Arbuscular mycorrhizal fungi regulate tomato silicon absorption

ABSTRACT

Arbuscular mycorrhizal fungi (AMF) can regulate the absorption of micro- and macro-nutrients and promote the growth and development of plants. The purpose of this study was to investigate the effect of AMF colonization on the Si absorption of tomato. Through potted planting experiment in the greenhouse, the symbiotic system of tomato seedlings and AMF, including Glomus versiforme (Gv), Funneliformis mosseae (Fm) and Acaulospora delicata (Ad), was established. The colonization rate, growth indicators and the Si content and concentration were analyzed. Different extractants were used to extract and determine the different Si-fractions in the rhizosphere soil. This study showed that the inoculation of Gv, Fm and Ad could promote the growth and increase the Si content in roots and shoots of tomato seedlings, and differences were found between AMF strains. Concentration of different binding states of Si in tomato rhizosphere soil was in descending order as citric acid-Si > NaOAc-HAc-Si > Oxalic acid-ammonium oxalate-Si > Na₂CO₃-Si > Na₂HCO₃-Si. When the tomato rhizosphere soils inoculated with various AMFs were extracted with the same extractant, the Si-fraction concentration changed with different AMF inoculations. AMF inoculation could regulate the conversion of different Si-fractions in the rhizosphere soil, which may be an important mechanism of AMF regulating the Si absorption of the tomato.

KEY WORDS:

• Arbuscular mycorrhizal fungi
• growth
• Solanum Lycopersicum L.
• silicon-fraction content

1. Introduction

It is generally believed that silicon (Si) plays an important role in relieving environmental stress and improving the growth of plants (Guntzer, Keller, and Meunier 2012; Ma 2014). However, the content of Si in soil is rich, accounting for 28%, most of which is not plant-available (Sommer et al. 2006). Si in soil includes organic-Si and inorganic-Si. Organic-Si has abundant forms but less content, therefore its biological significance is usually neglected. Inorganic-Si includes crystalline-Si and non-crystalline-Si. Crystalline-Si exists in the form of silicate mineral and silica crystalline, neither of which is insoluble in water and can be absorbed by plants (Liu and Zhang 2001). Non-crystalline-Si, as extractable Si (Zhao et al. 2012), is composed of amorphous-Si, active-Si and water-soluble-Si. Amorphous-Si consists of amorphous aluminosilicate and amorphous silica, with the latter being the majority. Amorphous-Si can be converted to active-Si. Active-Si consists of two types: exchangeable-Si and colloidal-Si. The exchangeable-Si is adsorbed on the surface of the solid phase of the soil and maintains a dynamic balance with the water-soluble-Si. Colloidal-Si exists in polysilicic acid, gel, or silica sol synthesized from monosilicic acid, which shows a high solubility in water but low content in soil (Zhang et al. 2010). Water-soluble-Si exists in the form of monosilicic acid (H₄SiO₄) (Epstein 1994), which can be directly absorbed by plants. The plant-available Si in soil is relatively less, mainly including water-soluble-Si, exchange-Si and colloidal-Si. The study shows that the non-crystalline-Si parts transform to each other (Liu and Zhang 2001; Zhao et al. 2012; Ning et al. 2016), which can be maintained at a relatively stable level for a certain period, but not constant (Ma 2014). It has been suggested that Si leaching and repeated monoculture of some Si-accumulator crops could lead to a reduction in monosilicic acid concentrations in the soil, which may be an important factor limiting the crop yield (Fox et al. 1967; Wang, Li, and Liang 2001; Raven 2003). There are two ways to solve this problem to improve the Si absorption capacity of plants and increase the plant-available Si content in soil. In order to increase the content of plant-available Si in soil and maintain the sustainable development of the agricultural ecosystem, Si fertilizer is widely used in Japan, Germany and South Korea, Argentina and other countries (Ma and Yamaji 2006; Frayssinet et al. 2019). Nevertheless, the soluble Si in Si fertilizer is easy to be fixed in soil and then the utilization rate is reduced. Studies have illustrated that arbuscular mycorrhizal fungi (AMF) inoculation can promote the absorption of mineral elements by plants (Sun et al. 2015; Karagiannidis et al. 2011), and the mycorrhizal symbiosis plays a vital role in changing the ecology of a given site and mycorrhizal promotes mineral cycling and are key component of efficient and closed the nutrient cycle of natural ecosystems (Abbasi et al. 2015). Therefore, using biology to improve the utilization efficiency of plant Si may be an ecological and sustainable approach.

Arbuscular mycorrhizal fungi (AMF) are important soil microorganisms, which can form symbionts with more than 80% of the plants – mycorrhiza (Hamel and Plenchette 2007; Smith and Read 2008; Li and Feng 2001). The foundation of the symbiotic relationship between plants and AMF is nutrient symbiosis. AMF needs carbohydrates and lipids synthesized by plants to survive, at the same time they absorb mineral nutrients in the soil and provide plants, especially under adverse conditions such as herbivory stress, acid rain stress and other stress factors (Frew et al. 2017; Ju et al. 2017a; Clark and Zeto 2000). Mycorrhizal has become an important driver of the agricultural ecosystem productivity. At present, there are many in-depth studies on the effect and mechanism of plant mineral nutrients promoted by AMF, mainly focusing on phosphorus (P) and nitrogen (N) (Li and Feng 2001; Smith and Read 2008; Sun et al. 2015), but few studies on Si (Garg et al. 2020; Kothari, Marschner, and Romheld 1990; Clark and Zeto 1996, 2000; Hammer et al. 2011; Frew et al. 2017, b). Studies have shown that AMF can help plants absorb Si, perhaps due to plant species, AMF population, soil physical and chemical properties (Clark and Zeto 1996, 2000), etc., their conclusions are not consistent. For certain host plants, under different environmental conditions, the same AMF has different infection and ecological functions than the same host (Clark and Zeto 1996; Hart et al. 2015; Ju, Wang, and Wang 2017b).

Tomato (Solanum lycopersicum L.) is a major vegetable worldwide. According to FAOSTAT (2020), in about 172 countries and territories, approximately 4.76 million ha are cultivated annually yielding around 3783.8 million tons. AMF can promote the absorption of mineral elements in tomato, such as P, N and so on (Hart et al. 2015; Sánchez‐Bel et al. 2018; Higo et al. 2020). Whether AMF can promote the absorption of Si in tomatoes has not been reported. Si can improve the resistance of tomato to biological and abiotic stress and promote the growth and development of tomato (Hart et al. 2015; Khalloufifi et al. 2017). Arbuscular mycorrhiza may be a way to improve the absorption of Si in tomatoes. The purpose of this study is to explore whether AMF inoculation can promote the absorption of Si in tomato, and to provide theoretical support for the development of AMF functional strains for promoting Si absorption in tomato agricultural ecosystem.

2. Materials and methods

2.1. Plant culture and treatments

Tomato seeds (Shanghai 906F₁, Xi’an Qin Vegetable Seed Industry Co. Ltd., China) were used to soak with tap water at 25 ± 2°C for 6 h, then germinated in an incubator at 25 ± 2°C. Seeds sprouted in uniform and full were sown, 15 seedlings per pot, at the same time, inoculated AMF: Glomus versiforme (GV), Acaulospora delicata (AD), and Funneliformis mosseae (FM), which were supplied by Bank of Glomales in China (BGC).150 g inoculum plus per pot, and 10 spores/g (Corn was the host on which inoculum was proliferated). The control treatment received killed inoculum plus microbial filtrate. The experiment consisted of 4 treatments (Table 1), each of which was repeated three times. Natural soil was used as the culture medium, the characteristics of which are shown in Table 2. In order to avoid the influence of AMF spores from the natural soil, it was sterilized at high-temperature steam before use.

Table 1. Total of 4 treatments

Directions	Treatments	AMF	BGC No.
The control treatment	CK	Control, without AMF
Inoculation Treatment with AMF	Gv	Glomus versiforme	XJ08F
	Ad	Acaulospora delicata	BJ02B
	Fm	Funneliformis mosseae	GZ01A

Table 2. The characteristics of natural soil

Soil type	pH	Organic material (g/kg)	Total N (g/kg)	Total P (g/kg)	K (g/kg)
Sandy loam	8.36	7.16	0.74	0.61	18.00

The experimental field is located at a nursery base, the Xuzhou Institute of Technology, in Xuzhou, Jiangsu province (34°15ʹN, 117°11ʹW), China. All plants were grown in a greenhouse with a natural photoperiod and light intensity, and the relative humidity was controlled at 50% to 70%. The temperature was regulated at 27 ± 6°C. The seedlings cultured for 60 days were collected for the determination of the test indices, including growth indicators, mycorrhizal infection rates, Si content in plant tissue and rhizosphere soil.

2.2. Quantification of the growth index

Growth was measured in terms of the plant height, root length and dry weight. Tomato seedlings were collected, cleaned, and washed three times with distilled water. The dry weight was determined after drying at 80°C for 12 h (Ju et al. 2017a). Five seedlings were randomly selected from each pot, and 15 seedlings were measured.

2.3. Quantification of mycorrhizal colonization

The percentages of roots exhibiting signs of AMF colonization at each site were determined using the trypan blue staining method of Koske and Gemma (1989) with modifications. Roots from each collected sample were cut into 2-cm-long segments and placed in tissue processing cassettes (Ted Pella, Redding, CA). Approximately 50 of these small root pieces per sample were cleared in 10% KOH at 90°C for 30 min in a water bath. Cleared pieces of roots were rinsed five times with tap water to remove KOH, and roots were immersed in 2% HCl at room temperature for 15–20 min to ensure the roots were immediately stained with 0.05% trypan blue by incubation overnight and then transferred to vials containing lactoglycerol at 4°C to allow excess stain to leach out of the roots. Stained root samples were stored in lactoglycerol solution for 48 h before being mounted in the same solution on a microscopic slide.

Mycorrhizal colonization by AMF structures was determined by assessing the root situation of five slides with 10 segments per slide from each sample and scoring the amount of colonization using the magnified intersection method of McGonigle et al. (1990) with minor modifications. A total of 50 stained root segments per sample were examined with a compound microscope (Olympus CH2, Tokyo, Japan) at 40 × to 100 × magnification for confirmation of mycorrhizal colonization of tomato plants. Root pieces showing the presence of blue-stained mycorrhizal structures including arbuscules, hyphae or vesicles were scored as positive for AMF. All microscopic examinations were carried out by the same individual. Root colonization percentage was averaged for the nine samples (three seedlings per pot) and calculated by the following formula: Root colonization percentage (%) = No. of segments colonized with AMF/No. of segments observed × 100.

2.4. Determination of Si concentration and content in tomato seedlings

Five tomato seedlings were randomly selected from each pot for each treatment, and washed, roots and stems separated, and dried separately. Three pots for each treatment as 3 replicates. The Si concentrations in tomato roots and shoots were determined according to the method described by Ming (2012) with some modifications. The fresh plant material was dried to constant weight at 80°C, then transferred to the dryer, after cooling, and about 100 mg of dried material was placed in a muffle furnace for incineration used for determination of total Si. The ash was fully ground in a mortar and added with 1 ml NaOH (50%); then, the ground mixture was put into the 50 ml high-temperature plastic pipe, then add 8 ml NaOH (50%), and mix. The plastic pipe was covered with a gap for high-pressure steam sterilization. The sterilized solution obtained by the above two methods was filtered, diluted with distilled water to 50 mL, and mix, then transfer 1 mL of this solution to a 50 ml volumetric flask, add 30 ml glacial acetic acid (20%), and 10 ml ammonium molybdate (54 g/L, pH 7.0), after 5 minutes, add 5 ml tartaric acid (10%), and 1 ml reducing reagent (8 g/l NaSO₃, 1.6 g/l 1-amino-2-naphthol-4-sulfonic acid, 25 g/l NaHSO₃), with glacial acetic acid (20%) to 50 mL. The absorbance was measured at 650 nm after 30 minutes. The standard curve was made with different concentrations of silica and corresponding absorbances, and the Si concentration in plant tissue was calculated according to the standard curve. Si content = Si concentration×Dry weight.

2.5. Determination of Si concentration in the rhizosphere soil of different extraction methods

Rhizosphere soil sampling was performed 1 time per pot for each treatment and 3 pots for each treatment as 3 replicates. The experiment selected five extractants to extract Si-fraction in rhizosphere soil, including (1) 0.19 M Na₂CO₃ (Na₂CO₃-Si pH 11.8), (2) 0.5 M NaHCO₃ (NaHCO₃-Si, pH 8.5) (3) 0.025 M citric acid (CA-Si), (4) Oxalic acid-ammonium oxalate butter (OA-Si, pH 3.2), and approximately 5 g of air-dried rhizosphere soil was extracted with 50 ml extractants in the polyethylene bottle, respectively, under mechanical shaking at 120 rpm and 40°C for 5 h. (5) Sodium acetate-acetic acid butter (NaOAc-HAc-Si pH 4.0), and approximately 5 g of air-dried soil was extracted with 50 ml NaOAc-HAc in polyethylene bottle under the water bath at 40°C for 5 h. the extraction solutions mentioned above were used for the determination of Si-fraction concentration. After filtration with Whatman No. 5 filter paper, the filtrates were refrigerated at 4°C for no more than 4 days prior to colorimetric analysis. Soil-test Si levels were determined using a modified blue silicomolybdous acid procedure (4500-SiO₂ E) that made use of 20% (w/w) tartaric acid instead of oxalic acid, and color absorption was measured at 660 nm (Xu et al. 2001; Sauer et al. 2006; Yu et al. 2016).

2.6. Statistical analysis

In this experiment, the number of rice seedlings per treatment was 3 boxes×15 seedlings. All data are presented as the means ± standard deviations. The difference significance between the different treatments was analyzed by the one-way analysis of variance (ANOVA) using SPSS 21 software (IBM, Armonk, New York, USA). Duncan’s multiple-range test was applied to determine the significance between different treatments (p < 0.05). Correlations between the measured indicators were analyzed using Origin 8.0 software (OriginLab, Northampton, USA).

3. Results

3.1. AMF root colonization

Overall, the non-AMF roots showed no evidence of root colonization (Table 3). There was a significant difference in root colonization in different AMF, and the colonization rates (CR) of Funneliformis mosseae (Fm), Glomus versiforme (Gv), and Acaulospora delicata (Ad) were 65.4%, 62.6%, and 45.6%, respectively. There was no significant difference in CR between Fm and Gv, the CRs of Gv and Fm were significantly higher than Ad.

Table 3. Effects of different AMF inoculations on tomato growth indicators and colonization rate

Treatments	PH (cm)	RL (cm)	DWS (g)	DWR (g)	TDW (g)	CR (%)
CK	10.00 ± 0.45 c	8.22 ± 0.26 c	0.184 ± 0.004 d	0.033 ± 0.002 c	0.217 ± 0.005 d	0.0 ± 0.0 c
Fm	13.44 ± 0.52 a	12.34 ± 0.34 a	0.267 ± 0.005 a	0.049 ± 0.004 a	0.326 ± 0.005 a	65.4 ± 1.1 a
Gv	13.16 ± 0.48 a	12.40 ± 0.42 a	0.248 ± 0.004 b	0.045 ± 0.002 a	0.293 ± 0.003 b	62.6 ± 4.0 a
Ad	11.50 ± 0.50 b	9.12 ± 0.33 b	0.207 ± 0.003 c	0.040 ± 0.002 b	0.247 ± 0.004 c	45.6 ± 2.3 b
F	63.0e16**	144.593**	520.837**	39.751**	540.858**	827.676**

PH: Plant height; RL: Root length; DWS: Dry weight of shoot; DWR: Dry weight of root; DW: Total dry weight; CR: Colonization rate

CK: control; Gv: Glomus versiforme; Ad: Acaulospora delicata; Fm: Funneliformis mosseae

The values represent the means ± standard deviation, n = 15(growth indicators), n = 9 (colonization rate). Significant differences are indicated by different letters in the same column. Significant at the 0.05 level.

3.2. Effect of AMF colonization on tomato seedling growth

Table 3 shows the effects of AMF colonization on the plant height (PH), root length (RL), dry weight of shoot (DWS), dry weight of root (DWR), and total dry weight (TDW) of tomato seedlings. The PH, RL, DWS, DWR, and TDW increased under different AMF treatments compared with those of the control (Table 3). Different AMFs have different effects, for DWS and TDW, Fm was obviously heavier than Gv. However, there was no significant difference between PH, RL and DWR in inoculation with Fm or Gv. When the tomato seedlings are inoculated with Fm and Gv, PH, RL, DWS, DWR, and TDW increased significantly compared to Ad. Overall, the effect of AMF inoculation on the tomato growth was in descending order as Fm > Gv > Ad.

3.3. Effect of AFM colonization on Si content in tomato seedling

Table 4 shows the effects of AMF colonization on the Si concentration and content in roots and shoots, and the total Si content of tomato seedling. When tomato seedlings were inoculated with three different AMFs, including Gv, Fm and Ad, Si concentrations and contents in roots and shoots, the total Si content of tomato seedlings increased significantly compared with the control. For Si concentrations in roots and shoots, Si content in shoots and total Si content, there were significant differences between Fm, Gv and Ad inoculation treatments, and the order of from high to low was Gv > Fm > Ad. However, for Si content in roots, Fm and Gv were significantly more effective than Ad and Gv was more effective than Fm, but the difference was not significant. The correlation coefficients indicated that the Si content in shoot was significantly positively correlated with DWS (R = 0.901, ρ = 0.033); however, the correlations of Si content in root to DWR (R = 0.815, ρ = 0.064), and total Si content to TDW (R = 0.846, ρ = 0.053) were not significant overall.

Table 4. Effects of different AMF inoculations on Si concentration and content in root and shoot of tomato seedling

Treatments	Si concentration in root (mg/g)	Si concentration in shoot (mg/g)	Si content in root (mg)	Si content in shoot (mg)	Total Si content (mg)
CK	1.831 ± 0.020 d	1.544 ± 0.031d	0.062 ± 0.002 c	0.287 ± 0.007 d	0.349 ± 0.008 d
Fm	2.747 ± 0.069 b	1.913 ± 0.028 b	0.136 ± 0.002 a	0.510 ± 0.002 b	0.646 ± 0.004 b
Gv	3.061 ± 0.105 a	2.117 ± 0.055 a	0.137 ± 0.002 a	0.525 ± 0.004 a	0.662 ± 0.006 a
Ad	2.002 ± 0.042 c	1.740 ± 0.019 c	0.079 ± 0.002 b	0.358 ± 0.006 c	0.438 ± 0.008 c
F	297.532**	114.368**	2251.484**	2392.124**	2472.603**

CK: control; Gv: Glomus versiforme; Ad: Acaulospora delicata; Fm: Funneliformis mosseae

The values represent the means ± standard deviation, n = 3. Significant differences are indicated by different letters in the same column. Significant at the 0.05 level.

3.4 Effect of AMF inoculation on the extracted Si concentration in rhizosphere soil of tomato seedlings

Table 5 shows the effect of AMF inoculation on the extracted Si concentration in rhizosphere soil of tomato seedlings. Five extractants, including NaHCO₃, Na₂CO₃, oxalate-ammonium and oxalate buffer (OA), sodium acetate-acetic acid butter (NaOAc-HAc), and citric acid (CA), were used to extract different Si fractions. Under the same AMF inoculation, the Si concentration of soil extracted by different extractors varied greatly, but the variation trend was the same, and the order was citric acid-Si (CA-Si) > sodium acetate-acetate buffer-Si (NaOAc-HAc-Si) > oxalate-ammonium oxalate buffer-Si (OA-Si) > Na₂CO₃-Si > NaHCO₃-Si.

Table 5. Effects of different AMF inoculations on the extracted Si-fraction concentration of rhizosphere soil (mg/kg)

Treatments	CA-Si	NaOAc-HAc-Si	OA-Si	Na2CO3-Si	NaHCO3-Si
CK	164.420 ± 2.217 c (100)	71.752 ± 1.799 d (43.64)	59.777 ± 0.826 a (36.36)	19.413 ± 0.974 b (11.81)	10.305 ± 1.145 b (6.27)
Fm	151.930 ± 1.238 d (100)	88.772 ± 0.826 c (58.43)	54.690 ± 1.892 b (36.00)	22.753 ± 1.799 a (14.98)	14.650 ± 0.715 a (9.64)
Gv	222.533 ± 1.483 a (100)	107.600 ± 1.892 a (48.35)	46.587 ± 1.799 c (20.93)	10.257 ± 1.319 d (4.61)	1.222 ± 0.145 c (0.55)
Ad	165.753 ± 1.092 b (100)	100.808 ± 2.699 b (60.81)	47.302 ± 0.413 c (28.54)	15.608 ± 0.675 c (9.42)	2.588 ± 0.675 c (1.56)
F	1158.785**	359.738**	680.088**	1002.249**	251.125**

CA-Si: 0.025 M citric acid (pH 2.5); NaoAc-HAc-Si: sodium acetate-acetate (pH 4.0); OA-Si: Oxalic acid-ammonium oxalate (pH 3.2); Na₂CO₃-Si:0.19 M Na₂CO₃ (pH = 11.8); NaHCO₃-Si: 0.5 M NaHCO₃ (pH = 8.5)

CK: Control; Gv: Glomus versiforme; Ad: Acaulospora delicata; Fm: Funneliformis mosseae

The values represent the means ± standard deviation, n = 3. Significant differences are indicated by different letters in the same column. Significant at the 0.05 level.

Under different AMF inoculation conditions, the Si concentration extracted by the same extraction agent was different. Extracting the Si with a more acidic solution, including NaOAc-HAc and CA, the soluble Si concentrations in rhizosphere soil of tomato seedlings inoculated with Gv, Fm and Ad were significantly higher than that of control, and there were also significant differences among AMF treatments, Gv > Fm > Ad. Extracting the Si with OA, the soluble Si concentrations in rhizosphere soil of tomato seedlings inoculated with Gv, Fm and Ad were significantly lower than that of control, and there were also significant differences among AMF treatments, Fm-treatment was significantly higher than Gv- and Ad-treatment, which had no significant difference. Extracting the Si with NaHCO₃ and Na₂CO₃, the soluble Si concentrations in rhizosphere soil of tomato seedlings inoculated with Fm were significantly higher than that of the control, and the soluble Si concentrations in rhizosphere soil of tomato seedlings inoculated with Gv and Ad were significantly lower than that of the control, Ad-treatment was significantly higher than Gv-treatment under Na₂HCO₃ solution extraction condition, but with the Na₂HCO₃ solution extraction, Gv-treatment and AD-treatment had not significant difference.

4. Discussion

4.1. Effects of AMF inoculation on plant growth

This study showed that three AMFs, including Glomus versiforme (Gv), Funneliformis mosseae (Fm) and Acaulospora delicata (Ad), had high colonization rates on tomato seedlings. To some extent, mycorrhizal colonization rate reflects the symbiotic affinity between AMF and host plants, and is an important index to evaluate mycorrhizal characteristics and ecological adaptability (Chenchouni, Mekahlia, and Beddiar 2020; Wang 2008). Gv, Fm and Ad inoculation promoted the growth of tomato seedlings, and significantly increased plant height, root length and dry weight. Ortas et al. (2013) also studied the effects of six different AMF inoculations on tomatoes. The results showed that all the inoculated plants were extensively colonized by all the AMF species, and the inoculated tomatoes produced significantly higher biomass than those of non-inoculated plants. Ortas et al. (2013) reported that colonization of the root system by AMF confers benefits directly to the growth of host plants and development by increasing nutrient uptake and also improves plant tolerance to stress conditions. Different species of AMF inoculation have different effects on the host (Ortas et al. 2013; Hart et al. 2015). In this study, inoculation with Fm and Gv had significantly better promoted growth effects than Ad. Researches have shown that the impact of AMF on the host plants was controlled by the genotypes of fungus (Baum, El-Tohamy, and Gruda 2015).

4.2. Effects of AMF inoculation on Si absorption in plants

This study showed Gv, Fm and Ad inoculation all improved the absorption capacity of Si in tomato seedlings, and the Si concentration and content in tomato roots and shoots was significantly higher than that in the control without AMF inoculation. Although there has been no relevant study on the effect of AMF inoculation on Si absorption of tomato, some studies have shown that AMF inoculation can promote the absorption of Si in other plants (Kothari, Marschner, and Romheld 1990; Clark and Zeto 1996, 2000; Garg and Bhandari 2016; Oye Anda, Opfergelt, and Declerck 2016; Garg and Singh 2018; Hammer et al. 2011; Frew et al. 2017, b). Different AMFs have different effects on Si absorption in plants (Clark and Zeto 2000). In the present study, the effect of Gv, Fm and Ad inoculation on Si absorption of tomato was significantly different, this might be due to the different ecological functions of AMF. Fm could obviously promote the growth of tomato seedlings, but the Si absorption effect was significantly lower than Gv. This might be because Fm promoted tomato seedling growth through other ways, for example, Fm might be more conducive to the absorption and utilization of other nutrients. Hart et al. (2015) reported that N and Cu were higher in tomatoes with R. irregularis inoculation, P was higher in tomatoes with Fm inoculation; however, Fm inoculation gave higher yields and shoot mass than R. irregularis inoculation. In this experiment, Gv, Fm and Ad inoculation increased the Si content in tomato seedlings under alkaline soil (pH 8.36) culture. This was inconsistent with the results of Clark and Zeto (1996), who believed that AMF could promote plant Si absorption under acidic conditions, while AMF had no effect on the plant Si absorption under alkaline conditions (Clark and Zeto 1996). This may be due to differences in host plants, environmental conditions, and the functioning of AMF functional strains (Clark and Zeto 1996; Moradtalab et al. 2019).

4.3. Effects of AMF inoculation on the concentration of extracted Si in rhizosphere soil

Based on the analysis of the different Si-fraction content, the study showed that different AMF inoculations had different regulation mechanisms on the Si-fraction content in tomato rhizosphere soil. Na₂CO₃ and NaHCO₃, as alkaline extractions, are used to analyze amorphous-Si in soils (Yu et al. 2016). Na₂CO₃-Si and NaHCO₃-Si content (i.e., amorphous-Si) in tomato rhizosphere soil inoculated Fm were significantly higher than those in control, indicating that Fm inoculation might inhibit the conversion of amorphous-Si to active-Si. Amorphous-Si contents in tomato rhizosphere soil inoculated Gv and Ad were significantly lower than the control, indicating that Gv and Ad inoculation might promote the conversion of amorphous-Si to active-Si. OA buffer is the most commonly used method to assess Si bound in allophanes and imogolite-type materials in soil (Snyder 2001; Sauer et al. 2006; Yu et al. 2016). OA-Si content in tomato rhizosphere soil inoculated Fm, Gv and Ad were significantly lower than those of the control, indicating that AMF inoculation might increase the conversion of bound-Si in allophanes and imogolite-type materials to active-Si, in descending order as Gv>Ad>Fm. The NaOAc-HAc-Si and CA-Si contents (i.e., plant-available Si) in tomato rhizosphere soil inoculated Gv and Ad were significantly higher than the control, indicating that AMF inoculation might be due to promoting the conversion of amorphous-Si to plant-available Si. However, the CA-Si content in tomato rhizosphere soil inoculated Fm was significantly lower than the control, Gv and Ad inoculation, which might have been attributed to inhibiting the conversion of amorphous-Si, bound-Si in allophanes and imogolite-type materials to plant-available Si and improving the Si absorption of tomato seedlings. Factors reported to influence soil Si availability or Si supplying power include parent material, historical land use change, soil pH, soil texture, soil redox potential, organic matter, and accompanying ions, such as Fe and Al (Struyf et al. 2009; Xu et al. 2001; Schaller et al. 2018). One of the mechanisms of AMF inoculation to activate soil nutrients is to regulate the form and content of mineral elements by secreting enzymes and organic acids, so as to improve the availability of soil nutrients (Qu et al. 2019; Gu 2014; Xue et al. 2019; Clark and Zeto 2000). AMF can secrete phosphatase and organic acids, mineralize organic phosphorus in soil, and promote the bioavailability of insoluble phosphorus (Gu 2014; Xue et al. 2019; Clark and Zeto 2000). Wang, Xia, and Wang (2009) and Wang (2008) reported that AMF could improve the contents of available iron through the changes in iron species in soil. Dehghanian et al. (2018) reported that Fm (Funneliformis Mosseae) inoculation could affect Fe, Mn, and Zn distributions in exchangeable, bound to carbonates, bound to iron–manganese oxides, bound to organic matter and residual fractions in root soil, and increase the uptake of Zn, Mn and Fe by maize probably with impact on bound to Fe-MnO fractions. Different AMFs produce different mycorrhizal exudates, so different AMFs inoculation activates mineral elements with different mechanisms (Wang, Xia, and Wang 2009; Qu et al. 2019; Liu et al. 2010; Hosseini and Gharghani 2015). In the present study, different AMFs inoculation led to different Si-fractions content in tomato rhizosphere soil, which might be due to different secretions produced by different AMF inoculations of tomato, and different secretions had different activation mechanisms for Si in soil.

5. Conclusion

We conclude that AMF inoculation could promote the growth of tomato seedlings and the absorption of Si element, and the effects of different AMF strains were different. According to the analysis of the concentration of different forms of Si in tomato rhizosphere soil and the content of Si in tomato plants, the regulation mechanism of different AMF inoculations on the absorption of Si by plant roots was different. This was mainly embodied in two aspects: first, different AMF inoculations have different effects on the Si absorption ability of tomato root, leading to the Si content in plants and plant rhizosphere soil were different, it had been verified in this study (Table 3). Second, different AMF inoculated plants might regulate mycorrhiza to form different metabolites, which may regulate the balance between different forms of Si and changing the content of different forms of Si in the rhizosphere, which requires further in-depth study.

Disclosure of potential conflicts of interest

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the National Spark Plan Project (No: 2013GA690441)the National Spark Plan Project [2013GA690441];