Changes in tannin and saponin components during co-composting of Camellia oleifera Abel shell and seed cake

Abstract

This study investigated the co-compost product of the shell and seed cake of Camellia oleifera Abel, which is a small evergreen tree in the family Theaceae. Tannin and saponin contents in compost samples at different time-points and their compositional changes, as well as their relationships with nutrients and compost maturity, were analyzed using an UPLC-triple-TOF/MS system. Our results showed that tannins in the compost samples mainly consisted of 11 phenolic acid compounds, of which four small-molecule phenolic acid compounds were found in low contents. Saponins mainly consisted of five saponin aglycones (A, B, C, D, E) and four of their derivatives in C. oleifera. Microbially secreted enzymes converted the large-molecule phenolic acid compounds into small-molecule compounds and their derivatives, while saponins were decomposed into saponin aglycones. Contents of tannin and saponin had correlations with the C/N ratio, germination index (GI), and the Solvita maturity index. After composting, the content of tannin was reduced to less than 1%, and the content of saponin was not more than 2%. And compost products were safe.

Introduction

Camellia oleifera Abel is an economically important woody plant of southern China having the highest production value among the Camellia oil tree species [1]. In southern China, C. oleifera is commercially grown in the Yangtze River Basin and Pearl River Basin, where the Hunan, Jiangxi, and Guangxi Provinces together account for 76.2% of China’s total area under cultivation. This prized plant is also cultivated in small amounts in some areas in Vietnam, Myanmar, Thailand, and Malaysia, as well as Japan [1–2]. Recent statistics show that the annual production of shells and seed cake of C. oleifera has reached approximately 8 million and 4 million tons, respectively. The C. oleifera shell mainly consists of cellulose, hemicellulose, and lignin, which each decompose rather slowly under natural conditions [3]. The seed cake is the residue that remains after the extraction of oil from C. oleifera seeds, and its main components are crude protein, crude fat, fiber, saponin, tannin, ash, and caffeine [4]. Much of C. oleifera shells and seed cakes are either burned or discarded, representing a large, needless waste of organic resources. However, when this waste type is untreated and directly used as an agricultural fertilizer, or is simply thrown away, it generally causes soil and environmental pollution.

But the co-composting of the C. oleifera shells and seed cakes enables biological decomposition to proceed in a controlled manner. Due to the heat produced during decomposition, their raw materials are rendered harmless and become stabilized, which can benefit the growth of plants that are fertilized with the co-compost product. Both the physical and chemical properties of the products after decomposition typically differ greatly from those of the initial raw materials (US Composting Council). The C. oleifera shell and seed cake contain tannins and saponins in certain amounts. Tannins are polyphenolic compounds that can inhibit microbial growth and resist biological decomposition [5], and C. oleifera saponins are pentacyclic triterpenes that are highly toxic to ectotherms. Low concentrations of C. oleifera saponins can promote plant growth [6] and may also control pests to a certain extent, including the early instars of Pieris rapae (L.) larvae and Plutella xylostella (L.) larvae [7]. During the composting process, the temperature of C. oleifera shell—seed cake mixture would heat up, and lots of microorganisms that degraded organic matter would grow and reproduce. For instance, Aspergillus niger and Paecilomyces varioti, they could degrade tannins by hydrolyzing ester and carboxylic acid bonds[8–9]. In addition, glycosidic bond and the ester bonds of saponin would be broken because of changes of pH of substrate. However, no such researches have been reported yet. Hence, studying the dynamics of changed tannin and saponin components during the co-composting of C. oleifera shell and seed cake is important if this waste product is to be used sustainably on a large scale.

Here, C. oleifera shell and seed cake were used as raw materials in a co-composting experiment. Their tannin and saponin contents were determined by UV spectrophotometry and gravimetric methods, respectively. An ultra-performance liquid chromatography-triple-time-of-flight/mass spectrometry system (UPLC-triple-TOF/MS) was used to analyze changes in the chemical composition and contents of the phenols and saponins in the compost product at different composting time-points. These findings provide a timely basis for the widespread application of the co-compost product of C. oleifera shell and seed cake.

Materials and methods

Compost materials and preparation

The C. oleifera shells were obtained from the Dongfanghong Forest Farm in Jinhua City, and its seed cakes from the Kangneng Tea Oil Co., Ltd., in Tiantai City (both in Zhejiang Province, China). The seed cake was formed after mechanical extraction of the oil from C. oleifera seeds, and further oil extraction by No. 6 solvent oil, which primarily consists of alkanes. Effective microorganisms (EMs) were obtained from the Henan Nanhua Qianmu Biotechnology Co., Ltd. in Henan Province, China, whose main components were Bacillus, Lactobacillus, Bifidobacterium, yeast, photosynthetic bacteria, acetic acid bacteria, Actinobacillus, and other original species. Except the addition of EM, laboratory-selected tannins-degrading microorganism agents (Aspergillus awamori) and saponin-degrading microorganism agents (Bacillus amyloliquefaciens and Meyerozyma guilliermondii) were also added. For experiment and collection locations no specific permits were required for the described field studies because the whole experiment process did not involve endangered or protected plant species or privately-owned locations. Basic properties of the co-composting raw materials are summarized Table 1.

Table 1. Basic properties of the co-composting raw materials.

	TOC (%)	TN (%)	C/N	TP (%)	TK (%)	Tannin (%)	Saponin (%)
C. oleifera shell	48.6	0.42	116.00	0.0169	0.854	2.26	4.80
C. oleifera seed cake	47.8	1.22	39.18	0.161	0.929	1.03	16.35

TN: total nitrogen; TP: total phosphorus; TK: total potassium.

Composting process

The composting experiment was conducted at the Research Institute of Subtropical Forestry, Chinese Academy of Forestry in Fuyang District, Hangzhou City, in Zhejiang Province, from February to April 2019. The experiment took place in an insulated and well-ventilated eco-composter (outer dimensions: 73 cm × 115 cm × 80 cm; volume = 220 L; manufacturer: Biolan, Finland). The dry mass ratio of C. oleifera shell (≤ 8 mm) and seed cake (≤ 3 mm) was 4:1. Urea was added to adjust the initial carbon-to-nitrogen ratio (C/N) to 30; water was added to adjust the initial moisture content to 55% (w/w); the EMs (weighing 3% of the dry mass of C. oleifera shell and seed cake) were also added. And tannins-degrading microorganism agents (Aspergillus awamori) and saponin-degrading microorganism agents (Bacillus amyloliquefaciens and Meyerozyma guilliermondii) were added in an amount of 1% of the dry mass of C. oleifera shell and seed cake, respectively. The raw materials were mixed well and deposited into the eco-composter to begin the composting experiment. After composting began, the temperature of the upper, middle, and lower parts of the compost pile and the ambient temperature were recorded daily at 3:00 pm. Once every 5 days, the compost was turned until the temperature of the compost core matched the room temperature. Samples (each weighing 500 g) were taken on the 0th, 20th, 40th, 60th, and 76th day of composting and were stored at 0°C for later use.

Analytical methods

For each sample, its tannin content was analyzed via spectrophotometry after extraction with a methanol-water mixture (1:1, v/v) [10]. The saponins were determined according to previous methods [11].

Total organic carbon (TOC), total nitrogen (TN), total phosphorus (TP), total potassium (TK), NO₃^--N, and NH₄⁺-N were determined following the methodology of Meng et al.[12]. The germination index (GI) was determined as described by Zhang et al.[13], and the Solvita maturity index was determined according to the “Guide to Solvita testing for compost maturity index” (Woods End Research, 2002).

Analysis using the UPLC-triple-TOF/MS system

Fresh samples of compost, each weighing exactly 10.0 g, were individually placed in a 200-mL conical flask. One hundred milliliters of 80% methanol were added to each flask, which was followed by ultrasonic extraction in a cold-water bath for 30 min (power = 500 W; frequency = 40 kHz). The flasks were centrifuged at 10 000 rpm for 20 min, and their supernatants obtained for analysis using the UPLC-triple-TOF/MS system. The processing of each of the above samples was replicated 3 times.

Liquid Chromatography (LC)

For all the chromatographic experiments, a Waters UPLC (Waters Corp., Milford, MA, USA), Agilent ZORBAX-SB C18 (100 mm × 4.6 mm i.d: 1.8 μm) was used, for which the mobile phases were 0.1% formic acid-water (A) and 0.1% formic acid-acetonitrile (B). The linear gradient pro-gram parameters were as follows: 0/5, 2/5, 25/50, 33/95 (min/B%); sample injection volume = 10 μL; column oven temperature = 30°C; flow rate = 0.8 mL min⁻¹; the UV detector was set to 254 nm.

Mass Spectrometry (MS)

For this, an AB TripleTOF 5600^plus system (AB SCIEX, Framingham, USA) was used, with these optimal MS conditions applied: scan range = 100–2000 m/z; for negative ion mode: source voltage = –4.5 kV and source temperature = 550°C; for positive ion mode: source voltage = +5.5 kV and source temperature = 600°C. The pressure of gas 1 (air) and gas 2 (air) was set to 50 psi; the curtain gas (N₂) pressure was set to 35 psi. The maximum tolerable error was ± 5 ppm, with a declustering potential (DP) = 100 V and collision energy (CE) = 10 V used. The IDA-based auto-MS2 was performed on the eight most intense metabolite ions in a full scan cycle (1 s). The m/z scan range of the precursor ion and product ion were set to 100–2000 Da and 50−2000 Da, respectively. The CE voltage was set to 20 V, 40 V, and 60 V in the positive ESI mode and, conversely, to −20 V, −40 V, and −60 V in the negative ESI mode. Finally, the ion release delay (IRD) used was 67, and the ion release width (IRW) was 25.

The exact mass calibration was performed automatically before each analysis, by using the Automated Calibration Delivery System. Spearman correlations (non-parametric) of the data were carried out in SPSS 20.0 (IBM Co., Armonk, NY, USA).

Results and discussion

Changes in contents and components of tannin and saponin during composting

As shown in Fig 1 and Tables 2 and 3, both the tannin and saponin contents continued to decrease during the composting. The decomposition rate of tannins on the 20th, 40th, 60th, and 76th day of composting was 17.27%, 31.48%, 44.88%, and 63.93%, respectively. In the first 20 days of composting, tannins and proteins combined to form a precipitate. At the same time, tannins suppressed the activity of some thermophilic microbes [5], and the microbial decomposition of organic compounds was hindered; this caused temperatures to rise slowly and the decomposition of tannins to proceed slowly during this initial period. But from the 20th to 60th day of composting, microbes evidently became adapted to the inhibitive effects of the tannins. High microbial activity caused the organic matter to decompose rapidly, producing a large amount of heat. The compost pile then entered a high-temperature stage in which the tannins decomposed quickly under the dual effects of microbes and high temperature [14]. The decomposition rate of saponins on the 20th, 40th, 60th, and 76th day of composting was 37.87%, 62.01%, 72.86%, and 86.84%, respectively. The saponins decomposed continuously throughout the experiment’s heating phase, high-temperature stage, and cooling periods. As Table 6 shows, the tannin and saponin contents were negatively correlated with the number of days of composting.

Fig 1. Changes in tannin and saponin contents during composting of C. oleifera shell and seed cake.

Table 2. Contents of typical phenolic acids in the compost samples of C. oleifera shell and seed cake at different composting time-points.

Chemical formula	Chemical name	Response intensity
		0 day	20 days	40 days	60 days	76 days
C7H6O4	3,4-dihydroxybenzoic acid	626	521	483	119	42
C7H6O3	4-hydroxy-benzoic acid	1363	316	269	169	143
C12H8O8	7-hydroxy-5-methoxy-2-oxo-2H-chromene-3,6-dicarboxylic acid	350	168	151	122	99
C7H6O2	p-hydroxybenzaldehyde	355	338	309	425	263
C21H18O13	4'-O-(β-D-glucopyranosyl)-3,3',4'-tri-O-methylellagic acid	59046	917	250	144	148
C21H20O10	vitexin	119277	10451	883	213	251
C14H6O8	ellagic acid	192869	63696	45799	69793	67233
C15H8O8	3-O-methylellagic acid	207028	33707	30107	8540	6989
C16H10O8	3,3'-di-O-methyl ellagic acid	116691	114925	38946	2174	893
C15H10O6	kaempferol	87553	6974	2622	1712	886
C17H12O8	3,3',4-tri-O-methylellagic acid	112799	81221	13895	841	320

The result of relative response was obtained by liquid chromatography (LC).

Table 3. Contents of typical saponins in compost samples of C. oleifera shell and seed cake at different composting time-points.

Chemical formula	Chemical name	Response intensity
		0 day	20 days	40 days	60 days	76 days
C53H86O26	desacyl-theasaponin A	480428	4319	1360	50	34
C53H84O25	desacyl-theasaponin E methyl ester	553167	60762	12632	496	538
C43H68O17	Prosapogenol	409005	5642	3800	52	39
C30H50O6	theasapogenol A	777130	24819	76356	2208	2194
C30H48O6	camelliagenin D	1373366	68201	232396	2101	1928
C30H50O5	camelliagenin C	345046	25071	51791	6425	4844
C29H46O5	camelliagenin D	490142	124771	763485	66368	27814
C30H48O5	camelliagenin B	822892	65810	82132	3827	3633
C30H50O4	camelliagenin A	165745	22051	26608	4131	3280

The result of relative response was obtained by liquid chromatography (LC).

Table 6. Spearman correlations among physical and chemical parameters during composting of C. oleifera shell and seed cake.

	Day	Tannin	Saponin	TN	TP	TK	NO3--N	NH4+-N	Total nutrient content
Day	1	–0.996**	–0.973**	0.981**	0.979**	0.958*	0.902*	–0.919*	0.992**
Tannin (%)		1	0.967**	–0.958*	–0.958*	–.0966**	–0.917*	0.906*	–0.979**
Saponin (%)			1	–0.957*	–0.977**	–0.983**	–0.789	0.984**	–0.984**
TN				1	0.995**	0.911*	0.846	–0.918*	0.992**
TP					1	0.936*	0.817	–0.951*	0.996**
TK						1	0.814	–0.958*	0.956*
NO3--N							1	–0.670	0.851
NH4+-N								1	–0.949*
Total nutrient content									1

TN: total nitrogen; TP: total phosphorus; TK: total potassium.

As Table 2 shown, in both the starting raw materials and final compost product, the tannins mainly consisted of 11 phenolic acid compounds, and their respective chemical formulas was shown in Table 4. The content of 3,4-dihydroxybenzoic acid, 4-hydroxy-benzoic acid, 7-hydroxy-5-methoxy-2-oxo-2H-chromene-3,6-dicarboxylic acid, p-hydroxybenzaldehyde, 4'-O-(β-D-glucopyranosyl)-3,3',4'-tri-O-methylellagic acid, vitexin, 3-O-methylellagic acid, 3,3'-di-O-methyl ellagic acid, kaempferol, and 3,3',4-tri-O-methylellagic acid in them declined with composting process. The ellagic acid content declined in the first 40 days of composting but increased on its 60th day. This was due to the hydrolysis of 3-O-methylellagic acid, 3,3'-di-O-methyl ellagic acid, and 3',4-tri-O-methylellagic acid, which led to the formation of ellagic acid [15] which replenished and increased its content overall. Since some of the ellagic acid was further metabolized by the microbial community, into urolithin [16–18], the ellagic acid content decreased again on the 76th day of composting. The continuous reduction in tannin content was mainly due to the cleavage of the ester or phenolic bonds of tannins catalyzed by microbially secreted tannase, respectively forming phenolic acids or polyols. Under the enzymatic actions of polyphenol oxidase and decarboxylase, these compounds were further decomposed into phloroglucinol and resorcinol, and eventually into 3-hydroxy-5-oxo-hexanoic acid, 5-oxo-6-methyl hexanoate, and other small molecules. Some of these molecules were utilized as carbon sources, while some were converted to phenol derivatives [19–20]. The raw materials of the co-compost of C. oleifera shell and seed cake had low contents of these four phenolic acid compounds: 3,4-dihydroxybenzoic acid, 4-hydroxy-benzoic acid, 7-hydroxy-5-methoxy-2-oxo-2H-chromene-3,6-dicarboxylic acid, and p-hydroxybenzaldehyde. By contrast, the initial contents of 4'-O-(β-D-glucopyranosyl)-3,3',4'-tri-O-methylellagic acid, vitexin, ellagic acid, 3-O-methylellagic acid, 3,3'-di-O-methyl ellagic acid, kaempferol, and 3,3',4-tri-O-methylellagic acid were all higher than those four compounds.

Table 4. Chemical names of typical phenolic acid compounds in the compost samples (of C. oleifera shell and seed cake) and their respective chemical formulas.

C. oleifera saponins are pentacyclic triterpenes and they contain ligands, glycosyl groups, and small-molecule organic acids. The organic acids consist mostly of angelic acid and acetic acid. Glycosyl groups include glucuronic acid, arabinose, xylose, and galactose [21,22]. Before the composting, saponins in the C. oleifera shell and seed cake mainly contained glycosides and glycosyl groups, yet after composting, saponins contained glycosyl groups chiefly. In particular, the content of the camelliagenin D derivative (C₂₉H₄₆O₅) was found to be highest after composting, a result consistent with the findings by Lin et al. [22]. C. oleifera saponins in the compost samples mainly included glycoside molecules, including desacyl-theasaponin A (C₅₃H₈₆O₂₆), desacyl-theasaponin E methyl ester (C₅₃H₈₄O₂₅), and prosapogeno (C₄₃H₆₈O₁₇), and aglycone molecules, including theasapogenol A (C₃₀H₅₀O₆), camelliagenin D (C₃₀H₄₈O₆), camelliagenin C (C₃₀H₅₀O₅), camelliagenin D derivative (C₂₉H₄₆O₅), camelliagenin B (C₃₀H₄₈O₅), and camelliagenin A (C₃₀H₅₀O₄) (Table 3 and Table 5). The content of glycoside molecules continued to decline as composting proceeded, whereas the aglycone molecules were already reduced by the 20th day, but increased on the 40th day, yet decreased again after the 60th day of composting. This pattern was driven by the hydrolysis of the glycosidic bond of C. oleifera saponins by acids, bases, or enzymes into sugars and aglycones. According to Hu et al. [23], under acidic conditions, the glycosidic bond at the 3-position was broken, while under alkaline conditions, ester bonds at the 16-, 18-, 21-, and 22-position were broken. Saponin aglycones and sugars were mineralized under the actions of microbial enzymes and were eventually converted to CO₂ and water. The increase in saponin aglycone content evident on the 40th day of composting was likely caused by the decomposition of glycoside molecules to saponin aglycones.

Table 5. Chemical names of typical saponins in compost samples (of C. oleifera shell and seed cake) and their respective chemical formulas.

Relationships between tannin and saponin contents and nutrients

As shown in Figs 2 and 3, the TN, TP, TK, NO₃^--N, and total nutrient contents increased with, and were all positively correlated with, the duration of composting time (Table 6). When composting began, the TN content was 1.27 ± 0.02 g/kg, which increased to 12.71 ± 0.22 g/kg on the 20th day, 14.92 ± 0.03 g/kg on the 40th day, 18.47 ± 0.06 g/kg on the 60th day, and 22.15 ± 0.04 g/kg, and 22.17 ± 0.16 g/kg on the 76th day. This pronounced increase was likely due to the net loss of dry matter, specifically in terms of CO₂ and evaporative loss of water, as driven by the heat produced during the oxidation of organic matter [24]. The TN, TP, TK, and total nutrient contents were all negatively correlated with the tannin and saponin contents (Table 6). During the composting process, the TP and TK contents increased slightly (Fig 2), possibly because of the net loss of dry mass.

Fig 2. Relationships between the tannin and saponin contents and the TN, TP, TK, and total nutrient contents.

TN: total nitrogen, TP: total phosphorus, TK: total potassium.

Fig 3. Relationships between the tannin and saponin contents and NO3^--N and NH4⁺-N.

Because NO₃^--N is the main source of nitrogen available for uptake by most plants [25], for a given compost product the higher its NO₃^--N content, the higher will be its fertilizer efficiency. As Fig 3 shows, the NO₃^--N content changed little over the first 40 days of composting but it increased rapidly during days 60–76. Some of the NH₄⁺-N formed during microbial deamination was probably volatilized as ammonia gas, while some would have been converted to NO₃^--N by ammonia-oxidizing microbes [26]. In the early stage of composting, when both its pH and temperature were relatively high, ammonia volatilization caused the rapid reduction of NH₄⁺-N in the compost pile. By the end of the composting period, the NH₄⁺-N content had been reduced to less than one-third of its starting value (i.e., at day zero of the experiment), whereas the NO₃^--N content had risen to more than 40 times its initial value. The tannin content was negatively correlated with NO₃^--N yet positively so with NH₄⁺-N. The saponin content was also positively correlated with NH₄⁺-N, but negatively correlated with NO₃^--N (Table 6).

Relationships between tannin and saponin contents and maturity

Compost maturity is related to the safety and stability of the compost product, so it serves as an important indicator for evaluating the quality of the compost product. In this study, the C/N ratio, GI, and Solvita maturity index were relied upon to evaluate maturity, and their relationships with tannin and saponin contents were investigated. The initial C/N of the raw compost materials was 30, but it fell to less than 20 on the 60^th day of composting. On the 76^th day, the compost temperature was now close to room temperature, so the C/N remained relatively stable at 16.2 (Fig 4). At this time, T = 0.53, a result consistent with by Morel et al. [27] finding that the compost reached its maturity when T [(C/N)_final / (C/N)_initial] < 0.6. The GI refers to the toxicity of the compost product to plants; when GI > 100%, the product is considered to nontoxic. As Fig 5 shown, the compost product attained a nontoxic status (>100%) on the 40^th day. The Solvita maturity index is a widely accepted index easily obtained through simple testing. From the start of the composting, this index followed a general trend of continual increase, in that it reached values of 3, 5, 7, and 8, on the 20^th, 40^th, 60^th, and 76^th days of composting, respectively (Table 7). According to the “Guide to Solvita testing for compost maturity index” (Woods End Research, 2002), when the Solvita maturity index is greater than 6, the compost is considered to have matured. Therefore, we may infer that on the 60^th day of composting, the compost of C. oleifera shells and seed cake had reached its state of maturity. We also found the tannin and saponin contents positively correlated with C/N but negatively correlated with both GI and Solvita maturity index (Table 8). In view of the C/N, GI, and Solvita maturity index results, we are confident that the compost had matured on the 60^th day of composting, and that it is more appropriate to have tannin and saponin contents respectively of ≤ 1% and ≤ 2% in the compost product.

Fig 4. Relationships between tannin and saponin contents and the C/N ratio.

Fig 5. Relationships between tannin and saponin contents and the GI.

Table 7. Solvita maturity index during the composting of C. oleifera shell and seed cake.

Days	20	40	60	76
Solvita maturity index	3	5	7	8

Table 8. Correlations of tannin and saponin contents with the C/N ratio, GI, and Solvita maturity index during composting of C. oleifera shell and seed cake.

	Day	Tannin (%)	Saponin (%)	C/N	GI	Solvita maturity index
Day	1	–0.996**	–0.973**	–0.983**	0.968**	0.991**
Tannin (%)		1	0.967**	0.980**	–0.956*	–0.981**
Saponin (%)			1	0.989**	–0.896*	–0.992**
C/N				1	–0.932*	–0.995**
GI					1	0.944*
Solvita						1

* Correlation is significant at 0.05

** Correlation is significant at 0.01

Conclusion

Tannins and saponins of C. oleifera shell and seed cake can be degraded, and the content of NO₃-N can be increased by composting technology with 60 days. Thus, qualified compost products that are beneficial to plant growth can be obtained. In the experimental compost samples, the identified tannins mainly consisted of 11 phenolic acid compounds, but four of these were small-molecule phenolic acids relatively low in content. The saponins identified mainly consisted of five saponin aglycones of C. oleifera (named A, B, C, D, and E) and four of their derivatives. With the progress of composting, the content of tannins and saponins gradually decreased to a safe range. Because microbially secreted enzymes converted most large-molecule phenolic acid compounds to small-molecule compounds and their derivatives, and saponins decomposed to saponin aglycones. In addition, the tannin and saponin contents were inversely related to the TN, TP, TK, GI, Solvita maturity index and total nutrient contents, and the negative correlation with Solvita maturity index was significant. Yet the tannin and saponin contents were significantly positively correlated with C/N and content of NO₃-N.

Data Availability

All relevant data are within the paper and its Supporting Information files.

Funding Statement

This work was financially supported by the Provincial Department of Science and Technology of Zhejiang, China, Grant NO.2017C02022