2.4.3. Analysis of Soil Fungi

Surface soils (0.04 g·mL−1) from the four aqueous extract treatments and the control from our experiment and surface soils of 0–5 cmfrom nine areas in which *C. migao* is distributed were used to analyze the diversity of soil fungi composition. Each treatment and field sample had three biological replicates. Soil genomic DNA was extracted using an extraction kit (FastDNA®tSPINKit for Soil, MP Biomedicals Co., Ltd., Shanghai, China). PCR amplification was performed using TaKaRa *rTaq* DNA polymerase in a 20 μL reaction system containing the following: 4 μL of 5× Buffer, 2 μL of 2.5 mmol/L dNTPs, 0.8 μL of forward primer ITS1F (5 -CTTGGTCATTTAGAGGAAGTAA-3 ) (5 μmol/L), 0.8 μL of reverse primer ITS2R (5 -GCTGCGTTCTTCATCGATGC-3 ) (5 μmol/L), 0.4 μL of rTaq polymerase, 0.2 μL of BSA, 10 ng of template DNA, and ddH2O to 20 μL. ABI GeneAmp 9700 was used for PCR. The PCR reaction parameters were as follows: 95 ◦C for 3 min; 35 cycles of 95 ◦C for 30 s, 55 ◦C for 30 s, and 72 ◦C for 45 s; 72 ◦C for 10 min; and, finally, 10 ◦C until stopping the reaction. The PCR products

were sequenced using an Illumina Hiseq high-throughput sequencing platform by Shanghai Ouyi Biomedical Information Co., Ltd.

#### *2.5. Statistical Analysis*

The homogeneity of variance was determined before performing ANOVA, and the data were logarithmically transformed when required. Duncan's test was used for assessing significant differences in all of the parameters using the SPSS 21.0 statistics package (Chicago, IL, USA). All presented data are indicated as the means and standard errors (SEs) of at least three replicates. The final results of fungal sequencing were collated, and SR statistics tools were used (http://www.mdts-erver.com/srst/) to analyze soil fungal data. Graphs were constructed with OriginPro 9.0 (Origin Lab, Northampton, MA, USA).

#### **3. Results**

#### *3.1. E*ff*ects of C. migao Litter Aqueous Extract on Seedling Growth*

The seedling height, biomass, and leaf area of *C. migao* were significantly decreased compared with those of the control when the concentration of leafAE was increased. At 0.04 g·mL<sup>−</sup>1, the seedling height, biomass, and leaf area decreased by 32.07%, 49.02%, and 42%, respectively. Within the experimental concentration range, pericarpAE, seedAE, and branchAE significantly promoted seedling growth. Among them, at 0.04 g·mL−<sup>1</sup> of pericarpAE, seedling height, biomass, and leaf area increased by 57.29%, 37.36%, and 50.88%, respectively. Similar increases with seedAE treatment of 44.89%, 25.62%, and 39.97% at 0.04 g·mL<sup>−</sup>1, respectively, and with branchAE treatment of 44.89%, 30.21%, and 47.31% at 0.04 g·mL<sup>−</sup>1, respectively, compared with those of the control were also identified (Table 2). The response of seedling growth to different litter aqueous extracts differed in the same concentration range.


**Table 2.** Effects of aqueous extract of litter from different anatomical parts of *C. migao* on seedling growth.

The indices of height, biomass, and leaf areas for *C. migao* seedlings after the application of leafAE, pericarpAE, seedAE, and branchAE treatments for 210 days. Values are expressed as mean ± SE. The values marked with different letters are significant at *p* < 0.05 (Duncan's test).

#### *3.2. E*ff*ects of C. migao Litter Aqueous Extract on the Antioxidant Systems of Seedlings*

Plants activate their own defense systems to resist toxicity caused by reactive oxygen species (ROS) when they are under stress, such as enhancing their antioxidant enzyme activities. Our results show that the MDA content in the leaves of *C. migao* seedlings increased as the concentration of leafAE increased (Figure 3a). Among the treatments, the MDA content was the highest when the concentration of leafAE was 0.04 g·mL−1, which was 3.3 times that of the control. In pericarpAE, seedAE, and branchAE treatments, the MDA content changed linearly and was higher than that of the control. This illustrates that different aqueous extracts could cause different degrees of peroxidation of the cell membrane lipid structure. The soluble protein content decreased as the concentration of the extract increased under different treatments. It is noteworthy that the soluble protein content of leafAE at 0.04 g·mL−<sup>1</sup> was the lowest, which was reduced by 55% compared with that in the control (Figure 3b). Moreover, the antioxidant enzyme activities of the seedlings were also affected by the concentration of aqueous extract. The activities of POD and SOD in the seedlings treated with pericarpAE, seedAE, and branchAE were higher than those of the control. This indicates that, under the synergistic action of POD and SOD, excessive free radicals in the cells could be effectively removed, thereby maintaining the structural stability of the cell membrane. Conversely, under a low concentration of leafAE (>0.01 g· mL<sup>−</sup>1), *C. migao* seedlings were placed under stress, which resulted in membrane lipid peroxidation, a sharp decrease in the soluble protein content, reductions in POD and SOD activity (Figure 3c,d), and significant inhibition of the antioxidant capacity of the seedlings.

**Figure 3.** The MDA (**a**), soluble protein (**b**), POD (**c**), and SOD (**d**) contents in *C. migao* seedlings after the application of litter aqueous extract treatments for 210 days. The *x* axis denotes the concentration of the aqueous extract and the *y* axis denotes the content. Values are expressed as mean ± SE. The values marked with different letters are significant at *p* < 0.05 (Duncan's test).

#### *3.3. E*ff*ects of C. migao Litter Aqueous Extracts on N, P, and K in Soils of Potted Seedlings*

The TN content of the soil was increased with increasing concentration of the aqueous extracts; the TN content increased by 69.68%, 42.26%, and 40.98% with leafAE, pericarpAE, and seedAE at 0.04 g·mL<sup>−</sup>1, respectively, compared with that of the control. The difference was the most significant when the concentration of leafAE was 0.04 g·mL−<sup>1</sup> (*p* < 0.05) (Figure 4a). The TP content in soil was higher than that in the control after all treatments. The most significant difference was found when the concentration of pericarpAE was 0.04 g·mL−<sup>1</sup> (*<sup>p</sup>* <sup>&</sup>lt; 0.05) (Figure 4b). The TK content after all treatments significantly increased (*p* < 0.05) compared with that of the control, which increased with increased

aqueous extract concentration (Figure 4c). The AN content showed an upward trend under leafAE and seedAE treatments, and the AN content significantly increased under seedAE (*p* < 0.05). The pericarpAE and branchAE treatments showed downward trends for the AN content, and this variable in soil was the lowest when branchAE was 0.01 g·mL−<sup>1</sup> (Figure 4d). The soil AP content showed an upward trend with the four kinds of aqueous extracts, with pericarpAE treatment causing the highest increase (*p* < 0.05) (Figure 4e). In addition, with increases of all aqueous extracts, the soil AK content significantly increased compared with that in the control group (*p* < 0.05) (Figure 4f).

**Figure 4.** The TN (**a**), TP (**b**), TK (**c**), AN (**d**), AP (**e**), and AK (**f**) contents in soil planted with *C. migao* seedlings after litter aqueous extract treatments for 210 days. The *x* axis denotes the concentration of the aqueous extract, and the *y* axis denotes the content. Values are expressed as mean ± SE. The values marked with different letters are significant at *p* < 0.05 (Duncan's test).

#### *3.4. E*ff*ects of C. Migao Litter Aqueous Extract on Soil Enzyme Activities of Potted Seedlings*

In the leafAE and branchAE treatments, the soil catalase activity was significantly decreased compared with that in the control (*p* < 0.05), with increasing extract concentration. The soil catalase activity in the pericarpAE and seedAE treatments first decreased and then increased (Figure 5a). This indicated that the leafAE and branchAE treatments had significant negative regulatory effects on soil catalase, whereas the pericarpAE and seedAE treatments showed a promoting effect. The soil urease activity showed an increasing trend in all the treatments, being significantly higher than that in the control (*p* < 0.05). All treatments promoted soil urease activity (Figure 5b). The soil phosphatase activity decreased with the increases in the concentrations of leafAE and pericarpAE; the value after the leafAE treatment was lower than that in the control, while that after the pericarpAE treatment was lower than that in the control at 0.02–0.04 g·mL<sup>−</sup>1. The seedAE and branchAE treatments could also promote soil phosphatase activity (Figure 5c). Moreover, under all the aqueous extract treatments, the soil sucrase activity was higher than that in the control, and the promoting effect of seedAE at 0.04 g·mL−<sup>1</sup> was the most pronounced (Figure 5d).

**Figure 5.** The catalase (**a**), urease (**b**), phosphatase (**c**), and invertase (**d**) contents in soil planted with *C. migao* seedlings after litter aqueous extract treatments for 210 days. The *x* axis denotes the concentration of the aqueous extract and the *y* axis denotes the content. Values are expressed as mean ± SE. The values marked with different letters are significant at *p* < 0.05 (Duncan's test).

#### *3.5. Composition of Fungi in Di*ff*erent Plots and Experimental Treatments*

Fungi in the surface soil of nine areas below *C. migao* canopy and upon the treatments with the highest concentration of the extracts were sequenced by high-throughput sequencing technology. As shown in Figure 6a, the common fungi (after the removal of unidentified fungi) in the surface soil below the canopy in the nine counties (Luodian, Ceheng, Libo, Zhenfeng, Wangmo, Napo, Funing, Tian'e, and Leye) were identified as *Archaeorhizomyces sp.*, *Chaetomium nigricolor*, *Chaetosphaeria vermicularioides*, *Chloridium sp.*, *Cladophialophora sp.*, *Cryptococcus podzolicus*, *Cylindrocladiella pseudoparva*, *Ilyonectria mors-panacis*, *Lophiostoma sp.*, *Microascaceae sp.*, *Mortierella biramosa*, *Myxocephala albida*, *Penicillifer martinii*, *Penicillium ochrochloron*, *Rozellomycota sp.*, *Trichoderma koningiopsis*, and *Trichosporon sporotrichoides*. Based on the relative abundance, the most abundant fungal species in Luodian, Ceheng, Libo, Zhenfeng, Wangmo, Funing, and Leye was *Cryptococcus podzolicus*, the most abundant fungal species in Napo was *Mortierella biramosa*, and the most abundant fungal species in Tian'e was *Penicillium ochrochloron*. Comparing the species' composition in terms of the top 30 species under the different treatments (Figure 6b), it could be found that the species overlapping between natural conditions and artificial controlled experimental conditions were *Cladophialophora* and *Lophiostoma*. The average abundance of *Cladophialophora* in each treatment group was in the order pericarpAE > leafAE > seedAE > control > branchAE, whereas that of *Lophiostoma* was branchAE > leafAE > pericarpAE > seedAE > control.

**Figure 6.** Relative abundance at the species level. Notes: (**a**) Relative abundance of the top 30 species under different treatments. (**b**) The common fungal species in nine areas.

The Adonis analytical method was used for analyzing the differences in the composition of fungi between the different treatment groups (Table 3). The results indicated that the composition of fungi significantly differed among the five treatments. It was also proven that the soil fungal community may have been changed by the extract treatment from different anatomical parts of *C. migao*.


**Table 3.** Composition of fungi among the four treatments by the Adonis analysis method.

*p* < 0.05 indicates that the feasibility of this test is high and that there are significant differences at an intergroup level.

Similarly, UPGMA sample similarity cluster analysis (BinaryJacCard distance) was performed on the three replicates of fungal communities (Figure 7). The results showed that the control, pericarpAE, and seedAE (1,2) treatments were grouped into a single branch, whereas the seedAE (3) and leafAE treatments were grouped together into another single branch. The clustering results showed that there was good similarity within the group, indicating that the change in the fungal composition among the different treatments was closely related to the type of litter.

#### **4. Discussion**

Obstacles to the natural regeneration of *C. migao* were always an important factor affecting its population continuity. Early studies also confirmed that *C. migao* has strong allelopathic effects on many plants [29]. Therefore, we hypothesized that *C. migao* is a medicinal plant with autotoxic allelopathy. One of the main methods to affect population regeneration is through the decomposition of litter and release of autotoxic compounds that alter the structure and properties of the ecosystems. Although our laboratory experiments have some limitations, the results indicate that the autotoxicity of *C. migao* is an important factor hindering its seedling growth and survival.

**Figure 7.** UPGMA sample similarity clustering results.

#### *4.1. Seedling Growth and Antioxidant System Responses to Aqueous Extracts from Di*ff*erent Anatomical Parts of C. migao*

Under autotoxicity, plants usually suffer a reduction in biomass, including inhibition of leaf, stem, and root growth [10]. When the concentration of autotoxic substances is low, it may promote plant growth, but if the concentration exceeds a threshold value, it will lead to unhealthy growth or other adverse effects [7]. Our study found that all growth indices of seedlings treated with leafAE were inhibited (Table 2). The height, biomass, and leaf area of seedlings were significantly decreased with an increase in the concentration of leafAE (*p* < 0.05), which may be the physio-morphological reaction to the autotoxic compounds contained in leaves on the seedlings [39]. This is similar to the findings in a study on biomass accumulation with the allelopathy of *Eucalyptus* [40]. Contrary to leafAE treatment, pericarpAE, seedAE, and branchAE treatments showed different degrees of promotion of seedling growth. The reason for this may be that the types of compounds are different and the levels of autotoxic substances in pericarps, seeds, and branches aqueous extracts at the same concentration are lower than that in leaves, which does not reach the threshold of seedlings' tolerance [7]. For instance, the *Pinus harcus* leave aqueous extract inhibited wheat growth, whereas aqueous extracts from other parts promoted it [41]. Meanwhile, preliminary tests were undertaken to clarify the mechanism of autotoxins in *C. migao* seedlings.

Plants may undergo secondary stress under autotoxicity, for example, oxidative stress, which results in the production of excessive ROS; causes accumulation of osmolytes, such as peroxide anion free radical O2 <sup>−</sup>, hydroxyl radical OH−, and hydrogen peroxide (H2O2) [42]; and leads to peroxidation of the membrane lipid structure in plant cells and, finally, MDA formation. The MDA content can be used as an important index for judging the degree of peroxidation damage in plant cell lipid membranes [43]. Our results showed that the content of MDA in leafAE was higher than that in other aqueous extracts (Figure 3a). Plants produce protective enzymes to cope with the damage caused by free radicals, including the cooperation of POD and SOD, which can effectively reduce and eliminate the damage caused by free radicals to cell membranes [44]. POD and SOD activities in seedlings were decreased with increasing applications of *C. migao* leafAE at the end of the experiments (Figure 3c,d). This indicated that the leafAE treatment had an inhibitory effect on the antioxidant system of seedlings; POD and SOD could not remove free radicals promptly, which made the seedlings

undergo membrane lipid peroxidation for a long time. However, the pericarpAE, seedAE, and branchAE treatments had no significant effect on seedling physiology (Table 2 and Figure 3), whereas the MDA content and antioxidant enzyme activity increased with increasing concentration, and the activity of antioxidant enzymes remained at a high level. This indicated that membrane lipid peroxidation occurs in seedlings after aqueous extract treatment [5] (Figure 3a). POD and SOD antioxidants can remove free radicals in seedlings over time, thereby preventing plant damage [9]. The increases in POD and SOD in our study also supported the notion that seedlings could cause biotic stress under autotoxicity. Similarly, soluble proteins are involved in the metabolism of various enzymes, the content of which could indirectly reflect the ability of plants to perform material synthesis and metabolism [45]. The decrease in the soluble protein content after leafAE treatment was higher than that after other aqueous extract treatments, indicating that the structure and function of membrane lipids in seedling leaves were damaged for a long time, and this might have led to the decreased activities of various enzymes involved in synthesis and metabolism. Studies have indicated that 30%–50% of the leaf soluble protein is Rubisco, which plays a decisive role in determining the plants' rate of photosynthesis [46]. Therefore, we considered that leafAE treatment affected the protective enzyme system and photosynthetic capacity of seedlings, which was consistent with the finding of inhibition of seedling growth after leafAE treatment at the end of the experiment. Although the effects of aqueous extracts from different anatomical parts of *C. migao* on the seedlings differed, the experimental results of leafAE treatment on the morphological and physiological inhibition of seedlings also validated our first hypothesis, i.e., autotoxicity is the main factor affecting seedling growth.

## *4.2. Soil Nutrition and Enzyme Activity Responses to Aqueous Extracts of Di*ff*erent Anatomical Parts of C. migao*

Soil nutrient elements and enzyme activities play an important role in plant growth. Generally, allelopathic (autotoxic) plant litter has a major impact on soil N, P, and K contents and soil enzymes [47]. Allelopathic substances could alter soil fertility and enzyme activity when they enter the soil, which indirectly leads to the occurrence of autotoxicity [48]. Soil enzymes are catalytically, biologically active substances secreted by fungi, plants, and living animals, which are released after the decomposition of animal and plant residues [49]. They are closely related to the transformation and utilization of soil N, P, and K [50]. Soil catalase mainly decomposes H2O2 produced by microorganisms to reduce toxicity; in the present study, the soil catalase content significantly decreased with an increase in the concentration of leafAE compared with that in the control (*p* < 0.05), indicating that leafAE treatment significantly inhibited the ability of soil catalase to scavenge H2O2, thereby indirectly increasing the toxicity to seedlings [51] (Figures 5a and 3a). Soil urease can catalyze the hydrolysis of urea in soil and improve the soil N supply capacity. Our study found that treatment with aqueous extracts from different anatomical parts of *C. migao* increased soil urease activity, which was similar to the changes in soil urease activity reported in studies of allelopathy on *Allium sativum* bulbs by *Astragalus mongholicus* root aqueous extracts [52]. Generally, invertase activity is positively correlated with soil urease and soil fertility. Compared with the control, the increase in soil fertility had positive effects on soil invertase and urease activities in our study, which illustrates this point [53]. In some autotoxic species, allelochemicals may cause deterioration of the soil substrate, thereby inhibiting the secretion and release of soil phosphatase by plant roots and soil microbial metabolic activities [11] and also affecting the conversion of soil organic P. Our results showed that the TP and AP contents upon different treatments were significantly higher than those in the control, whereas the soil phosphatase activity showed an opposite trend, with an increase in the concentration of the aqueous extract, and leafAE concentration was always lower than in the control at 0.01–0.04 g·mL−1. One potential reason for this is that the method of co-irrigation of extract and litter was adopted in this experiment, and the later decomposition of litter could supplement soil elements. Furthermore, it is possible that the aqueous extract has a negative effect on the metabolic activity of roots and the composition and activity of soil microorganisms, thus changing the intensity of secretion, release, and modification of

soil enzymes [11]. Although aqueous extract treatments supplemented the soil nutrients, they did not change the inhibitory effect of leafAE on *C. migao* seedling growth, which also confirmed that an improvement in soil fertility could not alleviate the negative effects of autotoxicity on plants [8]. Allelochemicals could affect soil physicochemical properties, microbial composition, and soil enzyme activity after entering the soil, thereby changing the growth of plants [54]. Our study found that litter aqueous extract caused certain changes in soil conditions compared with those in the control, particularly the leafAE treatment, which altered the soil conditions to indirectly inhibit seedling growth. Therefore, our second hypothesis was partially validated in this experiment. In addition, the inhibition of seedling growth by leafAE treatment indicates that there are indeed autotoxic substances in *C. migao* leaves, and the types and contents of these substances still need further study.

#### *4.3. Soil Fungal Component with Di*ff*erent Treatments and in Contrast to that in the Field*

Soil fungi are essential components of forest ecosystems and are considered to be key factors affecting plant growth. Many autotoxic substances bind in the form of acyl/glycosyl groups in plants. These toxic substances do not directly act on plants and do not exhibit autotoxicity, because of the lack of soil fungi participated [55]. However, when autotoxic compounds enter the soil in the form of litter, they lead to the degradation of acyl/glycosyl groups with the participation of fungi and conversion of inactive to active forms, which finally cause autotoxicity [56]. The soil fungal community is the main part of its microbial population and a decomposer of soil organic compounds [57]. It plays an important role in regulating soil enzyme activities [58] and is directly related to invertase, urease, and phosphatase activities [44]. In this study, soil fungal composition after different aqueous extract treatments were compared with that in the nine areas to explore the differences in fungal diversity in potted soils after different treatments. This could verify the possible effects of autotoxic substances on the soil after entering it. Moreover, the study could determine whether there are common fungi involved in the interaction of litter catabolism and soil substrate change during the process of autotoxicity. Our results showed that there were significant differences in the fungal composition of surface soil between different anatomical parts of aqueous extract treatments. Next, we performed comparisons with 30 common fungi found in the nine areas and the two fungi found in the pot experiments; both of these groups had a high abundance under leafAE treatment. This indicates that the different secondary metabolites of litter resulted in an interaction between fungi in soil, which resulted in the difference in the fungal composition after different treatments. It also indicates that allelochemicals could affect plant growth by changing soil fungal composition [59]. There are only two kinds of fungi in common between the pot experiment and the natural population. It may be that the pot experiment used high-temperature-sterilized soil, and the species and quantity of fungi carried by the litter itself are limited [13], which cannot fully reflect the fungal composition under natural conditions. In addition, the climate of the experimental site differs from the natural conditions; therefore, the main reason for these differences may be that the in situ environment is more suitable for related fungal growth [12]. From the perspective of revealing the relationship between fungi and autotoxicity, in situ experiments will be indispensable, which will be the focus of our work in the future.

## **5. Conclusions**

In summary, our results demonstrated that the leafAE treatment could inhibit seedling growth to different degrees, and the inhibitory effects mainly act via decreasing growth targets and damaging cell membrane lipid structures. Our results also revealed that soil enzyme activities were inhibited by leafAE, and the soil fungal composition after leafAE treatment was significantly different from that after the other treatments (including the control). Although the soil fertility was somewhat improved by leafAE treatment, it did not promote seedling growth, which indicated that the autotoxic substances contained in leaves are important factors affecting seedling growth. This is consistent with the long-term accumulation of leaf litter in the field and also supports the view that autotoxicity is an important obstacle to the natural regeneration of *C. migao*. From the perspective of species

protection, the litter under the canopy should be cleaned up in time to reduce accumulation of autotoxic substances in soils under canopy and conserve the natural population. Meanwhile, measures for ex situ conservation could be considered after artificial seedling cultivation in similar habitats.

**Author Contributions:** Experimental design, J.L. (Jiming Liu), X.H., and J.C.; field investigation, X.H., J.C., M.W., and B.T.; experiment and date analyzed, X.H., J.C., and J.L. (Jia Li); writing—review and editing, X.H., J.C., and J.L. (Jiming Liu).

**Funding:** This study was supported by the Guizhou Science and Technology Program "Source Screening and Mycorrhizal Seedling Breeding Technology of *Cinnamomum migao*" (Qiankehe Supporting [2019] 2774).

**Acknowledgments:** We thank all authors for their contributions to this study. We would like to thank Editage for the English-language revision.

**Conflicts of Interest:** The authors declare no conflicts of interest.
