**Xiaolong Huang** †**, Jingzhong Chen** †**, Jiming Liu \*, Jia Li, Mengyao Wu and Bingli Tong**

Department of Ecology, College of Forestry, Guizhou University, Guiyang 550025, China; guidah365@126.com (X.H.); chenjingzhong-2016@foxmail.com (J.C.); lijia22a@163.com (J.L.); mmmmengyao@126.com (M.W.); DCB0011@126.com (B.T.)

**\*** Correspondence: karst0623@163.com; Tel.: +86-13985015398

† Equal contribution to the study.

Received: 23 September 2019; Accepted: 16 October 2019; Published: 18 October 2019

**Abstract:** Autotoxicity is a widespread phenomenon in nature and is considered to be the main factor affecting new natural recruitment of plant populations, which was proven in many natural populations. *Cinnamomum migao* H. W. Li is an endemic medicinal woody plant species mainly distributed in Southwestern China and is defined as an endangered species by the Red Paper of Endangered Plants in China. The lack of seedlings is considered a key reason for population degeneration; however, no studies were conducted to explain its causes. *C. migao* contains substances with high allelopathic potential, such as terpenoids, phenolics, and flavonoids, and has strong allelopathic effects on other species. Therefore, we speculate that one of the reasons for *C. migao* seedling scarcity in the wild is that it exhibits autotoxic allelopathy. In this study, which was performed from the perspective of autotoxicity, we collected leaves, pericarp, seeds, and branches of the same population; we simulated the effects of decomposition and release of litter from these different anatomical parts of *C. migao* in the field; and we conducted 210-day control experiments on seedling growth, with different concentration gradients, using associated aqueous extracts. The results showed that the leaf aqueous extract (leafAE) significantly inhibited growth indicators and increased damage of the lipid structure of the cell membrane of seedlings, suggesting that autotoxicity from *C. migao* is a factor restraining seedling growth. The results of the analyses of soil properties showed that, compared with the other treatments, leafAE treatment inhibited soil enzyme activity and also had an impact on soil fungi. Although leafAE could promote soil fertility to some extent, it did not change the effect of autotoxic substances on seedling growth. We conclude that autotoxicity is the main obstacle inhibiting seedling growth and the factor restraining the natural regeneration of *C. migao*.

**Keywords:** *Cinnamomum migao*; autotoxicity; seedling growth; soil substrate; soil enzyme; soil fungi

## **1. Introduction**

Autotoxicity is a special type of intraspecific allelopathy of plants [1]. It is known to be widespread in nature, particularly in artificial agroforestry systems [2], leading to population deterioration and regeneration failure [3]. Autotoxicity refers to the phenomenon in which plants release their metabolites into the surrounding environment by volatilization, rainwater leaching, decomposition, and root excretion [4]; these metabolites then inhibit seed germination or other individuals' or their own seedlings' growth directly or indirectly. Canopy trees' autotoxicity to their seedlings may play an important role in forest species replacement [5], and the existence of autotoxicity was proved in many natural populations [4,6]. Current studies on plant autotoxicity have shown that allelochemicals mainly inhibit plant growth in the following ways: (1) Plant growth is directly inhibited by affecting photosynthesis and altering plant cell membrane structures and plant defense systems [7,8]. (2) Plant growth is indirectly affected by inhibiting nutrient absorption or changing soil enzyme activity [9,10]. (3) The rhizosphere microecosystems are changed through the interaction between plant metabolites and fungi to ultimately affect seedling growth [11] (Figure 1a).

**Figure 1.** Process of plant autotoxicity (**a**) and our experimental design (**b**).

Autotoxicity is regarded as a negative effect on plant growth, and secondary metabolites produced by maternal plants could hinder the growth of their seedlings [12]. Phenolics and terpenoids released by woody plants play a key role in these interactions by influencing the structure and diversity of plant and soil communities [13–15]. Soil microorganisms can be directly affected by plant phenolics [16], but they can also use these phytochemicals as carbon sources, thereby modifying the chemical plant–plant interactions [17]. Phenolics from litter can inhibit the symbiosis between trees and fungi [18], which may have important consequences on the seedling development of trees. Similarly, phenolics released by plants may also play a key role in influencing the soil microbial community structure [19] and litter decomposition process [20]. For instance, autotoxicity of canopy trees on their own seedlings probably plays a role in forest species turnover along succession in Mediterranean forests [5]. Some autotoxic compounds released by asparagus (*Asparagus o*ffi*cinalis*) probably inhibit its own growth and can also be a reason for "asparagus decline" [21]. Phenolics released by the understory dwarf shrub *Empetrum hermaphroditum* impair the regeneration of the dominant tree *Pinus sylvestris* in boreal forests [22]. In addition, studies by Alías et al. [23] on the soil properties of the invasive rock-rose (*Cistus ladanifer*) population also showed that the compounds released by the plant itself were involved in autotoxicity and regeneration of the rock-rose population. *Cinnamomum migao* contains substances with high allelopathic potential, such as terpenoids, phenolics, and flavonoids [24]; generally, species with high allelopathic potential tend to have strong autotoxicity [25], and medicinal plants are more likely to have autotoxicity than other plants [14]. Therefore, we speculate that one of the reasons for seedling scarcity in the wild population of the species *C. migao* is that this species has autotoxic allelopathy [14].

*C. migao* H.W. Li is a species of Lauraceae evergreen medicinal woody plant, which is defined as an endangered species by the Red Paper of Endangered Plants in China. It is endemic to China and mainly distributed in Southwestern China. The fruits of *C. migao*, which are effective in treating gastrointestinal and cerebrovascular diseases, are used as traditional medicine in Miao in China [24]. In the last 30 years, researchers have found that the natural population size of *C. migao* is very small, as the majority of populations contain only two or three individuals, and the tree age is relatively high;

the natural regeneration has some obstacles [26]. Moreover, recent investigations of this resource have found that many natural populations have disappeared, and the survival and reproduction of this species are greatly threatened [27]. However, to the best of our knowledge, no research was performed to explain the underlying cause of this phenomenon. Recent studies on the autotoxicity of medicinal plants mainly focused on medicinal herbs, such as *Codonopsis pilosula*, etc. [28]. Few studies have been performed on medicinal woody plants, and the mechanism underlying this autotoxicity has remained unclear [12].

Our previous studies confirmed that *C. migao* has a strong allelopathic effect [29]. To explain the difficulties associated with new recruitment of *C. migao*, the aim of this study was to understand the responses of the plants to autotoxicity and the possible mechanism of autotoxicity that inhibits plant growth. Accordingly, we proposed the following hypotheses: (1) autotoxicity is the main factor affecting seedling growth in *C. migao*, and (2) the main mechanism of autotoxicity is that decomposition and release of autotoxic substances from the litter of *C. migao* can be achieved by altering the soil environment (chemical property, soil enzyme activity, and fungi) to inhibit seedling growth and survival. To test these hypotheses, we adopted a method of irrigating the aqueous extract and simulated the effects of litter decomposition and release from different anatomical parts of *C. migao* on seedling growth under field conditions from the new perspective of plant autotoxicity [30]. The changes in morphology, physiological metabolism, soil substrate, and fungi, which are the four key factors affecting plant growth, were determined during seedling growth.

#### **2. Material and Methods**

#### *2.1. Experimental Site*

Experiments were conducted at the College of Forestry, Guizhou University, Huaxi District, Guiyang (106◦42 E, 26◦34 N, 1020 m a.s.l.). The climate in this region is subtropical monsoon, with a mean annual temperature of 15.3 ◦C and mean annual precipitation of 1129 mm.

#### *2.2. Plant Materials*

Fresh mature fruits of *C. migao* were collected from Luodian County, Guizhou Province (25◦26 N, 106◦31 E) from October to November 2017. They were then transported to the laboratory to remove their flesh immediately, followed by rinsing with water. Before germination, we cleaned the seeds, sterilized them by immersion in 0.5% KMnO4 for 2 h, and then rinsed them in sterile water five times. Seed germination was performed in an artificial climate chamber (RXZ-1500, Ningbo Jiangnan Instrument Factory, Ningbo, China) with a germination box. After germination of the seeds, we transferred them to nutrient soil for planting in January 2018, and seedlings with identical growth were transplanted into nutrient bags for slow seedling treatment in mid-March. The pot-culture experiment began in April after the seedlings (16.25 ± 1.41 cmin height; number of leaves 5 ± 0.58) had been transplanted into plastic plots. The soil in our experiments was selected from the section loess of nongrowing plants and uniformly mixed with humus and perlite (loess: humus: perlite = 7:2:1); each pot contained approximately 2.5 kg of soil. All tested soils were sterilized under high pressure at 121◦C twice, with each cycle lasting for 1 h. The litter of *C. migao* used in this experiment was directly collected from the forest floor. In addition, surface soil (0–5 cm) from under the canopy of nine natural populations of *C. migao* in Guizhou, Yunnan, and Guangxi provinces was sampled for sequence analysis to determine fungal diversity (Figure 2).

**Figure 2.** Experimental sites and soil sample collection sites for *C. migao* in Southwest China.

#### *2.3. Experimental Design*

Initially, we investigated and statistically analyzed the litter under the canopy of *C. migao* and found that the main components of the litter were leaves, fruits, and branches. Therefore, the materials were divided into four parts: leaf, pericarp, seed, and branch. To simulate natural conditions as much as possible, the litter was collected from under a canopy from November to December 2017 in Luodian County and classified. To prepare aqueous extracts, we cleaned all of the test materials and cut them into pieces (1 cm2); gathered them into weights of 0, 1, 2, 3, and 4 g; and immersed them in 100 ml of deionized water. Solutions with concentrations of 0.00, 0.01, 0.02, 0.03, and 0.04 g·mL−<sup>1</sup> were prepared in the dark. In accordance with the above method of aqueous extraction, five types of aqueous extract treatments were set up, namely, the control and treatment with the leaf aqueous extract (leafAE), pericarp aqueous extract (pericarpAE), seed aqueous extract (seedAE), and branch aqueous extract (branchAE). Each treatment had nine replicates, giving a total of 4 × 5 × 9 = 180 pots in this experiment. During the experiment, the aqueous extract was co-irrigated with litter (leaf, pericarp, seed, and branch) every 10 days, to better simulate the decomposition and release the effects of litter, and the positions of the pots were randomly moved. To maintain soil moisture content, quantitative deionized water was added to each pot at different times to supplement water. Through field monitoring, we found that seedlings usually withered or died within 8 months. Therefore, the experiment was designed to determine the growth indices after 210 days of treatment (Figure 1b).

#### *2.4. Chemical Analyses*

#### 2.4.1. Analysis of Seedling Growth and Physiology

At the end of the experiment, we determined seedling growth and physiology; the determination methods were as follows. (1) Seedling height: The seedling height for each treatment was measured using Vernier calipers and a tape measure. (2) Leaf area: Three pots of seedlings were selected for each treatment, and the leaf area of the third mature leaf below the apex of the seedlings was measured using a portable leaf-area meter (LI-3000C; LI-COR, Lincoln, NE, USA). (3) Biomass: At least three seedlings from pots undergoing different treatments were carefully removed from the soil to maintain their integrity, after which residual soil and impurities were removed by flushing with running water. After cleaning, the seedlings were oven-dried twice at 65 ◦C to a constant weight and then weighed.

Fresh plant materials (0.2 g) were collected from each treatment and homogenized with 5 mL of buffer (with 1% PVP), by grinding on ice, and then centrifuged at 4 ◦C. After centrifugation, the supernatants were used for measuring the levels of malondialdehyde (MDA), soluble protein, peroxidase (POD), and superoxide dismutase (SOD). In brief, the MDA content was estimated using the thiobarbituric acid method reported by Hodges et al. [31], with minor modifications. The soluble protein content was measured using the Folin's phenol reagent method [32]. POD activity was measured following the guaiacol method, with minor modifications [33]. SOD activity was analyzed using the nitroblue tetrazolium chloride method by Lacan and Baccou [34], with minor modifications.

#### 2.4.2. Analysis of Soil Physicochemical Properties and Soil Enzyme Activity

At the end of the experiment, soil samples were collected from each pot immediately after the different treatments and divided into two parts: One was used for soil nutrient analysis, and the other was stored at 4◦C for soil enzyme and soil fungal analyses.

Soil samples from the different aqueous extract treatments were naturally air-dried; the samples were sieved using a 2 mm diameter mesh to determine their chemical properties. In brief, the soil total nitrogen (TN) content was analyzed using the Kjeldahl method with a Foss–Kjeltec analyzer [35]. The soil total phosphorus (TP) content was determined by the Mo–Sb colorimetric method after soil was digested with a mixed acid solution of H2SO4 and HCLO4 [36]. The soil total potassium (TK) content was measured using an alkali melting-flame photometer. Soil available N (AN) content was determined by the method used by Dorich and Nelson [37]. Soil available P (AP) content was obtained by NaHCO3 extraction and analyzed by the sodium bicarbonate–molybdenum resistance colorimetric method [38]. The soil available K (AK) content was measured by the method of ammonium acetate leaching–flame photometry [38]. Soil enzyme activity, including catalase, urease, phosphatase, and invertase activity, was estimated using a soil enzyme activity kit (Beijing Solebo Biotechnology Co., Ltd.), in accordance with the manufacturer's instructions. The initial soil chemical composition and enzyme activity are detailed in Table 1.


**Table 1.** Initial chemical composition and soil enzyme activity of the tested soil samples.
