The fermentation of Camellia chrysantha leaves with Eurotium cristatum to yield biologically active compound

Abstract

Microfungi have appeared as one of the most effective and eco-friendly catalysers for derivatizing bio-active molecules in green chemistry. The use of Eurotium cristatum has been established as a method of post-harvest handling and value addition for tea products, especially in enhancing their health-protective properties. From our study, a consortium of E. cristatum was selected and introduced as inoculum on Camellia chrysantha leaves substrate, yielding transformed products with chemo-preventive and antidiabetic potential. Biological activity-guided fractionation incorporated with chromatographic separation of the fermented products gave out compound THV-LM03 as the most active isolate exhibiting inhibitory effect against lipase, α-glucosidase, and cancerous cell growth. By means of ¹H-NMR, ¹³C-NMR, HSQC and HMBC spectral analysis, THV-LM03 was elucidated as 22-O-tigloyloxycamelliagenin A – a triterpene which has not been isolated previously from Camellia materials. In vitro assays revealed THV-LM03 as a potent inhibitor against α-glucosidase (IC₅₀ value at 16.33 µg/mL) and against cancer cell lines LU-1 and MCF-7 (IC₅₀ values at 4.87 and 9.32 µg/mL, respectively). These findings brought out a new pharmacological aspect of microbial fermentation in C. chrysantha leaves, as well as contributed to highlight the potential of effective bio-transformation in a safer synthesis of target molecules.

Inoculation of Camellia chrysantha leaves with characterized Eurotium cristatum consortium led to the formation of fermented product that exhibited inhibition against cancerous cells and α-glucosidase. Triterpene compound 22-O-tigloyloxycamelliagenin A was isolated from the fermented leaves with enhanced cytotoxic, lipase inhibitory and α-glucosidase inhibitory properties.

Keywords

22-O-tigloyloxycamelliagenin A biotransformation fermentation

Introduction

Microbial fermentation is one of the oldest procedures for preserving agricultural products and producing beverages from organic materials. With enzyme activities to modify natural substrates and further release novel molecules, it has recently been considered a cost-effective catalysing method in environmentally friendly material transformation. The use of specific microorganisms, that is, lactic acid bacteria and filamentous fungi, was found as key factors in yielding characteristic aroma and taste of well-known fermented tea products, such as Yunnan pu’er, Toyama kurocha and Tteokcha dark teas.¹ Particular microbials of Aspergillus, Bacillus, and Penicillium were proved to play a role in supporting valuable medicinal properties for tea products, including reducing blood lipids and providing antioxidative effects.²–⁴ One of the most significant fungus for processing teas is Eurotium cristatum (Eurotiales, Asperillaceae, syn. Aspergillus cristatus) [https://www.mycobank.org/].⁵–⁷ To date, E. cristatum has extensively been utilized in herbal fermentation of miscellaneous plant taxa, such as Ginkgo biloba, Docynia indica, Eucommia ulmoides and Acanthopanax senticosus, to facilitate value-added products.⁸ A comparison of chemical compositions between non-fermented and fermented Cyclocarya paliurus leaves revealed 67 differential metabolites, including 31 terpene compounds.⁹ Interestingly, while the flavonoids amount increased by 25.66%, the total contents of polyphenols and triterpenoids decreased by 53.04% and 27.91%, respectively, after the fermentation.⁹ The components of herbal leaves fermented by E. cristatum were supposed to undergo structural modifications, correlating with changes in the taste, aroma, and bioactivities of the products.⁸

Section Chrysantha (Ericales, Theaceae) of Camellia comprises more than 24 species with unique yellow flowers and was popularly regarded as ‘golden camellias’.¹⁰ Plants of this section attracted attention due to their unique importance in either floriculture or the production of valuable bio-active compounds.¹¹ In the last decade, the number of publications on the first discovered golden camellia plant Camellia chrysantha has significantly increased, mostly regarding the scientific validity, the medicinal and commercial importance of its flowers. Recently, vitexin, aromadendrin, quercetin, kaempferol, and phenolic acid were reported as components of C. chrysantha’s leaves.¹² Impressively, they were claimed to play a supportive role in controlling cellular activities and protecting cells from harmful free radicals.¹⁰ Another phytochemical investigation showed four triterpenoids of β-amyrin hexanoate, friedelin, friedelanol, and β-amyrin and a steroid chondrillasterol as components of C. chrysantha’s leaves.¹³ In addition, 17 types of amino acids from its leaves with averaged amounts ranging from 3.73 to 5.93 g/100 g were recorded.¹⁴ In general, the chemical composition and pharmaceutical potential of C. chrysantha’s leaves have been thoroughly provided, laying the basis for further studies of their development as beneficial bio-transformed products.

In geographical areas with natural distribution of C. chrysantha, systematic studies to preserve and sustainably exploit the plant treasure are recommended. As parts of commercial activities, C. chrysantha is presently cultivated for limited seasonal harvest of the flowers, while the leaves are abundant in all seasons but generally unfavoured for utilization. Methods for improving the tea leaves’ quality to take advantage of this resource are thus required. In these circumstances, microbial fermentation for manufacturing tea beverage is among the most conceivable approaches to be considered. Our study aimed to employ E. cristatum strains in the transformation of golden C. chrysantha’ leaves to yield fermented tea with enhanced biological properties, as well as to characterize the most biologically active component from the tea product. For this purpose, extracts and fractions from the fermented tea leaves were evaluated for their in vitro chemo-preventive and antidiabetic potential on cancerous cell growth and α-glucosidase, respectively. Isolated component from the fermented tea leaves was additionally assessed for its inhibitory activity against pancreatic lipase, the key lipid-digesting enzyme in the intestine.

Materials and methods

Materials

Fungal strains

Six fungal strains of Eurotium spp. were used in this study, inherited from the laboratory of experimental biology – Institute of Natural Products Chemistry. Three of these were isolated from commercial Anhua dark tea (Hunan Provincial Baishaxi Tea Industry Co., China), while other strains were isolated from plants (Sonneratia sp., Zingiber officinale).

Camellia leaves collection and pre-processing

Fresh leaves of C. chrysantha were collected from tea plantations in Ba Che, Quang Ninh (21°19′34.2″N 107°06′07.7″E), taxonomical considered,¹⁵ washed and air dried. The leaf samples are stored at the laboratory of experimental biology – Institute of Chemistry.

Chemicals, reagents and apparatus

Ellipticine, Acarbose, Orlistat, 4-Nitrophenyl α-D-glucopyranoside (pNPG), Sulforhodamine B (SRB) and α-glucosidase were purchased from Sigma-Aldrich Co. (St Louis, MO, USA). Methanol and dimethyl sulfoxide (DMSO) with a purity of over 98% were purchased from Xilong Scientific Co., Ltd (Guangdong, China).

Cell lines

Two human cancer cell lines and a normal cell line were provided by the Department of Experimental Biology–Institute of Natural Products Chemistry, including MCF7 (Human breast adenocarcinoma, ATCC no. HTB-22), LU-1 (Human lung adenocarcinoma, ATCC no. HTB-57) and Vero cells (Chlorocebus aethiops kidney epithelial cells, ATCC no. CCL-81.4).

Methods

Identification of fungal strains

Morphological identification

For macromorphology observations, including colonies’ pigmentation, size and texture, fungal strains were grown on different growth media, including Dichloran Glycerol Agar (DG-18), Czapek’s Agar (CZA) and Sabouraud Dextrose Agar (SDA) plates, at 25°C in 7 days. The fungal specimens were examined by microscope for determination of hyphal and reproductive structures as previously described by Klich.¹⁶

Molecular identification

Fungal genomic DNAs were extracted from biomass using CTAB method.¹⁷ Primer pair ITS1 (5′-TCC GTA GGT GAA CCT GCG G-3′) and ITS4 (5′-TCC TCC GCT TAT TGA TAT GC-3′) was employed for amplification of the internal transcribed spacer (ITS) region from isolated fungal DNAs.¹⁸ PCR was performed with a 25-μL reaction mixture consisting of the following (per reaction): 4 µL of template, 1 µL of each primer, 2 µL of 10 mM dNTPs, 0.25 µL of Taq polymerase, 5 µL of 10X Taq polymerase buffer, 1.5 µL of 50 mM MgCl₂ and 10.25 µL of deionized water. After an initial denaturation at 94°C for 3 mins, thirty cycles of amplification were performed. Each cycle composed of a denaturation step at 94°C for 1 min, an annealing step at 55°C for 1 min, and an extension at 72°C for 2 min. PCR products were purified and subsequently sequenced by Perkin Elmer ABI Prism 377 DNA Sequencer (USA). By Basic Local Alignment Search Tool (BLAST) in the National Centre for Biotechnology Information (NCBI) database, identified regions of high similarity to ITS fungal sequences were determined. The phylogenetic tree was constructed using a neighbour-joining method with 1000 bootstrap values available in MEGA 7.0 software.¹⁹

Physiological and biochemical identification

Certain characteristics of isolated strains, including optimal temperatures, pHs and salinities for growth, were identified through experimental procedures. The production of enzymes by fungal strains was determined by measurement of halo size on agar plates containing specific substrates.²⁰

Preparation of fermented C. chrysantha leaves

Air-dried tea leaves were subjected to an enzyme-inactivation process (121°C, 20 min) by which microbial spores and infusion factors on the surface of tea leaves were eliminated.

Selected strains of Eurotium spp. were activated by cultivation in Sabouraud Dextrose broth (25°C, 3 days) before inoculating into C. chrysantha leaves.

The fermentation process was performed in 500 mL Erlenmeyer flasks for both non-inoculated (NF) and microbial inoculated fermentation (IF). In brief, 100 grammes of pre-processed (121°C for 20 min) raw tea leaves were adjusted to moisture level of 60% to generate C. chrysantha tea infusion. The infusion was inoculated with either 10 mL sterile water or the same volume of fungal suspension (5 × 10⁶ CFU/mL) to produce portions of NF and IF, respectively. The whole fermentation process was taken for 21 days at 25°C and protected from light. Each fermentation mode (NF/IF) was repeated three times for consistency and accuracy determination.

Extraction and chemical isolation

Fermented leaves of C. chrysantha (3 kg) were extracted thrice (each 10 L, 3 days) with ethanol at 50°C using a conventional ultrasound-assisted technique. The solvent was then in vacuo evaporated (EYELA, Japan) to yield ethanol extract (600 g). The ethanol extract residue was suspended in water (2.0 L) and successively partitioned with n-hexane and ethyl acetate (EtOAc) to obtain n-hexane (THV-LM/H, 120 g) and EtOAc (THV-LM/E, 250 g) extracts, respectively.

Ethyl acetate extract THV-LM/E was chromatographed on a normal phase silica gel column eluting with a dichloromethane: methanol:water gradient system (4:1:0.05 to 0:100:0, v:v:v) to give rise three fractions (THV-LM/E1→THV-LM/E3). Using a normal phase silica gel column eluting with gradient solvents of dichloromethane: methanol:water (10:1:0.05, v:v:v), fraction THV-LM/E1 was chromatographed to give rise five subfractions (THV-LM/E1.1→THV-LM/E1.5). Further chromatographic separation of subfraction THV-LM/E1.1 (normal phase silica gel column, gradient solvent system of n-hexane:EtOAc with ratios v/v of 100%, 50:1, 20:1, 10:1, 100%) led to isolation of THV-LM 03 (white powder, 10.0 mg).

Structure elucidation

The nuclear magnetic resonance (NMR) spectroscopy, including ¹H NMR (600 MHz) and ¹³C NMR (150 MHz), COSY, HMQC and HMBC of isolated compound were measured on Bruker spectrometer (Germany) at the Institute of Chemistry. Thin layer chromatography (TLC) was performed on pre-coated silica gel DC-Alufolien 60 F254 (Merck), combined with an ultraviolet lamp (UV) at 254 nm and 365 nm, 10% aqueous H₂SO₄ solution and heating.

Assessment of in vitro cytotoxicity

The cytotoxicity of extracts and compounds was determined by SRB colorimetric method as described by Skehan et al.²¹ Briefly, dissolved test samples in DMSO were diluted and incubated with cells (seeded at 0.8 × 10⁴ cells per well) in 96-well microtiter plates (37°C, 5% CO₂, 48 h). Cultures were then fixed with trichloroacetic acid and stained by 0.4% (w/v) SRB. The protein-bound dye was extracted with 10 mM Tris-base for the determination of optical density (OD) at a single wavelength of 540 nm in a 96-well plate reader (Tecan F150 microplate reader, Switzerland).

Assessment of inhibitory effects against lipase and alpha-glucosidase

The lipase and α-glucosidase inhibitory activities of extracts were evaluated as mentioned in our previous publication²² following Lewis and Liu²³ and Tadera et al.²⁴ with minor modifications.

The absorbance of released p-nitrophenol was measured at 405 nm using a spectrometer system (Tecan F150 microplate reader, Switzerland). Acarbose was included as a standard in α-glucosidase inhibition assay, while orlistat was employed as a positive reference in lipase inhibitory assay. Each experiment was performed in triplicates. IC₅₀ values of test samples were calculated by interpolation from linear regression analysis.

Results

Fungal characterization

In order to select E. cristatum candidates for effective fermenting C. chrysantha leaves, six strains from our microbial collection (E1, E2, E3, E6, E7, and E8) were characterized. These strains were found to exhibit great identities to ITS sequences of published Aspergillus and Eurotium species and typical morphological characteristics of E. cristatum on agar plates. The characterization of fungal candidates in detail was presented in Supplementary file 1.

A screening process of in vitro antioxidative and antidiabetic activities, as well as safety test in Balb/c mice (data not shown) suggested E. cristatum strains E1, E3 and E8 as potential bio-active inoculums. These characterized fungi were then chosen to inoculate in C. chrysantha leaves. Crude extracts yielding from the fermentation of C. chrysantha tea leaves by E. cristatum strains E1, E3 and E8 either as monocultures or associated as a consortium of three strains, were symbolized as IF-E1, IF-E3, IF-E8 and IF-LM, respectively. The biological activities of these extracts were assayed and compared to their non-fermented counterpart (NF) to determine fermenting efficiency.

Biological activities of crude extracts from E. cristatum fermented tea leaves

Cytotoxicity

The effect of tea leaves extracts on the viability of cells was in vitro assessed both on adenocarcinoma cell lines LU-1 and MCF-7 and normal cells (Vero cells). Results depicted in Table 1 indicated cytotoxic effects on cancer cells of fermented extracts (IF-E1, IF-E3, IF-E8 and IF-LM) in comparison to unfermented NF. Remarkably, while NF exhibited no obvious change in survival rates of tested cells, the application of E. cristatum fungal strains, especially as a fungal consortium (IF-LM) seemed to yield extracts with elevated cytotoxic effects against both cancer cell lines.

Table 1.

Cytotoxicity of tea leaves’ extracts in cancer cell lines LU-1 and MCF-7.

No.	Sample name	Test conc. (µg/mL)	cell survival (CS, %)			IC₅₀ (μg/mL)
No.	Sample name	Test conc. (µg/mL)	LU-1	MCF-7	Vero	LU-1	MCF-7	Vero
	Negative control^*	-	100	100	100	-	-	-
	Positive control ^* *	5	0.86 ± 0.51	2.72 ± 0.45	18.03 ± 1.02	0.27 ± 0.01	0.49 ± 0.02	1.93 ± 0.01
1	IF-E1	50	38.46 ± 0.50	90.82 ± 1.56	99.63 ± 0.13	46.59 ± 0.03	-	-
2	IF-E3	50	40.84 ± 0.36	81.65 ± 2.03	98.22 ± 0.38	47.32 ± 0.05	-	-
3	IF-E8	50	40.12 ± 1.07	91.84 ± 0.86	99.82 ± 0.80	47.16 ± 0.02	-	-
4	IF-LM	50	24.52 ± 2.05	70.27 ± 2.33	98.69 ± 1.15	40.57 ± 0.04	-	-
5	NF	50	99.82 ± 0.12	92.86 ± 2.15	99.21 ± 0.74	-	-	-

Negative control: DMSO. * Positive control: Ellipticine.

Alpha-glucosidase inhibitory activity

The inhibitory activity of ethanolic crude extracts from unfermented (NF) and fermented tea leaves by E. cristatum strains (IF-E1, IF-E3, IF-E8 and IF-LM) against α-glucosidase was investigated and compared. The results were shown in Table 2.

Table 2.

Inhibitory activities of tea leaves’ extracts against α-glucosidase.

No.	Sample name	Test conc. (μg/mL)	Inhibition (%)	IC₅₀ (μg/mL)
	Positive control (Acarbose)	100	67.48 ± 0.38	90.75 ± 0.04
	Negative control (DMSO)	-	0	-
1	IF-E1	500	85.74 ± 0.22	111.14 ± 0.01
2	IF-E3	500	82.95 ± 0.38	187.84 ± 0.01
3	IF-E8	500	80.29 ± 0.16	95.08 ± 0.03
4	IF-LM	500	85.28 ± 0.09	87.51 ± 0.02
5	NF	500	81.12 ± 0.25	135.17 ± 0.04

It was observed that not only unfermented tea leaves’ extract (NF) but also microbial fermented leaves’ extracts exhibited inhibition against α-glucosidase, with inhibitory rates ranging from 80.29% ± 0.16% to 85.74% ± 0.22% at the test concentration of 500 μg/mL. Among them, IF-LM extract showed an enhanced inhibitory activity with IC₅₀ at 87.51 ± 0.02 μg/mL, indicating the advantage of using microbial consortium over applying single fungal strain in the fermentation process.

Determined biological activities of extracts (Tables 1 and 2) demonstrated significant differences between non-fermented and fermented leaves and suggested the use of the fungal consortium composed of E. cristatum E1, E3 and E8 in enhancing biological activities. Further steps in our study were undertaken to characterize the most bio-active components from fermented tea leaves when applying the fungal consortium by stepwise bioactivities guided chemical extraction, fractionation and isolation.

Biological activities of extracts and fractions from fermented tea leaves

For selecting effective portion in fermented tea leaves’ extracts, n-hexane (THV-LM/H) and ethyl acetate (THV-LM/E) extracts from IF-LM were evaluated for in vitro anticancer and antidiabetic properties. Obtained results (Table 3) indicated more potent effects in inhibiting cancer cells and an enhanced inhibitory effect against α-glucosidase of THV-LM/E over THV-LM/H. At 50 µg/mL, THV-LM/E exhibited cytotoxicity against LU-1 and MCF-7 cells, with cell survival rates at 43.95%±2.06% and 29.15%±2.26%, respectively. The α-glucosidase inhibitory assay revealed significant activity of THV-LM/E, with exceeding 60% inhibited enzyme activity at test concentration of 200 µg/mL (Table 3).

Table 3.

Biological activities of n-hexane and ethyl acetate extracts from C. chrysantha fermented leaves.

No.	Sample *	Cytotoxicity against cancer cells and normal cells				Inhibitory effect against α-glucosidase
No.	Sample *	Test conc. (µg/mL)	LU-1–cell survival rates (CS, %)	MCF-7–cell survival rates (CS, %)	Vero–cell survival rates (CS, %)	Test conc. (µg/mL)	Enzyme inhibition (%)
1	THV-LM/H	50	98.39 ± 1.54	95.71 ± 0.87	98.83 ± 0.39	200	32.84 ± 0.36
2	THV-LM/E	50	43.95 ± 2.06	29.15 ± 2.26	78.52 ± 1.58	200	64.13 ± 0.11
	Ellipticine	5	2.11 ± 0.30	4.42 ± 1.05	21.65 ± 0.42		-
	Acarbose	-	-	-	-	100	60.37 ± 0.21

THV-LM/H, THV-LM/E: n-Hexane and Ethyl acetate extracts from fungal fermented C. chrysantha leaves, respectively.

Results indicated ethyl acetate as the appropriated solvent for extraction of active fractions from fermented golden tea leaves. Further chemical fractionation of THV-LM/E by chromatography (normal phase silica gel) and elution with a dichloromethane:methanol:water gradient system gave rise to three fractions (THV-LM/E1→THV-LM/E3). Determined biological activities of these were concluded in Table 4.

Table 4.

Biological activities of fractions from extract THV-LM/E of fermented tea leaves.

No.	Sample	Cytotoxicity against cancer cells and normal cells				Inhibitory effect against α-glucosidase
No.	Sample	Test conc. (µg/mL)	LU-1–cell survival rates (CS, %)	MCF-7-cell survival rates (CS, %)	Vero–cell survival rates (CS, %)	Test conc. (µg/mL)	Enzyme inhibition (%)
1	THV-LM/E1	20	36.10 ± 1.82	39.73 ± 1.50	57.25 ± 1.31	200	56.33 ± 0.26
2	THV-LM/E2	20	64.69 ± 2.41	99.50 ± 1.93	99.07 ± 0.28	200	58.29 ± 0.18
3	THV-LM/E3	20	79.76 ± 1.54	84.16 ± 2.27	97.33 ± 0.92	200	33.88 ± 0.15
	Ellipticine	5	0.86 ± 0.58	2.36 ± 0.79	21.65 ± 0.42		-
	Acarbose	-	-	-	-	100	62.31 ± 0.10

Among three investigated fractions, THV-LM/E1 appeared as the most active fraction, with significant cytotoxicity against two cancer cell lines LU-1 and MCF-7, and positive inhibition against α-glucosidase (Table 4). THV-LM/E2 was found as an α-glucosidase inhibitory but inadequate cytotoxic fraction, while THV-LM/E3 exhibited negative results in both assays. Fraction THV-LM/E1 was thus selected for chemical purification, where the combination of column chromatography and thin-layer chromatography methods was applied and led to the isolation of compound THV-LM03.

Chemical elucidation of THV-LM03

Compound THV-LM03 was isolated as white powder. The ¹H-NMR spectrum of THV-LM03 (Figure 1) showed seven singlet signal of methyl groups at δ_H 0.80 (3H, s, CH3-23), 1.0 (3H, s, CH3-24), 1. 0 (3H, s, CH3-25), 0.98 (3H, s, CH3-26), 1.50 (3H, s, CH3-27), 0.93 (3H, s, CH3-29), 1.92 (3H, s, CH3-30) ppm, an olefinic proton at δ_H 5.36 (1H, t, J = 3.6 Hz, H-12), an oximethine group at δ_H 3.18 (1H, dd, J = 4.8/11.4 Hz, H-3). Moreover, in the ¹H-NMR spectrum of THV-LM03, two methyl groups at δ_H 1.99 (3H, d, J = 7.2 Hz) and 1.92 (3H, s), an olefinic proton at δ_H 6.08 (1H, dq), two oximethine at δ_H 4.13 (1H, br s) and 5.45 (1H, dd, J = 6.0/12.0 Hz) were indicated.

Figure 1.

¹H-NMR spectrum of compound THV-LM03.

The ¹³C-NMR and HSQC spectroscopic data of compound THV-LM03 (Figure 2, Figure 3) exhibited the presence of thirty carbons characterizing for an oleanolic acid triterpene with seven methyl signals at δ_C 16.31 (C-23), 28.72 (C-24), 16.15 (C-25), 17.30 (C-26), 27.72 (C-27), 33.59 (C-29) and 25.25 (C-30); two sp² carbon atoms at δ_C 124.51 (C-12), 144.0 (C-13) revealed a double bond between C-12 and C-13 of the oleanane skeleton; a carbinol carbon at δ_C 64.84 (C-28). Chemical shift at δ_C 79.70 (C-3) and δ_H 3.18 (1H, dd, J = 4.8/11.4 Hz) showed a β-OH at C-3. Besides, the ¹³C-NMR and HSQC spectroscopic data of THV-LM03 revealed the presence of five other carbons of tigloyl group, including a carbonyl carbon at δ_C 169.75 (C-1′), a methine carbon at δ_C 138.08 (C-3′) which corresponded to δ_H 6.08 (1H, m, H-3′) in HSQC spectrum, two methyl groups at δ_C 15.92 (C-4′), 20.91 (C-5′) corresponding to δ_H 1.99 (3H, d, J = 7.2, CH₃-4′) and 1.92 (3H, s, CH₃-5′), respectively, in HSQC spectrum, and a carbon at δ_C 129.98 (C-2′).

Figure 2.

¹³C-NMR spectrum of compound THV-LM03.

Figure 3.

HSQC spectrum of compound THV-LM03.

Correlation signals between δ_H 5.45 (1H, dd, J = 6.0/12.0 Hz, H-22) and δ_C 169.75 (C-1′) on HMBC confirmed the attachment of tigloyl group at C-22 position. Similarly, correlation signals between δ_H 3.08 (1H, m, CH₂-28a)/3.29 (1H, m, CH₂-28b) and δ_C 45.32 (C-17) indicated an attachment of hydroxymethyl group at C-17 position (Figure 4).

Figure 4.

HMBC spectrum of compound THV-LM03.

From these spectral data (Table 5) and comparison to those in literature,²⁵ compound THV-LM03 was determined to be 22-O-tigloyloxycamelliagenin A (M = 556.82 g/mol) (Figure 5).

Table 5.

¹H-NMR and ¹³C-NMR data of 22-O-Tigloyloxycamelliagenin A and compound THV-LM03.

Pos.	δ _C ^#	δ _C ^{a, b}	DEPT	δ _H ^{a, c} mult. (J in Hz)	HMBC
1	38.6	40.03	CH₂	1.03 (1H, m),1.67 (1H, m)	C-3, C-5, C-10
2	27.0	27.91	CH₂	1.58 (1H, m),1.65 (1H, m)	C-3, C-4
3	79.0	79.70	CH	3.18 (1H, dd, 4.8/11.4)	C-1, C-4, C-23, C-24
4	38.8	41.00	C	-
5	55.5	56.79	CH	0.78 (1H, br s)	C-1, C-23, C-24, C-25
6	18.2	19.47	CH₂	1.45 (1H, m),1.59 (1H, m)	C-4
7	32.7	33.98	CH₂	1.37 (1H, m),1.62 (1H, m)	C-26
8	39.7	39.86	C	-
9	46.6	48.09	CH	1.70 (1H, m)	C-25, C-26
10	36.9	38.05	C	-
11	23.3	24.63	CH₂	1.95 (2H, m)	C-9
12	124.1	124.51	CH	5.36 (1H, t, 3.6/7.2)	C-9, C-14, C-19, C-27
13	141.6	144.0	C	-
14	41.4	42.45	C	-
15	33.1	35.28	CH₂	1.36 (1H, m),1.77 (1H, m)	C-27
16	71.0	70.99	CH	4.13 (1H, br s)	C-22, C-28
17	45.0	45.32	C	-
18	42.4	41.76	CH	2.55 (1H, m)	C-22, C-28
19	46.9	48.04	CH₂	1.08 (1H, m),2.48 (1H, m)	C-30
20	31.6	32.50	C	-
21	42.1	42.07	CH₂	1.58 (1H, m),2.28 (1H, m)	C-29, C-30
22	72.4	73.92	CH	5.45 (1H, dd, 6.0/12.0)	C-1′, C-16, C-17, C-18, C-21, C-28
23	15.6	16.31	CH₃	0.80 (3H, s)	C-4, C-24, C-25
24	28. 1	28.72	CH₃	1.00 (3H, s)	C-4, C-25
25	16.7	16.15	CH₃	1.00 (3H, s)	C-4, C-8, C-9
26	16.1	17.30	CH₃	0.98 (3H, s)	C-7
27	27.0	27.72	CH₃	1.50 (3H, s)	C-13, C-14, C-15
28	64.6	64.84	CH₂	3.08 (1H, m),3.29 (1H, m)	C-17, C-18, C-22
29	33.0	33.59	CH₃	0.93 (3H, s)	C-19, C-20, C-21
30	24.6	25.25	CH₃	1.92 (3H, s)	C-19, C-20, C-21
1′	168.9	169.75	C	-
2′	128.5	129.98	C	-
3′	138.1	138.08	CH	6.08 (1H, m)	C-5′
4′	12.1	15.92	CH₃	1.99 (3H, 3H, d, 7.2)	C-2′, C-3′
5′	14.5	20.91	CH₃	1.92 (3H, s)	C-3′

Recored in CD₃OD, ^b150 MHz, ^c600 MHz. δ_C^#: 22-O-Tigloyloxycamelliagenin A (in CDCl₃).

Figure 5.

Chemical structure of THV-LM03.

Biological activities of compound THV-LM03

Studies on the chemical composition based on biological activity of fermented leaves of C. chrysantha resulted in the isolation of compound 22-O-tigloyloxycamelliagenin A (THV-LM03) – an oleanane pentacyclic triterpenoid. Results of the evaluation on its cytotoxicity in cancer cells and the inhibitory activity against α-glucosidase were concluded in Table 6. Collected data indicated potent inhibitory activities of 22-O-tigloyloxycamelliagenin A against cancer cells of LU-1 and MCF-7 with IC₅₀ values at 4.87 ± 0.03 and 9.32 ± 0.02 µg/mL (~8.75 and ~16.74 µM equivalent), respectively, as well as against α-glucosidase (IC₅₀=16.33 ± 0.01 µg/mL, ~29.33 µM equivalent) (Table 6).

Table 6.

Biological activities of compound THV-LM03.

Sample	Cytotoxicity against cancer cells and normal cells			Inhibitory effect against α-glucosidase		Inhibitory effect against lipase
Sample	LU-1–cell inhibitory concentration (IC₅₀, µg/mL)	MCF-7–cell inhibitory concentration (IC₅₀, µg/mL)	Vero–cell inhibitory concentration (IC₅₀, µg/mL)	Enzyme inhibition at 100 µg/mL (%)	IC₅₀ (µg/mL)	Enzyme inhibition at 100 µg/mL (%)	IC₅₀ (µg/mL)
THV-LM03	4.87 ± 0.03	9.32 ± 0.02	8.30 ± 0.01	90.26 ± 0.15	16.33 ± 0.01	73.46 ± 0.20	25.74 ± 0.08
Ellipticine	2.18 ± 0.01	2.46 ± 0.01	1.97 ± 0.02	–	–	–	–
Acarbose	–	–	–	61.57 ± 0.24	90.05 ± 0.01	–	–
Orlistat	–	–	–	–	–	91.22 ± 0.14	2.92 ± 0.12

Notwithstanding previous results concerning cytotoxic property of 22-O-tigloyloxycamelliagenin A,²⁵ the inhibition against cancer cells of LU-1 and MCF-7 is a new contributing data for its profile. Most importantly, our present study revealed 22-O-tigloyloxycamelliagenin A as a product of fungal biotransformation on tea leaves which exhibited diabetes and obesity associated enzymes inhibition. The IC₅₀ value (~29.33 µM) of THV-LM03 against α-glucosidase was roughly lowered in comparison to the corresponding concentration of acarbose (IC₅₀ = 90.05 ± 0.01 µg/mL, ~139.61 µM equivalent) (Table 6), contributed to highlight a great potential of the compound as a α-glucosidase inhibitor.

The results indicate the potential of THV-LM03 as a dual-action therapeutic agent in the management of both metabolic disorders, including diabetes and obesity, and in cancer therapy, though its cytotoxicity to normal cell line (Vero cells, IC₅₀ at 8.30 ± 0.01 µg/mL) warned us of possible risk and mandatory further studies to confirm its safety profile as well as consumption thresholds to the body.

For the first time, the compound was discovered as a potential pharmaceutical ingredient for treating cancer and weight management. However, further experiments and trials are required to develop biomedical preparations from the transformed terpenoid.

Discussion

Hypothetical novel fungi-driven biotransformation pathway of triterpenoids

Previously, 22-O-tigloyloxycamelliagenin A has been reported as a constituent from the bark of Barringtonia asiatica.²⁵ Our present result brought out the first isolation of such triterpene as a plant leaves derived compound. Interestingly, this compound has not been isolated from either leaves of C. chrysantha or E. cristatum fungi. Possibly, the fungi transformed triterpenes on tea leaves substrate and led to the production of novel compounds, including 22-O-tigloyloxycamelliagenin A (THV-LM03).

In Camellia sinensis, Xiao et al.²⁶ reported a significant increase of aroma quality and terpene alcohols during E. cristatum driven fermentation, indicating the fungal involvement in the metabolism of terpenoids. Among four previously isolated triterpenoids from C. chrysantha leaves,¹³ β-amyrin shared the most similar skeleton to THV-LM03, and thus far could be considered the parent compound in the biotransformation process. As results of our private experiment (data not shown), not only β-amyrin but also three other isolated triterpenoids from C. chrysantha leaves (β-amyrin hexanoate, friedelin, and friedelanol)¹³, were found negative in the cytotoxic assay against cancer cells of Hep-G2 and MCF-7 with cell survival rates ranging from 75.42% to 97.45% at test concentration of 20 µg/mL. Furthermore, their enzyme inhibitory property against glucosidase was at a neglectable level (under 50% of inhibition at test concentration of 100 µg/mL). Therefore, the present structural modification by E. cristatum consortium is considered in parallel with the enhancement of triterpenoid’s biological activities.

Possibly, during the process, two major transformations occurred, allowing the insertion of hydroxyl groups and stepwise tigloyl group at β-amyrin’s specific sites to yield the corresponding derivative THV-LM03. Under the catalysation of the E. cristatum fungal consortium, an intermediate product was synthesized via the hydroxylation at C-16, C-22 and C-28 positions of the parent compound, bearing additional hydroxyl groups at these sites. Subsequently, the compound combined with tiglic acid (2-methyl-2-butenoic acid) at C-22 through an esterification reaction (Figure 6). Previous studies emphasized the improvement of solubility and thereby altering physicochemical properties and thus modifying biological activities of natural products by hydroxylation.²⁷ The hydroxylated products were reported to be extremely diverse in structure and properties. For instance, novel hydroxylated derivatives of diterpenoids by using filamentous fungi Mortierella ramanniana and Gibberella fujikuroi were found with reduced cytotoxicity.²⁸

Figure 6.

Hypothetical biotransformation of β-amyrin to THV-LM03 in fermented tea leaves.

The formation of THV-LM03 was suppositional dedicated by a combination of fungal oxygenase and lipase activities, by which oxidation and transesterification reactions were catalysed, respectively. Fungal enzymes, especially lipases, have commonly been applied in the transesterification process of biomass.²⁹ As presented in Supplementary File 1, all used E. cristatum strains (E1, E3 and E8) possessed lipase activity, suggesting the formation of THV-LM03 as a sequence of the microbial enzymatic catalysis. Miettinen et al.³⁰ reviewed the involvement of CYP716A subfamily cytochrome P450 (CYP716s) monooxygenases of eudicot plants derivation in C16/C22/C28 hydroxylation of triterpenoids, however fungal factors with similar catalytic manner heretofore have not been mentioned. This is seemingly the first evidence for the presence of such CYP716s-like system in filamentous fungi.

For determining accurate parameters of novel triterpenoid biosynthesis using the microbial consortium, further studies on optimizing fermentation conditions in herbal substrates, that is, C. chrysantha leaves, are required. As proposed in our hypothetical conversion pathway (Figure 6), β-amyrin and 22-O-tigloyloxycamelliagenin A could serve as chemical signatures for the determination as well as evaluation of fermentation efficiency in extended optimization processes.

An effective microbial consortium for enhancing tea leaves’ value

In the past decades, microbial transformation has been demonstrated to be a versatile tool to enlarge the structural diversity of triterpenoids. For the purpose, the selection of appropriate organisms serving for a desired modification of particular substrates played a key role.³¹ Our results shed a light on the use of an E. cristatum fungal consortium for specific hydroxylation and esterification of triterpenes to more bio-active products. In detail, characterized E. cristatum strains E1, E3 and E8 of different origins showed mostly similar bioactivities when inoculated in C. chrysantha leaves’ substrate as monocultures, but even an enhanced effect when fused as a consortium (Table 3, Table 4). As a fungal consortium, the efficiency of catalytic engineering might be more advantageous due to the potential activation of silent metabolic pathways, which are normally not expressed under monoculture condition.³²

Our study lays foundation for the development of fermented tea beverage production using E. cristatum fungi in the form of a consortium. On one hand, the existence of E. cristatum in dark teas might play a role as probiotics for modulation of gut microbiota,⁶ on the other hand, the fungal consortium could help improve the beneficial ingredients’ bio-synthesis efficiency in the value-added products.

Conclusion

In our work, three fungal strains of E. cristatum were characterized and successfully inoculated in C. chrysantha leaves. In terms of biological assessments, the fermentation by a fungal consortium including these strains was found with advantages over monocultural and uninoculated counterparts. Further in vitro evaluation results indicated enhanced effects in inhibiting cancer cells and against α-glucosidase of ethyl acetate extract over n-hexane extract from the microbial fermented tea leaves. The biological activities guided chemical isolation of C. chrysantha leaves’ ethyl acetate extract gave out 22-O-tigloyloxycamelliagenin A as the most active ingredient. The compound was found as a potential α-glucosidase inhibitor, α-lipase inhibitor and cytotoxic triterpene against cancer cells and thus worthy for further studies on application aspects. Furthermore, a hypothetical two-step transformation of 22-O-tigloyloxycamelliagenin A from plant triterpenoid component under the catalysis of E. cristatum’s enzymes was proposed, supporting for more comprehensive understandings about the conversion of triterpenoids during microbial fermentation of herbal substrates.

In conclusion, this study brought out the realization of E. cristatum fungal strains as an effective consortium in the fermentation of C. chrysantha leaves and an expanded understanding of us about the biotransformation of triterpenoids in organic materials in general. These might provide a neoteric option for processing golden camellias’ leaves, especially for promoting value-added products with enhanced health benefits. Equally important, our study contributed to establish an environmentally friendly synthesis method of bio-active molecules which is cost-effective with reducing energy use and chemical waste in the future.

Supplemental Material

sj-docx-1-chl-10.1177_17475198261421696 – Supplemental material for The fermentation of Camellia chrysantha leaves with Eurotium cristatum to yield biologically active compound

Supplemental material, sj-docx-1-chl-10.1177_17475198261421696 for The fermentation of Camellia chrysantha leaves with Eurotium cristatum to yield biologically active compound by Hang TN Tran, Quan T Nguyen, Van TH Nguyen, Tuyen T Tran, Bach C Pham, Cuong H Le, Nghi H Do, Ha TH Tran and Chi K Hoang in Journal of Chemical Research

Footnotes

Acknowledgements

We would like to thank Dr Luu Van Chinh and other co-workers for helpful discussion and technical works.

ORCID iD

Chi K. Hoang

Ethical considerations

The submitted manuscript contains original and new results which are not submitted for publishing to other publications or published elsewhere.

All research was conducted to the highest possible ethical standards.

All authors are involved in the work. Authors and co-authors reviewed and ensured the accuracy and validity of all the results prior to submission.

All authors guarantee that the research findings have not been previously published.

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Vietnam Academy of Science and Technology [grant number THTE00.10/24-25].

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

Data supporting this study are included within the article and supporting materials.

Supplemental material

Supplemental material for this article is available online.

References

Assad

Ashaolu

Khalifa

, et al. Dissecting the role of microorganisms in tea production of different fermentation levels: a multifaceted review of their action mechanisms, quality attributes and future perspectives. World J Microbiol Biotechnol 2023; 39: 265.

Zhang

Deng

Wan

XC.

Advantage of LC-MS metabolomics to identify marker compounds in two types of Chinese dark tea after different post-fermentation processes. Food Sci Biotechnol 2014; 23: 355–360.

Shanenko

Efremenkova

Mukhamedzanova

, et al. Using the Eurotium cristatum fungus for preparing fermented herbal teas. J Pharm Nutr Sci 2020; 10(6): 341–361.

Zhao

Tang

Fang

, et al. Effects of Urolithin A-producing streptococcus thermophilus FUA329 fermentation on the composition and antioxidant bioactivities of Black Tea. Food Med Homol 2025; 2(1): 9420041.

Jiang

Liu

, et al. Non-volatile metabolic profiling and regulatory network analysis in fresh shoots of tea plant and its wild relatives. Front Plant Sci 2021; 12: 746972.

Kang

Duan

, et al. Eurotium cristatum, a potential probiotic fungus from Fuzhuan brick tea, alleviated obesity in mice by modulating gut microbiota. Food Funct 2019; 10: 5032–5045.

Zhang

Wang

Shen

, et al. Antioxidant phenolic compounds from Pu-erh tea. Molecules 2012; 17(12): 14037–14045.

Zou

Cao

Zhang

, et al. Improvement of quality of Ginkgo biloba seeds powder by solid-state fermentation with Eurotium cristatum for developing high-value ginkgo seeds products. J Bioresour Bioprod 2022; 7(2): 135–144.

Tang

Han

Zhao

, et al. From forest to cup: a comprehensive review of health benefits and quality control of Cyclocarya paliurus tea. Food Rev Int 2025; 1: 1–26.

10.

Zhuang

Oil-tea camellia in China. Beijing, China: China Forestry Publishing House, 2008.

11.

Song

Wang

Zheng

, et al. Polyphenolic antioxidant profiles of yellow camellia. Food Chem 2011; 129: 351–357.

12.

Shi

, et al. Chemical constituents from leaves of Camellia nitidissima and their potential cytotoxicity on SGC7901 cells. Chin Herb Med 2016; 8: 80–84.

13.

Van

NTH

Nghi

Bach

, et al. Triterpenoids from the leaves of Camellia chrysantha growing in Quang Ninh (Vietnam) and their activities on main protease (Mpro) and ACE2. Vietnam J Chem 2023; 61: 140–147.

14.

, et al. Analysis of amino acids composition and evaluation of nutritional value in flowers of 6 species of Camellia sect. Chrysantha Chang. Southwest China J Agri Sci 2020; 33: 2062–2068.

15.

Hakoda

Kirino

Tran

New species of genus Camellia in Vietnam. Int Camellia J 2007; 39: 54–57.

16.

Klich

MA.

Identification of common Aspergillus species. Washington, DC: ASM Press, 2002.

17.

Tripathy

Maharana

Ithape

, et al. Exploring rapid and efficient protocol for isolation of fungal DNA. Int J Curr Microbiol Appl Sci 2017; 6(3): 951–960.

18.

White

Bruns

Lee

, et al. PCR protocols a guide to methods and applications. London: Academic Press, 1990.

19.

Kumar

Stecher

Tamura

MEGA7: molecular evolutionary genetics analysis version 7.0 for bigger datasets. Mol Biol Evol 2016; 33: 1870–1874.

20.

Bhatt

Bhatia

Bhalla

TC.

Basic Biotechniques for Bioprocess and Bioentrepreneurship. London: Academic Press, 2023; p313-318.

21.

Skehan

Storeng

Scudiero

, et al. New colorimetric cytotoxicity assay for anticancer-drug screening. J Natl Cancer Inst 1990; 82(13): 1107–1112.

22.

Hoang

Nguyen

, et al. Steroid components of marine-derived fungal strain Penicillium levitum N33.2 and their biological activities. Mycobiol 2023; 51(4): 246–255.

23.

Lewis

Liu

DJ.

Direct measurement of lipase inhibition by orlistat using a dissolution linked in vitro assay. Clin Pharmacol Biopharm 2012; 1: 1000103.

24.

Tadera

Minami

Takamatsu

, et al. Inhibition of α-glucosidase and α-amylase by flavonoids. J Nutr Sci Vitaminol 2006; 52: 149–153.

25.

Ragasa

Espineli

Shen

CC.

Cytotoxic triterpene from Barringtonia asiatica. Pharm Chem J 2014; 48: 529–533.

26.

Xiao

Liu

, et al. The effect of Eurotium cristatum (MF800948) fermentation on the quality of autumn green tea. Food Chem 2021; 358: 129848.

27.

Sun

Zeng

Chen

, et al. Recent advances and challenges in the production of hydroxylated natural products using microorganisms. Ferment 2024; 10(12): 604.

28.

Cao

Liu

, et al. Regio-and stereo-selective hydroxylations of ingenane diterpenoids by Mortierella ramanniana and Gibberella fujikuroi. Chin J Nat Medicines 2016; 14(12): 939–945.

29.

Bajaj

Lohan

Jha

, et al. Biodiesel production through lipase catalyzed transesterification: an overview. J Mol Catal B Enzym 2010; 62(1): 9–14.

30.

Miettinen

Pollier

Buyst

, et al. The ancient CYP716 family is a major contributor to the diversification of eudicot triterpenoid biosynthesis. Nat Commun 2017; 8(1): 14153.

31.

De Carvalho

da Fonseca

MMR

. Biotransformation of terpenes. Biotechnol Adv 2006; 24(2): 134–142.

32.

Qian

Chen

Sui

, et al. Biotechnological potential and applications of microbial consortia. Biotechnol Adv 2020; 40: 107500.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.53 MB