Pharmacological evaluation of 1-acetyl-β-carboline, a naturally occurring compound with anti-skin cancer potential

Abstract

The human microbiome comprises microbial communities that reside in the human body and contribute to host health through molecular mediators. Lactobacillus spp. are frequently used as probiotics to restore microbial balance, and L. gasseri has been reported to exert a wide range of beneficial effects. In this study, 1-acetyl-β-carboline (ABC) was identified in L. gasseri cultures and subsequently synthesized via the Pictet-Spengler reaction followed by palladium-catalyzed oxidation. ABC exhibited significant anticancer activity by reducing colony formation and growth of epidermal growth factor-induced JB6 cells and by inhibiting the proliferation of SK-MEL-5 and SK-MEL-28 melanoma cells. Mechanistic studies revealed that ABC induced G2/M phase cell-cycle arrest and promoted apoptosis by regulating related markers, including p27 and caspases-3 and -7. Additionally, ABC significantly inhibited the mitogen-activated protein kinase kinase (MEK)-extracellular signal-regulated kinase (ERK) signaling pathway by reducing phosphorylated MEK and phosphorylated ERK levels. ABC also downregulated cyclooxygenase-2 expression, targeting inflammation-related pathways in melanoma cells. In a mouse model, ABC effectively mitigated 12-O-tetradecanoylphorbol-13-acetate (TPA)-induced epidermal hyperplasia and reduced inflammation. These findings highlight the pharmacological significance of ABC, independent of its origin, and suggest that this naturally occurring compound possesses preventive and therapeutic potential against skin cancer.

Graphical abstract

The naturally occurring compound 1-acetyl-β-carboline (ABC) shows selective anticancer and anti-inflammatory activity in melanoma models, suggesting its potential as a pharmacological agent for skin cancer treatment.

Introduction

The human microbiome comprises a diverse collection of microorganisms that inhabit the human body.¹ In particular, the human gut hosts a substantial and dynamic microbial community.² Several studies have demonstrated an association between gut microbial imbalance and various diseases,^3,4 underscoring the close relationship between the gut microbiome and host health.

Cancer patients often exhibit gut microbial dysbiosis,⁴ which can affect cancer initiation, progression, and treatment response through local or systemic interactions with cancer cells.⁵ Examples include the significantly higher abundance of Fusobacterium nucleatum and Bacteroides fragilis in stool samples from patients with colorectal cancer (CRC) compared with that in healthy individuals.^6,7 F. nucleatum invades CRC cells and promotes carcinogenesis through the activation of β-catenin signaling pathway,⁸ whereas B. fragilis produces carcinogenic toxins that induce DNA damage.⁹

The human gut microbiome also influences skin conditions such as acne, rosacea, atopic dermatitis, and skin cancer.^10,11,12,13 A distinctive difference in gut microbial composition has been observed in melanoma patients receiving anti-PD-1 immunotherapy.^14,15 Responders showed a markedly higher abundance of the gut bacteria Ruminococcaceae and Faecalibacterium compared to non-responders.^14,16 Therefore, modulating gut bacterial composition represents a potential strategy for enhancing cancer treatment, including skin cancer therapies.

Interest in probiotics, which are live microorganisms capable of providing health benefits, has increased recently owing to their potential as therapeutic options for restoring gut microbiome imbalance.^{17,18,19,20,21,22} Notably, Lactobacillus and Bifidobacterium are well-established probiotics known for their therapeutic effects in treating various gastrointestinal disorders, such as irritable bowel syndrome (IBS), gastroenteritis, diarrhea, and inflammatory bowel disease.^22,23,24 The administration of L. gasseri has been shown to improve symptom severity and stool characteristics in patients with IBS.²⁵ In addition to gastrointestinal effects, probiotics containing L. gasseri have been reported to stimulate interleukin-12 production,²⁶ alleviate endometriosis-associated pain,²⁷ and reduce visceral fat^28,29 in clinical trials.

Although numerous studies have examined the biological efficacy of L. gasseri, the underlying molecular mechanisms remain unclear. Recent human microbiome studies suggest a possible role for microbial small molecules in mediating the biological effects of L. gasseri, which has been shown to inhibit the growth of pathogenic bacteria through the production of bacteriocins and lactic acid.^30,31 Messenger metabolites produced by other human microbial species include lipoteichoic acid from L. rhamnosus^32,33 and 6-N-hydroxyaminopurine from the skin commensal bacterium Staphylococcus epidermidis,³⁴ both of which have been demonstrated to suppress skin cancer growth in mouse models. Nevertheless, further biochemical investigations are required to elucidate the molecular mechanisms underlying the biological effects of the human microbiome.

Melanoma is one of the most aggressive forms of skin cancer, with incidence rates continuing to rise globally.³⁵ Melanoma is characterized by rapid metastasis, distinguishing it from other types of skin cancer, and is often associated with high mortality rates, particularly when treatment options are limited. Although conventional chemotherapies using xenobiotic compounds have proven effective, they are frequently accompanied by the development of drug resistance.³⁶ This reinforces the urgent need for novel therapeutic agents with enhanced efficacy and reduced adverse effects. In this context, the endogenous human microbiome has emerged as a promising source for cancer prevention and treatment.

We performed chemical analysis of L. gasseri (KCTC 3163) culture extract using liquid chromatography-mass spectrometry (LC-MS) and nuclear magnetic resonance (NMR) and identified 1-acetyl-β-carboline (ABC) as a component. Independent of its origin, ABC is one of the β-carboline alkaloids known for diverse biological activities,³⁷ and its pharmacological potential warrants direct investigation. Therefore, chemical synthesis of ABC was achieved via the Pictet-Spengler reaction, followed by oxidation using Pd/C, enabling the biological evaluation of its anticancer effects in vitro and in vivo. We examined the effects of ABC on viability, colony growth, cell-cycle distribution, and apoptosis in melanoma cells, as well as its anti-inflammatory effects in a 12-O-tetradecanoylphorbol-13-acetate (TPA)-induced skin hyperplasia mouse model. This study aimed to evaluate the biological effects of ABC—a naturally occurring compound identified in microbial culture—against skin cancer, and to elucidate its underlying molecular mechanisms.

Results

Isolation and structure determination

L. gasseri was cultured under anaerobic conditions at 37°C to simulate the human intestinal environment, and the culture extract was chemically evaluated using LC-photodiode array-MS. One component of the extract showed a pseudomolecular ion at m/z 211 [M + H]⁺ upon mass spectrometric analysis. Additionally, the UV spectrum of the compound displayed absorption bands with λ_max at 218, 283, and 378 nm, suggesting the presence of indole³⁸ with extended conjugation. Therefore, the compound was inferred to be more intricate than a simple indole-containing amino acid or its analog such as tryptophan or tryptamine and was subsequently isolated using preparative high-pressure liquid chromatography (HPLC). The ¹H NMR data of the compound at δ_H 8.20 (1H, d, J = 7.9, H-5), 7.69 (1H, d, J = 8.3, H-8), 7.59 (1H, dd, J = 7.6, 8.3, H-7), and 7.29 (1H, dd, J = 7.9, 7.6, H-6) indicated the presence of a 1,2-disubstituted benzene as part of an indole. Also, the chemical shift values of the signals at δ_H 8.45 (1H, d, J = 4.9, H-3) and 8.29 (1H, d, J = 4.9, H-4) were suggestive of pyridine protons. The downfield-shifted methyl singlet at δ_H 2.82 (3H, s, CH₃) indicated the attachment of an acetyl group to an aromatic ring system. The heteronuclear multiple bond correlation (HMBC) correlations of H-3 with C-1, C-4, and C-4a and of H-4 with C-4a, C-4b, and C-8b established a connection between the indole and pyridine rings at positions C-4a and C-8b. Furthermore, the HMBC cross-peak from the methyl protons to C-1 and the carbonyl carbon confirmed the attachment of the acetyl group to C-1, completing the structure of ABC, as illustrated in Figure 1. The observed NMR data (Figures S1–S4) matched the values in the literature.^39,40

Figure 1.

Schematic representation of the synthesis and chemical structure of 1-acetyl-β-carboline

Synthesis

ABC was synthesized on a large scale to enable its biological and pharmacological evaluation. Initially, tryptamine (TRA) was reacted with methylglyoxal (MGO) via the Pictet-Spengler reaction, in which intramolecular substitution of the indole with the electrophilic iminium ion resulted in cyclization to form piperidine (Figure 1). Subsequently, intermediate product 1 was refluxed in DMSO with Pd/C to oxidize piperidine to pyridine, ultimately yielding the extended aromatic compound ABC with a total yield of 8.8%, including purification steps.

Cytotoxicity of ABC on normal human dermal fibroblast cells

L. gasseri has been reported to have beneficial effects, including antimicrobial, anti-biofilm, antioxidant, and anti-inflammatory activities in various skin-related conditions such as atopic dermatitis.^41,42,43 β-Carboline derivatives are considered promising agents as potential anticancer therapeutics.⁴⁴ Prior to evaluating the biological effects of ABC on the skin, we assessed its cytotoxicity by treating normal human dermal fibroblast (NHDF) cells with different concentrations of ABC (0, 25, 50, and 100 μM) for 24 and 48 h. Cell viability was determined using the WST-8 assay. The results demonstrated that ABC exerted minimal cytotoxicity on NHDF cells, with viability above 87% even at the highest concentration of 100 μM after 48 h of treatment (Figure 2A). These results suggest that ABC is not significantly cytotoxic to normal cells at the concentrations tested.

Figure 2.

Cytotoxicity of ABC in normal human dermal fibroblasts and its effects on the viability of melanoma cell lines SK-MEL-5 and SK-MEL-28

(A) NHDF cells were treated with different concentrations of ABC (0, 12.5, 25, 50, and 100 μM) for 24 and 48 h. SK-MEL-5 cells were treated with (B) ABC (0, 12.5, 25, 50, and 100 μM), (C) 5-FU (0, 1.25, 2.5, 5, and 10 μM), or (D) DTIC (0, 25, 50, 100, and 200 μM) for 24, 48, 72, and 96 h. SK-MEL-28 cells were treated with (E) ABC, (F) 5-FU, or (G) DTIC using the same concentrations and treatment durations as in panels B–D. Cell viability was assessed using the WST-8 assay. Data are presented as mean ± SD. Statistical significance was determined using Student’s t test. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001 compared to the control group.

Effect of ABC on melanoma cell viability

The effect of ABC on the viability of SK-MEL-5 and SK-MEL-28 melanoma cells was assessed using the WST-8 assay following treatment of the cells with various concentrations of ABC (0, 12.5, 25, 50, and 100 μM) for 24, 48, 72, and 96 h. The commonly used chemotherapeutic agents 5-fluorouracil (5-FU) and dacarbazine ([DTIC] 5-(3,3-dimethyl-1-triazeno)imidazole-4-carboxamide) served as positive controls.^45,46 The results showed a dose- and time-dependent decrease in SK-MEL-5 (Figures 2B–2D) and SK-MEL-28 (Figures 2E–2G) cell viability following ABC treatment. Both cell lines showed a significant reduction in viability after 96 h of treatment with 100 μM ABC, with SK-MEL-5 and SK-MEL-28 cells exhibiting inhibitory effects of 57.3% and 44.9%, respectively (p < 0.001). These findings indicate that ABC effectively reduced the viability of melanoma cells, comparable with the 5-FU- or DTIC-mediated effects.

Inhibition of anchorage-independent growth by ABC in soft agar assay

The inhibitory effect of ABC on epidermal growth factor (EGF)-induced cell transformation and colony formation was evaluated in JB6 cells. JB6 cells were cultured in soft agar for 2 weeks after being treated with EGF alone or EGF in combination with different concentrations of ABC (0, 25, 50, and 100 μM). The results demonstrated that ABC significantly reduced colony formation in a dose-dependent manner. Specifically, in the presence of EGF, colony counts decreased by 37.8% (p < 0.001), 54.5% (p < 0.001), and 75.7% (p < 0.001) in the 25, 50, and 100 μM ABC groups, respectively, compared to that in the EGF alone group (Figures 3A and 3B). This inhibition of colony formation suggests that ABC effectively suppressed EGF-induced cell transformation, a critical process in tumorigenesis.

Figure 3.

Inhibition of colony formation and growth in soft agar following ABC treatment

(A) Representative images of colonies formed in soft agar by JB6 cells treated with EGF (10 ng/mL) and different concentrations of ABC (0, 25, 50, and 100 μM). (B) Quantitative analysis of colony formation by JB6 cells treated with EGF and different concentrations of ABC. Results are expressed as mean ± SD. Statistical analysis was performed using Student’s t test. ^###p < 0.001 vs. control; ∗∗∗p < 0.001 vs. EGF-treated group. (C) Representative images of colonies formed in soft agar by SK-MEL-5 and SK-MEL-28 cells treated with different concentrations of ABC (0, 25, 50, and 100 μM). Quantitative analysis of colony formation by SK-MEL-5 (D) and SK-MEL-28 (E) cells treated with different concentrations of ABC. The number of colonies after ABC treatment was normalized to the control and presented as a percentage. Data are presented as mean ± SD. Statistical analysis was performed using Student’s t test. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001 compared to the control group.

Soft agar assay was also performed to assess the effect of ABC on the anchorage-independent growth of melanoma cells. In both SK-MEL-5 and SK-MEL-28 cells, ABC treatment at concentrations of 25, 50, and 100 μM significantly reduced the number and size of colonies in a dose-dependent manner (Figure 3C). In SK-MEL-5 cells, ABC treatment resulted in 28.8% (p < 0.05), 45.3% (p < 0.01), and 53.1% (p < 0.001) reduction in the number of colonies at 25, 50, and 100 μM concentration, respectively (Figure 3D). Similarly, in SK-MEL-28 cells, ABC treatment led to 34.9% (p < 0.01), 43.7% (p < 0.01), and 56.4% (p < 0.001) reduction in colony numbers at the same concentrations (Figure 3E). These results indicate that ABC inhibited the malignant growth of melanoma cells in an anchorage-independent environment.

ABC induces G2/M phase cell-cycle arrest in melanoma cells

We examined the effects of ABC on cell-cycle progression using flow cytometry. The results revealed that treatment with 25, 50, and 100 μM ABC induced G2/M phase arrest in both SK-MEL-5 and SK-MEL-28 cells (Figure 4A). In SK-MEL-5 cells, the proportion of cells in the G2/M phase increased by 1.4%, 3.6%, and 5.4% with 25, 50, and 100 μM ABC, respectively, compared to the control (Figure 4B). Furthermore, in SK-MEL-28 cells, the percentage of cells in the G2/M phase increased by 3.3%, 5.8%, and 7.2% with 25, 50, and 100 μM ABC, respectively, compared to the control (Figure 4C). These results indicate that ABC disrupts cell-cycle progression by causing G2/M phase arrest in a dose-dependent manner, thereby inhibiting cell proliferation.

Figure 4.

Effects of ABC on cell-cycle distribution in SK-MEL-5 and SK-MEL-28 cells

(A) Representative flow cytometric histograms showing cell-cycle distribution in SK-MEL-5 and SK-MEL-28 cells treated with ABC (0, 25, 50, and 100 μM) for 24 h. Quantitative analysis of cell-cycle distribution in SK-MEL-5 (B) and SK-MEL-28 (C) cells treated with different concentrations of ABC (0, 25, 50, and 100 μM). Western blot analysis of SK-MEL-5 (D) and SK-MEL-28 (E) cells treated with ABC displaying the levels of the cell-cycle regulator p27. β-actin served as a loading control. Quantitative densitometric analysis of p27 normalized to β-actin in SK-MEL-5 (F) and SK-MEL-28 (G) cells. Data are presented as mean ± SD. Statistical significance was determined using Student’s t test. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001 compared to the control group.

Western blotting was performed to further investigate the molecular mechanisms underlying the ABC-mediated G2/M phase arrest. ABC treatment increased the expression of p27, a key cell-cycle regulator (Figures 4D and 4E). Quantitative densitometry revealed that p27 abundance increased to 1.9 ± 0.2-fold relative to control in SK-MEL-5 cells after 100 μM ABC treatment (Figure 4F), whereas the same dose elevated p27 to 1.4 ± 0.2-fold in SK-MEL-28 cells (Figure 4G). These elevated p27 levels support the flow cytometric findings and suggest that ABC blocks cell-cycle progression at the G2/M checkpoint, thereby suppressing melanoma cell growth.

ABC induces apoptosis in melanoma cells

ABC-induced apoptosis was assessed in SK-MEL-5 and SK-MEL-28 cells using Annexin V/propidium iodide (PI) staining, followed by flow cytometry. Treatment of cells with 25, 50, and 100 μM ABC for 72 h significantly increased the percentage of apoptotic cells in a dose-dependent manner (Figure 5A). In SK-MEL-5 cells, the percentage of total apoptotic cells increased from 6.1% in the control group to 13.4%, 17.1%, and 35.4% (p < 0.001) in the 25, 50, and 100 μM ABC groups, respectively (Figure 5B). Similarly, in SK-MEL-28 cells, ABC treatment increased apoptosis from 6.6% in the control group to 17.0%, 22.1%, and 29.9% (p < 0.001) at the same doses (Figure 5C). These observations indicate that ABC actively induces apoptosis in melanoma cells, resulting in reduced cell viability.

Figure 5.

Effects of ABC on apoptotic induction in SK-MEL-5 and SK-MEL-28 cells

(A) Representative flow cytometric histograms showing apoptotic levels in SK-MEL-5 and SK-MEL-28 cells following treatment with ABC (0, 25, 50, and 100 μM) for 72 h. Apoptosis was assessed using Annexin V/PI staining. Quantitative analysis of apoptosis in SK-MEL-5 (B) and SK-MEL-28 (C) cells treated with different concentrations of ABC. Western blot analysis of cleaved caspase-3 (c-Cas-3) and cleaved caspase-7 (c-Cas-7) in SK-MEL-5 (D) and SK-MEL-28 (E) cells after 72-h treatment with ABC. β-actin served as a loading control. Quantitative densitometric analysis of cleaved caspase-3 (c-Cas-3) and cleaved caspase-7 (c-Cas-7) normalized to β-actin in SK-MEL-5 (F and G) and SK-MEL-28 (H and I) cells. Data are presented as mean ± SD. Statistical significance was determined using Student’s t test. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001 compared to the control group.

To confirm the induction of apoptosis at the molecular level, we analyzed the levels of the key apoptotic proteins caspase-3 and caspase-7 using western blotting. As shown in Figures 5D and 5E, 72-h treatment with ABC increased the levels of both cleaved caspase-3 and cleaved caspase-7, further supporting the conclusion that ABC promotes apoptosis in melanoma cells. Densitometric analysis revealed that 100 μM ABC treatment increased cleaved caspase-3 and cleaved caspase-7 levels to 2.2 ± 0.3-fold and 5.4 ± 0.6-fold, respectively, in SK-MEL-5 cells (Figures 5F and 5G), and to 1.8 ± 0.3-fold and 5.7 ± 1.8-fold, respectively, in SK-MEL-28 cells (Figures 5H and 5I), relative to control. These findings indicate that ABC induces caspase activation in both melanoma cell lines, thereby promoting apoptotic cell death. The observed increases in cleaved caspase-3 and cleaved caspase-7 across SK-MEL-5 and SK-MEL-28 cells confirm activation of the apoptotic cascade and support the pro-apoptotic effect of ABC.

ABC modulates the mitogen-activated protein kinase kinase signaling pathway and cyclooxygenase-2 expression in melanoma cells

To examine the regulatory effects of ABC on the mitogen-activated protein kinase kinase (MAP2K aka MEK) signaling pathway and cyclooxygenase-2 (COX-2) expression in melanoma cells, SK-MEL-5 and SK-MEL-28 cells were treated with different concentrations of ABC. Western blotting was conducted using antibodies specific to both phosphorylated and total forms of key signaling molecules, including MEK, extracellular signal-regulated kinase (ERK), and COX-2. Western blot analysis revealed a marked decrease in phosphorylated-MEK (p-MEK) and -ERK (p-ERK) levels in ABC-treated SK-MEL-5 and SK-MEL-28 cells compared to those in control cells. Because the MEK/ERK pathway is critical for regulating cell growth and survival, its inhibition by ABC suggests a broader disruption of oncogenic signaling. ABC also significantly reduced COX-2 expression in both melanoma cell lines (Figures 6A and 6B). Quantitative densitometric analysis demonstrated a marked reduction in the expression levels of p-MEK, p-ERK, and COX-2 following treatment with 100 μM ABC. Specifically, in SK-MEL-5 cells, the expression levels of p-MEK, p-ERK, and COX-2, normalized to β-actin, were reduced to 73.4% ± 2.2%, 42.3% ± 2.7%, and 73.4% ± 0.8% of control, respectively (Figures 6C–6E). In SK-MEL-28 cells, the corresponding values were 68.7% ± 9.6%, 65.8% ± 3.9%, and 16.3% ± 2.0%, respectively (Figures 6F–6H). These findings indicate that ABC simultaneously targets the MEK/ERK pathway and COX-2 expression, potentially enhancing its effectiveness in inhibiting melanoma cell proliferation, survival, and inflammation.

Figure 6.

ABC downregulates MEK-related signaling pathways and COX-2 expression in melanoma cells

SK-MEL-5 (A) and SK-MEL-28 (B) cells were treated with different concentrations of ABC (0, 25, 50, and 100 μM) for 48 h. Protein lysates were probed with antibodies specific to p-MEK, total MEK, p-ERK, total ERK, and COX-2. β-actin was used as a loading control. Quantitative densitometric analysis of p-ERK, p-MEK, and COX-2 normalized to β-actin in SK-MEL-5 (C–E) and SK-MEL-28 (F–H) cells. Data are presented as mean ± SD. Statistical significance was determined using Student’s t test. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001 compared to the control group.

ABC reduces TPA-induced epidermal hyperplasia

To evaluate the effects of ABC on TPA-induced epidermal hyperplasia, dorsal skin tissues were analyzed following H&E staining (Figure 7A). The data show that TPA treatment resulted in an approximately 2-fold increase in epidermal thickness from 21.4 ± 2.4 μm in the vehicle control to 41.8 ± 7.8 μm in the TPA-treated group (Figure 7B). ABC treatment effectively mitigated TPA-induced hyperplasia at medium and high doses. Notably, high-dose ABC (2 μmol/mouse) reduced the epidermal thickness to 25.4 ± 4.6 μm, a reduction of approximately 39.2%. In contrast, treatment with high-dose ABC alone resulted in an epidermal thickness of 20.3 ± 1.3 μm, comparable to that of the vehicle control. These results suggest that ABC significantly inhibits TPA-induced hyperplasia in the skin tissue.

Figure 7.

ABC reduces TPA-induced epidermal hyperplasia and COX-2 expression

(A) Representative H&E-stained images of dorsal skin tissues from mice treated with TPA and/or ABC, showing the effects of different treatments on epidermal thickness. (B) Quantification of epidermal thickness of dorsal skin tissues. (C) Representative images showing immunohistochemical staining of COX-2 in dorsal skin tissues. (D) Quantification of COX-2 expression. Data are presented as mean ± SD. Statistical analysis was performed using Student’s t test. ^##p < 0.01; ^###p < 0.001 vs. vehicle; ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001 vs. TPA-treated group.

ABC inhibits TPA-induced COX-2 expression in mouse skin

Immunohistochemical staining for COX-2 revealed differences in the staining intensity across the experimental groups. Enhanced COX-2 expression was observed in the TPA-treated group, indicating increased inflammation in both the epidermis and dermis. However, in the ABC-treated groups, COX-2 staining intensity decreased in a dose-dependent manner (Figure 7C). Quantitative analysis confirmed that COX-2 levels were markedly elevated in the TPA group compared to the vehicle group, showing an 11.2-fold increase (p < 0.001). In contrast, COX-2 levels were significantly reduced in the ABC group compared to those in the TPA group, with reductions of 65.9% at 0.02 μmol (p < 0.001), 85.4% at 0.2 μmol (p < 0.001), and 88.5% at 2 μmol (p < 0.001) (Figure 7D). Additionally, in the positive control group treated with 5-FU, COX-2 levels decreased by 58.2% at 0.1 μmol. These findings suggest that ABC inhibits COX-2 expression, which may effectively reduce TPA-induced inflammation.

Discussion

The human microbiome comprises diverse microorganisms that inhabit the human body. Regulation of the gut microbial composition is increasingly recognized as a potential clinical approach for treating diseases,^23,24 highlighting the close relationship between the human microbiome and host health. Beyond short-chain fatty acids,⁴⁷ small molecules associated with the human microbiome are gaining attention as mediators of this relationship.^31,32,33,34 L. gasseri, a well-known probiotic, offers numerous health benefits.^{26,27,28,29,31} In this biochemical study, we sought such compounds in L. gasseri culture to explore their pharmacological properties and possible contributions to microbiome-host interactions.

The culture of L. gasseri KCTC 3163 under anaerobic conditions at body temperature yielded an extract containing small molecules potentially relevant to its biological activity. The chemical profiles of these compounds were evaluated using LC-MS and NMR. One of the components in the extract was identified as ABC. The synthesis of ABC, which involved the Pictet-Spengler reaction and oxidative aromatization, allowed for structural confirmation of the molecule and facilitated its biological analyses. This enabled us to focus more on the pharmacological actions of ABC rather than its origin, thereby expanding the molecular therapeutic understanding of β-carboline compounds.

ABC was originally isolated from Streptomyces kasugaensis,⁴⁸ and its biosynthesis has been studied under various media conditions.⁴⁹ Antibacterial activity of ABC against methicillin-resistant Staphylococcus aureus has been demonstrated using the compound obtained from marine bacteria, including Streptomyces sp. and Pseudomonas sp.^39,40 Recently, ABC derived from Lactobacillus spp. was reported to block Candida albicans filamentation through inhibition of a DYRK1-family kinase.⁵⁰ Another study also identified ABC and structurally related metabolites in the human commensal Lactobacillus crispatus.⁵¹ However, other reports have described that ABC is not a direct metabolite of Lactobacillus species, but rather a non-enzymatic reaction product formed from tryptophan and MGO present in the culture medium^52,53 and has also been recently found in foods.⁵³

To further explore this issue, we quantitatively analyzed ABC in both L. gasseri cultures and De Man, Rogosa and Sharpe (MRS) broth alone as a control (Figure S5). While ABC was detectable in autoclaved MRS, its concentration in the L. gasseri cultures increased in parallel with microbial growth (OD₆₀₀) and exceeded that of the control. However, given that the maximum absolute difference was less than 0.5 μM, these data alone are insufficient to conclusively determine that ABC is produced by L. gasseri. Further studies, such as isotope tracing or genetic knockout experiments, are required to establish ABC biosynthesis by L. gasseri under physiological or culture conditions. Contrary to the previously described antimicrobial activities of ABC,^39,40,50 disk diffusion assays performed by our group using ABC concentrations of 0.24 and 0.05 M in a volume of 20 μL/disk did not show any clear zone of inhibition against C. albicans, Bacillus subtilis, Gardnerella vaginalis, Escherichia coli, and S. aureus.

Multiple studies have indicated a link between gastrointestinal disorders and various skin diseases.^54,55 For instance, 10%–25% of patients with ulcerative colitis, Crohn’s disease, or celiac disease also exhibit skin-related symptoms, particularly psoriasis and cutaneous ulcers.⁵⁴ Furthermore, oral administration of Lactobacilli has been reported to reduce the symptoms of psoriasis, cutaneous wounds, ulcers, and photoaging,⁵⁴ drawing significant attention to the concept of the gut-skin axis. Because the L. gasseri strain used in this study was isolated from the human intestine,⁵⁶ and this species is one of the most prevalent Lactobacilli spp. on the skin,⁵⁷ we evaluated the anticancer potential of ABC—identified in L. gasseri culture—in the context of skin health. Our investigations focused on the effects of ABC on anchorage-independent growth, cell viability, cell-cycle progression, and apoptosis in melanoma cells. The results of our study provide strong evidence that ABC exerts significant antitumor activity in melanoma cells.

In addition to showing dose- and time-dependent cytotoxic effects on melanoma cells, ABC exhibited markedly lower toxicity in NHDF cells, indicating selective cytotoxicity (Figure 2). The progressive decline in cell viability observed at higher concentrations and longer exposure times suggests that ABC interferes with essential cellular processes and inhibits key signaling pathways, ultimately leading to apoptotic cell death. Although ABC (82.5 μM) exhibited a higher IC₅₀ value in SK-MEL-5 cells compared to 5-FU (3.19 μM), this should not be interpreted as an indication of inferior efficacy. Therapeutic effectiveness should be evaluated not only by IC₅₀ but also by mechanistic action, safety, and bioavailability. In this study, ABC elicited apoptosis via activation of cleaved caspase-3 and caspase-7, induced cell-cycle arrest through p27 upregulation, suppressed COX-2 expression, and exhibited lower cytotoxicity toward normal cells. In contrast, DTIC showed only limited activity at its maximal soluble concentration. These findings suggest that ABC possesses a favorable therapeutic window, with both selectivity and safety, highlighting its potential for further investigation as a therapeutic candidate for melanoma.

EGF-induced transformation is a crucial process in the malignant progression of normal cells, during which EGF binds to the EGF receptor on the cell surface to activate signaling pathways that promote cell growth, differentiation, survival, and motility.^58,59,60,61 Dysregulation or overactivation of these pathways leads to uncontrolled cell proliferation and malignant transformation. In the present study, ABC inhibited anchorage-independent growth in soft agar assays (Figure 3A) and significantly reduced colony formation in EGF-induced JB6 cells (Figure 3B), indicating its potential to suppress EGF-driven transformation. Additionally, the inhibition of anchorage-independent growth of melanoma cells by ABC supports its potential as a therapeutic agent against melanoma initiation and progression (Figures 3C–3E). Thus, our results suggest that ABC not only inhibits the initial stages of melanoma development but also hinders the subsequent growth of established tumors, underscoring its broader application in melanoma treatment.

Mechanistically, a key finding of this study is that ABC induced G2/M phase cell-cycle arrest in SK-MEL-5 and SK-MEL-28 cells. Flow cytometric analysis revealed a significant increase in the G2/M phase cell population following ABC treatment, suggesting that ABC blocked cell-cycle progression by arresting cells in the G2/M phase (Figures 4A–4C). This G2/M phase arrest may serve as the primary mechanism of cell proliferation inhibition via disruption of cell division and growth.

The expression of p27, a well-known tumor suppressor protein, is decreased in various types of cancer.⁶² As a cyclin-dependent kinase (CDK) inhibitor, p27 plays a critical role in cell-cycle regulation by interacting with CDKs to control cell-cycle arrest or progression at multiple stages, thereby maintaining cellular stability and survival.^63,64 In the present study, western blot analysis confirmed an increase in p27 expression following ABC treatment (Figures 4D and 4E). This finding supports the flow cytometric results, indicating that p27 is involved in cell-cycle arrest at the G2/M phase. Thus, ABC-induced upregulation of p27 expression is closely associated with G2/M phase cell-cycle arrest, which results in inhibition of cancer cell division and growth. This mechanism reduces the abnormal proliferative capacity of melanoma cells and may act as a crucial factor in enhancing the anticancer effects of ABC.

Additionally, ABC induces apoptosis in melanoma cells. Annexin V/PI flow cytometric assays revealed a significant dose-dependent increase in the proportion of apoptotic cells following ABC treatment. The increase in apoptotic rate was most pronounced at 100 μM ABC, indicating that ABC activates apoptotic pathways, which contribute to its potent anticancer effects. In SK-MEL-5 cells, ABC markedly increased the total apoptotic cell populations by 29.3%, whereas 5-FU and DTIC induced comparatively lower increases of 17.5% and 4.2%, respectively (Figure S6). Similarly, in SK-MEL-28 cells, ABC elevated apoptosis by 23.3%, while 5-FU and DTIC resulted in increases of only 3.1% and 6.8%, respectively (Figure S7). These findings indicate that ABC exerts significantly greater pro-apoptotic activity than both positive controls under equivalent conditions. Concordantly, western blot analysis showed a marked accumulation of the cleaved forms of caspase-3 and caspase-7 (Figures 5D and 5E). This increase in cleaved caspase-3 and caspase-7 reflects the activation of executioner caspases, which are key effectors in the terminal phase of the apoptotic pathway.^65,66,67,68 Therefore, the observed cleavage of these caspases provides direct molecular evidence supporting the pro-apoptotic effect of ABC and further explains its ability to reduce melanoma cell viability.

Furthermore, our study demonstrated that ABC inhibits the MEK/ERK signaling pathway and COX-2 expression in melanoma cells. The MEK/ERK pathway is critical for regulating melanoma cell proliferation, survival, and resistance to apoptosis, all of which are key processes in tumor progression and metastasis.^69,70 Recent studies have emphasized the therapeutic potential of targeting the MEK/ERK pathway in melanoma.^71,72 Our findings indicate that ABC markedly suppresses the phosphorylation of MEK and ERK (Figure 6), effectively attenuating the MEK/ERK signaling cascade in melanoma cells. Inhibition of the MEK/ERK pathway was associated with a significant reduction in cell viability and in the expression of downstream oncogenic markers, suggesting that MEK/ERK inhibition contributes to the anticancer effects of ABC. Moreover, ABC significantly downregulated COX-2 expression, a key mediator of inflammation and tumor progression. This result is consistent with previous reports showing that MEK/ERK inhibition leads to reduced COX-2 expression.^73,74 COX-2 is frequently overexpressed in melanoma and contributes to an inflammatory tumor microenvironment, angiogenesis, and apoptosis resistance.^75,76 By reducing COX-2 levels, ABC may impair these tumor-promoting processes and enhance its antitumor efficacy. The inhibition of both the MEK/ERK pathway and COX-2 expression by ABC suggests that it targets multiple key regulators of melanoma progression.

In addition, in vivo experiments confirmed that ABC effectively inhibits TPA-induced skin hyperplasia, a well-established model of cutaneous inflammation and epidermal thickening.^77,78,79 5-FU was included as a reference compound to evaluate the efficacy of ABC in vivo.⁸⁰ H&E staining revealed that TPA treatment significantly increased epidermal thickness compared to that in the control group, resulting in an approximately 39.2% reduction (p < 0.05) following high-dose ABC treatment. In contrast, high-dose ABC treatment alone resulted in an epidermal thickness comparable to that of the control group. These results suggest that ABC inhibits TPA-induced hyperplasia. Additionally, ABC treatment effectively inhibited TPA-induced COX-2 expression in the mouse skin, demonstrating its anti-inflammatory potential. COX-2 is a well-known pro-inflammatory mediator and tumor promoter in TPA-induced epidermal hyperplasia models.^77,81,82 TPA stimulation significantly upregulates COX-2 expression in the skin, contributing to inflammation and epidermal proliferation. In this study, ABC treatment notably reduced COX-2 expression in a dose-dependent manner, with high-dose ABC reducing TPA-induced COX-2 expression by up to 88.5% (Figures 7C and 7D). This reduction in COX-2 expression suggests that ABC mitigates TPA-induced inflammatory responses by attenuating COX-2-mediated inflammatory pathways.

Overall, the findings of this study suggest that ABC, a biologically active compound detected in L. gasseri culture extracts, warrants further investigation as a potential drug lead for melanoma treatment. ABC selectively inhibits melanoma cell viability, suppresses anchorage-independent growth, and induces cell-cycle arrest and apoptosis, although the concentrations used in the present assays may not directly reflect physiologically relevant levels. Moreover, both in vitro and in vivo results indicate that ABC may hold therapeutic potential not only for melanoma but also for other skin-related disorders, suggesting a previously unrecognized connection between L. gasseri probiotics and their health benefits. Nevertheless, the concentrations of ABC detected in L. gasseri cultures (maximum approximately 0.7 μM) were below the levels required to elicit biological activity in our in vitro assays. Further research is required to confirm ABC production in human tissues, determine whether it occurs at biologically active concentrations and under relevant physiological or pathological conditions, and elucidate the molecular mechanisms underlying its biological effects. In conclusion, ABC is presented here as a naturally occurring small molecule identified in the culture extract of the well-known commensal microbe L. gasseri, with favorable pharmacological properties, as well as potential biocompatibility and reduced toxicity attributable to its natural origin.

Methods

General experimental procedures

Flash column chromatography was conducted using a CombiFlash Retrieve system connected to a RediSep prepacked C₁₈ reverse phase column (30 × 130 mm) (Teledyne ISCO, Lincoln, NE, USA). HPLC separations were performed using a Waters system (Waters Corporation, Milford, MA, USA) consisting of a 1526 binary HPLC pump and a 996 photodiode array (PDA) detector, equipped with a Phenomenex Kinetex C₁₈ column (5 μm, 21 × 250 mm). LC-MS data were recorded on another Waters system, which included an ACQUITY Arc UPLC with a 2998 PDA detector and a ZQ single quadrupole detector, using a Waters XBridge C₁₈ column (2.5 μm, 2.1 × 150 mm). The NMR data were collected using a JEOL JNM-ECZ 500 MHz spectrometer (Tokyo, Japan). NMR spectra were recorded in deuterated chloroform or methanol, with chemical shifts reported in ppm and residual solvent peaks used as internal reference standards (δ_H/δ_C: 7.26/77.2 for chloroform; 3.31/49.0 for methanol). Dichloromethane (DCM) was obtained from Samchun Pure Chemical (Gyeonggi-do, Korea). Trifluoroacetic acid (TFA), TRA, MGO, DMSO, and Pd/C were purchased from Sigma-Aldrich (St. Louis, MO, USA) for chemical synthesis. MRS broth was obtained from BD Biosciences (San Jose, CA, USA). Eagle’s minimum essential medium (MEM), DMEM, Medium 199, fetal bovine serum (FBS), penicillin/streptomycin (P/S), MEM non-essential amino acids (NEAA), sodium pyruvate, L-glutamine, and trypsin-EDTA were purchased from Gibco (Grand Island, NY, USA). EGF, basal medium Eagle (BME), 5-FU, and DMSO were supplied by Sigma-Aldrich for biological evaluation, and DTIC was obtained from ChemFaces (Hubei, China). FxCycle PI/RNase staining solution, fluorescein isothiocyanate (FITC)-conjugated Annexin V/PI, antibodies against ERK1/2, phosphorylated ERK1/2, and β-actin were purchased from Thermo Fisher Scientific (Rockford, IL, USA). Antibodies against caspase-3, caspase-7, MEK1/2, and phosphorylated MEK1/2 were acquired from Cell Signaling Technology (Danvers, MA, USA). The p27 antibody was sourced from Santa Cruz Biotechnology (Santa Cruz, CA, USA), and antibodies against COX-2 (human) and COX-2 (mouse) were purchased from Cayman Chemical (Ann Arbor, MI, USA). The mouse epithelial cell line JB6 Cl41 (JB6) and NHDF were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). The melanoma cell lines SK-MEL-5 and SK-MEL-28 were provided by the Korean Cell Line Bank (KCLB, Seoul, Korea).

Bacterial source

L. gasseri ATCC 33323, isolated from human intestine,⁵⁶ was obtained from the Korea Collection for Type Cultures (accession number: KCTC 3163). Taxonomic assignment was performed by analyzing partial 16S rRNA gene sequences using PCR primers, and the sequence data were deposited in GenBank (accession number: M58820).

Culture and extraction

L. gasseri was cultured in 1 L of MRS broth under anaerobic conditions at 37°C. After 14 days of cultivation, sterile Amberlite XAD7HP resin (Sigma-Aldrich, St. Louis, MO, USA) was added to the medium (40 g/L), followed by shaking at 120 rpm for 2 h to allow for complete adsorption of organic molecules. The resin was then filtered through a filter cloth to remove any remaining broth, transferred to another flask, and immersed in acetone. The mixture was then incubated with shaking at room temperature for 24 h. After removing the resin by filtration, L. gasseri extract (6.3 g) was obtained by vacuum evaporation. The acetone extract was dissolved in 20 mL water and partitioned with chloroform (3 × 20 mL) to yield an organic material (538 mg).

Isolation of 1-acetyl-β-carboline

The chloroform-soluble fraction of L. gasseri was dissolved in methanol (50 mg/mL) to prepare a sample for preparative HPLC and then injected in 50-μL aliquots. The sample was eluted with 30% MeCN/H₂O containing 0.1% formic acid for 5 min, followed by a linear gradient to 100% over 25 min using a C₁₈ HPLC column (5 μm, 10.0 × 250 mm, 2.0 mL/min). The isolated compound (0.6 mg; retention time [t_R], 26.3 min) was identified as ABC through NMR and LC-MS data analyses, and its structure was confirmed by comparing the data with corresponding values in the literature.^39,40

Synthesis of 1-acetyl-β-carboline

MGO (1.4 mL, 9.37 mmol) was added to a solution of TRA (1 g, 6.25 mmol) in DCM (10 mL), followed by slow addition of TFA (1 mL) at room temperature. After stirring for 48 h, the reaction mixture was neutralized by adding a saturated solution of sodium hydrogen carbonate and then partitioned between chloroform (70 mL) and deionized water (3 × 70 mL). The synthesis and partitioning processes were repeated in the same manner, and the organic solvent-soluble portions were combined and concentrated under vacuum to obtain a crude product containing reaction intermediate 1 (1.2 g, 46.8%). Aromatization was performed directly on the crude intermediate (1.2 g) using 10% Pd/C (0.3 g) in DMSO (20 mL), without prior purification, to maximize synthetic yield. After refluxing for 6 h, the mixture was filtered through a bed of Celite (7.0 g). The organic solvent was removed, and the crude product was purified by flash column chromatography using a C₁₈ reverse phase column (30 × 130 mm) with water/methanol (4:6) to afford ABC (231 mg, 8.8%). ¹H NMR (methanol-d₄, 500 MHz) δ 8.45 (1H, d, J = 4.9, H-3), 8.29 (1H, d, J = 4.9, H-4), 8.20 (1H, d, J = 7.9, H-5), 7.69 (1H, d, J = 8.3, H-8), 7.59 (1H, dd, J = 7.6, 8.3, H-7), 7.29 (1H, dd, J = 7.9, 7.6, H-6), and 2.82 (3H, s, CH₃); ¹H NMR (CDCl₃, 500 MHz) δ 8.54 (1H, d, J = 4.8, H-3), 8.15 (2H, m, H-4, H-5), 7.59 (2H, m, H-7, H-8) 7.33 (1H, ddd, J = 1.8, 6.4, 8.0, H-6), and 2.89 (3H, s, CH₃); ¹³C NMR (CDCl₃, 150MHz) δ 203.3 (carbonyl carbon), 141.3 (C-8a), 138.2 (C-3), 136.0 (C-1), 135.5 (C-8b), 131.7 (C-4a), 129.4 (C-7), 122.0 (C-5), 120.9 (C-6), 120.7 (C-4b), 119.2 (C-4), 112.1 (C-8), and 26.1 (methyl carbon).

LC-MS quantification of ABC in L. gasseri cultures

MRS liquid broth was prepared in 300-mL Erlenmeyer flasks, each containing 98 mL of the medium for L. gasseri cultures (n = 8) or 100 mL of the medium alone for negative controls (n = 8). All media were autoclaved, UV-sterilized for 1 h, and then pre-conditioned under anaerobic conditions for 24 h. An L. gasseri inoculum was prepared by adjusting the optical density at 600 nm (OD₆₀₀) of the culture in MRS broth to 0.1 ± 0.02. A 2 mL aliquot of this suspension was inoculated into each flask containing 98 mL of the prepared MRS medium. At each time point, 200 μL samples from the L. gasseri cultures and the negative controls were collected and centrifuged at 10,000 rpm for 2 min. The resulting supernatants were used for LC-MS analysis. Additionally, OD₆₀₀ values of a subset of the L. gasseri cultures (n = 4) were monitored.

Quantification of ABC (t_R, 9.25 min) was performed using LC-MS (2.5 μm, 2.1 × 150 mm, 0.4 mL/min), eluted with 15% MeCN/H₂O (0.1% formic acid) for 1.5 min, followed by a linear gradient to 100% MeCN/H₂O (0.1% formic acid) over 8.5 min. Electrospray ionization MS was operated in positive ion mode with a capillary voltage of 3.30 kV, cone voltage of 47 V, desolvation gas flow of 700 L/h, cone gas flow of 100 L/h, desolvation temperature of 400°C, and source temperature of 120°C. A calibration curve was generated by plotting the MS peak areas of ABC standard solutions against their corresponding concentrations. A stock solution was prepared by accurately weighing 1.0 mg of ABC and dissolving it in 100% methanol, followed by serial dilution to produce calibration standards at 0.2, 0.15, 0.1, 0.05, and 0.03 μg/mL. Three replicates were prepared for each concentration. The resulting calibration curve, which exhibited a slope of 364,292, a y intercept of −30,577, and a coefficient of determination (r²) of 0.9991, was used to determine ABC concentrations in the samples by interpolating their MS peak areas.

Cell culture and cell viability assay

NHDF cells were cultured in a mixture of Medium 199 and DMEM supplemented with 10% FBS and 100 U/mL P/S. The melanoma cell lines SK-MEL-5 and SK-MEL-28 were cultured in MEM supplemented with 10% FBS, 100 U/mL P/S, 1× MEM NEAA, and 1× sodium pyruvate. JB6 cells were cultured in MEM supplemented with 5% FBS, 100 U/mL P/S, 1× MEM NEAA, and 1× sodium pyruvate. All cells were maintained at 37°C under 5% CO₂, with medium changes every 2–3 days. To assess the cytotoxicity of ABC, NHDF cells (5,000 cells/well) were seeded in 96-well plates and incubated for 24 h. After treatment with ABC (0, 25, 50, and 100 μM) for 24 and 48 h, 10 μL of WST-8 (BIOMAX, Guri, Korea) was added to each well. To evaluate the effect of ABC on melanoma cell viability, SK-MEL-5 and SK-MEL-28 cells (1,000 cells/well) were seeded in 96-well plates. Following a 24-h incubation, the cells were treated with ABC. In parallel, cells were treated with 5-FU and DTIC as positive controls under the same culture conditions. Cell viability was assessed using the WST-8 assay kit after 24, 48, 72, and 96 h of treatment. Absorbance was measured at 450 nm using a Multiskan SkyHigh spectrophotometer (Thermo Fisher Scientific, Vantaa, Finland).

Soft agar assay

To evaluate the effects of ABC on the anchorage-independent colony growth of melanoma cells (SK-MEL-5 and SK-MEL-28) and EGF-induced colony formation in JB6 cells, we performed soft agar colony formation assays following established protocols.⁸³ For JB6 cells, the assay involved solidifying a mixture of 0.5% agar containing BME, 10% FBS, 10% PBS, 2 mM L-glutamine, 5 μg/mL gentamicin, and different concentrations of ABC (0, 25, 50, and 100 μM) in the presence of EGF (10 ng/mL) in 6-well plates (3 mL per well). JB6 cells (8,000 cells/well) were suspended in 1 mL of BME medium with 0.3% agar containing EGF (10 ng/mL) and/or ABC and then layered onto the solidified bottom agar layer. For SK-MEL-5 and SK-MEL-28 cells, 3 mL of bottom agar containing 1× BME supplemented with 10% FBS, 2 mM L-glutamine, 25 μg/mL gentamicin, 0.6% agar, and varying concentrations of ABC (0, 25, 50, and 100 μM) was solidified in 6-well plates. Subsequently, SK-MEL-5 or SK-MEL-28 cells (8,000 cells/well) were suspended in 1 mL of top agar containing 1× BME medium supplemented with 10% FBS and 0.3% agar, along with the different concentrations of ABC, and layered onto the bottom agar. The plates were incubated at 37°C with 5% CO₂ for 2 weeks. After incubation, images of colonies were captured using an optical microscope (Leica Microsystems, Wetzlar, Germany), and the number of colonies were counted using Image-Pro Plus software (v.6.1; Media Cybernetics, Rockville, MD, USA).

Cell-cycle analysis

To assess the effect of ABC on cell-cycle progression in SK-MEL-5 and SK-MEL-28 cells, flow cytometric analysis using PI staining was conducted according to previously validated methods.⁸⁴ The cells were seeded in 6-well plates at a density of 1.5 × 10⁵ cells/well and cultured overnight. The cells were then treated with different concentrations of ABC (0, 25, 50, and 100 μM) for 24 h. After treatment, the cells were harvested by trypsinization, washed with cold PBS, and fixed in 70% ethanol at −20°C for more than 24 h. The fixed cells were washed with cold PBS and stained with FxCycle PI/RNase staining solution in the dark for 15 min at room temperature. Cell-cycle distribution was analyzed by flow cytometry (CytoFLEX; Beckman Coulter, CA, USA), and the data were processed using CytExpert software (v.2.2, Beckman Coulter).

Analysis of apoptosis

To evaluate ABC-induced apoptosis in SK-MEL-5 and SK-MEL-28 cells, Annexin V-FITC/PI staining was performed according to the manufacturer’s instructions and previously published protocols.⁸⁵ Cells were seeded in 6-well plates at a density of 1.5 × 10⁵ cells/well. Following 24 h of culture, the cells were treated with ABC (0, 25, 50, and 100 μM) for 72 h. After treatment, both the adherent and floating cells were collected by trypsinization, washed with cold PBS, and resuspended in 1× binding buffer. The cells were then stained with Annexin V/PI according to the manufacturer’s instructions and incubated for 15 min at room temperature in the dark. The samples were immediately analyzed using flow cytometry (CytoFLEX; Beckman Coulter, CA, USA) to quantify the percentage of apoptotic cells. Data were acquired and analyzed using the CytExpert software (v.2.2, Beckman Coulter).

Western blot analysis

Western blotting was performed to investigate the changes in the expression of cell-cycle- and apoptosis-related proteins, including p27, caspase-3, and caspase-7. Additionally, the analysis was conducted to evaluate the potential inhibition of the MEK-ERK pathway by detecting p-MEK, MEK, p-ERK, and ERK, as well as to assess the modulation of COX-2 expression, a key mediator of inflammation and tumor progression. β-actin served as a loading control. SK-MEL-5 and SK-MEL-28 cells were seeded in 100-mm dishes at a density of 1 × 10⁶ cells. After 24 h of incubation, the cells were treated with different concentrations of ABC (0, 25, 50, and 100 μM) for an additional 48 and 72 h. Following treatment, the cells were washed twice with cold PBS and lysed using PRO-PREP Protein Extraction Solution (iNtRON Biotechnology, Gyeonggi-do, Korea) containing a protease and phosphatase inhibitor cocktail to extract total protein. The protein concentration was determined using a bicinchoninic acid assay protein assay kit (Thermo Fisher Scientific, Waltham, MA, USA). The protein samples were resolved on 10%–12% SDS-PAGE gels and subsequently transferred onto polyvinylidene difluoride membranes using an iBlot 3 Dry Blotting System (Thermo Fisher Scientific). After transfer, the membranes were blocked with 3% nonfat dry milk in Tris-buffered saline containing 0.1% Tween 20 (TBS-T) for 1 h at room temperature to block the nonspecific detection. The membranes were then incubated with primary antibodies against p27 (1:1,000), caspase-3 (1:1,000), caspase-7 (1:1,000), p-MEK (1:1,000), MEK (1:1,000), p-ERK (1:1,000), ERK (1:1,000), COX-2 (Cayman #160107, 1:500), and β-actin (1:5,000) diluted in 3% skimmed milk in 1× TBS-T overnight at 4°C. The membranes were subsequently washed three times with TBS-T before incubation with secondary antibodies. Following incubation with a horseradish peroxidase (HRP)-conjugated secondary antibody for 1 h at room temperature, the membranes were washed three times with TBS-T, and the protein bands were detected using an enhanced chemiluminescence HRP substrate (Thermo Fisher Scientific) according to the manufacturer’s instructions. Protein bands were visualized using a chemiluminescent imaging system (LAS-Amersham Imager 600; GE Healthcare, Uppsala, Sweden). Band intensities were quantified with ImageJ (v.1.54) and normalized to β-actin.

In vivo experiments

All animal experiments were conducted in accordance with the guidelines of the Institutional Animal Care and Use Committee and approved by the Dongshin University Animal Ethical Committee (approval number DSU2022-03-03). Female ICR mice (5 weeks old) were obtained from a certified laboratory animal supplier (Samtako, Osan, Korea), and housed under controlled conditions with a 12/12 h light/dark cycle, regulated temperature and humidity, and free access to food and water. The TPA-induced epidermal hyperplasia model used in this study is a well-established in vivo system for evaluating skin inflammation and proliferation, as demonstrated in previous reports.^78,79 After 1 week of acclimatization, the mice were randomly assigned to seven treatment groups as follows: vehicle (n = 7), TPA (10 nmol, n = 8), TPA/low-dose ABC (0.02 μmol/mouse, n = 7), TPA/medium-dose ABC (0.2 μmol/mouse, n = 7), TPA/high-dose ABC (2 μmol/mouse, n = 8), TPA/5-FU (0.1 μmol/mouse, n = 7), and high-dose ABC alone (2 μmol/mouse, n = 5). Four days before treatment, the dorsal skin of each mouse was depilated using a shaver and depilatory cream. On the day of the experiment, ABC (0.02, 0.2, or 2 μmol/mouse) or 5-FU (0.1 μmol/mouse) was diluted in acetone and topically applied to the depilated dorsal area. After 30 min, 10 nmol TPA in acetone was applied topically to the same area. The vehicle group was treated with a mixture of DMSO and acetone without active compounds, and the high-dose ABC-alone group was treated with 2 μmol ABC without TPA application. The mice were sacrificed 4 h post-TPA treatment, and skin tissues were fixed in 10% formalin.

H&E staining

The mice were sacrificed, and the dorsal skin tissues were excised and fixed in 10% formalin. The fixed tissues were dehydrated and cleared using an automated tissue processor (Leica TP1020; Leica, Wetzlar, Germany). After processing, the tissues were embedded in paraffin wax using a paraffin-embedding station (Leica EG1150 H; Leica). Paraffin-embedded tissue blocks were sectioned at a thickness of 5 μm using a rotary microtome (HistoCore MULTICUT, Leica). The sections were rehydrated and stained with hematoxylin to visualize the nuclei, followed by eosin staining of the cytoplasm. After staining, the sections were gradually dehydrated in ethanol, cleared in xylene, and covered with coverslips using a mounting medium. The slides were scanned at 20× magnification using a slide scanner (Pannoramic Digital Slide Scanner; 3DHistech Ltd., Budapest, Hungary). The images were enlarged, viewed, and saved using the Slide Viewer 2.6 program (3DHistech Ltd.). The thickness of the epidermis was measured using Image-Pro Plus software (v.6.1; Media Cybernetics, Rockville, MD, USA).

Immunohistochemical staining

Tissue sections were deparaffinized and rehydrated through a series of xylene and graded ethanol washes, followed by heat-induced antigen retrieval in 10 mM sodium citrate buffer. To block endogenous peroxidase activity, the sections were treated with 0.3% hydrogen peroxide. Subsequently, a serum-blocking solution was applied to minimize nonspecific antibody binding. The sections were then incubated overnight at 4°C with a primary antibody against COX-2 (Cayman #160106, 1:300). The following day, the secondary antibody was applied using the ImmPRESS HRP Universal Antibody Polymer Detection Kit (Vector Laboratories), and the DAB peroxidase substrate (Vector Laboratories) was used to visualize the staining. Hematoxylin was applied as a counterstain. The stained sections were scanned using the Pannoramic SCAN II system (3DHistech Ltd., Budapest, Hungary), and COX-2 expression was quantified using Image-Pro Plus software (v.6.1; Media Cybernetics, Rockville, MD, USA).

Statistical analyses

All experiments were performed in triplicate unless otherwise stated. Data are presented as the mean ± SD. Statistical differences between groups were evaluated using the Student’s t test, with significance defined as p < 0.05.

Data and code availability

The data that support the findings of this study are available on request from the corresponding author.

Acknowledgments

This research was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (grant nos. 2020R1C1C1007535 and RS-2025-25401670 to I.H.H. and 2022R1A5A2029546 to M.-H.L.).

Author contributions

Conceptualization and supervision, M.-H.L. and I.H.H.; investigation and methodology, M.L., S.-Y.H., K.W., and J.K.; data curation and formal analysis, M.L. S.-Y.H., M.-H.L., and I.H.H.; resources, M.L.; writing – original draft, M.L., S.-Y.H., M.-H.L., and I.H.H.; writing – review & editing, M.-H.L. and I.H.H.; funding acquisition, M.-H.L. and I.H.H.; all authors read and approved the final manuscript.

Declaration of interests

The authors declare no competing interests.