The Dynamics of Allelochemicals and Phytotoxicity in Eisenia fetida during the Decomposition of Eucalyptus grandis Litter

Abstract

Allelopathy is an underlying and controversial mechanism for detrimental environmental effects in the management of Eucalyptus plantations. However, little attention has been paid to the dynamics of allelochemicals and phytotoxicity in soil fauna during litter decomposition. To explore the relationship between the dynamics of phytotoxicity and allelochemicals, a decomposition experiment was conducted using 4-year-old and 8-year-old Eucalyptus grandis litter (0, 10, 20, 30, and 45 days). The acute toxicity of Eisenia fetida was assessed, and a chemical analysis of the eucalyptus leaves was performed. Biochemical markers, including total protein, acetylcholinesterase (AChE) activity, and oxidative stress levels (SOD and MDA) were measured. A comet assay was used to determine DNA damage in E. fetida cells. The results showed that after 20–30 days of decomposition, E. grandis litter exhibited stronger phytotoxic effects on E. fetida in terms of growth and biochemical levels. After 20 days of decomposition, the weight and total protein content of E. fetida first decreased and then increased over time. SOD activity increased after 20 days but decreased after 30 days of decomposition before increasing again. MDA content increased after 20 days, then decreased or was stable. AChE activity was inhibited after 30 days of decomposition and then increased or stabilized with further decomposition. Soluble allelochemicals, such as betaine, chlorogenic acid, and isoquercitrin, significantly decreased or disappeared during the initial decomposition stage, but pipecolic acid significantly increased, along with newly emerging phenolic fractions that were present. More allelochemicals were released from 8-year-old litter than from 4-year-old E. grandis litter, resulting in consistently more severe phytotoxic responses and DNA damage in E. fetida. Scientific management measures, such as the appropriate removal of leaf litter in the early stages of decomposition, might help support greater biodiversity in E. grandis plantations.

1. Introduction

Eucalyptus species have been introduced in many countries because of their rapid growth, broad adaptability, and high productivity for use in wood, paper, and charcoal industries [1,2]. Over 5.46 million ha of Eucalyptus plantations have been established in southern China [3]. However, the extensive area of Eucalyptus plantations and frequent harvesting during a short rotation period has led to various environmental issues, such as soil degradation, reduced biodiversity, decreased productivity, and allelopathy [1,4,5]. Allelopathy is considered to be one of the key mechanisms contributing to the decreased biodiversity associated with the development of Eucalyptus plantations [6,7,8].

Allelopathy is the effect of chemicals released by plants or microbes on the growth of other organisms [9]. Allelopathic compounds can be actively released by plants into the environment through volatilization, leaching, root exudation, or passive production during litter or residue decomposition [10,11,12]. Litter decomposition is one of the most important allelochemical release pathways after root exudation. Allelochemicals have undergone several abiotic and biotic processes in the soil, such as sorption and polymerization by soil organic matter and clay minerals and chemical transformation and degradation, maintaining sufficient concentrations to affect plants and other biota [13,14]. The allelopathic effects of plant litter on germination and seedling growth have been studied extensively [15,16,17]. However, limited information is available on the toxic effects of allelochemicals released by plants on the soil biota. Allelochemicals entering soils can be directly toxic to soil faunal communities or indirectly toxic by changing their food resource patterns and feeding habitats by affecting the soil microbial community [12,18,19]. Allelochemicals can also exert a positive influence on the soil biota community since they can serve as a carbon for the soil microbial community. Macrofauna, such as earthworms, occupy a large amount of soil biomass and play a vital role in the turnover of soil organic matter, nutrient cycling, and plant growth [20,21]. Earthworms can facilitate litter fragmentation and aerate soils by burrowing, increasing water infiltration, transporting soil organic matter, and reducing soil erosion by bioturbation [22]. Earthworms are ideal bioindicators of soil quality because of their sensitivity to the environment and contaminants [23,24].

Leaf litter decomposition is a major ecosystem process that determines nutrient cycling, primary productivity, and soil fertility [25]. Several studies have focused on the effects of litter quality, diversity, and environment on decomposition rate and nutrient release [26,27,28]. Different litter components have different mechanisms of litter decomposition [27,28]. Most previous studies have focused on the effects of complex phenolic compounds, such as polyphenols and tannins, on their retarding effects at the late stage of litter decomposition [29,30,31]. However, allelochemicals mainly consist of simple phenolic compounds such as phenolic acids and flavones, which can be produced through litter decomposition [32]. Although the allelopathic effects of plant litter have been extensively studied, limited attention has been paid to the dynamics of allelochemicals and their kinetic phytotoxicities during decomposition. Existing studies indicate that most of the allelochemicals are released rapidly and are degraded during the initial stage of field decomposition [33,34,35,36]. The resistant and newly emerging fractions of allelochemicals produced by litter decomposition undergo several physical and biochemical processes that can either increase or decrease phytotoxicity on other biota [34]. Previous studies have indicated that the decaying litter of Eucalyptus inhibits the germination and growth of target plants at the early stages of decomposition; however, phytotoxicity declines as decomposition proceeds [36]. However, little information is available on the allelochemical dynamics and phytotoxicity of soil fauna during the decomposition of Eucalyptus litter.

Eucalyptus has been widely studied for its high allelochemicals content, including water-soluble phenolics and VOCs released from decomposing litter, such as benzoic acid, hydroxybenzoic acid, vanillic acid, and terpenoids [7,37]. Many previous studies have evaluated the allelopathic effects of Eucalyptus litter extracts on the germination and seedling growth of target plant species [16,17]. Few studies have been conducted to evaluate the phytotoxicity of Eucalyptus litter to soil fauna. In addition, most previous studies have been based on laboratory, and few studies have been conducted to survey the phytotoxicity of Eucalyptus litter during decomposition under field conditions. Eucalyptus grandis is one of the main introduced, fast-growing tree species used for afforestation in Southwest China and is usually managed with a short rotation period (5–7 years) [1]. However, large areas of E. grandis plantations have also resulted in adverse ecological effects, such as reduced biodiversity, soil degradation, and decreased productivity [4,10,38,39]. Our previous results at the study sites showed that plant and soil biodiversity in E. grandis remained stable or decreased over 4–5 years, followed by a significant increase with the age of the plantation. Consistently, higher levels of VOCs and phenolic allelochemicals were detected in younger E. grandis litter (approximately four years old). In addition, a critical increase in allelochemicals in E. grandis litter was observed at the eight-year-old mark [1,7]. In this study, four- and eight-year-old E. grandis plantations were selected to represent the critical time point for the increase in allelochemicals. Eisenia fetida, a common soil fauna species used in toxicological and standard tests (OECD), was selected as the target organism, as it has been found in E. grandis plantations at our study sites [1,40]. The purpose of this study was to address the dynamics of phytotoxicity during the decomposition of E. grandis litter by measuring the growth and biochemical responses (malondialdehyde [MDA], superoxide dismutase [SOD], acetylcholinesterase [AChE], and DNA damage) of E. fetida. In addition, the dynamics of the identified allelochemicals were evaluated during the decomposition of the eucalyptus litter. We hypothesized that (1) allelochemicals are rapidly released and degraded in the initial stages of field decomposition, influencing the phytotoxicity dynamics of litter extracts on E. fetida, and (2) more allelochemicals are released from 4-year-old E. grandis litter than from 8-year-old litter.

2. Materials and Methods

2.1. Study Site

This study was conducted at the Experimental Station of Sichuan Agricultural University, located in the western Sichuan Province (102°57′–103°04′ E, 29°55′–29°59′ N, 570–592 m altitude). The climate is subtropical, with an average annual rainfall and temperature of 1397 mm and 17.5 °C, respectively [1]. The soil is a ferralsol with an old alluvial yellow loam and a granular structure [41]. Large E. grandis plantations were established at various growth stages. We selected four- and eight-year-old E. grandis plantations; both of the two sites were larger than 5 hectares. The study sites were planted with crops before afforestation with E. grandis. Cropping system, intensity, and management were typical and similar in the study region. Weed control treatments were not applied after afforestation. Three sites in each age group were selected for sampling (

Table 1. Soil characteristics of two differently aged E. grandis plantations.

Stand	Soil Bulk Density	Soil pH	Soil Water Content	Soil Organic Matter	Soil Total N	Soil Available p
4a	1.33 ± 0.08	3.99 ± 0.03 *	29.38 ± 0.01 ***	15.90 ± 0.88	0.27 ± 0.03	2.54 ± 0.24
8a	1.22 ± 0.12	3.90 ± 0.03	28.99 ± 0.02	24.56 ± 0.58 ***	0.91 ± 0.06 ***	6.89 ± 0.26 ***

* indicates significant differences between soil properties depending on forest ages. *, p < 0.05; ***, p < 0.001; mean ± standard error (SE).

).

2.2. Biodiversity Investigation

Vegetation surveys were conducted within each plot. Three 5 × 5 m quadrats containing one 1 × 1 m sub-plot were established to quantify the woody and herbaceous plants and record the species name, number of plants, and coverage of shrubs and herbaceous plants, respectively. Three 2 × 2 m plots were randomly selected, and soil macrofauna in the litter layer and 0–5 cm and 5–10 cm soil layers were hand-sorted within a sampling area of 0.25 m² (50 × 50 cm). Three sampling points within each plot were selected, litter samples (10 × 10 cm) were collected, and soil samples from 0 to 5 cm and 5 to 10 cm were collected using a cylindrical sampler (5 cm in diameter, 10 cm in height). Tullgren funnels (24 h, 38 °C) were used to extract microarthropods. All the fauna specimens extracted were preserved in 75% ethanol. Individuals were counted and identified at the family level. Rhizosphere soil samples within 2 mm of the root surface of E. grandis were collected and stored at −80 °C for soil microbial high-throughput sequencing [40].

2.3. Leaf Litter and Soil Sampling

Intact and newly fallen leaves, recognizable by their color, were collected from two mature E. grandis plantations using litter traps (1 m × 1 m, 1 m above the ground) in June 2021. The E. grandis fallen leaves were air-dried and used in laboratory and field experiments. Fifty litter bags made of plastic with a 1.5 mm mesh, allowing access to soil microbes and fauna and containing 20 g of newly fallen air-dried litter, were placed on the soil surface in the plantations. The litter bags were recovered after 0, 10, 20, 30, and 45 days of E. grandis litter decomposition and air-dried.

Air-dried leaf litter was cut into small pieces (<5 mm). Twenty grams of the leaf litter was shaken for 24 h in 160 mL of distilled water at a mass: volume ratio of 1:8 at room temperature. The mixture was then centrifuged (Beckman Coulter, Inc., Brea, CA, USA), filtered, and diluted to concentrations of 10 mg mL⁻¹, 25 mg mL⁻¹, 75 mg mL⁻¹, and 125 mg mL⁻¹ based on the original solution. Sterilized distilled water was used as control. Natural soil samples from surface horizons (0–15 cm) were collected from abandoned croplands near E. grandis plantations. The soil samples were thoroughly mixed and sieved through a 2 mm mesh to measure soil physicochemical properties.

2.4. Earthworms

E. fetida was cultivated according to OECD guidelines [42]. E. fetida was identified according to the pictorial keys to the soil animals of China. Earthworms were reared in a container with a substrate composed of peat moss, manure, and CaCO₃ (the pH value was adjusted to 6–7). The earthworms were fed twice per month. Adult worms with a clitellum and similar weights (300–400 mg) were selected. The earthworms were removed from the culture, rinsed with water, placed on dampened filter paper, kept in the dark at 18 °C for depuration, and acclimated to the test soil before treatment.

2.5. Identification of Secondary Metabolites

Ten grams of air-dried pure E. grandis litter samples at the initial stage and two months after decomposition from each plantation were subjected to extraction with 95 mL of 80% methanol (Thermo Fisher Scientific, Waltham, MA, USA). Before ultrasonic extraction (Zhengzhou Shengyuan Instrument Co., Ltd., Zhengzhou, China) (30 kHz for 40 min), the solutions stored in 250 mL brown flasks were shaken (Changzhou Tianrui Instrument Co., Ltd., Suzhou, China) for 24 h for a more complete extraction. The suspension was then filtered through a 0.45 μm micro-porous membrane and stored at −70 °C until subsequent ultra-performance liquid chromatography–tandem mass spectrometry (UPLC-MS/MS) analysis [32].

Metabolites of all thawed litter samples were extracted using methanol at 4 °C. Extraction solution (180 μL) from each sample and 20 μL of 100 ppm internal standards (naringenin and baicalein) were added to a 2 mL centrifuge tube. Subsequently, 20 μL of each sample was mixed with a quality control (QC) sample, and the remaining samples were subjected to UPLC-MS detection. Liquid chromatography was performed using a Thermo Ultimate 3000 system (Thermo Fisher Scientific) equipped with an AXQUTTY UPLC^® HSS T3 column (150 × 2.1 mm, 1.8 μm; Waters, Milford, MA, USA). Mass spectrometry was performed using a Thermo Q Exactive Focus mass spectrometer (Thermo Fisher Scientific). High-energy collision-dissociation cells were used for secondary dissociation to acquire MS/MS information, and unnecessary information was removed by dynamic exclusion.

After converting the raw data into XCMS format with Proteowizard software (V3.0.8789), the identification, filtration, and alignment of peaks were processed using the XCMS package in R (V3.3.2). To compare different magnitudes within the data, we obtained a dimensionless unit, that is, the normalized intensity, by normalizing the peak areas. All data were processed for QC to obtain reliable and high-quality metabolomic data. The peaks were matched using Metlin (http://metlin.scripps.edu, accessed on 20 August 2021), MoNA (http://mona.fiehnlab.ucdavis.edu, accessed on 20 August 2021), and a proprietary database (BioNovoGene; Suzhou, China). The OPLS-DA model was used to determine the differential metabolites (DMs) between groups based on thresholds of variable importance in projection (VIP) ≥ 1 and p < 0.05 [1,32].

2.6. Acute Toxicity

Brown glass bottles (200 mL) filled with 200 g of air-dried and sieved abandoned cropland soils were prepared. The litter extracts (20 mL) were mixed into the soil samples and stirred using glass sticks, with a range of concentrations (CK and C1-C4) as follows: 0.00, 1.00, 2.50, 7.50, and 12.50 g kg⁻¹ sets for the acute toxicity tests. In total, the toxicity test involved five litter decomposition times × two forest ages × five test concentrations × three replicates = 150 bottles. Ten healthy E. fetida individuals were randomly placed into each bottle. These bottles were sealed with sterile, breathable sealing films. The bottles were then placed in a cabinet (Zhengzhou Shengyuan Instrument Co., Ltd.) maintained at 20 °C, with 1000 illumination and a 12/12 h photoperiod. Water was added to the samples, and the soil water content was adjusted to 30% of the water-holding capacity. Soil samples dried at 105 °C for 24 h were used to determine the soil water content. The water-holding capacity was determined as described in the ISO guidelines [43]. Water was added to ensure constant substrate water content by weighing the containers.

The mortality of the earthworms was determined after 7 and 14 days of exposure. Earthworms were sorted by hand and considered dead if they did not respond to a gentle mechanical stimulation. Living earthworms were rinsed with distilled water, dried on a filter paper, and weighed.

2.7. Physiological Properties

Two live E. fetida from each replicate of each treatment were randomly selected for the measurement of physiological properties. Normal saline (0.9%) (determination of total protein, MDA content, and SOD activity) or extracting solution (determination of AChE activity) was added, as well as the remains of the earthworms (to 1:9 weight-to-volume ratio). The mixture was then homogenized under ice-cold conditions and centrifuged at 8000 rpm for 15 min at 4 °C [44]. The supernatant was collected and stored to evaluate enzymatic activity and protein content. Physiological properties were determined using test kits following the manufacturer’s instructions. Test kits for total protein content, superoxide dismutase (T-SOD) activity, and MDA content were purchased from Nanjing Jiancheng Bioengineering Institute (Nanjing, China). Test kits for AChE) activity were purchased from Beijing Solarbio Science & Technology Co., Ltd. (Beijing, China).

Total protein content was determined using the bicinchoninic acid (BCA) method [45]. Under alkaline conditions, the protein reduces Cu²⁺ to Cu⁺, forming purple complexes with BCA. Supernatants and various reagents were pipetted into 2 mL centrifuge tubes in an orderly fashion, swirled well, and allowed to stand for 5 min. Subsequently, 200 μL was pipetted from the tubes into wells of a 96-well transparent plate. The absorbance of the microplate was measured at 562 nm (AU_562nm) using a microplate reader [46].

SOD activity was determined using the xanthine oxidase–hydroxylamine method; O₂⁻ was produced by the xanthine and xanthine oxidase reaction, which oxidized hydroxylamine to form nitrite and presented purple red under the action of a chromogenic agent. Color was read at 550 nm using a microplate reader [47]. MDA reacts with thiobarbituric acid at 95 °C for 80 min to form a red product, and the color was read using a spectrophotometer at 532 nm [48]. AChE catalyzes the hydrolysis of Ach to produce choline, which reacts with 5,5’-dithiobis-2-nitrobenzoic acid (DTNB) to produce TNB, and the absorbance was then measured at 412 nm using a microplate reader [49].

2.8. Comet Assay

Based on the following results, the effects of litter extracts on the growth and physiological properties of earthworms were stronger after 30 days of litter decomposition. Earthworms were cultivated in brown bottles treated with E. grandis litter extracts for 30 days of decomposition. After 14 days, earthworm coelomocytes were extracted. Three E. fetida individuals were selected from each bottle, rinsed with distilled water, and soaked in chilled extrusion medium (5% ethanol, 95% saline, 2.5 mg mL⁻¹ Na₂ EDTA, and 10 mg mL⁻¹ guaiacol glyceryl ether, pH 7.3) for 3 min [50]. The obtained coelomocyte cells were released, washed three times in phosphate-buffered saline, and centrifuged at 4 °C for 5 min. The final cell concentration was adjusted to approximately 10⁶ cells mL⁻¹ with phosphate-buffered saline.

The cell suspension was mixed with melted comet agarose (37 °C) at 1:10 (volume ratio) in an EP tube. The mixture (20 μL) was added to the wells of a 6-Well Comet Slide, covered with a coverslip, and incubated at 4 °C for 15 min under yellow light to mitigate UV-induced DNA damage. Alkaline electrophoresis was conducted in a buffer of 300 mA NaOH and 1mM Na₂EDTA (pH > 13) for 30 min at 25 V and 300 mA, using a horizontal electrophoresis apparatus. After washing twice with distilled water for 5 min each, the slides were dehydrated in ethanol (10 min) and dried. Then, 40 μL of dyeing solution was added to each agarose spot and stained in a humidified dark box for 10 min at room temperature. A fluorescence microscope (BX41 Olympus, Tokyo, Japan) was used for observation and capture of comet images. The captured images were analyzed using Comet A 1.0 software (Cell Biolabs, San Diego, CA, USA). The obtained data of tail DNA (%), tail length, tail moment, and OTM data were utilized for the assessment of DNA damage in earthworms. OTM refers to the tail moment (distance between the center of mass of the tail and that of the head) multiplied by the percentage of DNA within the tail. It is also the most sensitive parameter reflecting the quality and quantity of DNA damage [14].

2.9. Statistical Analysis

The survival rate of E. fetida was calculated as follows:

SR = (S₀ − D_t)/S₀ × 100%

where S₀ is the initial number of earthworms, and D_t is the number of dead earthworms after t days.

The rate of weight inhibition of E. fetida was calculated as follows:

IR = (W₀ − W_t)/W₀ × 100%

where W₀ is the initial weight of earthworms and W_t is the weight of earthworms after t days [44].

The diversity indices were calculated as follows:

$$\mathrm{Shannon}–\mathrm{Wiener} \mathrm{diversity} \mathrm{index} (H)=-{\sum }_{i=1}^n{P}_i\ln P_i$$

$$\mathrm{Pielou} \mathrm{evenness} \mathrm{index} \left(E\right)=H/\ln S$$

Berger–Parker index (D_B-P) = N_max − N

$$\mathrm{Simpson} \mathrm{index} \left(D\right)=1-\sum P_i^2$$

where N_i is the number of individuals in the i-th group, N is the number of individuals in all groups in the community, N_max is the number of individuals of the species with the highest abundance in the sample unit, P_i = N_i/N, and S is the number of groups.

All statistical analyses were performed using the SPSS software (version 25.0, SPSS Inc., Chicago, IL, USA), PAST 4.02, and R 4.3.1. Graphics were plotted using Origin 2021 and R version 4.3.1. The exposure time was taken as a within-subjects variable, and the decomposition time, age, and dose were taken as between-subjects factors to conduct repeated measurement analysis of variance and interaction analysis for all indicators. One-way analysis of variance (ANOVA) and Tukey’s test were used to analyze the significance of the differences in growth, physiological properties, and DNA damage depending on litter decomposition time and concentration. The data normality and homogeneity of the variations were tested using Shapiro–Wilk and Levene’s test, respectively. Data sets failing the tests were Ln transformed or analyzed using non-parametric tests before further analysis to help satisfy the requirement of normality and variance homogeneity. An independent sample t-test was used to analyze the significance of the differences in the above-mentioned properties of earthworms depending on forest age and exposure time. Analysis of similarity based on Bray–Curtis distances was used to determine differences in the composition of metabolic compounds in litter in the two aged E. grandis plantation stands during litter decomposition. Non-metric multidimensional scaling (NMDS) ordination was used to plot differences. RDA analysis was used to determine the correlations between the main differential potential allelochemicals in litter and understory environmental factors in the two aged plantations, such as pH, water content, organic matter, and N and P contents, and biodiversity was determined before and after two months of litter decomposition. RDA analysis was also used to determine the correlations between the growth and physiological properties of E. fetida and the main differential potential allelochemicals in litter.

3. Results

3.1. Soil Properties and Plant and Soil Biodiversity

The soil bulk density did not differ significantly between the two mature plantations. Soil water content and soil pH were higher in the 4-year-old stands than in the 8-year-old stands. The soil organic matter, total N, and available P contents were higher in the 8-year-old stands than in the 4-year-old stands (Table 1). Regarding woody species, the species number, density, and Shannon–Wiener index were higher in 8-year-old stands than in 4-year-old stands, whereas the Berger–Parker indices showed the opposite trend. For macrofauna and microfauna, the Shannon–Wiener index was higher in 8-year-old stands than in 4-year-old stands; however, the Berger–Parker indices of macrofauna exhibited the opposite trend (

Table 2. Plant, soil microbial, and soil fauna diversity in the two different aged E. grandis plantations.

		Species Number	Density (Ind m−2)	Shannon– Wiener	Pielou	Berger–Parker
Plant’s diversity	Shrub (4a)	3.00 ± 0.87	0.51 ± 0.18	0.75 ± 0.14	0.76 ± 0.08	0.68 ± 0.03 ***
	Shrub (8a)	6.50 ± 0.43 **	1.11 ± 0.18 *	1.65 ± 0.05 ***	0.82 ± 0.02	0.35 ± 0.00
	Herbaceous (4a)	7.50 ± 2.18	48.00 ± 18.99	1.57 ± 0.21	0.73 ± 0.06	0.40 ± 0.03
	Herbaceous (8a)	7.75 ± 1.30	54.25 ± 4.63	1.64 ± 0.20	0.79 ± 0.07	0.39 ± 0.03
Soil faunal diversity	Macrofauna (4a)	6.42 ± 0.14	54.00 ± 11.27	1.58 ± 0.09	0.80 ± 0.05	0.40 ± 0.05 *
	Macrofauna (8a)	7.50 ± 1.30	62.33 ± 19.73	1.79 ± 0.09 *	0.86 ± 0.07	0.31 ± 0.02
	Microfauna (4a)	9.08 ± 0.95	8441.67 ± 1203.21	1.66 ± 0.02	0.64 ± 0.07	0.40 ± 0.02
	Microfauna (8a)	10.67 ± 0.14	12141.67 ± 2546.61	1.79 ± 0.07 *	0.59 ± 0.05	0.37 ± 0.04
		OTUs	Shannon–Wiener	Simpson	ACE	Chao1
Soil microbial diversity	(4a)	1623.08 ± 20.12	5.94 ± 0.07	0.99 ± 0.00	2474.72 ± 35.69	2312.38 ± 16.72
	(8a)	1537.58 ± 91.82	5.72 ± 0.17	0.98 ± 0.01	2398.02 ± 127.48	2208.14 ± 145.15

* indicates significant differences between soil properties depending on forest ages. *, p < 0.05; **, p < 0.01; ***, p < 0.001; mean ± standard error (SE).

).

3.2. Secondary Metabolites during E. grandis Litter Decomposition

A total of 328 and 381 primary and secondary metabolites, respectively, were identified in the litter of the two mature E. grandis plantations before and after two months of litter decomposition. The identified metabolites mainly included low-molecular-weight primary compounds, such as carbohydrates, amino acids, peptides, nucleotides and purines, cofactors, and vitamins. In addition, a wide variety of secondary metabolites, such as phenolic acids, flavones, terpenoids, alkaloids, and non-protein amino acids, were observed in E. grandis litter (Figure 1). The composition of metabolites in the litter varied between the two E. grandis plantations during litter decomposition (Figure 2).

A total of 154 and 120 DMs were identified in the litter of the two mature E. grandis plantations during the litter decomposition, respectively. A total of 301 and 310 DMs were detected before and after two months of decomposition in the 4- and 8-year-old stands, respectively. Notably, the levels of amino acids and peptides were higher in the litter of 4-year-old E. grandis than in 8-year-old E. grandis, a pattern also observed for lipids after two months of decomposition. Conversely, other primary metabolites were more abundant in the litter of 8-year-old E. grandis than in the 4-year-old E. grandis. Among the secondary metabolites, higher levels of terpenoids and amines were observed in the initial litter of the 4-year-old than in the 8-year-old E. grandis. Other potential allelochemicals, such as phenolic acids, flavonoids, alkaloids, and non-protein amino acids, were more abundant in 8-year-old stands than in 4-year-old stands. A similar trend was observed for amines in the litter after two months of decomposition. Notably, no significant variation in terpenoid levels was observed between the two mature E. grandis after two months of decomposition. Except for lipids in the 4-year-old E. grandis and amines in the 8-year-old E. grandis, there were no significant differences during decomposition; nucleotides, alkaloids, and non-protein amino acids exhibited a significant increase after two months of litter decomposition, whereas the other primary and secondary metabolites decreased significantly after litter decomposition for the same months (Figure 3). SIMPER analysis revealed that betaine, pipecolic acid, phenol, gamma-aminobutyric acid, and the other 13 potential compounds contributed to the variation in allelochemical composition during E. grandis litter decomposition and between the two mature plantations (

Table 3. Results of analysis of similarity (ANOSIM) for differential potential allelochemicals of E. grandis between forest ages and decomposition times and analysis of similarity percentage (SIMPER).

Effects	R and Significance	Main Contributive Compounds for Variation
Two-way ANOSIM
Forest age	(1) ***	Betaine, pipecolic acid, phenol, gamma-aminobutyric acid, catechin, chlorogenic acid, eucalyptol, isoquercitrin, kaempferol, pulegone, sinapic acid, sinapoyl aldehyde, and sphinganine
Decomposition time	(1) ***	Gamma-aminobutyric acid, phenol, pipecolic acid, betaine, catechin, chlorogenic acid, eucalyptol, isoquercitrin, kaempferol, sinapoyl aldehyde, pulegone, sinapic acid, sphinganine, and phytosphingosine
One-way ANOSIM of forest age effects
Initial	(1) *	Betaine, methyleugenol, catechin, isoquercitrin, pulegone, sinapic acid, Sphinganine, 4-hydroxycinnamic acid, eucalyptol, Taraxerol, chlorogenic acid, sinapoyl aldehyde, and pipecolic acid
After 2 months	(1) *	Betaine, pipecolic acid, phenol, gamma-aminobutyric acid, and saccharopine
One-way ANOSIM of decomposition time effects
4 years	(1) *	Betaine, 2-aminophenol, Taraxerol, gamma-aminobutyric acid, phenol, pipecolic acid, chlorogenic acid, eucalyptol, catechin, isoquercitrin, and pulegone
8 years	(1) *	Betaine, 2-aminophenol, phenol, gamma-aminobutyric acid, Taraxerol, catechin, pipecolic acid, chlorogenic acid, isoquercitrin, and eucalyptol

Forest ages: 4 and 8 years old. Data in parentheses are the R value. Significant effects are indicated in superscript: *, p < 0.05; ***, p < 0.001. The rightmost column is the contributive compounds (SIMPER analysis, contributive rate ≧ 2%) to the variation in differential potential allelochemicals composition among study sites or decomposition times.

).

The levels of some key differentially contributing metabolites in E. grandis litter during decomposition, such as betaine, chlorogenic acid, isoquercitrin, and kaempferol, generally decreased during decomposition. However, the pipecolic acid content increased after two months of decomposition. Catechin, eucalyptol, pulegone, sinapic acid, sinapoyl aldehyde, and sphinganine disappeared as the decomposition progressed, whereas phenol, phytosphingosine, and gamma-aminobutyric acid appeared as the decomposition process (Figure 4).

3.3. Effects of Leaf Litter Extract on Earthworms

3.3.1. Survival Rate and Growth of E. fetida

Repeated ANOVA revealed that decomposition, exposure time, and dose significantly affected the survival and weight inhibition rates of E. fetida (

Table 4. Repeated measures ANOVA of the effects of forest age, decomposition time, exposure time, and dose on growth and physiological characteristics of E. fetida.

	Survival Rate		Weight Inhibition Rate		Total Protein Content		SOD Activity		MDA Content		AChE Activity
	F	p	F	p	F	p	F	p	F	p	F	p
Decomposition time	17.476	0.000 ***	5.498	0.000 ***	41.199	0.000 ***	69.155	0.000 ***	29.708	0.000 ***	43.625	0.000 ***
Age	0.129	0.720	0.088	0.767	5.130	0.026 *	10.673	0.001 ***	2.596	0.110	2.815	0.097
Dose	0.587	0.673	4.680	0.002 **	3.034	0.021 *	7.769	0.000 ***	2.708	0.034 *	6.349	0.000 ***
Exposure time	134.20	0.000 ***	420.00	0.000 ***	22.767	0.000 ***	2.395	0.125	94.713	0.000 ***	8.834	0.004 **
Age × Dose	1.863	0.123	2.372	0.057	0.816	0.518	2.345	0.060	0.532	0.712	1.100	0.361
Age × Decomposition time	0.662	0.620	2.455	0.051	0.905	0.464	0.865	0.488	1.065	0.378	3.667	0.008 **

* indicates significant differences. *, p < 0.05; **, p < 0.01; ***, p < 0.001.

). After 14 days of exposure to the 4-year-old litter treatment, the survival rate of E. fetida at C2 significantly increased after 10 days of decomposition and then remained stable over time (Table S1). The weight inhibition rate of E. fetida at C3 significantly increased after 20 days of decomposition and subsequently decreased over time. The weight inhibition rate of E. fetida decreased with increasing doses of litter extract under several treatments (Table S2).

3.3.2. Physiological Properties of E. fetida

The content of total protein, MDA, and activities of SOD and AChE changed significantly with decomposition time, forest age, dose, and exposure time. Significant interactive effects on AChE activity were observed (Table 4). Under 4-year-old litter treatments, after seven days at C1 and C3 and 14 days of exposure at C1, C3, and C4 doses, the total protein content of E. fetida decreased significantly after 10–20 days of decomposition but remained stable or increased after 30 days and then decreased with time. Under 8-year-old litter treatments and seven days of exposure at C1 and C2, the total protein content of E. fetida decreased after 10 days of decomposition and then remained stable over time. After 14 days of exposure to C4, the protein content was initially stable but decreased after 45 days of decomposition (Figure S1).

After 7 or 14 days of exposure at C1 and C4 under the two mature litter treatments, the SOD activity of E. fetida fluctuated. It increased after 10–20 days of decomposition, decreased after 30 days, and then increased again after 45 days of decomposition. However, after seven days of exposure at C2 and C4, the SOD activity of E. fetida increased after 10 days of decomposition and then remained stable over time (Figure S2). After 14 days of exposure, the MDA content of E. fetida stimulated under 4-year-old litter treatments at C1, C2, and C4 increased after 20 days but decreased after 30 days of decomposition and then increased with time. For the treatments of the litter of 8-year-old trees, the MDA content at C3 after seven days of exposure was stimulated after ten days of decomposition and then stabilized over time. After 14 days of exposure at C1 and C2, the MDA content of E. fetida fluctuated and was stimulated after 20 or 45 days of decomposition (Figure S3).

After seven days of exposure, the AChE activity of E. fetida under 4-year-old litter treatments at C1 and C2 decreased after ten days of decomposition and then stabilized over time. After 14 days of exposure at C1 and C4, AChE activity decreased after 20–30 days, then stabilized with decomposition time. Under the treatment of 8-year-old litter at C2 after seven days of exposure, AChE activity decreased after 10 days and increased after 20 days of decomposition. However, after 14 days of exposure to C1, AChE activity was initially stable and then decreased after 30 or 45 days of decomposition (Figure S4).

The effects of dose and forest age on E. fetida were observed in several treatments. After 20 days of decomposition, the weight inhibitory rate of E. fetida increased with increasing doses of litter extracts of 4-year-old E. grandis and reached a maximum at C3. SOD activity of earthworms treated with 8-year-old litter increased at C1 and then stabilized with increasing doses. The total protein content of E. fetida was higher in the litter extracts of 4-year-old trees than in 8-year-old trees. However, SOD activity was higher in the litter of 8-year-old E. grandis than in 4-year-old E. grandis litter.

3.4. Effect of E. grandis Leaf Litter Extract on DNA Damage in E. fetida

Based on the above results, the toxic effects of E. grandis on E. fetida were stronger after 20–30 days of litter decomposition. We selected E. grandis litter after 30 days of decomposition to evaluate DNA damage in E. fetida using a comet assay. After 14 days of exposure, the litter extracts of 8-year-old E. grandis caused DNA damage to E. fetida, and no significant DNA damage to E. fetida was observed after treatment with litter extracts of 4-year-old trees. Compared with the control, the tail DNA content, tail moment, and OTM decreased significantly at C1 and then increased with increasing doses. At the highest concentration (C4), tail DNA content, tail length, and OTM were significantly higher in the litter of 8-year-old E. grandis than in the 4-year-old E. grandis litter (

Table 5. DNA damage effects of E. grandis litter extracts from different aged stands evaluated via a comet assay on coelomocytes from E. fetida.

Dose	Tail DNA (%)		Tail Moment		Olive Tail Moment		Tail Length
	4a	8a	4a	8a	4a	8a	4a	8a
CK	1.51 ± 0.81	1.51 ± 0.81 (ab)	0.19 ± 0.12	0.19 ± 0.12 (ab)	0.46 ± 0.21	0.46 ± 0.21 (ab)	7.46 ± 2.31	7.46 ± 2.31
C1	1.63 ± 0.48	1.17 ± 0.29 (b)	0.34 ± 0.2	0.12 ± 0.03 (b)	0.50 ± 0.17	0.29 ± 0.07 (b)	6.28 ± 0.69	5.02 ± 0.49
C2	2.16 ± 0.87	1.52 ± 0.13 (ab)	0.52 ± 0.32	0.52 ± 0.18 (ab)	0.75 ± 0.33	0.56 ± 0.04 (ab)	9.04 ± 4.48	7.98 ± 1.44
C3	1.50 ± 0.54	2.84 ± 1.32 (ab)	0.24 ± 0.1	0.79 ± 0.56 (a)	0.46 ± 0.21	0.97 ± 0.55 (a)	6.85 ± 2.02	11.75 ± 6.19
C4	1.03 ± 0.24	3.20 ± 0.74 (a)(*)	0.12 ± 0.07	0.82 ± 0.58 (a)	0.27 ± 0.08	1.04 ± 0.27 (a)(*)	5.33 ± 1.34	11.70 ± 3.28 (*)

Lowercase letters in parentheses indicate significant differences among various concentrations; (*) indicates significant differences between two aged E. grandis stands; p < 0.05; mean ± standard error (SE).

).

3.5. Correlation Analysis

Environmental changes between the two aged E. grandis affected the potential allelochemicals released by the litter during litter decomposition. The RDA showed that the first ordinate axes represented changes in most environmental factors, except for the diversity of herb species. At the initial decomposition stage, the second ordinate axes were closely correlated with the diversity of herb species. The contents of alkaloids, non-protein amino acids, and flavonoids in the initial litter were positively correlated with soil available P, total N, organic matter, shrub, and soil faunal diversity in the two-aged E. grandis plantations. The contents of terpenoids and phenolic acids were positively correlated, but that of amines was negatively correlated with changes in herb diversity. After two months of litter decomposition, the content of alkaloids and non-protein amino acids was positively correlated with changes in soil organic matter, available P, total N, soil faunal diversity, and shrub diversity. The phenolic acid content in E. grandis litter was positively correlated with changes in herb diversity. The amine, terpenoid, and flavonoid contents were not significantly correlated with the environmental factors (Figure 5, Table S3).

RDA also indicated that the responses of growth and physiological properties of E. fetida were correlated with the release of potential allelochemicals after two months of decomposition. For the 4-year-old stands, the first ordinate axis represents the changes in potential allelochemicals, whereas the second ordinate axis represents the changes in amines during litter decomposition in 8-year-old stands. Under the treatments of 4-year-old litter, the responses of SOD, AChE, and MDA in E. fetida were closely related to the changes in amines, phenolic acids, flavonoids, and terpenoids during litter decomposition. For the litter treatments of 8-year-old E. grandis, the changes in amines were responsible for the changes in the total protein and weight inhibitory rate of E. fetida (Figure 6, Table S4).

4. Discussion

Earthworms are epidermal respiratory soil macrofauna that can be easily absorbed into the body through respiration or can cause toxic effects through direct skin contact [51]. In this study, under most treatments, the survival rate of earthworms did not change significantly. The results indicated that allelochemicals in E. grandis litter were not maintained at sufficient concentrations, and earthworms activated the surface defense system in response to stimuli to avoid death [52]. However, inhaled and absorbed toxic chemicals inhibited this increase in earthworm weight. The weight inhibition rate of E. fetida increased significantly when treated with litter extracts after 20 days of decomposition. Earthworms reduce their food intake while avoiding the intake of toxic substances that are involved in the inhibitory effects on the weight of E. fetida [53].

Earthworms also adapt to environmental stress by adjusting their physiological activities. In this study, the total protein content and activities of SOD and AChE decreased, but the MDA content of E. fetida increased after 20 or 30 days of decomposition. The results indicated that the phytotoxicity of the litter extracts of E. grandis on E. fetida was severe at the initial stage of approximately 20–30 days of decomposition, followed by weaker toxicity as decomposition progressed. These results were consistent with previous studies, indicating that the most inhibitory effects occurred in the early stage of decomposition and then weakened in phytotoxicity with decomposition [11,33]. The initial decomposition phase of litter consists of comminution and release of the soluble fraction, followed by the degradation of cellulose and hemicellulose and retarded decomposition caused by inhibitory compounds such as tannins and lignin at the late stage of decomposition [26,54,55]. Several studies have reported a rapid loss of soluble phenolics from the Eucalyptus litter within the first week or month [33,36]. The rapid degradation of phenolics might be attributed to chemical processes after leaf fall, such as higher concentrations of oxidizing enzymes, particularly phenolases [34]. Subsequently, the rapid loss of phenolics from the Eucalyptus litter could be ascribed to leaching and sugar and microbial degradation [34,36,56]. A delay has been demonstrated between nutrient release from decomposing organic matter and root colonization of decomposing litter, which could be related to the occurrence of phytotoxic “windows” during the early stages of decomposition, which inhibits the development of the soil microbial community [11,57]. Despite a rapid decrease in phenolic compounds within a few days of leaf fall, Puig et al. (2018) observed that lettuce germination was completely inhibited at every sampling time [36], which demonstrated that E. globulus litter showed a continuous release of phenolic and volatile compounds during the 30 days of decomposition. As the decomposition period was prolonged, the phytotoxin levels declined when allelochemical inactivation was not balanced by new production. Previous studies have shown that there is a more resistant phenolic fraction after the initial rapid disappearance of phenolics and volatile compounds [36,58]. The relationship between allelochemicals and biological features, such as soil fauna, should refer to the resistant fraction that might be leached into the soil [11,34]. These soluble phenolics and volatiles entering the soil might be directly or indirectly toxic to soil fauna by changing the soil microbial community.

A total of 328 and 381 primary and secondary metabolites, respectively, were identified in the litter of the two mature E. grandis, and their presence, quantities, and composition varied significantly, depending on individual compounds as well as the decomposition process. Except for lipids in 4-year-old litter and amine in the litter of 8-year-old trees, which were stable, and the levels of nucleotides, alkaloids, and non-protein amino acids, which increased significantly, the levels of other primary and secondary metabolites reduced significantly after litter decomposition for two months. The secondary metabolites identified, such as catechin, eucalyptol, sinapic acid, and epicatechin, were partly in agreement with the allelochemicals previously identified in Eucalyptus litter [32,36,58]. In contrast, a few compounds, such as betaine, aminobutyric acid, and isoquercitrin, were not detected in previously identified E. grandis litter. This can be explained by the fact that allelochemicals are subjected to various biotic and abiotic processes, reduce their concentrations, and induce chemical transformation into other compounds [59,60]. The quantities of some of the main differentially contributing metabolites in E. grandis litter, such as betaine, chlorogenic acid, isoquercitrin, and kaempferol, generally decrease during decomposition. However, the quantity of pipecolic acid increased after two months of decomposition. Catechin, eucalyptol, pulegone, sinapic acid, sinapoyl aldehyde, and sphinganine disappeared as the decomposition proceeded, whereas phenol, phytosphingosine, and gamma-aminobutyric acid appeared as the decomposition proceeded. Although metabolites from E. grandis litter have not been consistently identified at each specific time point of decomposition, we also observed that the dynamics of allelochemical release were closely aligned with the phytotoxicity of E. grandis litter throughout the decomposition process. The main allelochemicals, including amines, phenolic acids, flavonoids, and terpenoids, explain the variation in the phytotoxicity of E. fetida during decomposition. Despite the insolubility of volatile organic compounds, such as terpenoids, their lower solubility and retention in soil particles have also been reported, which partly explains the phytotoxicity of eucalyptus litter on E. fetida [14,36].

Biochemical responses of earthworms have been regarded as a warning system for environmental stress [61]. Proteins are important components of an organism and regulate various physio-metabolic processes [62]. In this study, the total protein content of E. fetida decreased significantly after 20 days of litter decomposition. After the initial fragmentation of the litter, the rapid release of potential allelochemicals promoted the secretion of protein enzymes from the cell wall and intestinal tract and altered membrane permeability, consuming total protein from E. fetida [63,64]. Subsequent release of the soluble C fraction and nutrients gradually reduced the stimulation of earthworms, provided food resources, and increased total protein [20]. SOD is an important enzyme in the antioxidant system that prevents lipid peroxidation [65]. The SOD activity of E. fetida was initially increased but decreased after 30 days of decomposition and then increased. The increase in SOD activity implies that earthworms suffered from oxidative stress, as reflected by the production of superoxide anion radicals to reduce the cellular stress response [66,67]. The MDA content was also stimulated after 20 days of decomposition and then stabilized or decreased with decomposition time. MDA is the final product of lipid peroxidation, caused by an increase in oxygen free radicals, reflecting the degree of lipid peroxidation and damage to the organism [68,69]. The AChE is a specific filament amino acid hydrolase that promotes neuronal development and regeneration [70]. AChE activity of E. fetida was initially stable but decreased after 30 days of decomposition and then stabilized or increased with decomposition. Inhibition of AChE activity and behavioral alterations stimulate cholinergic receptors, ultimately leading to uncoordinated movements and neuromuscular paralysis [71,72]. In this study, a lack of avoidance response was observed after treatment with litter after 30 days of decomposition. Therefore, leaf litter of E. grandis after decomposition for 20–30 days had a stronger effect on the physiological properties of E. fetida.

In addition, the dynamics of allelochemicals and phytotoxicity of E. grandis litter on E. fetida varied considerably depending on plantation forest age. In this study, the total protein content of E. fetida was higher in the litter of 4-year-old trees than in 8-year-old trees, whereas the SOD activity of earthworms showed the opposite trend. Natural antioxidant defenses can be overwhelmed when enzymes cannot scavenge free radicals from pollutants, causing serious cellular DNA damage [73,74]. Comet assays have been used to detect DNA damage in E. fetida exposed to E. grandis litter extract. Olive tail moment (OTM) is the most sensitive comet parameter, which reveals the severity and quantity of DNA damage [75,76]. In this study, the increment of tail DNA, comet tail length, and OTM indicated that litter extracts of 8-year-old E. grandis caused more DNA damage to E. fetida than 4-year-old trees. DNA damage in earthworms leads to compromised immunity, rendering E. fetida vulnerable [74]. In addition, the litter extracts of E. grandis showed more severe toxic effects on E. fetida with increasing doses [77]; however, DNA damage decreased at the lowest dose. Lower concentrations of allelochemicals can weaken phytotoxicity and enhance the growth and survival of the earthworm.

Previous studies at the study sites indicated that 4-year-old and 8-year-old trees were the critical turning points for the increment of potential allelochemicals in litter across a range of E. grandis plantations. In the present study, except for the higher amount of amino acids and peptides, terpenoids, and amines in the initial litter and lipids after 2-month decomposition, the other primary and secondary metabolites were in greater abundance in the litter of 8-year-old E. grandis than in 4-year-old E. grandis litter. In general, mature leaves produce more secondary metabolic substances than young leaves, whereas young leaves have more abundant amino and organic acids [78]. Given the higher growth rate of E. grandis at four years, intra- and inter-specific competition for soil resources was severe, and E. grandis would invest more resources in chemical defense mechanisms [7,39]. However, the rapid growth rate of the tree created a relatively open canopy before the rotation period (approximately four years), which facilitated the development of soil microbial communities and, to some extent, might promote the degradation of the accumulated production of allelochemicals. Previous studies at the study sites also observed a slight difference in the diversity of the soil microbial community between the two-aged stands; however, they were higher in 4-year-old stands than in 8-year-old stands [79]. External environmental factors could also explain the between-age differences in potential allelochemicals [12,80]. Soluble compounds may have been efficiently leached out of young E. grandis litter under the relatively open canopy of 4-year-old stands. The closed canopy and developed understory plant structure and diversity in 8-year-old E. grandis stands increased the interception of rainfall, and phenolics were less efficiently leached out and consequently persisted longer in the mature leaf litter. As the decomposition proceeded, more abundant terpenoids were released quickly in the young leaf litter but retained for longer periods in mature E. grandis litter [14,81]. In addition, pH can affect the solubility of allelochemicals with increasing toxicity under acidic conditions in 8-year-old stands [11,82]. All of these factors explain the severe phytotoxicity of litter extracts of 8-year-old E. grandis compared to 4-year-old E. grandis.

5. Conclusions

In conclusion, this study demonstrated the occurrence of litter phytotoxicity in E. fetida with clear, dynamic patterns during the initial decomposition stage. The litter of E. grandis, after decomposition for 20–30 days, exerted stronger phytotoxicity on the growth and biochemical characteristics of E. fetida, and the toxicity was weakened by decomposition. E. fetida individuals exhibited significant inhibition of weight and the activities of AChE under litter treatment after 20–30 days of decomposition, whereas the activities and MDA content were stimulated at this stage. The phytotoxicity dynamics of decaying litter on E. fetida might be involved in the resistance of phenolic and volatile fractions after the initial fast disappearance of allelochemicals. Compared with the litter of 4-year-old trees, E. fetida exhibited a more severe phytotoxicity response to protein content, stimulation of SOD activity, and DNA damage in the litter extracts of 8-year-old E. grandis. Consistently, an increased abundance of potential allelochemicals was observed in the litter of 8-year-old E. grandis. However, the expression of allelopathy in 8-year-old litter does not necessarily need to outcompete 4-year-old litter under natural conditions because of the developed stand structure, environmental heterogeneity, and abundant precipitation in this region. Overall, our results suggest that the use of scientific management measures, such as the appropriate removal of leaf litter in the early stages of decomposition, may help support greater biodiversity in E. grandis plantations.