The fresh leaves of Clinacanthus nutans (Burm. f.) Lindau (cultured plants) were collected from Mae Rim District, Chiang Mai province in compliance with relevant institutional, national, and international guidelines and legislation. The plant specimens were identified by a botanist, Dr. Wittaya Pongamornkul, and deposited at the Queen Sirikit Botanic Garden Herbarium (QSBG herbarium) under a voucher number; WP8603. To comply with the ConPhyMP checklist (Heinrich et al., 2022), multiple chemical profile characterization methods should be employed. In this study, the C. nutans extract from the same source has been previously characterized through TLC, HPLC-UV/DAD, LC-MS/MS (Panya et al., 2020) and GC-MS/MS (Thongyim et al., 2023). It should be noted that direct overlay of the chromatograms with the commercial standards of potential bioactive compounds was performed to confirm the presence of the compounds. Moreover, we have performed the characterization of the extract thoroughly and to make sure that the extracts yield reproducing profiles. These analyses revealed several bioactive compounds, including glyceryl 1,3-distearate, identified as a glycolipid, and kaempferol 3-O-feruloyl-sophoroside 7-O-glucoside, identified as a C-glycoside flavone. These compounds are listed as active constituents in C. nutans according to the Thai Herbal Pharmacopoeia. Previously, the C. nutans used in this work has also been investigated using sequence-related amplified polymorphism (SRAP) technique in comparison with other sources of C. nutans in Thailand. The DNA patterns from seven pairs of SRAP markers for this C. nutans were reported (Chiangchin et al., 2023). The leaves of C. nutans were washed and dried at 60°C in a hot air oven for 72 h. The sample was ground to powder before resuspending in 70% ethanol at a ratio of 1:20 (w/v) and shaking at 160 rpm/min 25°C for 72 h. The total solution was evaporated in a rotary evaporator to collect the 70% ethanolic crude extracts. The 70% ethanolic crude extracts were dried in a chemical fume hood and stored at 20°C until use. According to the Thai Herbal Pharmacopoeia, the traditional method for preparing C. nutans extract involves crushing the leaves and soaking them in 40% (v/v) ethanol or Thai rice whiskey.

The antibacterial activity was judged using a minimum inhibitory concentration (MIC) and a minimum bactericidal concentration (MBC). The MIC was conducted by the broth dilution method. Clinacanthus nutans extracts and glyceryl 1,3 distearate were prepared in Brain-heart infusion (BHI) broth (BD DifcoTM, Franklin Lakes, NJ, United States) at the indicated concentrations (two-fold dilution) where antibiotic gentamicin (Sigma-Aldrich, St. Louis, MO, United States) was used as a positive control. The bacterial suspension including P. gingivalis, P. intermedia, and Streptococcus mutans was adjusted for the turbidity to be matched with a 0.5 McFarland standard. The 50 µL of standardized bacteria suspension was added to 50 µL of prepared C. nutans extracts, glyceryl 1,3 distearate, or gentamycin in a 96-well plate. The plate was incubated at 37°C, under anaerobic conditions for 24 h. The lowest concentration with no visible growth after macroscopic evaluation was considered as MIC. After then, the culture was then streaked on BHI Agar (BD DifcoTM, Franklin Lakes, NJ, United States) and incubated at 37°C, under anaerobic conditions for a further 24 h. The lowest concentration of the plate with no visible growth was considered as MBC.

HGF-1 cell line (Human Gingival Fibroblasts cells; ATCC®CRL-2014 Manassas, VA, United States) was cultured in the low glucose Dulbecco’s Modified Eagle Medium (DMEM; glucose 1 g/L) (Gibco, Thermo Fisher Scientific, MA, United States) supplemented with 10% (v/v) FBS and 1% (v/v) 200 mM L-glutamine (Gibco, Thermo Fisher Scientific, MA, United States) at 37°C in a 5% CO₂ incubator.

The effect of LPS from P. gingivalis (InvivoGen, San Diego, CA, United States), C. nutans extract, and glyceryl 1,3 distearate on cell viability of HGF-1 cells was determined using cell viability assay. In brief, HGF-1 cells were plated a day before the experiment in 96-well plates (5,000 cells/well). The samples were prepared in media with low glucose (1 g/L) and high glucose (4.5 g/L) as followed concentrations; 0.78–100 μg/mL (2-fold dilution) of LPS derived from P. gingivalis (LPS-PG) or 4–2,500 μg/mL (5-fold dilution) of C. nutans extracts, or 0.63–20 μg/mL (2-fold dilution) of glyceryl 1,3 distearate (Sigma-Aldrich, St. Louis, MO, United States). The 100 µL of the sample was added to the cells and incubated for 24 h. The cell viability was measured by using PrestoBLUE™ cell viability reagent (Thermo Fisher Scientific, MA, United States) following the manufacturer’s protocol. The percentage of cell viability relative to the non-treatment control (set as 100%) was calculated as the equation below. %  Cell viability = OD 570 ‐ OD 595  treated cells / OD 570 ‐ OD 595  non ‐ treated cells × 100

To determine the effect of glucose on inflammation induction, the HGF-1 cells were treated with the LPS-PG in the presence of low-glucose (1 g/L) and high-glucose (4.5 g/L) media. The HGF-1 cells were plated at 50,000 cells/well in 12-well plates 1 day before the experiment. The cells were treated with LPS-PG at concentrations of 0.1, 1, and 10 μg/mL, which were prepared in low-glucose and high-glucose media for 24 h. The inflammation in these two conditions was compared by monitoring the alteration of pro-inflammatory and inflammatory cytokines including IL-6, IL-8, and TNF-a judged by real-time PCR. In brief, total RNA was isolated using TRIzol^® reagent (Invitrogen, Carlsbad, CA, United States). The total RNA (500 ng) was used as a template for cDNA synthesis using Tetro cDNA Synthesis Kit (Bioline USA Inc., United States). Real-time PCR was then conducted in QIAquant real-time PCR cyclers (Qiagen, Venlo, Netherlands) using Luna^® Universal qPCR Master Mix (New England Biolabs, MA, United States) and specific primers (Supplementary Table S1). Fold changes of gene expression were calculated with the 2^−ΔΔCT method where GADPH was used as the housekeeping gene for the normalization of gene expression.

To measure the effect of C. nutans extract and glyceryl 1,3 distearate on lowering the expression of inflammatory genes, the cells were treated with LPS-PG at a concentration of 10 μg/mL in high-glucose media in the presence or absence of C. nutans extracts (25, 50, and 100 μg/mL) or glyceryl 1,3 distearate (5, 10, and 20 μg/mL). After 24 h of treatment, the cells were harvested followed by real-time PCR following the protocol described above.

The HGF-1 cells were plated at 50,000 cells/well in 12-well plates 1 day before the experiment and the cells were treated with LPS-PG (10 μg/mL) in the absence or presence of C. nutans extract (100 μg/mL) in a high-glucose condition. The cells were incubated at 37°C in a 5% CO₂ humidified atmosphere for 24 h. The culture supernatant was collected and used for determining the cytokine IL6 and IL8 by using a Human ELISA kit (ImmunoTools, Friesoythe, Germany). In brief, the capture antibody was added to the plate one night before the experiment at room temperature. The blocking buffer (1X PBS containing 2% bovine serum albumin and 0.05% Tween-20) was added to the well for 1 h. The buffer was removed and then the collected culture supernatant and the dilute standard protein were added to the wells. The plate was incubated at room temperature for 2 h before being washed with a wash buffer (1X PBS containing 0.05% Tween-20) for 5 times. The biotinylated detector antibody was added before adding the HRP-conjugated streptavidin. After 5-time washing, the 3–3′,5,5′-tetramethylbenzidine (TMB) substrate (Thermo Fisher Scientific, MA, United States) was added. After stopping the reaction with 2 N sulfuric acid, the absorbance was measured at OD450. The absorbance values were used to calculate the protein concentration relative to the corresponding standard protein curve.

The HGF-1 cells were plated at 25,000 cells/well in 24-well plates 1 day before the experiment. The cells were treated with 10 μg/mL of LPS-PG in the absence or presence of C. nutans extract (100 μg/mL) in a high glucose media condition for 24 h. The treated cells were harvested and fixed with 3.6% formaldehyde for 15 min before being permeabilized with 0.2% Triton X for 15 min. The plate was blocked with 2% bovine serum albumin for 1 h before monoclonal antibody specific to NF-κB p65 clone D14E12 (dilution of 1:500, Cell Signaling, MA, United States) was added. The plate was incubated for 2 h followed by a washing step and the addition of a secondary antibody conjugated with Alexa-Fluor 488 (dilution of 1:1,000, Thermo Fisher Scientific, MA, United States) and Hoechst 33342 for nuclear staining (dilution 1:1,000, Thermo Fisher Scientific, MA, United States). The nuclear translocation of NF-κB protein was monitored under a confocal microscope (Nikon, Amsterdam, Netherlands), and NISelements AR program was used for fluorescence detection and image analysis.

The cells were plated at 25,000 cells/well in 24-well plates 1 day before the experiment and were treated with LPS of P. gingivalis (10 μg/mL) in the absence or presence of C. nutans extract (300 μg/mL) in a high glucose media condition. The cells were incubated at 37°C in a 5% CO₂ humidified atmosphere for 30 min to evaluate IKKα/β, IκBα, and NF-κB p65 and their phosphorylated forms whereas COX2 was detected at 24 h after treatment. The SDS-PAGE was performed using 12% polyacrylamide and then transferred to a nitrocellulose membrane. Membranes were blocked with 5% skim milk in 0.1%TBST for 1 h at room temperature before adding the monoclonal antibodies specific to phospho-IKKα/β, phospho-IκBα, and phospho NF-κB p65, COX2, GAPDH (Cell Signaling Technology, Boston, MA, United States) (1:1,000 dilution) followed by a goat anti-rabbit or mouse secondary antibody (ABclonal, Woburn, MA, United States) (1:1,000 dilution). The protein bands were detected using SuperSignal™ West Pico PLUS Chemiluminescent Substrate (Thermo scientific, IL, United States) and visualized by ImageQuant LAS 500 chemiluminescence CCD camera (GE, Boston, MA, United States). The membranes were stripped with 0.2 M NaOH for 3–5 min and were washed with 0.1%TBST for 5 min three times and used to detect IKKα/β, IκBα, and NF-κB p65 protein as the protocol described above. The protein band intensity was analyzed by ImageJ analysis software and normalized by GADPH (ABclonal, Woburn, MA, United States).

Total RNA extraction was carried out from HGF-1 cells in three different groups: DMSO-treated, LPS-treated, and LPS-treated in the presence of C. nutans extract (LPS + C. nutans) (n = 3), using Trizol reagent (Invitrogen, Carlsbad, CA, United States). Subsequently, samples from each group were combined. RNA sequencing was performed using the custom service of Novogene Corporation Inc. (Sacramento, CA, United States). mRNA was isolated using poly-T oligo-attached magnetic beads and fragmented using a fragmentation buffer. Random hexamers were added for reverse transcription to generate the first strand. Subsequently, the addition of dNTPs and DNA polymerase I resulted in the formation of double-stranded cDNA. The construction of the cDNA library consisted of a series of steps: end repair, A-tailing, adapter ligation, size filtering, USER enzyme digestion, amplification, and purification. Concentrations of the cDNA library were determined using both real-time PCR and the Qubit HS Assay kit (Thermo Fisher Scientific, Waltham, MA, United States), following the manufacturer’s instructions. The size distribution of the library was assessed using an Agilent Bioanalyzer (Agilent Technologies). Sequencing took place using an S4 flow cell and an Illumina NovaSeq 6000 (Illumina Inc.). Raw reads were processed to remove adaptor, poly-N and low-quality sequences. The resultant clean reads were aligned to the reference genome (GRCh38) using HISAT2 (v2.0.5). Using featureCounts (v1.5.0-p3) and gene annotations, gene expression was quantified and normalized as FPKM (Fragments per kilobase of exon model per million mapped reads). Next, DESeq2 (1.10.1) was used to analyze differential gene expression across the conditions. Genes were identified as differentially expressed if |log2(FoldChange)| > 1 and p.adj < 0.05. Following this, downregulated differentially expressed genes between the conditions of LPS + C. nutans and LPS were subjected to Gene Ontology (GO) enrichment analysis using the clusterProfiler package (version 4.6.2) in R. Enriched GO terms with a false discovery rate (FDR) < 0.05 were considered significant. Pathway analysis on the downregulated differentially expressed genes between the LPS + C. nutans and LPS conditions was carried out using KEGG analysis on the DAVID platform (https://david.ncifcrf.gov).

The bar graphs representing the mean and standard error of the mean (SEM) were plotted from three independent measurements. The statistical analyses were performed using Student’s t-test (Xing et al., 2024) of GraphPad Prism Software version 9 (GraphPad Software, Inc., La Jolla, CA, United States), where significant differences were indicated as follows (* indicates p < 0.05, **p < 0.01, ***p < 0.001).

The antibacterial activity of C. nutans has been previously reported but none of these reports have addressed its activity against anaerobic bacteria species that cause periodontitis. Therefore, we conducted novel experiments to assess the effect of C. nutans and its identified bioactive compound glyceryl 1,3 distearate, on inhibiting the growth inhibition of oral anaerobic bacteria including P. gingivalis, P. intermedia, and S. mutans. We treated the bacteria cultures with varying concentrations of C. nutans extract (ranging from 15.62 to 250 mg/mL) and glyceryl 1,3 distearate (ranging from 31.25 to 500 μg/mL) and determined their inhibition effects using minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) assays. The results indicated that both C. nutans extract and glyceryl 1,3 distearate had a limited impact on bacterial propagation (Table 1). The MIC and MBC for C. nutans extract against P. gingivalis, P. intermedia, and S. mutans was 250 mg/mL. Glyceryl 1,3 distearate exhibited less inhibitory effect, with MIC and MBC values of up to 500 μg/mL against P. intermedia and S. mutans, and 250 μg/mL against P. gingivalis. Notably, the positive control, gentamicin treatment, resulted in MIC and MBC values ranging from 7.81 to 31.25 μg/mL. When comparing the potency to glyceryl 1,3 distearate to gentamicin, gentamicin showed approximately 15.99-fold higher MIC against P. gingivalis, a 64.02-fold higher MIC against P. intermedia, and a 16-fold higher MIC against S. mutans. On the other hand, C. nutans extract exhibited more than a 1,000-fold higher MIC, demonstrating limited activity in controlling the growth of periodontitis-causing bacteria.

We investigated the effect of LPS-PG on the upregulation of key inflammatory gene expression (IL6 and IL8) and its potential to induce cell death in HGF-1 cells. We compared two treatment conditions, high-glucose and low-glucose conditions, to assess the effect of LPS on individuals with and without diabetes. The result revealed that LPS treatment elicited an inflammation response but had no effect on cytotoxicity (Figure 1). Notably, following LPS treatment (0.78–100 μg/mL), the cell viability of HGF-1 cells remained consistently above 95% (Figure 1A). Expectedly, LPS treatment significantly induced an inflammatory response (Figure 1B). Treatment of the HGF-1 cells with LPS concentrations of 0.1, 1, 10 μg/mL resulted in the upregulation of mRNA expression of the inflammatory genes, IL6 and IL8. Importantly, this upregulation was significantly pronounced under high-glucose conditions compared to low-glucose conditions. Under high-glucose conditions, a concentration of 10 μg/mL led to a remarkable 72.43-fold increase in mRNA expression of IL8 and 14.02-fold increase in IL6, whereas under low-glucose conditions, these values were 17.42-fold and 4.08-fold, respectively (Figure 1B). These results suggest that LPS stimulation in high-glucose conditions exerts a strong induction of the inflammatory response in HGF-1 cells.

In a previous study, we reported the anti-inflammatory activity of C. nutans extract and glyceryl 1,3 distearate in a bovine mastitis model (Thongyim et al., 2023). Building on this, we hypothesized that C. nutans extract and glyceryl 1,3 distearate would exhibit anti-inflammatory properties in HGF-1 cells following LPS stimulation. We conducted a cell viability assay to evaluate the toxicity of C. nutans extract and glyceryl 1,3 distearate on HGF-1 cells (Figures 2A,B). The results demonstrated that concentrations of up to 500 μg/mL of C. nutans extract and up to 20 μg/mL of glyceryl 1,3 distearate had no adverse effects on cell viability. Subsequently, we treated the cells with varying concentrations of C. nutans extract (25–100 μg/mL) and glyceryl 1,3 distearate (5–20 μg/mL) following LPS stimulation at a concentration of 10-µg/mL in high-glucose conditions. Real-time PCR analysis revealed that the C. nutans extract had a potent dose-dependent effect in reducing mRNA expression of IL6 and IL8 (Figures 2C,D). Treatment with a concentration of 100 μg/mL resulted in a significant reduction of relative expression, lowering IL6 to 0.43-fold and IL8 to 0.29-fold (Figures 2C,D). Conversely, glyceryl 1,3 distearate treatment showed no significant difference (Figures 2E,F). In summary, C. nutans extract, but not glyceryl 1,3 distearate, demonstrated potential in inhibiting inflammation induced by LPS stimulation.

Furthermore, we validated the inhibitory effect of C. nutans extract on the production of IL6 and IL8 at the protein levels. Upon LPS stimulation, the secreted levels of IL6 and IL8 increased significantly, reaching 424.21 pg/mL and 302.25 pg/mL, respectively. However, in the presence of C. nutans extract, these levels were markedly suppressed (Figures 2G,H). Specifically, the extract reduced IL6 to 236.74 pg/mL and IL8 level to 234.50 pg/mL, highlighting the effectiveness of C. nutans extract in attenuating LPS-induced inflammation under high-glucose conditions.

The NF-κB signaling pathway is recognized as a key driver of inflammation, particularly in response of LPS stimulation. Consequently, we aimed to investigate whether the inhibitory effect of C. nutans extract operated through the NF-κB signaling pathway. HGF-1 cells were subjected to treatment with LPS (10 μg/mL) in the presence or absence of C. nutans extract (100-µg/mL). As expected, we observed a reduction in NF-κB nuclear translocation in cells treated with C. nutans extract compared to those subjected to LPS stimulation alone (Figure 3A). To further validate the effect of C. nutans extract on NF-κB signaling, we employed an immunoblotting technique. In agreement with the observed increase in NF-κB nuclear translocation, the levels of phosphorylated IκBα and NF-κB p65 were significantly elevated following LPS stimulation, but these increases were notably attenuated in the presence of C. nutans extract (Figure 3B).

Furthermore, we accessed changes in the protein levels of COX2, a pivotal regulator of an inflammatory process, after treatment with C. nutans extract in LPS-exposed cells (Figure 3C). LPS significantly increased the COX2 expression, validating the inflammatory response in the cells. However, the overexpression of COX2 was inhibited in the presence of C. nutans extract. Taken together, these results showed the effectiveness of C. nutans extract in attenuating inflammation by suppression of NF-κB signaling pathway.

To further assess the anti-inflammatory activity of C. nutans extract, we conducted RNA-seq analysis on HGF-1 cells subjected to different treatments: DMSO-treated control cells, LPS-treated cells (10 μg/mL for 24 h), and LPS-treated cells in the presence of C. nutans extract (100 μg/mL for 24 h). RNA sequencing was performed using the Illumina NovaSeq 6000 platform, generating over 40 million raw reads for each sample. Following trimming, we obtained over 40 million clean reads with a Q30 score exceeding 94%, indicating excellent data quality (Table 2). Subsequently, the reads were aligned to the reference genome, annotated, and quantified. The counts were then normalized as fragments per kilobase of exon model per million mapped reads (FPKM). Differential gene expression analysis was carried out using DESeq2, where genes with an adjusted p-value (p.adj) less than 0.05 and an absolute log2 fold change (|log2FC|) greater than 1 were considered statistically significant (Figures 4A–C). In comparison between LPS-treated cells and control cells, we identified 1,359 differentially expressed genes (DEGs) with 1,038 upregulated and 321 downregulated (Figure 4B). When comparing cells treated with LPS in the presence of C. nutans extract (LPS + C. nutans) to LPS-treated cells, we found 622 DEGs comprising 85 upregulated and 537 downregulated genes (Figure 4C). To evaluate the relationship in gene expression profiles among the three conditions, we performed the Principal Component Analysis (PCA) (Figure 4D). LPS treatment led to a substantial separation from the control cells along principal component 1, accounting for a total variance of 89.56%. However, when treating the LPS-exposed cells with C. nutans extract, this separation was reduced to cluster closer to the control cells. Along principal component 2, the cells treated LPS + C. nutans extract showed a slight separation from both the control and LPS-treated cells, which clustered together, accounting for 10.44% of the total variance. In summary, these finding suggest that LPS induced a significant perturbation in gene expression which is partially alleviated by C. nutans treatment.

To elucidate the biological processes and pathways overrepresented among these DEGs, we conducted GO-BP and KEGG analysis, focusing on the DEGs that exhibited downregulation upon C. nutans treatment. The top fifteen enriched terms identified through GO-BP analysis primarily encompassed processes related to immune function and inflammation (Figure 5A). These processes included cytokine-mediated signaling, regulation of response to biotic stimulus, and response to LPS. These findings were validated by KEGG analysis (Figure 5B), which revealed the top enriched pathways including cytokine-cytokine receptor interaction, the TNF signaling pathway, and the NF-κB pathway. Collectively, these data suggest that C. nutans treatment results in the downregulation of inflammation and immune signaling pathways stimulated by LPS. To explore these changes in more detail, we individually examined the gene expression of the downregulated DEGs within GO term “cytokine-mediated signaling” (Figure 6). Within this gene set, we observed the upregulation of numerous cytokines, chemokines, and interleukins, as well as proteins of the NF-κB, TNF and STAT families following LPS treatment. Some of those upregulations could be significantly suppressed by C. nutans treatment. Notably, the expression of CXCL10, which exhibited the highest expression levels following LPS stimulation, was validated by the qPCR data (Supplementary Figure S1). The result showed that treatment with 100 µg/mL of C. nutans led to a significant reduction in CXCL10 expression, down to 0.03-fold. Taken together, C. nutans treatment was observed to downregulate inflammatory and immune signaling pathways activated by LPS treatment.