Arbuscular mycorrhizal fungi improve the growth and drought tolerance of Cinnamomum migao by enhancing physio‐biochemical responses

Abstract Drought is the main limiting factor for plant growth in karst areas with a fragile ecological environment. Cinnamomum migao H.W. Li is an endemic medicinal woody plant present in the karst areas of southwestern China, and it is endangered due to poor drought tolerance. Arbuscular mycorrhizal fungi (AMF) are known to enhance the drought tolerance of plants. However, few studies have examined the contribution of AMF in improving the drought tolerance of C. migao seedlings. Therefore, we conducted a series of experiments to determine whether a single inoculation and coinoculation of AMF (Claroideoglomus lamellosum and Claroideoglomus etunicatum) enhanced the drought tolerance of C. migao. Furthermore, we compared the effects of single inoculation and coinoculation with different inoculum sizes (20, 40, 60, and 100 g; four replicates per treatment) on mycorrhizal colonization rate, plant growth, photosynthetic parameters, antioxidant enzyme activity, and malondialdehyde (MDA) and osmoregulatory substance contents. The results showed that compared with nonmycorrhizal plants, AMF colonization significantly improved plant growing status; net photosynthetic rate; superoxide dismutase, catalase, and peroxidase activities; and soluble sugar, soluble protein, and proline contents. Furthermore, AMF colonization increased relative water content and reduced MDA content in cells. These combined cumulative effects of AMF symbiosis ultimately enhanced the drought tolerance of seedlings and were closely related to the inoculum size. With an increase in inoculum size, the growth rate and drought tolerance of plants first increased and then decreased. The damage caused by drought stress could be reduced by inoculating 40–60 g of AMF, and the effect of coinoculation was significantly better than that of single inoculation at 60 g of AMF, while the effect was opposite at 40 g of AMF. Additionally, the interaction between AMF and inoculum sizes had a significant effect on drought tolerance. In conclusion, the inoculation of the AMF (Cl. lamellosum and Cl. etunicatum) improved photosynthesis, activated antioxidant enzymes, regulated cell osmotic state, and enhanced the drought tolerance of C. migao, enabling its growth in fragile ecological environments.


| INTRODUC TI ON
Extreme weather, which is caused by global climate change, seriously threatens the survival of several living species on the whole Earth. Drought, one of the most serious natural disasters, has become an important concern for governments and scientists over the world (Wang et al., 2021). It greatly affects plant growth and distribution, and the degree of damage it causes is closely related to its severity and duration (Liu et al., 2011). However, the severity and duration of drought are unpredictable due to many reasons, such as precipitation occurrence and distribution, evaporation, and soil water storage capacity (Farooq et al., 2009). Therefore, drought has severe impacts on forest ecosystems, particularly fragile ecological environments like karst rocky desertification areas.
Karst lands, which are fragile ecological environments, are characterized by slow soil formation, shallow and discontinuous soil, low water holding capacity, low vegetation coverage, high soil erosion, and even serious rock desertification (Wang, Zhang et al., 2019;Zhang et al., 2021). They account for approximately 10% of the world's land surface (Hartmann et al., 2014). Additionally, the precipitation in karst areas is unevenly distributed in time and space , causing regional and seasonal droughts . Altogether, the mechanism of drought in karst regions is complex and unpredictable , which makes it difficult to study drought in such regions (Huang et al., 2008;Jiang et al., 2009). However, recently, under the background of global warming, the degree of drought in karst areas has shown an increasing trend, making drought stress an important factor limiting the survival and growth of plants (Guo, 2012).
Furthermore, drought stress adversely affects the physiology, biochemistry, growth, and development of plants worldwide. It leads to the accumulation of reactive oxygen species (ROS) in plants, destroys cell membranes, and disrupts the dynamic balance of active oxygen content (Komivi et al., 2017). These physiological and biochemical responses of plants under drought stress cause growth inhibition and even death (Fathi & Tari, 2016;Khan et al., 2020).
Many studies have shown that the inoculation of mycorrhizal fungi initiates morphological, nutritional, and physiological changes in host plants to counter biotic and abiotic stresses and enhance plant growth and vigor (Ahmad et al., 2018;Brundrett & Tedersoo, 2018;Orians et al., 2017;Rajtor & Piotrowska-Seget, 2016;Tamayo-Velez & Osorio, 2017;Tran et al., 2019). Approximately 72% of the known vascular plants can act as hosts for arbuscular mycorrhizal fungi (AMF), and such mutually beneficial mycorrhizal associations have key roles in maintaining plant productivity in natural and agricultural habitats (Brundrett & Tedersoo, 2018;Huang et al., 2020;Smith & Read, 2008). AMF affect the vegetative roots of host plants that have not yet been lignified, penetrate and colonize the root to form highly differentiated symbiotic structures known as arbuscules, and form epitaxial mycelia and other fungal structures (Brundrett & Tedersoo, 2018;Genre et al., 2017;Harrison, 2005;Liu et al., 2006). This increases the root surface area (RSA), thereby improving the absorption range and capacity of water and nutrient elements (Guo et al., 2019;Li et al., 2018). Additionally, AMF can activate the defense mechanisms of plants against oxidative damage, induce increased antioxidant enzyme activity, clear harmful substances such as ROS and malondialdehyde (MDA) (Hou et al., 2018;Li et al., 2018), improve chlorophyll content, and promote photosynthesis (Sohrabi et al., 2012). Furthermore, AMF can adjust osmoregulatory substances and molecular signals and alter water metabolism (Golparyan et al., 2018;Lehto & Zwiazek, 2011;Sebastiana et al., 2018). These changes directly or indirectly enhance plant drought tolerance.
Considering the characteristics of drought and nutrient deficiency in degraded karst ecosystem (Liu et al., 2008), inoculating AMF enhances plant growth by improving drought tolerance (Sebastiana et al., 2018), gradually restoring the function of the fragile karst ecosystem (Jiang et al., 2009). Notably, although AMF are beneficial to the growth of symbiotic plants (Sanders, 2003), there is a certain limit to the inoculum size. Because AMF promote nutrient absorption and growth in host plants, they also need to obtain carbohydrates from host plants (Brundrett & Tedersoo, 2018). Thus, excessive AMF inoculation can make host plants lose a substantial amount of nutrients, thereby affecting plant growth. On the contrary, inoculum size below the optimal level does not induce good mycorrhizal infection . Generally, the optimal inoculum size is determined by the mycorrhizal colonization rate, plant growth, photosynthesis, and other physiological and biochemical indicators (Geng et al., 2016;Li et al., 2019;Wang, 2008;Xiong et al., 2009).
Cinnamomum migao H. W. Li, a species of the Lauraceae family, is one of the endemic evergreen woody plants in the karst areas of southwestern China. The bark and wood of C. migao can be used as raw materials for making paper and wood slab. The fruit of C. migao contains a large amount of volatile oil and monoterpene or sesquiterpene chemicals; it is a traditional medicinal material of the state, and enhanced the drought tolerance of C. migao, enabling its growth in fragile ecological environments.

K E Y W O R D S
antioxidant system, arbuscular mycorrhizal fungi, Cinnamomum migao, drought stress, osmotic adjustment, photosynthesis, plant growth

T A X O N O M Y C L A S S I F I C A T I O N
Agroecology; Applied ecology; Botany; Ecosystem ecology; Microbial ecology Miao people in China and is used to cure gastrointestinal and cerebrovascular diseases (Huang et al., 2019;Liao et al., 2021). However, researchers discovered that the allelopathic and autotoxic nature of C. migao kept its population size small and range of species distribution narrow. Moreover, the low seed viability and germination rate, weak seedling formation, slow natural reproduction rate, and poor drought tolerance resulted in a low dispersal ability (Chen, 2020; Figure 1). Many natural populations of C. migao have disappeared, greatly threatening the survival and reproduction of this species (Huang et al., 2019). Additionally, seasonal and geological droughts in the main distribution areas lead to poor growth in the seedling stage ( Figure 1), aggravating the survival crisis of C. migao in fragile karst areas (Chen, 2020;Cheng et al., 2018;Dai & Zhong, 2021).
To improve the drought tolerance of plants in karst forest ecosystems with frequent droughts, scientists have conducted substantial research on the mechanism of interaction between AMF and plants (Lehnert et al., 2018;Sanders, 2003;Sohrabi et al., 2012). However, to date, the influence of AMF on the growth and stress resistance of C. migao has not been studied, and the symbiotic effect is not clear.
We propose the hypotheses that the inoculation of AMF strains promotes the growth and drought tolerance of C. migao. Therefore, in this study, two strains of Glomus were inoculated individually and collectively to analyze any changes in the drought tolerance of C. migao in karst soil (calcareous soil). The effects of inoculating AMF, Claroideoglomus etunicatum and Claroideoglomus lamellosum, on C. migao seedlings in four inoculum sizes (20, 40, 60, and 100 g) were also evaluated. In particular, we studied the effect of inoculation on the physiological processes in C. migao by analyzing several physiological indices, such as colonization rate (CR); biomass (DW); growth indices; photosynthetic indices; chlorophyll relative content value (SPAD); relative water content (RWC); total soluble sugar (TSS), soluble protein (SP), proline (Pro), and MDA contents; and catalase (CAT), peroxidase (POD), and superoxide dismutase (SOD) activities. We believe that this study can help elucidate the mechanism underlying AMF-induced drought tolerance in C. migao and provide a scientific basis for AMF inoculation for artificial vegetation restoration in karst mountains. China) with controlled conditions of 25°C/20°C day/night temperature, >60% relative humidity, and a photoperiod of 14 h of artificial light (8000-10,000 Lux) and 10 h of darkness. The seeds were watered once per week with 60 ml of sterile Hoagland's nutrient solution (low phosphorus). After 128 days of culture, the roots of the T. repens seedlings were densely packed in the entire container, and the pot was dried for 1 week in a shady and ventilated room with stable temperature (20-25°C) and humidity (55-60%) after the shoots of the T. repens seedlings were removed. A 1-cm layer of substrate was removed, and all the cultures (including T. repens roots, hyphae, spores, and substrates) were poured onto clean copier papers on a clean table top (all wiped with 70% alcohol). The plant parts were subsequently chopped with a flame-sterilized hatchet. Finally, the chopped cultures were rolled up in paper and poured into sterile ziplock bags (wiped with 70% alcohol), which were sealed and stored in a refrigerator (4°C). These were used as fresh mycorrhizal inocula.

| Experimental site
Sterile gloves were worn throughout the processes of cultivation and propagation to ensure that the cultures were not handled with bare hands. The inocula of Cl. lamellosum contained 19 spores per gram, whereas those of Cl. etunicatum contained 21 spores per gram, as the spores were isolated by wet screening combined with sucrose centrifugation and determined by stereomicroscopy (Liu & Li, 2000).

| Nursery substrate material
The seedling culture medium was mixed with forest soil (screened using a 10-mesh sieve), river sand (with particle size of 0.7-1 mm), and perlite (with particle size of 2-4 mm) at a volume ratio of 5:1:1.
Subsequently, it was sterilized continuously for 2 h under 0.14 MPa pressure and 125°C temperature. The basic physical and chemical properties of the soil were as follows: pH 6.6, soil organic matter 220.9 g/kg, total phosphorus 2.6 g/kg, total nitrogen 7.96 g/kg, total potassium 18.4 g/kg, alkali-hydrolyzable nitrogen 31.4 mg/kg, available phosphorus (extracted with 0.5 mol/L NaHCO 3 ) 1.6 mg/kg, and available potassium (extracted with 1 mol/L NH 4 OAc) 568.0 mg/kg.
The culture container was soaked in 75% ethanol for 30 min and then rinsed with sterile water for later use.

| Plant materials
From October to November 2018, fresh fruits of C. migao were collected from Luodian County, Guizhou Province (25°260 N, 106°310 E, 761 M A.S.L.). Subsequently, they were transported to the laboratory, and their flesh was immediately removed and washed with water. Before germination, the seeds were cleaned, disinfected by soaking in 0.5% KMnO 4 for 2 h, and then washed five times with sterile water. The seeds were treated with 200 mg/L gibberellin solution for 48 h, sterilized by stirring with 5% sodium hypochlorite for 10 min, washed four times with sterile water, planted in sterilized river sand, and then transferred to an artificial climate chamber (RXZ-1500) with a germination box under the controlled conditions of 25°C/20°C day/night temperature, >40% relative humidity, and a 12/12-h light cycle comprising 12 h of light (8000-10,000 Lux) and 12 h of darkness; they were regularly irrigated with deionized water.
After the seeds germinated, the seedlings were moved out of the artificial climate chamber and grown for 30 days in a greenhouse.

| Greenhouse study
The pot-culture experiment was started in March 2019. The seedlings were grown under natural light in a plastic greenhouse (annual average temperature: 18.7°C, annual mean humidity: 60%, and annual average illumination: 1148.3 h) after the seedlings had been transplanted into sterilized plastic plots. Cinnamomum migao seedlings with relatively consistent growth were selected, rinsed with detergent, immersed in 1% NaClO solution for 2-3 min, and rinsed with sterile water. First, 2.5 kg of sterilized soil was added to the pot and flattened, following which the microbial inocula were evenly added.
Subsequently, a sterilized seedling was added, followed by the addition of 1 kg of sterilized soil to complete the planting. For a total of 64 pots, each pot had one seedling.
After 120 days of growth under normal water supply conditions, C. migao seedlings were subjected to drought stress for 30 days and then the physiological and biochemical parameters were measured. Soil moisture content was measured by a soil moisture meter (ECA-SW1) and maintained at 80-90% of the field capacity by weighing the pot and watering daily. Watering was discontinued after 120 days, and drought stress treatment was initiated for 30 days when the soil moisture content in the pots naturally reached 50%. The soil moisture content was measured and maintained at 50% daily.

| Plant harvest and analysis of physiological and biochemical parameters
Before harvest, shoot height (Sh) and stem diameter (Sd) were measured using a ruler and vernier calipers, and leaf area (La) of the fourth and fifth leaves were measured using a La meter (LI-3100, LI-COR, USA), a total of eight leaves were measured for each treatment. The SPAD value of the relative content of chlorophyll was measured using a portable SPAD-502 chlorophyll meter (Konica Minolta, Japan using a portable photosynthesis system (LI-6800, LI-COR). The measurements were performed at a light intensity of 1000 μmol (photon)/ m 2 /s from 9:00 to 11:00 a.m. The CO 2 concentration in the sample chamber was 400 μmol/mol with a flow rate of 500 μmol/s. The leaf temperature was 25 ± 0.8°C, and relative humidity was 60%. Wateruse efficiency (WUE) was calculated as the ratio between Pn and Tr.
Following this, the plants were harvested and divided into roots, stems, and leaves. Their fresh weights were subsequently measured using a scale. Roots of C. migao seedlings were subjected to a root image scanner (Epson7500, resolution: 600 dpi), and the root length (Rl), root surface area (RSA), and the number of connections (Ncon), tips (Ntip), furcation (Nf), and crossing (Ncro) of roots were analyzed using WinRHIZO Pro LA2400 root analysis system (Regent Instruments, QC, Canada). Some fresh samples were immediately put in liquid nitrogen and then stored in a low-temperature refrigerator (−80°C) for further use, whereas the remaining were dried in an oven to measure dry weight. Biomass was calculated in terms of the dry weight of the whole plant.
The mycorrhizal root CR was determined according to the method described by Phillips and Hayman (1970). Four plants were determined for each treatment. Briefly, 1-cm segments were cut from the middle part of the root (180 root segments for each treatment), washed with 10% (w/v) KOH at 90°C for 30 min, decolorized with 10% H 2 O 2 , and stained with 0.05% (w/v) trypan blue and 0.05% and MD ≥300%, indicating strong dependence on mycorrhiza.
The frozen plant leaves were taken out of liquid nitrogen, and the corresponding indices were determined according to the following methods: SOD activity was determined by measuring the amount of reduced nitroblue tetrazolium (Gao, 2006); CAT activity was determined by measuring the reduction in hydrogen peroxide absorbance at 240 nm using the ultraviolet absorption method (Becana et al., 1986;Britton & Machlly, 1956); POD activity was determined by tracking the alteration in POD activity at 470 nm for 5 min using the guaiacol method ; proline content was determined colorimetrically via the ninhydrin method previously described by ; TSS content was determined colorimetrically via the anthrone method (Gao, 2006); SP content was determined using the Coomassie Brilliant Blue G-250 method (Bradford, 1976); and MDA content was determined using the thiobarbituric acid method (Gao, 2006). Relative water content (RWC) was determined using the following formula (Gao, 2006): where FW, DW, and SFW represent fresh weight, dry weight, and saturated fresh weight, respectively. SFW was measured after the leaves were completely immersed in water for 24 h in darkness at 4°C.

| Statistical analysis
Data were tested for normality (Shapiro-Wilk test, p > .05) and homogeneity of variances. Values are expressed as mean ± SE (plot replicates, n = 4). The data were subjected to analysis of variance (ANOVA) and factor analysis using SPSS 21.0 statistical software package (Chicago, IL, USA), and differences between the mean values were compared using the least significant differences (LSD) post hoc test. p Values of <.05 were considered statistically significant. Correlation coefficients between variables were tested using Pearson's correlation. Graphs were created using OriginPro 9.0 (Origin Lab in Northampton, MA, USA), Photoshop (Adobe, USA), and R language software (R version 4.1.2).   Among the three AMF inoculation groups, the highest biomass was observed in Cl-S2 (10.20 ± 1.53 g), Ce-S3 (9.01 ± 1.44 g), and HA-S3

| Growth of C. migao seedlings
(10.21 ± 0.73 g) ( Table 2). The Sh, Sd, and La of all AM plants were greater than those of NM plants, except the Sh of the NM-S4 was greater than the that of Cl-S4 and Ce-S4, and the La of NM-S1 was greater than that of Cl-S1 ( Table 2). With an increase in the inoculum size, these parameters first increased and then decreased. For the underground parts, except the RSA of Cl decreased with increasing the inoculum size, Rl, RSA, and root dry weight (RDW) ( Table 2) as well as the Ncon, Ntip, Ncro, and Nf of the roots (Table 3) showed the same trend, indicating that within appropriate inoculum size, the inoculation of AMF promoted root elongation and bifurcation to form root networks, which was beneficial in terms of retaining soil moisture and absorbing nutrients.

| Photosynthetic parameters
Single

| Antioxidant enzyme activity
Superoxide dismutase activity in AM plants ranged from 153.9 ± 11.1 to 230.56 ± 11.9 U/g FW/h, which was significantly higher than that in NM plants by 13.9-74.5% (p < .05; Figure 4a). SOD activity in HA-S2 plants (230.6 ± 11.9 U/g FW/h) was the highest among AM plants, whereas that in Ce-S4 plants (153.9 ± 11.1 U/g FW/h) was the lowest

| MDA content
As shown in Figure 4d

| RWC and TSS, SP, and Pro contents
Under drought stress, TSS, SP, Pro, and RWC contents in C. migao seedlings increased to varying degrees depending on the differences in AMF and inoculum sizes. The contents of the four osmoregulatory substances first increased and then decreased with an increase in the inoculum size (Figure 5a-d).
TSS content in AM plants ranged from 0.3 ± 0.01% to 0.4 ± 0.01% (Figure 5a), which was significantly higher than that in NM plants by 8.3-90.5% (p < .05). Differences in AMF treatments had no significant effects on TSS content, whereas differences in inoculum sizes and the interaction between AMF and inoculum sizes significantly affected TSS content (p < .001; Figure 5a). SP content in AM plants ranged from 8.6 ± 0.2 to 13.5 ± 0.8 mg/g FW (Figure 5b). SP content in Cl-S3 (12.1 ± 0.4 mg/g FW), Ce-S3 (12.8 ± 0.3 mg/g FW), HA-S2 (12.6 ± 0.3 mg/g FW), HA-S3 (13.5 ± 0.8 mg/g FW), and HA-S4 (12.5 ± 0.4 mg/g FW) plants was significantly higher than that in NM plants (p < .05; Figure 5b). Furthermore, SP content in plants receiving other treatments was also higher than that in NM plants, but the difference was not significant (Figure 5b). AMF and inoculum sizes had significant effects on SP content (p < .05), but the effect of their interaction was not notable in the two-factor ANOVA (Figure 5b) Figure 5d). Differences in AMF treatments had no significant effects on RWC, whereas differences in inoculum sizes F I G U R E 4 Superoxide dismutase (SOD) activity (a), catalase (CAT) activity (b), peroxidase (POD) activity (c), and malondialdehyde (MDA) content (d) in Cinnamomum migao seedlings treated with different inoculation treatments (NM, Cl, Ce, and HA) and different inoculum sizes (S1, S2, S3, and S4). Note: Vertical bars represent the standard errors of the means based on four replicates. Different lowercase letters indicate significant differences between different inoculum sizes in the same AMF treatment, whereas different capital letters indicate significant differences between different AMF treatments with the same inoculum size (LSD test; p < .05). Significant differences at **p < .001 and *p < .05, and NS indicates no significant differences and the interaction between AMF and inoculum sizes significantly affected SP content, as revealed by ANOVA (p < .05; Figure 5d). Furthermore, studies have shown that plants could enhance their access to various resources and occupy diverse environments by altering functional organ morphology, biomass allocation, and physiological characteristics (Ahmad et al., 2018;Brouwer, 1983). In this study, the inoculation of AMF facilitated root elongation ( Table 2) and branching (Table 3), which enabled root growth in the soil and facilitated nutrient and water uptake. Numerous studies have also shown that modifications in the root architecture induced by AMF help maintain water status and essential nutrients (Aroca et al., 2013;Rajtor & Piotrowska-Seget, 2016;Tamayo-Velez & Osorio, 2017).

| Correlation between measurement indices
Additionally, leaf characteristics and biomass were significantly improved ( Table 2). Root and leaf structures are very important for the plant to resist drought stress (Ekblad & Högberg, 2001). Similar results were reported in previous studies - Boutasknit et al. (2020) and Sebastiana et al. (2018) found that with AMF inoculation, Ceratonia F I G U R E 5 Total soluble sugar (TSS) (a), soluble protein (SP) (b), proline (Pro) (c), and relative water contents (RWC) (d) in Cinnamomum migao seedlings treated with different inoculation treatments (NM, Cl, Ce, and HA) and different inoculum sizes (S1, S2, S3, and S4). Note: Vertical bars represent the standard errors of the means based on four replicates. Different lowercase letters indicate significant differences between different inoculum sizes in the same AMF treatment, whereas different capital letters indicate significant differences between different AMF treatments with the same inoculum size (LSD test; p < .05). Significant differences at **p < .001 and *p < .05, and NS indicates no significant differences siliqua and cork oak root biomass and growth status were significantly improved in drought stress conditions.
The drought tolerance of a plant is determined by comprehensive physiological indices. Photosynthesis is the basis of plant productivity and crop yield and is highly sensitive to environmental factors (Yin & Struik, 2017). Under water-deficit conditions as well as stomatal (reduced CO 2 supply due to stomatal closure) and nonstomatal (decreased photosynthetic activity of mesophyll cells) factors lead to a decrease in the net Pn of plants (Deng et al., 2020;Farquhar & Sharkey, 1982). In this study, the inoculation of AMF studies have shown that oxidative stress occurs as a result of increased ROS and/or decreased capacity of the antioxidant system (Nawaz et al., 2017;Zhao et al., 2018). MDA content is an important index of plasma membrane damage (Loutfy et al., 2012). When plants were subjected to drought stress, the accumulation of ROS in the cell disrupted the metabolic balance, increased the MDA content and cell membrane permeability, damaged the cell structure, and affected the growth and development of plants (Krasensky & Jonak, 2012;Lopes et al., 2017). To survive under adverse environmental conditions, plants maintain an equilibrium between the formation and detoxification of ROS by activating the antioxidant defense system (comprising enzymatic and nonenzymatic antioxidants) to mitigate oxidative damage (Hossain et al., 2020).
The antioxidant enzymes include CAT, POD, and SOD, which play important roles in mitigating oxidative damage (Halo et al., 2015;Parviz et al., 2015). SOD catalyzes the transformation of accumulated ROS into H 2 O 2 and molecular oxygen species, whereas CAT and POD convert H 2 O 2 into water and molecular oxygen for the elimination of ROS Ragupathy et al., 2016;Sousa et al., 2020). In addition, the variation in SOD activity (Figure 4a) in AM plants was lower than that in POD ( Figure 4c) and CAT ( Figure 4b) activities at the inoculum sizes of 40 and 60 g, potentially because POD and CAT had a higher capacity for the decomposition of H 2 O 2 generated by SOD (Cai & Gao, 2020) or because SOD, CAT, and POD had different sensitivities to different strains and inoculum sizes . Hu et al. (2016) proposed that the maintenance of higher POD and CAT activities may provide further oxidative protection by detoxifying H 2 O 2 produced due to stress by weakening the SOD enzyme system. Certainly, further research is warranted to reveal the mechanism by which AM symbiosis alters the adaptability of C. migao to drought tolerance, the relationship between AMF and the metabolic pathways of antioxidant enzymes, and the mechanism by which AM symbiosis regulates antioxidant enzyme gene expression.
During water shortage, plants altered the osmotic potential and increased the contents of osmoregulatory substances through osmotic regulation to sustain the balance between water content and cell swelling pressure, maintaining normal growth and metabolism to resist damage from drought (Zhang et al., 2017). The accumulation of activated osmotic molecular compounds and ions in plant cells may reduce osmotic potential, causing water to move into the cell, thereby increasing cell turgor to resist water deficit (Farooq et al., 2009). The main osmoregulatory substances, such as Pro, TSS, and SP, were involved in the regulation (Jadrane et al., 2020;Lin et al., 2019) and could be used as indices to evaluate drought tolerance in many plant breeding programs (Yu et al., 2017;Zhao et al., 2021). The effect of AMF colonization on host plants depends on many factors, including the host specificity and local diversity of AMF, the selectivity of host plants for AMF, and the adaptability of fungal strains to different environmental conditions, which affect inoculation effectiveness (Wei et al., 2020;Yang et al., 2012;Zhang et al., 2011). In this study, the inoculation of Cl. lamellosum and Cl.
etunicatum increased the drought tolerance of C. migao seedlings.
The three AMF inoculation treatments had significantly different effects on Sh, La, Rl, RSA, Ncon, Pn, WUE, Gs, SPAD, Ci, SOD, CAT, MDA, SP, and Pro, as revealed by ANOVA (  (Smith & Read, 1997). Some studies showed that coinoculation produced better growth-promoting effects than single inoculation (Takács et al., 2018) and that the inoculation of various types of fungi produced positive effects on target plants (Lioussanne et al., 2009).
Similar results were obtained in the present study. At the 60 and 100 g inoculum size, coinoculation had a more positive effect on the drought tolerance of C. migao seedlings, but the effects were opposite at 20 and 40 g inoculum size. Meanwhile, the effects of the three AMF treatments at 40 and 60 g were significantly higher than those at 20 and 100 g (Figure 7). This might be caused by the interaction between the two fungi in antagonism of growth inhibition and competitive growth (Ansari & Ahmad, 2019). Coinoculation had a better inoculation effect as the inoculum size increased to 60 g, which might mean that the coexistence of the two microorganisms achieves a balance between survival and reproduction, and generates positive feedback to the plant symbiosis (Akiyama et al., 2002), thus showing a synergistic effect. However, when the inoculum size continued to increase, intraspecific competition as well as interspecific competition might exist between the two fungi, affecting substrate action (Morón-Ríos et al., 2017;Zhang et al., 2010). The symbiosis between AMF and host plants was affected by intraspecific, interspecific, and nutrient competition between AMF and host plants (Johnson et al., 2012;Kiers et al., 2011;Lang et al., 2013;Morón-Ríos et al., 2017), which was closely related to the quantity of AMF in the soil. Furthermore, inoculum sizes significantly affected CR, MD, plant growth indices, photosynthetic parameters, antioxidant enzymes, and osmoregulatory substances (Table 4), resulting in differences in drought tolerance. A low inoculum size (20 g) only partially induced growth, whereas appropriate inoculum sizes (40-60 g) had a better effect on plants and promoted drought tolerance (Figure 7).
Conversely, a high inoculum size (100 g) inhibited plant growth. This was consistent with the results of Li et al. (2019). As the total fungal biomass increases, the net costs to the plant also increase because the mycorrhizal fungi can outcompete the plant for carbon to meet their energy demands (He et al., 2017;Johnson et al., 1997;Kiers et al., 2011). A symbiotic imbalance occurs when the net cost of symbiosis exceeds the net benefit , and in such situations, the mycorrhizal fungi may be considered parasitic to plants (Johnson  , 1997). This nutrient competition affected the normal growth of plants (Huang, 2020). In addition, the interaction between AMF and inoculum sizes was obvious for all indices, excepting Gs and SP (Table 4). Therefore, it is necessary to investigate the appropriate concentration and inoculum size of different fungi types to promote the drought tolerance of C. migao seedlings.

| CON CLUS IONS
In this study, the inoculation of both Cl. lamellosum and Cl. etunicatum had positive effects on the drought tolerance of C. migao seedlings. Furthermore, a high inoculum size (100 g) may inhibit plant growth.
When the inoculum size was ≥60 g, coinoculation had a more positive effect on the drought tolerance of C. migao seedlings than single inoculation, but the effect was opposite when the inoculum size was ≤40 g. Thus, it is of great significance to study the correlation between the inoculation density of AMF and drought tolerance of plants, which should be explored further in future studies. The results of this study revealed the mechanism by which AMF promote the drought tolerance of C. migao seedlings and provided a basis for further understand-

TA B L E 4 (Continued)
F I G U R E 7 Ranking of comprehensive evaluation scores of drought tolerance of Cinnamomum migao seedlings with different treatments. Note: Vertical bars represent the standard errors of the means based on four replicates. Different lowercase letters indicate significant differences between different inoculum sizes in the same AMF treatment, whereas different capital letters indicate significant differences between different AMF treatments with the same inoculum size (LSD test; p < .05) (supporting); writing -review and editing (equal). Xuefeng Xiao: Data curation (supporting); formal analysis (supporting); investigation (supporting); methodology (supporting); writing -review and editing (supporting). Jingzhong Chen: Formal analysis (supporting); investigation (supporting); methodology (supporting).

ACK N OWLED G M ENTS
We are very grateful to the financial support from Guizhou Science We appreciate the invaluable, thoughtful feedback from the reviewers, which helped improve our research.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data used in this study are stored in Dryad (https://doi.org/10.5061/ dryad.5dv41 ns88).