mRNA transcriptome profiling of human hepatocellular carcinoma cells HepG2 treated with Catharanthus roseus-silver nanoparticles

Abstract

BACKGROUND

The demand for the development of cancer nanomedicine has increased due to its great therapeutic value that can overcome the limitations of conventional cancer therapy. However, the presence of various bioactive compounds in crude plant extracts used for the synthesis of silver nanoparticles (AgNPs) makes its precise mechanisms of action unclear.

AIM

To assessed the mRNA transcriptome profiling of human HepG2 cells exposed to Catharanthus roseus G. Don (C. roseus)-AgNPs.

METHODS

The proliferative activity of hepatocellular carcinoma (HepG2) and normal human liver (THLE3) cells treated with C. roseusAgNPs were measured using MTT assay. The RNA samples were extracted and sequenced using BGIseq500 platform. This is followed by data filtering, mapping, gene expression analysis, differentially expression genes analysis, Gene Ontology analysis, and pathway analysis.

RESULTS

The mean IC₅₀ values of C. roseusAgNPs on HepG2 was 4.38 ± 1.59 μg/mL while on THLE3 cells was 800 ± 1.55 μg/mL. Transcriptome profiling revealed an alteration of 296 genes. C. roseusAgNPs induced the expression of stress-associated genes such as MT, HSP and HMOX-1. Cellular signalling pathways were potentially activated through MAPK, TNF and TGF pathways that are responsible for apoptosis and cell cycle arrest. The alteration of ARF6, EHD2, FGFR3, RhoA, EEA1, VPS28, VPS25, and TSG101 indicated the uptake of C. roseus-AgNPs via both clathrin-dependent and clathrin-independent endocytosis.

CONCLUSION

This study provides new insights into gene expression study of biosynthesised AgNPs on cancer cells. The cytotoxicity effect is mediated by the aberrant gene alteration, and more interestingly the unique selective antiproliferative properties indicate the C. roseusAgNPs as an ideal anticancer candidate.

Key Words: Catharanthus roseus, HepG2, Silver nanoparticles, Transcriptome, oxidative stress, Apoptosis, Cell cycle

Core Tip: Despite the increased attention on cancer nanomedicine which is advantageous to overcome the limitations of conventional cancer treatment, the information on the selectivity and detailed mechanisms at the cellular and molecular level remain unclear. To evaluate its selectivity effects, the proliferative activity of both liver cancer cells HepG2 and normal liver cells THLE-3 in response to Catharanthus roseus-silver nanoparticles (C. roseus-AgNPs) was assessed. To determine the possible signalling pathways induced by the C. roseus-AgNPs, the mRNA transcriptome profiling of hepatocellular carcinoma cell line HepG2 was performed, highlighting the expression of genes associated with oxidative stress, apoptosis, and cell cycle arrest. The elucidation of its selectivity effects and detailed wide genome screening would enlighten the cellular and molecular signalling pathways and provide a strong basis towards the development of C. roseus-AgNPs as an anticancer drug for liver cancer.

INTRODUCTION

Nanoparticles are materials or discrete clusters of atoms having dimensions within 1-100 nm[1]. Having a large surface area-to-volume ratio with unique biological properties, nanoparticles have gained immense usage in the early diagnosis and treatment of cancer, the application of which is termed nano oncology[2,3]. Nanoparticles can offer an alternative to the current conventional chemotherapeutic agents which although exhibit high efficacy in killing cancer cells, still suffer from significant drawbacks due to the poor specificity in causing severe damage to healthy cells[4]. Amongst various nanoparticles, silver nanoparticles (AgNPs) have been reported to demonstrate a significant biological effect, particularly in the healthcare industry[5,6]. Concurrently, the market demand for eco-friendly, hazard-free, and cost-effective synthesis of AgNPs was higher as many of the common nanoparticle production methods involved hazardous chemicals and high energy- consumption[7]. One of the most effective biogenic approaches is to use plant extract that contains metabolites, which can enhance the reduction of silver ions. Plant extract-mediated silver nanoparticle synthesis is found to have a promising anticancer property. Plant extract-based synthesis is largely favoured due to the lower degree of adverse effect as well as the low cost of synthesis that enables large-scale production. Moreover, biologically active ingredients or phytomolecules in the plant extract act as reducing agents to promote the synthesis of AgNPs[8]. A previous study has corroborated the anticancer property of AgNPs, for example, biosynthesised AgNPs using Acalypha Indica, which exhibited anti-cancer activity against human breast cancer cell line MDA-MB-231[9]. In another study, AgNPs synthesised using leaf extract of Tropaeolum majus L. also demonstrated anti-cancer properties on the MCF7 cell line[10]. These findings cumulatively proved the anti-cancer property of the biogenic AgNPs.

Previously, an herbal plant Catharanthus roseus (C. roseus) G. Don has demonstrated its ability as a reducing agent to synthesise AgNPs. This plant is commonly known as periwinkle which belongs to the Apocynaceae family[11]. This plant is very synonymous with its content, indolomonoterpenic alkaloids vincristine and vinblastine[12]. These compounds are commonly used in the treatment of several malignant conditions, such as Hodgkin’s and non-Hodgkin’s lymphomas, acute lymphoblastic leukaemia, neuroblastoma and breast carcinoma[13]. These alkaloids may be responsible for the reduction of the silver ions to AgNPs and at the same time exert their function by disrupting the mitotic spindle apparatus of microtubules through tubulin interaction, thus blocking the mitosis process, and arresting the cancer cells during metaphase[14].

An understanding of the anti-cancer mechanisms of AgNPs at the molecular level would provide detailed insight into various physiological processes involved. This is achievable via transcriptome analysis, a holistic view of gene expression. An overview or snapshot of the gene expression landscape could reveal the intricate molecular network that underlies the myriad of biological processes in a cell. As compared to hybridisation-based RNA quantification methods such as microarray analysis, this sequencing-based transcriptome detection can perform well within a wide range of circumstances, where this method could quantify gene expression with low background, high accuracy, and high reproducibility levels with significant dynamic range transcriptome analysis can detect subtle changes in gene expression, mutations, splice variants and fusion genes that cannot be identified by microarrays[15].

Fuelled by the intriguing capacity of the transcriptome analysis, in this study, we endeavoured to carry out an mRNA transcriptome profiling of the human hepatocellular carcinoma cell (HepG2) treated with AgNPs synthesised using an aqueous extract of C. roseus G. Don. The human hepatocellular carcinoma cell (HepG2) was used as a representative in vitro cancer cell line model, due to its known well-characterised property of cell line and its wide usage in many toxicity studies for screening hepatotoxic compounds[16]. To the best of our knowledge, there was no study reported on the transcriptome profiling of cancer cells treated with plant extract-mediated synthesised AgNPs. As such, this study is the first study that focuses on the transcriptome profiling of cancer cells treated with AgNPs synthesised using plant extract. This study can be a significant step in identifying potential genes that are regulated by the treatment of C. roseus-AgNPs on HepG2 cells, which will lead to the establishment of the underlying molecular network of the mechanistic actions of the AgNPs.

MATERIALS AND METHODS

Preparation of cell line

The hepatocellular carcinoma cell line HepG2 used in this study was purchased from American Type Culture Collection (ATCC, Cat. HB-8065™, Rockville, MD, United States). Complete RPMI-1640 medium supplemented with 10% heat-inactivated foetal bovine serum, 1% penicillin-streptomycin (v/v) and 1% L-glutamine (v/v) was used to culture and maintain the cells. All the reagents were purchased from Nacalai Tesque (Kyoto, Japan). Meanwhile, a normal liver cell line (THLE-3) (ATCC) was cultured in Bronchial Epithelial Cell Growth Basal Medium (Lonza, Basel, Switzerland) supplemented with frozen additives without gentamycin/Amphotericin and Epinephrine, 5 ng/mL EGF, 70 ng/mL Phosphoethanolamine and 10% fetal bovine serum. The incubator used for the cell culture work was set at 37 °C with 5% CO₂ (Shellab, Cornelius, OR, United States). Upon reaching 80% confluency, the cells were subcultured and transferred into new cell culture flasks. The cells were seeded at a concentration of 1 × 10⁵ cells/mL.

Preparation of C. roseus G. Don aqueous extract

The C. roseus aqueous extract was prepared according to our previous study[17]. A voucher specimen of C. roseus plant was deposited at the Herbarium of Universiti Sains Malaysia with reference number 10933. The leaves were washed using free-flowing clean water and left dried in an oven at 40 °C. The leaves were first ground before mixing with double distilled water with a ratio of 50 g: 1 L in a conical flask. Following overnight incubation in a water bath at 40 °C, the mixture was centrifuged at 2000 rpm for 15 min. The filtered supernatant was freeze-dried and ready to be used for the preparation of C. roseus-AgNPs.

Preparation of C. roseus G. Don-AgNPs

The C. roseus-AgNPs used in this study have been successfully synthesised, optimised, and characterised in our previous study[5]. The optimised C. roseus-AgNPs consist of 10% of C. roseus aqueous extract and 5 mmol/L of silver nitrate (AgNO₃) solution. The mixture was allowed to react in a dark environment at room temperature for 24 h until the colour changes from light yellowish to dark brownish. The mixture was then collected and centrifuged for 15 min at 10000 rpm. The supernatant was discarded while the pellet was collected and freeze-dried.

Cell viability

The proliferative activity of HepG2 and THLE-3 cells was assessed using Cell Titer 96® AQueous Non-Radioactive Cell Proliferation Assay Kit (Promega, Madison, WI, United States) which consists of (3-(4,5-Dimethylthiazol-2-yl)-2,5-Diphenyltetrazolium Bromide) dye solution or also known as MTT and solubilisation solution. The method was performed according to the manufacturer’s protocol. HepG2 cells and THLE-3 cells were seeded in a 96-well plate (Eppendorf, Hamburg, Germany) at a concentration of 1 × 10⁵ cells/mL. Cells were treated with C. roseus-AgNPs (Merck, Billerica, MA, United States) in serial dilution manner which was 1.96 μg/mL, 3.91 μg/mL, 7.82 μg/mL, 15.63 μg/mL, 31.25 μg/mL, 62.5 μg/mL, 125 μg/mL, 250 μg/mL, 500 μg/mL, and 1000 μg/mL. The cells were incubated for 24, 48, and 72 h at 37 °C, 5% incubator. Untreated cells were used as a control. Each sample size was prepared in triplicate. Following the indicated incubation time, each well was added with 20 μL of MTT reagent and further incubated for 4 h in a humidified 5% CO₂ incubator at 37 °C. After 4 h of incubation, 100 μL of stop solution was added to each well and incubated for 1 h to solubilise the formazan. The absorbance at 570 nm was recorded using a microplate reader (Bio Tek, Winooski, VT, United States). The half-maximal inhibitory concentration (IC₅₀) values were calculated based on the following formula:

% Cell Viability = [Mean OD _sample - OD _blank]/[Mean OD _control - OD _blank] × 100

OD = Optical Density

Treatment of HepG2 cells with C. roseus-AgNPs and total RNA extraction

The HepG2 cells were seeded approximately at 1 × 10⁵ cells/mL. The seeded cells were treated with C. roseus-AgNPs at a concentration of 4.95 μg/mL, which is the IC₅₀ value used in our previous study[5] and incubated for 72 h at 37 °C in a humidified atmosphere of 5% CO₂. Untreated HepG2 cells were used as a control. After 72 h of exposure, the cells were washed with PBS and immediately lysed and homogenised in TRIzol™ Reagent (Thermo Fisher, Waltham, MA, United States). Total RNA extraction was carried out using the manufacturer’s protocol. The resulting pellet was solubilised in RNAse-free water and was kept at -80 °C until further processing. The purity and concentration of RNA (260/280 ratio) were determined using Nanodrop 2000 spectrophotometer (NanoDrop Products, Wilmington, DE, United States). The integrity of the total isolated RNA was assessed by Agilent 2100 Bioanalyser (Agilent RNA 6000 Nano Kit, Santa Clara, CA, United States).

Beijing Genomics Institute sequencing

All RNA samples were sent to Beijing Genomics Institute (BGI, Shenzen, China) for sequencing. The total RNA extracted was pre-processed for transcriptome sequencing. The poly-A-containing mRNA molecules were captured and purified using a technique probe. The purified RNA molecules were reverse-transcribed into the first-strand cDNA, subsequently followed by the second-strand cDNA synthesis using Polymerase I and treatment with RNase H. The resulting product was purified and enriched with PCR amplification. The PCR amplicon was quantified by Invitrogen Qubit 2.0 Fluorometer (Thermo Fisher). The amplicon libraries were pooled together to make a single-strand DNA circle (ssDNA circle). DNA nanoballs (DNBs) were generated from the ssDNA circle by rolling circle amplification and loaded into a flow cell in which DNB binding sites are patterned nano-arrays. Sequencing was carried out using a paired-end 100 bp sequencing strategy on the BGIseq500 platform.

Bioinformatics analysis

High-quality genome sequencing data was developed by removing the adapter, poor quality and low complexity reads. The cleaned sequences were mapped onto the reference genome (hg19), subsequently followed by the identification of the novel genes, SNP (single nucleotide polymorphism), InDels (insertions and deletions) and the detection of gene splicing. Differential Gene Expressions were obtained by applying a paired, two-tailed t-test to the calculated expression data of the treated and untreated samples. Gene Ontology (GO) analysis was used to analyse the enrichment of gene sets associated with biological processes, molecular functions, and cellular components. Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis was carried out to permit the pathway annotation to the differentially expression genes (DEGs). A P value less than 0.1 is considered a statistically significant difference.

RESULTS

Cytotoxicity of C. roseus-AgNPs on HepG2 Cells

In this study, the cytotoxic effects of C. roseus-AgNPs were assessed on HepG2 cells and normal liver epithelial cells, THLE-3 cells. Figure 1A shows the cytotoxicity effects of HepG2 cells after treatment with C. roseus-AgNPs. In comparison to the untreated cells, C. roseus-AgNPs significantly (P < 0.001) inhibited the proliferation of HepG2 cells at all concentrations and incubation periods in time- and dose-dependent manner, indicating the cytotoxic effect of C. roseus-AgNPs towards HepG2, with 7.79 %, 21.59%, and 30.15% of cells were inhibited at the lowest concentration of C. roseus-AgNPs at 24, 48 and 72 h, respectively. HepG2 cells showed a consistent percentage decrement of cell viability upon the treatment, and only 1.78 % average of the cells survived between 24 to 72 h of incubation at the highest concentration of C. roseus-AgNPs. The percentage of C. roseus-AgNPs cytotoxicity compared to the untreated cells was used to determine the IC₅₀ values as illustrated in Figure 1B where the IC₅₀ were 7.81 ± 0.02 μg/mL, 3.87 ± 0.02 μg/mL, and 3.20 ± 0.04 μg/mL at 24, 48, and 72 h of incubation, respectively. Figure 1C shows the effect of THLE-3 treated with C. roseus-AgNPs. The results demonstrate an intriguing finding, where the C. roseus-AgNPs demonstrated a substantial (P < 0.001) increment in THLE3 proliferation at concentrations of 1.96 μg/mL and 7.81 μg/mL for all incubation times. On the contrary, during all incubation times, there was no significant difference at concentrations of 15.63 μg/mL. However, at concentrations 31.25 μg/mL to 1000 μg/mL, C. roseus-AgNPs significantly (P < 0.001) inhibited the proliferation of THLE3 cells as compared to untreated THLE3 cells. After 72 h, approximately 55.78% of cells survived at the highest concentration of C. roseus-AgNPs. There were no IC₅₀ values at concentrations for 24 and 48 h, but at 72 h, the IC₅₀ was recorded at 615 ± 0.05μg/mL μg/mL, as depicted in Figure 1D. Based on the results, we observed that C. roseus-AgNPs was found to inhibit the growth of the HepG2 cell line with a mean IC₅₀ value of 4.95 ± 0.03 μg/mL. Contrarily, C. roseus-AgNPs showed very weak inhibition activity toward THLE3 cells with IC₅₀ value of 615 ± 0.05 μg/mL.

Figure 1 Cytotoxicity evaluation of Catharanthusroseus-silver nanoparticles on HepG2 and THLE-3 cells. A: The cytotoxicity of HepG2 cell lines treated with different concentrations of Catharanthus roseus-silver nanoparticles (C. roseus-AgNPs); B: The IC₅₀ of C. roseus-AgNPs on HepG2 cells; C: The cytotoxicity of THLE3 cell lines treated with different concentrations of C. roseus-AgNPs; D: The IC₅₀ of C. roseus-AgNPs on THLE-3 cells. All experiments were done in triplicate, and the data represent means ± standard deviations. The comparison between each concentration with untreated cells was done using two-way ANOVA with Dunnet post-test to detect any significant differences (^bP < 0.01; ^cP < 0.001; ns not significant). C. roseus-AgNPs: Catharanthus roseus-silver nanoparticles.

Quantitative and qualitative measurement of total RNA

Total isolated RNA was quantified using Bioanalyser. As depicted in Supplementary Figure 1, the representative electropherogram indicated two intact bands that are visible in each sample. These two bands represent 28s and 18s ribosomal RNA, respectively. RNA integrity number (RIN) was then determined, which is the value of the RNA integrity. The value that falls within a range between 8 to 10 showed an acceptable value of RIN[18]. The RIN and ribosomal ratio values acquired for both untreated and C. roseus-AgNPs treated HepG2 cells were 9.6 and 9.4, respectively. Both RIN values were within the acceptable range.

mRNA Transcriptome sequencing

Sequencing data filtering: Two samples were sequenced using the DNBseq platform and the result was about 6.98 Gb bases per sample. The distribution of the base quality was shown in Supplementary Figure 2. As observed in both Supplementary Figure 2A and B, the percentage of clean reads was 93.69% and 94.03%, respectively.

Genome mapping: After read filtering, the clean reads were mapped to the reference genome using HISAT2[19]. On average, 95.88 % of reads were found to be mapped to the population of human genomes and the consistency of the mapping result for each sample suggests that the samples were comparable. The mapping details are shown in Table 1.

Table 1 Summary of genome mapping.

Sample	Total clean reads	Total mapping ratio	Uniquely mapping ratio
Untreated HepG2	70025052	95.94%	76.60%
C. roseus-AgNPs treated HepG2	72598578	95.82%	75.94%

C. roseus-AgNPs: Catharanthus roseus-silver nanoparticles.

Gene expression analysis: To get a complete reference for the gene mapping and expression, novel coding transcripts were merged with the reference transcripts, and clean reads were mapped to them using Bowtie2[20]. The gene expression level for each sample was calculated with RSEM[21]. The result of this analysis is summarised in Table 2.

Table 2 Summary of gene mapping ratio.

Sample	Total clean reads	Total mapping ratio	Uniquely mapping ratio
Untreated HepG2	70025052	66.79%	63.92%
C. roseus-AgNPs treated HepG2	72598578	62.94%	60.15%

C. roseus-AgNPs: Catharanthus roseus-silver nanoparticles.

The sufficiency of sequencing data for bioinformatics analysis was approached using sequencing data saturation analysis. As the number of sequenced reads increased, the number of identified genes also increased. On the other hand, when the number of sequenced reads reached a certain amount, the determining gene growth curve flattens, indicating the identified gene reached saturation. Supplementary Figure 3 displays the saturation analysis for each sample.

Reads coverage and distribution of each detected transcript are shown in Supplementary Figure 4 and Supplementary Figure 5, respectively. This approach allows access to the excellent quality of the samples and sequencing data sufficiency by showing the completely covered transcripts and evenly distributed reads throughout the transcript. These results suggest that both untreated and treated HepG2 had excellent sample quality and sufficient sequencing. Correlation between samples was assessed by Pearson correlation coefficient calculations for all gene expressions between the samples, as shown in Figure 2.

Figure 2 The correlation analysis between samples. The colour represents the correlation coefficient.

The identification of differentially expressed genes: DEGs were determined by using DEseq2 and passion Dis algorithms. The distribution of DEGs is summarised using the volcano plot as shown in Figure 3. The treatment of HepG2 cells with C. roseus-AgNPs revealed 296 DEGs, with 182 genes were upregulated while 114 genes were downregulated (Figure 3A).

Figure 3 Volcano plot of differentially expression genes. Red points represent upregulated differentially expression genes (DEGs). Blue points represent down-regulated DEGs. Grey points represent non-DEGs. DEGs: Differentially expression genes.

Gene ontology analysis of DEGs: The identified DEGs were subjected to Gene ontology analysis. GO unveiled three ontologies which are related to molecular biological function, cellular components, and biological processes. The classification result is depicted in Figure 4.

Figure 4  Gene Ontology classification of upregulated and downregulated differentially expression genes.

Pathway analysis of DEGs: KEGG pathway classification and functional enrichment were generated based on DEGs. Pathway enrichment result is shown in Table 3 and the network enrichment is depicted in Figure 5.

Figure 5 Network enrichment result. The darker the colour indicates the highest enrichment pathways. The larger the area, the higher the degree of enrichment.

Table 3 Pathway functional enrichment results.

Pathway ID	Pathway	Genes	Hits	Adj P value
K05200	Pathway in cancer	FAS, GADD45A, BAX, PMAIP, BID, JUN, CXCL8, HMOX1, STAT1, FOS, CEBPA, VEGFA, FGF5, EGF, RHOA, FADD, FH, SMAD, MTOR, NFκBIA, CDKN1A, WNT4a, WNT7Aa, FGFR3a, BMP4a, CDK4a, CDK2a, MDMa	29/530	3.28E-12
K04115	p53 signaling pathway	FAS, GADD45A, SERPINE, THBS1, CDK4, BAX, CDK2, PERP, SESN1, SESN2, PMAIP1, BID, IGFBP3a, MDM2a	14/72	1.28E-13
K04210	Apoptosis	FAS, GADD45A, BAX, PMAIP1, BID, JUN, FOS, FADD, NFKBIA, BCL2A1, ATF4, MCL1, TNFSF10a, RIPK1a	14/136	9.61E-10
K04144	Endocytosis	ARF6, RHOA, EHD2, FGFR3, HSPA6, HSPA1L, VPS28, EEA1, VPS25, TSG101, STAM, SMAD, EHD4, LDLR, TFRCa, MDM2a	16/244	4.08E-08
K04010	MAPK signaling pathway	FAS, GADD45A, JUN, FOS, VEGFA, FGF5, CDKN1A, EGF, ATF4, HSPA6, HSPA1L, CDKN1A, HSPB1, DUSP1, NR4A1, EFNA4a, FGFR3a, ERBB3a, SKP2a	16/295	5.58E-07
K04668	TNF signalling pathway	FAS, JUN, FOS, FADD, NFKBIA, ATF4, RIPK1a, CCL2, NOD2, CCL20a	10/110	9.07E-07
K04350	TGF beta signaling pathway	THBS1, RHOA, SMAD2, SMAD4, BMP4, SMAD7, BAMBI, BMP6	8/90	1.66E-05
K02010	Cell cycle	GADD45A, SOX15, CDK4a, CDK2a, MDM2, SMAD2, SMAD4, CDKN1A, PCNAa, MCM3a, SKP2a	9/124	2.09E-05
K04978	Mineral absorption	HMOX1, MT1F, MT1X, MT1H, MT1B, FTH1	6/51	3.58E-05

^aDownregulated genes.

DISCUSSION

C. roseus-AgNPs exhibited anti-cancer properties with negligible effect on normal cells

Our group has previously demonstrated the anti-cancer properties of C. roseus-AgNPs on cancer cells[22]. The anti-cancer properties of the C. roseus-AgNPs were estimated by IC₅₀, which represents the concentration of C. roseus-AgNPs required to inhibit 50% of the total cells[23]. According to the IC₅₀ value (800 ± 1.55 μg/mL) observed at 72 h, the THLE3 cells substantially (P < 0.001) inhibited only at very high concentrations of C. roseus-AgNPs. On the other hand, the IC₅₀ of the C. roseus-AgNPs on the HepG2 cells was 4.38 ± 1.59 μg/mL. This study revealed that C. roseus-AgNPs showed a significant (P < 0.001) cytotoxicity towards HepG2 cells as compared to THLE3 cells. C. roseus-AgNPs can inhibit the progressive development of HepG2 while causing very insignificant toxicity to normal cells at low concentrations. Several studies have also shown that biosynthesised AgNPs show no toxicity against normal cells while demonstrating cytotoxic effects against cancer cells[24-29]. For example, a study by Halkai et al[24] showed that fungal-derived AgNPs exerted minimal cytotoxicity against human gingival fibroblast cell line. Additionally, Sriram et al[25] also reported similar observations in their experiments, where AgNPs acted as an anti-proliferative agent by effectively inhibiting the development of Dalton’s lymphoma ascites cell lines without causing toxicity on normal cell lines. The findings from our study agreed with the previous reports, corroborating the potentiality of C. roseus-AgNPs as an anti-cancer agent.

mRNA transcriptome analysis identified 296 protein-coding genes

An in-depth understanding of the anti-cancer properties of the C. roseus-AgNPs entails the identification of the genes that act in concert in orchestrating the effect. As transcriptome analysis can provide an overarching view of the gene expression profile under a certain condition or state, it was adopted in our effort to comprehend the underlying mechanisms of the anti-cancer activity of C. roseus-AgNPs against HepG2 cells. In the present study, the untreated HepG2 cells and C. roseus-AgNPs treated HepG2 cells were subjected to mRNA transcriptome analysis using the BGI DNBseq Platform. As revealed by the mRNA transcriptome analysis, it was found that the treatment of HepG2 cells with C. roseus-AgNPs has resulted in the regulation of 296 protein-coding genes, of which 182 genes were upregulated while 114 genes were downregulated, as shown in Figure 3.

GO analysis exhibited that the highest fraction of the regulated genes were involved in “cellular and signalling response” followed by “biological regulation”, “regulation of biological process”, “metabolic process” and “response to stimulus” (Figure 4). The underlying pathways regulated by the genes are the p53 signalling pathway, pathway in cancer, apoptosis pathway, endocytic pathway, MAPK signalling pathway, TNF signallingg pathway, TGF signallingg pathway, cell cycle pathway and mineral absorption pathway.

C. roseus-AgNPs induced the expression of stress-associated genes such as MT, HSP and HMOX-1

C. roseus-AgNPs treatment of the HepG2 cells was found to upregulate several members of the gene isoforms that encode metallothionein (MT), such as MT1F, MT1X, MT1H, and MT1B. MTs are intracellular proteins that contain approximately 30% thiol-containing cysteine residues, which can bind several cytotoxic agents, including platinum compounds, alkylating agents, and metal ions such as zinc and copper[30]. MTs also regulate various pathophysiological processes such as apoptosis, and angiogenesis and could also act as radical scavengers by protecting the cells from free radicals[31]. As such, an increased level of MT is an indicator that the cells were undergoing ‘stress’ and the cells are striving to mitigate the cytotoxic effect of the anticancer drug, in this case, C. roseus-AgNPs[32]. This finding is also in agreement with the findings by Woo et al[33], who reported that Javanese medaka, a type of seawater organism showed MT upregulation upon exposure to AgNPs. On the other hand, heat-shock genes such as HSPA1L, HSPB1, and HSPA6 were also found to be upregulated in HepG2 cells exposed to C. roseus-AgNPs. HSPs are upregulated by stress signals such as high temperature, decreased availability of oxygen, infectious agents, and inflammatory mediators[34]. The increased expression level of HSPs is needed to counteract the stress, which is induced by C. roseus-AgNPs in this study. Furthermore, the up-regulation of oxidative stress-related genes HMOX-1 was also documented in our experiment. HMOX-1 is a reactive oxygen species (ROS) sensor that has antioxidant and anti-inflammatory properties[35]. During stress conditions, HMOX-1 catalyse the degradation of the Heme group into biliverdin, carbon monoxide, and iron[36]. Similar increased expression of HMOX-1 was also observed by Gurunathan et al[37], in mouse embryonic fibroblast cells upon treatment with AgNPs. Collectively, the upregulation of stress-response genes such as MTs, HSPs, and HMOX-1 in this study indicates that C. roseus-AgNPs exposure invokes the cell’s defensive response in negating effects of cellular stresses caused by C. roseus-AgNPs. The increased expression of stress-response genes indirectly reflects the cytotoxic effect of C. roseus-AgNPs. We have also observed significant production of NO and ROS in our previous study upon treatment of HepG2 cells with C. roseus-AgNPs[38]. These findings are substantial and in agreement with the previous findings, whereby upregulation of MTs, HSPs, and HMOX-1 was observed in cells exposed to AgNPs[39-41].

C. roseus-AgNPs increased expression of tumour suppressor genes and apoptotic genes

The most intriguing finding in our study is that C. roseus-AgNps treatment on HepG2 cells induces the expression of growth arrest and DNA damage-inducible alpha (GADD45A) gene, which is a type of tumour suppressor gene that regulates processes such as DNA repair, cell cycle control, senescence, and genotoxic stress[42]. The expression of the GADD45A gene in cell cycle inhibition is also regulated by p53. p53 protein is involved in maintaining genetic integrity and regulating the cellular response towards genotoxic stress by inducing cell cycle arrest or apoptosis to prevent tumorigenesis[43]. p53 is negatively regulated by MDM2. Interestingly, our experimental findings demonstrated that MDM2 was downregulated in C. roseus-AgNPs treated HepG2 cells, suggesting that its inhibitory effect against p53 was ameliorated, causing the upregulation of the p53 gene. As such, the expression of the p53 gene elevates, causing the suppression of the proliferation of cancer cells. Sahu et al[44] reported similar observations in their study where under normal conditions, p53 was constitutively expressed, but inactivated by its negative regulator, MDM2. However, during cellular stresses, MDM2 was downregulated which in turn caused the upregulation of p53 gene. The upregulation of the p53 gene is indicative of the anticancer effect of C. roseus-AgNPs in amplifying the tumour suppressor activity of the cancer cells. Besides the upregulation of the tumour-suppressor genes, apoptotic-related genes BAX and FAS were also found to be upregulated, suggesting the anticancer efficacy of the C. roseus-AgNPs in promoting apoptosis in cancer cells.

C. roseus-AgNPs activated signal transduction pathways such as MAPK signalling pathway

The MAPK pathway is a series of protein kinase cascade essential in regulating numerous physiological functions including inflammation, cell stress response, cell differentiation, cell division, cell proliferation, metabolism, motility, and apoptosis[45]. Treatment with C. roseus-AgNPs activated MAPK signalling pathway in HepG2 cells. In this study, several genes that are involved in MAPK pathways were found to be regulated such as FAS, GADD45A, P53, JUN, and FOS. As indicated previously, GADD45A, a tumour suppressor gene which could also be involved in the MAPK signalling pathway was found to be upregulated upon treatment of the HepG2 cells with C. roseus-AgNPs. Increased expression of GADD45A conduces to baicalein-induced apoptosis and activation of MAPK signalling pathway[46]. In this study, activation of MAPK signalling pathway also upregulates the p53 gene as mentioned previously, as MAP kinase phosphorylates and activates the p53 protein in response to stressful stimuli induced by C. roseus-AgNPs[47]. Taken together, activation of MAPK pathway prepares the cell for counteracting actions such as inflammation, cell stress response, and apoptosis upon treatment with C. roseus-AgNPs, which indirectly implies the anticancer properties harboured by these nanoparticles.

C. roseus-AgNPs activated TNF signalling pathway

TNF alpha is a pro-inflammatory cytokine that acts by binding to TNF-R1 and TNF-R2 receptors, resulting in the recruitment of signal transducers that activate the effector, leading to the activation of caspases and two transcription factors, NF-κB, as well as MAPKs such as ERK, p38, and JNK, which will induce apoptosis and necrosis[48]. In this study, the treatment of C. roseus-AgNPs caused the upregulation of several genes related to TNF signalling pathway such as FADD, NF-κbia), ATF4, CCL2, NOD2. FADD protein interacts directly with TRADD, which are signal transducers that activate NF-κB and trigger apoptosis[49]. The overexpression of the FADD genes in our study suggests that treatment with C. roseus-AgNPs eventually promotes apoptosis. Similar overexpression was also reported in the previous study, whereby AgNPs treated MDA-MB-436 cells showed an increase in the level of FADD gene[50]. In this study, the upregulation of ATF4 was also found. The overexpression of ATF4 was reported by Iwasaki et al[51], which happens in response to metabolic stresses caused by SFAs and ER stressors. RIPK1 gene is involved in the system that controls cell survival, signalling nodes in cell death and inflammation and cytokine production. The downregulation of the RIPK1 gene in this study upon treatment with C. roseus-AgNPs can induce apoptosis via the cleavage activity of the caspase 3 associated pathway[52]. Qiu et al[53] reported similar observations in their experiments. CCL20 is known to enhance cancer cell progression[54]. The downregulation of the CCL20 gene in this study suggests that C. roseus-AgNPs are able to induce inflammation through TRAIL as reported by a previous study[55].

C. roseus-AgNPs elicited the activation of TGF-β signalling pathway

TGF-β signalling pathway plays a crucial role in controlling various fundamental aspects of cellular activities such as cellular growth, development, differentiation, and apoptosis[56]. As a secreted polypeptide, TGF-β functions via receptor serine/threonine kinases and intracellular SMAD effectors[57]. TGF-β acts as a tumour suppressor at the early stage of cancer while it also acts as a pro-metastatic factor in the later stages of cancer[58]. Exposure of HepG2 cells to C. roseus-AgNPs activates TGF-β signalling pathway. The effect of C. roseus-AgNPs is analogous to a previous study, whereby ellagic acid was found to exert anti-proliferation effects by activating TGF-β/Smad3 signalling pathway[59]. Transcriptome analysis also showed that isoforms of SMAD, which are part of TGF-β pathway were also upregulated. Moreover, BMPs such as BMP4 and BAMP6, which are extracellular signalling molecules that belong to the TGF-β pathway, were also upregulated. The tumour suppressor effect mediated by TGF-β pathway was imparted upon treatment with C. roseus-AgNPs, which corroborates its anticancer property.

The uptake of C. roseus-AgNPs occurred via endocytosis

Endocytosis involves the formation of small membrane vesicles (60-120 nm) that transports various molecules or cargo from the plasma membrane to the cytoplasm. Though there are several types of endocytosis, previous studies have shown that clathrin-dependent endocytosis and macropinocytosis are the major routes of transportation of AgNPs into the cells[60]. The observations also agree with the results reported by Treuel et al[61] that endocytosis has been demonstrated to be a key mechanism in driving the cellular uptake of AgNPs, with NPs entering cells via early endosomes, late endosomes, and lysosomes. In this study, a few genes such as ARF6, EHD2, FGFR3, RhoA, EEA1, VPS28, VPS25, and TSG101, were upregulated, suggesting that the uptake of C. roseus-AgNPs can occur via the clathrin-dependent or clathrin-independent endocytosis pathway. ARF6 gene, also known as ADP-ribosylation factor 6, is a small GTPase that regulates endocytic membrane trafficking and actin remodelling[62]. The upregulation of ARF6 gene in this study is consistent with the findings of Tanabe et al[62], which suggest that ARF6 gene regulates the membrane trafficking between the plasma membrane and endosome via clathrin-dependent or clathrin-independent endocytosis[63]. A previous study by Morén et al[63] showed that the overexpression of EHD2 gene inhibited the formation of caveolae. Interestingly, our study demonstrated an upregulation of EHD2 gene, which encodes a member of the EH domain-containing protein family. EHD2 protein has an N-terminal domain that interacts with the actin cytoskeleton and a C-terminal EH domain that binds to an EH domain-binding protein[64]. This interaction appears to link clathrin-dependent endocytosis and actin, implying that this gene is involved in the endocytic pathway, particularly clathrin-dependent endocytosis[65]. These findings suggest that clathrin-dependent endocytosis was one of the major uptake mechanisms of C. roseus-AgNPs while ruling out the involvement of possible involvement of caveolin-dependant endocytosis.

Another interesting finding in this study is the potential involvement of macropinocytosis, attributable to the upregulation of RhoA gene macropinocytosis. This gene encodes a member of the Rho family of small GTPases, which regulates macropinocytosis via active and inactive GTP-binding while simultaneously playing an important role in the remodelling of the actin skeleton during macropinocytosis[66,67]. According to Patel et al[67], after the macropinocytic cups closed to form macropinosomes, the expression of another Rho subtype, RhoA, increased significantly. This corroborates our findings in this study on the RhoA gene upregulation, which suggests that C. roseus-AgNPs uptake also could occur via macropinocytosis. The overexpression of EEA1 in this study indicated that the formation of early endosomes occurs during the uptake of C. roseus-AgNPs. This finding is consistent with prior work, which demonstrated the high frequency of the EEA1 gene in early endosomes that are the primary sorting station in the endocytic pathway[68]. On the other hand, TSG101 and VPS28 genes are involved in late endosomal trafficking[69]; the upregulation of TSG101 and VPS28 genes in this study suggests that late endosome was formed during the C. roseus-AgNPs uptake. This finding was in line with a previous study, where the expression of TSG101 and VPS28 was found to be increased[70]. The upregulation of the stress-responsive genes as mentioned previously in our study is also indicative of the successful uptake of C. roseus-AgNPs into the HepG2 cells, as the generation of free radicals that induced stress in the cells can be caused by the leaching of Ag⁺ from AgNPs into the cytosol, because of high acidic lysosome rupture. TFRC gene encodes a cell surface receptor necessary for cellular uptake by the process of receptor-mediated endocytosis[71]. The downregulation of the TFRC gene in this study suggests that the expression of the gene was induced to reduce endocytosis via negative feedback regulation as a response to cellular homeostasis. Our finding also is in tandem with the findings of Wang et al[72], who have also noticed a drop in the expression of the TFRC gene, which could be due to negative feedback for defensive actions.

The uptake of C. roseus-AgNPs arrested cell cycle

Cancer progression is associated with aberrancy in the cell cycle, such as the anomalous expression of CDKs[73]. CDKs are usually highly expressed, causing the uncontrolled proliferation of cancer cells[74]. Upon the treatment of the HepG2 cells with C. roseus-AgNPs, CDK4, and CDK2 were found to be downregulated, implying the antagonistic effect of the nanoparticles against the cell cycle protein. A previous study reported that the blockage of Go/G1 was accompanied by the downregulation of the cell cycle regulators CDK4 and CDK2[75]. In this study, the downregulation of CDK4 and CDK2 suggested that C. roseus-AgNPs was arrested at Go/G1. Another important observation is the upregulation of GADD45A, which caused a decrease in the SKP2 expression. Overexpression of SKP2 is associated with the cell cycle progression and as the SKP2 expression was found to be reduced in the present study, it is surmised that the cell proliferation is being forestalled. Moreover, the reduced expression of SKP2 is also associated with the increased expression of CDKN1A, which is an inhibitor of cell cycle progression by inhibiting the activity of cyclin-dependent kinase expression[76]. The upregulation of CDKN1A corroborates the anticancer activities of C. roseus-AgNPs that can induce cell cycle arrest. MCM3 is a member of minichromosome maintenance family that is associated with tumour invasiveness[77]. The treatment of the C. roseus-AgNPs caused the downregulation of MCM3, in HepG2 cells, ratifying the anticancer activities of C. roseus-AgNPs in alleviating tumour aggressiveness.

The overall proposed mechanism as depicted in Figure 6, consists of clathrin-dependent and clathrin-independent endocytosis. The signalling pathways indicate the involvement of the up and down-regulated genes in various cellular organelles. The understanding of cellular and molecular mechanisms would provide a strong justification of the rationale of C. roseus G. Don-AgNPs as anticancer compounds for liver cancer therapy.

Figure 6 Model of cytotoxicity mechanism in HepG2 cell treated with C.roseus-AgNPs that involves the upregulated genes (in red) and downregulated genes (in green).

CONCLUSION

In this study, the treatment of HepG2 cells with C. roseus-AgNPs has resulted in the increase of the expression of tumour suppressor genes, apoptotic genes, and activation of signal transduction pathway such as mitogen-activated protein kinase (MAPK) signalling pathway, endocytosis signalling pathway, TNF signalling pathway, TGF-Beta signalling pathway as well as cell cycle arrest. Collectively, the findings from our study have demonstrated the anti-cancer properties of C. roseus-AgNPs, with insignificant effects on normal cells. The therapeutic property of the C. roseus G. Don-AgNPs should be further explored in the future as part of the endeavours to surrogate or complement the current conventional chemotherapeutic-based intervention.

ARTICLE HIGHLIGHTS

Research background

Conventional chemotherapy and radiotherapies based on x-ray and gamma-ray radiations are the most widespread techniques in the world for the treatment of malignant diseases due to their ability to penetrate tissues and thus allow them to reach deep sites. The only limitation of these treatments is the lack of selectivity between the tumour and the healthy surrounding tissues. Interestingly, previous studies have shown that silver nanoparticles (AgNPs) have the ability to selectively induce cytotoxic effects on cancer cells, as compared to normal cells. Therefore, the present study aims to evaluate the cytotoxic effects of AgNPs synthesised by C. roseus aqueous extract against liver carcinoma cells HepG2 and normal liver cells THLE-3, by assessing the proliferative activity followed by the mRNA transcriptome profiling analysis.

Research motivation

Due to the limitations of the conventional treatment like non-specificity and less effectiveness, novel strategies are in demand to solve these issues. Amongst all, the use of plant-synthesised silver nanoparticles has gained attention as they are known for non-toxic properties, are cost-effective, are easily assessable and environmentally friendly. The unique properties of nano-sized nanoparticles have been reported can penetrate cancer cells effectively. In this study, the anticancer activity was evaluated at both cellular and molecular levels to gain insight into its mechanisms.

Research objectives

To evaluate the proliferative activity of the human hepatocellular carcinoma cells HepG2 in response to the Catharanthus roseus-silver nanoparticles (C. roseus-AgNPs), in comparison to the normal liver cells THLE-3 cells.

Research methods

To evaluate the proliferative activity, the hepatocellular carcinoma cells HepG2 and normal human liver cells THLE3 were treated with standardised Catharanthus roseus-silver nanoparticles (C. roseusAgNPs) in a double dilution manner and analysed using MTT assay. To elucidate the gene expression study, the RNA samples were extracted and sequenced using BGIseq500 platform. This is followed by data filtering, mapping, gene expression analysis, DEGs analysis, GO analysis, and pathway analysis.

Research results

The proliferative activity revealed selective effects, indicating that the Catharanthus roseus-silver nanoparticles were cytotoxic on hepatocellular carcinoma cells HepG2 cells but not on the normal liver cells THLE3 cells. The transcriptome analysis has resulted in the regulation of 296 protein-coding genes, of which 182 genes were upregulated while 114 genes were downregulated. The most intriguing finding is the expression of tumour suppressor gene GADD45A, responsible for the regulation of DNA repair, cell cycle control and genotoxic stress. The expression of this gene is regulated by p53. The upregulated GADD45A was supported by the downregulated MDM2, which is the negative regulator for p53. Our findings revealed the activation of several signalling pathways including the mitogen-activated protein kinase signalling pathway, TNF signalling pathway and TGF-β signalling pathway. These pathways are the main regulator in fundamental intracellular activities such as apoptosis, cell cycle and cellular growth. The upregulation of ARF6, EHD2, FGFR3, RhoA, EEA1, VPS28, VPS25 and TSG101 indicated that the C. roseus-AgNPs were taken up by HepG2 cells via both clathrin-dependent and clathrin-independent.

Research conclusions

The selective proliferative activity between cancerous and normal liver cells indicates a promising potential of Catharanthus roseus-silver nanoparticles (C. roseus-AgNPs) as an effective anticancer agent. The understanding of the molecular signalling pathways induced by the genes associated with oxidative stress, apoptosis and cell cycle arrest provides the novelty towards the development and establishment of C. roseus-AgNPs as an anticancer drug for hepatocellular carcinoma. Moreover, we propose that the uptake was via both clathrin-dependent and clathrin-independent endocytosis. These findings would explain the cytotoxicity mechanisms of the C. roseus-AgNPs at cellular and molecular level towards hepatocellular carcinoma cells HepG2.

Research perspectives

While the endocytic pathways emphasise the action of the selectively permeable plasma membrane on the nanomaterials, cytotoxicity of silver nanoparticles (AgNPs) generally involves the cells’ downstream activity, including reactive oxygen species (ROS)- dependent pathway, cell cycle arrest and genotoxicity. Moreover, the small-sized AgNPs can easily penetrate the cells and bind to macromolecules including proteins and DNA, either directly or indirectly although the exact mechanism for this interaction has not been clarified. The physicochemical characteristics that make AgNPs so useful can be the main reason they might be dangerous to cells, and at a higher level to human health. Therefore, to avoid these problems, the AgNPs must be engineered from either biocompatible, nontoxic, biodegradable material or materials have with minimal toxic effects.

ACKNOWLEDGEMENTS

We would like to offer special appreciation to all laboratory staff and postgraduate students under the laboratory of the Advanced Medical and Dental Institute for their guidance, assistance, and support. A preprint has previously been published by Azhar et al, 2018 (Available from: Preprints:2021090431, [DOI: 10.20944/preprints202109.0431.v1]).