table S3: DrugBank (Wishart et al., 2018), Human Protein Atlas (Uhlén et al., 2015), STITCH (Szklarczyk et al., 2016), Surfaceome (Bausch-Fluck et al., 2018), Tumor Suppressor Gene Database v2.0 (TSGene) (Zhao et al., 2016), pepBDB (Wen et al., 2019), and Comparative Toxicogenomics Database (Grondin et al., 2021). First, entered the DrugBank ID for each gene to navigate the details of known drugs from the complete database xml file. Second, downloaded the FDA-approved potential drug candidate list from the Human Protein Atlas database. Third, searched STITCH to observe small-molecule drug interactions. Fourth, used Surfaceome to understand the cell surface proteins. Fifth, used TSGene to obtain the cancer therapeutic gene candidates. Sixth, observed the peptide-binding protein interactions.

The detection of nitric oxide (NO) levels has been described previously (Ramadhania et al., 2022). The RAW 264.7 cells (1 × 10⁴) were placed in 24-well culture plates and incubated for 24 h at 37 °C in a humidified environment with 5% CO₂. Then, they were treated with different concentrations of DDK-401 or DDK-204 (0, 25, 50, 100, 250, and 500 μg/mL) for 1 h. In the presence of the samples, 1 μg/mL lipopolysaccharide (LPS) was used as the stimulator, and the treated cells were placed in an incubator for one day. The nitrite levels in the cell media were determined using the Griess reagent: 100 μL of the stimulated supernatant was mixed with an equivalent volume of the Griess reagent. A microplate reader was used to compare the absorbance at 540 nm with a standard curve obtained using sodium nitrite (BioTek Instruments, Inc.). L-NMMA (50 μM), a standard inhibitor, was used as the positive control in this experiment. Each assay was repeated three times, and the results are expressed in terms of percentage of NO production.

The total RNA was extracted using QIAzol lysis reagents (QIAGEN, Germantown, MD, USA), and the reverse transcription reactions were performed using 1 µg of total RNA in 20 µL of the reaction buffer with an amfiRivert reverse transcription kit (GenDepot, Barker, TX, USA), according to manufacturer instructions. The obtained cDNA was amplified with primers, as shown in Table 1. The reaction was cycled 35 times: 30 s at 95 °C, 30 s at 60 °C, and 50 s at 72 °C. Using 1% agarose gels, the amplified RT-PCR products were analyzed, visualized using Safe-Pinky DNA Gel Staining (GenDepot, Barker, TX, USA), and imaged under ultraviolet light.

All experiments were performed at least in triplicate (n = 3) unless stated otherwise. The experimental data are reported as mean ± standard error (SEM). Statistical significances between the control and sample groups were evaluated by Student’s t-test with a two-tailed distribution and two-sample equal variances. A greater extent of statistical significance is indicated by an increasing number of asterisks (*p < 0.05, **p < 0.01, and ***p < 0.001) and hash markers (#p < 0.05, ##p < 0.01 and ###p < 0.001). The hash marker (#) indicates significance between the normal and stimulated controls, and the asterisk (*) indicates significant differences between the stimulation groups (DDK-204 or DDK-401).

The cytotoxicities of DDK-401 and DDK-204 to HaCaT cells was determined for safety purposes. The HaCaT cells represent normal cell conditions, and lung cancer cells (A549) were used to examine the apoptosis signaling pathway. Each sample was evaluated at various sample concentrations (62.5, 125, 250, and 500 μg/mL in HaCaT and 25, 50, 100, 250, and 500 μg/mL in A549 cells). As shown in Figure 5, at concentrations less than 500 μg/mL, both DDK-401 and DDK-204 were nontoxic to HaCaT cells. In the A549 cells, DDK-401 demonstrated minimal toxicity after 24 h at 250 μg/mL. At a concentration of 500 μg/mL after 24 h, DDK-401 showed significantly decreased cancer cell proliferation than DDK-204. Moreover, A549 cell viability was reduced by DDK-401 in a dose-dependent manner. The cytotoxicity results in this investigation match those in a previous report (Yu et al., 2018).

The results shown in Figure 5 indicate that at 500 μg/mL, DDK-401 and DDK-204 were only mildly toxic, from which it can be concluded that both substances are relatively safe when cell conditions are normal. On the other hand, in lung cancer A549 cells, which represent cell damage and imbalanced conditions, DDK-401 had a much higher toxicity than DDK-204, producing apoptosis of the cancer cells.

Living cells produce ROS as a normal metabolic byproduct. Under excessive stress, the cells generate excess ROS, so living organisms have evolved a series of response mechanisms to adapt to ROS exposure and use ROS as a signaling molecule. ROS molecules cause oxidative stress in a feedback mechanism involving numerous biological processes, including apoptosis, necrosis, and autophagy (He et al., 2017). Apoptosis is a normal process that occurs during development and aging as well as functions as a homeostatic mechanism to maintain cell populations in tissues. Apoptosis can even occur as a defense mechanism during immune responses or when cells are damaged by disease or toxins (Elmore, 2007). In this study, we investigated several gene markers that we selected through a whole-transcriptome search for differential expressions. We found differentially expressed genes related to apoptosis and immune responses to inflammation, such as FOXO3, TLR4, caspase-8, MMP-9, and p38 MAP kinase (p38) (Cuadrado and Nebreda, 2010; van der Vos and Coffer, 2011; Zheng et al., 2021). In addition, we investigated the effects of DDK-401 and DDK-204 without any stimulation (UV-B irradiation) in HaCaT cells to observe whether our samples could trigger inappropriate apoptosis or inflammation under normal conditions showed in (Figure 7). The RT-PCR analysis (Supplementary Figure S3) showed that in HaCaT cells, neither compound regulated FOXO3, TLR4, MMP-9, or p38 expression, indicating that DDK-401 did not trigger inappropriate apoptosis or inflammation under normal conditions. The ability to control cellular living or death has enormous therapeutic potential. However, upregulation of caspase-8 was observed, which is similar to the RNA-seq data indicating differential expression (Figure 4). Although the activation of caspase-8 is mainly associated with death receptor signaling cascades, it is also activated downstream of the mitochondria. The roles of caspase-8 in the shift from autophagy to apoptosis in cisplatin-resistant MCF7 cells and in TRAIL-mediated autophagy in HCT 116 cells have already been studied (de Vries et al., 2006; Hou et al., 2010). In the lung cancer A549 cells, DDK-401 treatment downregulated the expressions of MMP9 and TLR4 while upregulating the expressions of p38 and caspase-8 genes, compared with the cells treated with DDK-204. The MMPs are a group of zinc-dependent metalloenzymes that regulate various cellular processes, including tumor cell proliferation and metastasis (Guo et al., 2019). Several studies have noted that MMPs are overexpressed in malignant tissues than the adjacent normal tissues in a range of tumors, including lung, colon, breast, and pancreatic carcinomas (Radad et al., 2011; Nakhjavani et al., 2019). Previous research has shown that downregulating the expressions of intracellular MMP-9 can increase invasion and metastasis processes several cancers (Zhang et al., 2017, Liu et al., 2019). Furthermore, oxidative stress can induce receptor-dependent apoptosis and damage the mitochondria of normal cells. Mitochondrial dysfunction then further increases ROS accumulation and activates the p38 MAPK pathway. ROS can continuously activate p38 MAPK by activating MAPK kinase and inhibiting MAPK phosphatase. In A549 cells, ROS can regulate the expressions of Bax and Bcl-2 by activating p38 MAPK, which increases the level of cytochrome c in the cytoplasm and triggers the caspase cascade reaction leading to apoptosis (Jin et al., 2019; Nguyen and Nguyen, 2019). The caspases are a family of cysteine-containing proteolytic enzymes that play a central role in the execution phase of cell apoptosis. It has been reported that the effects of the caspase 8 pathway on cancer cells involve inducing apoptosis (Huang et al., 2019). The apoptosis induced by most anticancer drugs occur by the activation of caspases (González-Burgos et al., 2015; Kim and Kim, 2018). Caspase-8 is important in the death receptor-mediated extrinsic pathway, and DDK-401 promotes the activation of caspase-8 in A549 cells in this study, showing that it can cause apoptosis by activating the extrinsic caspase pathway (Wu et al., 2019; Yi, 2019). TLR4 is an important member of the type I transmembrane protein family. Recently, growing evidence has shown TLR4 in various tumors (Gao et al., 2017; Li and Ji, 2018; Zhou et al., 2019), including head and neck, lung, gastrointestinal, liver, pancreatic, skin, breast, ovarian, cervical, and prostate cancers. TLR4-mediated cancer growth is involved in breast tumor progression, and the downregulation of TLR4 prevented breast cancer progression and improved survival (Zhou et al., 2019c). According to our findings, DDK-401 could induce cell apoptosis by upregulating and downregulating various transcriptional factors under cancerous conditions.