Arbuscular mycorrhizal fungi increase Pb uptake of colonized and non-colonized Medicago truncatula root and deliver extra Pb to colonized root

Arbuscular mycorrhizal (AM) fungi form symbiosis with terrestrial plants and improve lead (Pb) tolerance of host plants. The AM plants accumulate more Pb in root than their non-mycorrhizal counterparts. However, the direct contribution of the mycorrhizal pathway to host plant Pb uptake was less reported. In this study, the AM fungi colonized and non-colonized root of Medicago truncatula was separated by a split-root system, and their differences in responding to Pb application was compared. Inoculation of Rhizophagus irregularis increased shoot biomass accumulation and transpiration, and decreased both colonized and non-colonized root biomass accumulation. Application of Pb in the non-colonized root compartment increased the colonization rate of R. irregularis and up-regulated the relative expressions of MtPT4 and MtBCP1 in the colonized root compartment. Inoculation of R. irregularis increased the Pb uptake in both colonized and non-colonized plant root, while R. irregularis transferred Pb to the colonized root. The Pb transferred through the mycorrhizal pathway had low mobility move from root to shoot, and might be sequestrated and compartmented by R. irregularis. water ow that facilitated aquaporin The quantication of Pb transfer via mycorrhizal and deserve further


Introduction
Heavy metal contamination in soil is a worldwide issue due to rapid urbanization, mining, sewage sludge, application of fertilizers, and other anthropogenic activities (Fan et al. 2020;Gonzalez-Alcaraz et al. 2018;Sidhu et al. 2017). Lead (Pb) is one of the most common heavy metal pollutants in China (Li et al. 2014b) and is a non-essential element that have an immense risk for human beings especially for children (Baloch et al. 2020;Wang et al. 2020). Phytoremediation is an e cient and noninvasive way to remediate soils (Chang et al. 2018;Ma et al. 2019). The application of microorganisms in phytoremediation of Pb has received extensive attention (Gonzalez-Chavez et al. 2009;Jan and Parray 2016;Jia et al. 2016). Arbuscular mycorrhizal (AM) fungi can establish mutualistic symbioses with more than 80% of terrestrial plants in different ecosystems (Davison et al. 2015;Parniske 2008) including Pb polluted areas (Faggioli et al. 2019;Zhang et al. 2020a). With AM fungal colonization, plants usually have higher biomass (Chen et al. 2005;Huang et al. 2017), increased antioxidant enzymes activities and photosynthetic rates, and showed improved Pb tolerance Zhang et al. 2019).
Establishment of AM symbiosis leads to enhancement of host plant photosynthetic rates, transpiration ow, and water uptake (Gavito et al. 2019;Huang et al. 2017;Kaschuk et al. 2009;Mortimer et al. 2008;Puschel et al. 2020). The water transport from soil to plant leaves require participation of aquaporins (AQPs), which are a class of membrane intrinsic proteins (MIPs) that mediate water transport across membranes following an osmotic gradient (Li et al. 2014a) and participate in hydraulic conductance regulation (Maurel et al. 2015;Watts-Williams et al. 2019). Plant AQPs include plasma membrane intrinsic proteins (PIPs), tonoplast intrinsic proteins (TIPs), NOD26-like intrinsic proteins (NIPs), small basic intrinsic proteins (SIPs), and uncategorized intrinsic proteins (XIPs) (Maurel et al. 2015). In AM plants, the uptake of water and nutrients was suggested via two pathways, of which one pathway (plant root pathway) relies on plant root and the other pathway (mycorrhizal pathway) relies on the AM fungal hyphae (Ferrol et al. 2016;Smith and Smith 2011). It was con rmed that nutrients transportation via mycorrhizal pathway to plant needs participation of aquaporin (Kikuchi et al. 2016).
In previous studies, the AM plants accumulated more Pb in root and less Pb in shoot than their nonmycorrhizal counterparts Sudova and Vosatka 2007;Yang et al. 2016). However, the direct contribution of mycorrhizal pathway to host plant Pb uptake has not been reported. In this study, we used a split-root system (Fig. 1a) to separate the AM fungi colonized and non-colonized root, and investigated the in uence of AM fungi on the root Pb uptake and Pb transfer from root to shoot. We hypothesized that: (1) the AM fungi increase the Pb uptake in both colonized and non-colonized root part through improvement of plant transpiration; (2) the AM fungi deliver Pb to colonized root part through the mycorrhizal pathway; and (3) the plant Pb uptake involves participation of plant aquaporin. To our knowledge, this is the rst study use split-root system to verify the contribution of AM fungi to plant Pb uptake.

Materials And Methods
Plant material, growth substrate, and AM fungal inoculum Seeds of M. truncatula (Jemalong A17) were kindly provided by Prof. Philipp Franken (Plant Physiology Department, Humboldt University of Berlin). The seeds were soaked in concentrated sulfuric acid for 10 min and washed 5 times using sterile distilled water, then incubated in 3% (v/v) sodium hypochlorite solution for 3 min. After surface sterilization, seeds were washed 3 times with sterile distilled water.
Sterilized seeds were germinated in Petri dishes with water agar (0.7%; w/v) at 4 °C for 4 days, and at room temperate in darkness for 2 days. Germinated seeds were transplanted into plastic pots (10 cm in diameter, 12 cm in height) with sterilized sands to grow roots. After 8 weeks, the roots were washed with tap water and divided into two halves of a similar size. Then, the roots were planted in a split-root system consisting of two adjoining compartments with one root half in each (as Fig. 1a). Each root compartment was lled with 0.8 kg growth substrate. The split-root system was made of acrylic plate and was stuck by ABS plastic adhesive. Two compartments were separated by an acrylic plate to prevent Pb spill. The splitroot system was raised 4 cm above the growth substrate on each side to prevent transfer of Pb and AM fungal inoculum between compartments.
The growth substrate was a mixture of sand and vermiculite (1: 1; v: v). The sand was sieved through a 2 mm sieve, thorough washed with tap water, and sterilized at 170 °C for 4 h. The vermiculite was autoclaved at 121 °C for 2 h for sterilization. The vermiculite was clay mineral with 2:1 crystalline structure which contain two silica tetrahedral sheets with a central alumina octahedral layer (dos Anjos et al. 2014).
The AM inoculum of Rhizophagus irregularis (Bank of Glomales in China, No. BGC BJ09), which consisted of a sandy substrate that contained spores (approximately 21 spores per gram), mycelium, and colonized root fragments, was provided by the Beijing Academy of Agriculture and Forestry Sciences (Beijing, China) and multiplied in pot cultures of Zea mays Linn.

Experimental design
The experiment consisted of 5 treatments (Fig. 1b) which include neither AM inoculum nor Pb application in root compartment (CK), only AM inoculum application in one root compartment (OA), only Pb application in one root compartment (OP), AM inoculum and Pb applications in separated root compartments (SE), and AM inoculum and Pb applications in the same root compartment together (TO). The seedling roots in different root compartments were also denominated according to its position, AM status, and Pb status (as Fig. 1b). Ten-gram inoculum was applied underneath the root of M. truncatula seedlings at transplanting into split-root system in mycorrhizal treatment, while sterilized inoculum (170°C for 4 h) was applied in the non-mycorrhizal treatment. The Pb application was 4 weeks post the AM inoculum application to ensure AM fungal colonization, and was accomplished by applying 32 mL 20 g L -1 Pb (NO 3 ) 2 solution to the junction of root and growth substrate by syringe to reach 800 mg kg -1 Pb in growth substrate. Each treatment contained 3 replicates and each replicate included 4 seedlings.
Seedlings were grown in a greenhouse with 28 °C/24 °C day/night temperatures under 16 h daylight and 40-60% humidity. Twenty milliliters of modi ed Hoagland's nutrient solution (Hoagland and Arnon 1950) containing 10% phosphate (0.1 mM KH 2 PO 4 ) was added twice a week to each root compartment before Pb application. After Pb application, only water (20 mL) was added to root compartment of all treatments once in 2 days to avoid direct precipitation of Pb.

Plant sampling, biomass, and AM fungal colonization
At harvest (8 weeks after Pb treatment), biomass of shoots and roots and fresh-to-dry mass ratio (Ma et al. 2014) were measured. After measuring fresh weights, part of leaves was dried in an oven at 105 °C with forced air circulation for 15 min to inactivate enzymes and then turned to 65 °C until they reached a constant weight for Pb content measurement. The remaining part of leaves were immediately frozen in liquid nitrogen and stored at -80 °C. Roots were soaked with water for the root structure scanning (EPSON EXPRESSION 1680, Seiko Epson Corporation, Japan). After root structure scanning, part of roots was xed in FAA solution (37% formaldehyde: glacial acetic acid: 95% ethanol, 9: 0.5: 0.5, v: v: v) for assessment of the AM colonization as Koske and Gemma (1989). The total colonization and arbuscule colonization were measured using magni ed cross sections method as McGonigle et al. (1990). Part of roots were dried in an oven at 105 °C for 15 min and then at 65 °C with forced air circulation until they reached a constant weight for Pb concentration measurement. The remaining part of roots were immediately frozen in liquid nitrogen and stored at -80 °C.

Pb concentration and content
The dry sample was ground in a mortar and placed in the digestion tube (50 mL) with 5 mL mixture of HNO 3 + HClO 4 (4:1) to digest at a temperature that gradually increased to 220 °C. Pb concentration was measured using ame atomic absorption spectrometry (PinAAciie 900F, American). Pb content was calculated using Pb concentration, fresh-to-dry mass ratio, and plant biomass (Ma et al. 2014 Table S1) for 20 s, extension at 72 °C for 20 s; followed by heating from 60 to 95 °C to check the speci city of the PCR ampli cation. All samples were technically replicated twice. Negative controls without cDNA were run within each analysis. The relative quantity of transcripts was determined using the 2 −△CT method (Livak and Schmittgen 2001).

Statistical analysis
Statistical analysis was performed using the SPSS 19.0 statistical programmed (SPSS, American). Data used for one-way ANOVA complied with the assumption of a normal distribution, and the variance equality was also tested by LSD test. Correlation analysis were analyzed by Spearman's (Supplementary

Biomass and colonization
Eight weeks after Pb application in root compartments, the biomass of shoot and root in different compartments was recorded (Fig. 2a). In CK treatment, the biomass of root in two compartments showed no difference, which indicated that the spilt-root system divided roots into two part evenly. Inoculation of R. irregularis in one root compartment (comparing treatment OA with treatment CK) increased the shoot biomass (not signi cantly), but reduced root biomass both locally (OA-RAN) and systemically (OA-LNN). Application of Pb in one root compartment (comparing treatment OP with CK) signi cantly reduced shoot biomass and root biomass in the other root compartment (OP-LNN). When plants received both inoculation of R. irregularis and Pb (together and separated), the shoot biomass was not affected signi cant and the root biomass was reduced.
No AM fungal feature was observed in roots from non-mycorrhizal treatment and root compartments (Fig. 2b, c). Over 60% of root (OA-RAN) was colonized and the typical feature (arbuscules) was observed in treatment which received only AM fungal inoculum. Application of Pb showed little limitation on the total and arbuscular colonization of R. irregularis in the together treatment (TO), but promoted the colonization in the separated treatment (SE). The relative expression of MtPT4 and MtBCP1 that were used as the indicator of functional and numeric of arbuscules (Javot et al. 2007;Parádi et al. 2010) resembled the colonization results (Fig. 3a, b).

Root structure, Pb concentration and content
In CK treatment, root surface, length, and average diameter in two compartments showed no difference ( Fig. 4a, b, c). Inoculation of R. irregularis or Pb application only in one root compartment (treatment OA and treatment OP) did not show local and systemic in uence on the root surface area, length, and average diameter, except for OA-RAN treatment in which the average root diameter was locally reduced. It indicated that inoculation of R. irregularis directly reduced average root diameter. When plant roots received both inoculation of R. irregularis and Pb application separately (treatment SE) and jointly (treatment TO), the root surface area, length, and average diameter were reduced (comparing with treatment CK).
Environmental Pb existed and was unable to eliminate as in previous study (Zhang et al. 2020b). The lowest concentration and content of Pb in root and shoot was shown in the CK treatment (Fig. 5a, b).
Solely Pb application increased local root Pb concentration and shoot Pb concentration. The highest concentration and content of Pb in shoot was shown in separate treatment (SE). The highest concentration of Pb in root was shown in the SE-LNP root compartment, and the highest content of Pb in root was shown in the TO-RAP root compartment. Inoculation of R. irregularis in one root compartment increased Pb concentration in roots from the other compartment in which extra Pb solution was added (comparing SE-LNP with OP-RNP) or not (comparing OA-LNN and TO-LNN with CK-RNN). The Pb concentrations and contents in root from compartment that received Pb were increased compared with CK. Especially, the increment of Pb concentrations and contents in root was much higher when inoculation of R. irregularis involves (comparing SE-LNP and TO-RAP with OP-RNP). Besides, the Pb contents in root that have direct contact with R. irregularis (TO-RAP) was much higher than that in root that have indirect contact with R. irregularis (SE-LNP).
The ratio of root Pb content to root surface area was calculated to evaluate the contribution of plant root surface to Pb uptake (Fig. 5c). Compared with CK, the increased ratios were observed in the root compartment that received extra Pb (SE-LNP and TO-RAP). The highest ratio was shown in the root compartment (TO-RAP) which received both R. irregularis inoculum and Pb application.
Application of Pb increased Ci while R. irregularis inoculation decreased Ci (Fig. 6b). When Pb and R. irregularis inoculum were applied together in the same root compartment, the Gs was higher than they were applied separately in different root compartments.

Relative expression of aquaporins
In order to test the hypothesis that inoculation of R. irregularis improves the capacity of Pb uptake in M. truncatula with the help of aquaporins, the relative expression of aquaporins were detected (Fig. 7). Inoculation of R. irregularis locally increased the relative expression of MtAQP1 in root compartment SE-RAN, MtPIP2 in root compartment TO-RAP, and MtNIP1 in root compartment OA-RAN and TO-RAP (comparing with CK). Inoculation of R. irregularis also systemically increased the relative expression of MtPIP1 in root compartment OA-LNN. Application of Pb locally increased the relative expression of MtAQP1 in root compartment OP-RNP, MtPIP2 in root compartment TO-RAP, and MtNIP1 in root compartment TO-RAP (comparing with CK). The relative expression of MtNIP4 was higher in root compartment OA-RAN than that in root compartment TO-RAP.

Correlation analysis
From the Spearman correlation analysis (Supplementary Table S2), Pb concentration and content of shoot and root showed negative correlations with root biomass, root surface area, and root length, but positive correlations with the relative expression of MtPIP2 in root. Moreover, root Pb concentration showed a positive correlation with the relative expression of MtPT4. Shoot Pb content showed positive correlations with the relative expressions of MtPT4 and MtBCP1 in root. The ratio of root Pb content to root surface area showed positive correlations with Pb concentration and content of shoot and root. In addition, the ratio of root Pb content to root surface area showed positive correlations with the relative expression of MtPT4, MtBCP1, and MtPIP2 in root.

Discussion
AM fungi can survive in various environments including Pb polluted areas, improve growth and stress tolerance of host plants (Faggioli et al. 2019;Yang et al. 2015;Zhang et al. 2020b). Under Pb stress, AM fungi colonized plants were reported to have better growth (Chen et al. 2005;Dhawi et al. 2016;) and accumulate more Pb in root than in shoot (Yang et al. 2016). To verify the in uence of AM fungi on plant root Pb uptake, a spilt-root system was established to separate the colonized and noncolonized root and compare their differences in Pb uptake.
The evenly distributed root biomass in two root compartments of CK treatment demonstrated a success of the split-root system. The similar split-root system was used in other studies for the systemic in uence of AM fungi in M. truncatula (Liu et al. 2007;Zhang and Franken 2014). When roots of M. truncatula were only colonized in one root compartment, the AM fungi showed improvement of plant shoot growth and systemic reduction of plant root growth (Fig. 2) as reported previously (Liu et al. 2007). It might be due to the diluted mycorrhizal effect on nutrient and water uptake (Weissenhorn and Leyval 1995). The systemic in uence of AM fungi on root growth reduction was due to the carbon investment of plant in AM fungal hyphae, which require less carbon than root ). Application of Pb had a negative effect on the biomass accumulation of non-AM plants con rming a sensitivity of M. truncatula to Pb toxicity in the treated concentration . The bene cial effect and alleviation of Pb toxicity by AM fungi, which indicated by the higher shoot biomass and photosynthetic parameters of AM plants than those of the non-AM plants, was consistent with previous studies Sudova and Vosatka 2007;Zhang et al. 2020b).
Inoculation of R. irregularis successfully established AM symbiosis in root compartments, and set the basis for this study. The colonization rate showed similar tendency with the relative expression of MtPT4 and MtBCP1 in root, and this support the view that the expression of these two genes was the indicator of AM symbiosis in root of M. truncatula (Javot et al. 2007;Parádi et al. 2010;Zhang and Franken 2014). When Pb and R. irregularis were applied in different root compartments, the colonization rate of R.
irregularis increased (Punamiya et al. 2010) and the relative expressions of MtPT4 and MtBCP1 upregulated. This might be due to the increased reliance of plant on AM symbiosis, which maintains the balance of plant mineral elements uptake (Ferrol et al. 2016;Zheng et al. 2015), and Pb disturbs plant ion homeostasis through hindering permeability of root cell plasma membrane (Sharma and Dubey 2005;Yadav 2010). Nevertheless, when Pb and R. irregularis were applied in the same root compartment, the extension of extraradical hyphae was inhibited (Weissenhorn et al. 1993), the reliance of plant on AM symbiosis was limited, and the colonization rate was restored.
Although Pb is a non-essential element, the uptake of Pb by plant is inevitable (Baloch et al. 2020;Wang et al. 2020;Yabe et al. 2018). Compared with non-mycorrhizal plants, the mycorrhizal plants usually had higher shoot biomass and Pb content in shoot, which was explained by the bio-dilution effect (Chen et al. 2005;Gonzalez-Chavez et al. 2004;Joner et al. 2000;Weissenhorn and Leyval 1995;Zhang et al. 2020b). Similar result was observed in this study, the Pb content in shoots of treatment TO was higher than that of treatment OP and the relative expression of MtPT4 and MtBCP1 was positive correlated with Pb content in shoot, yet the Pb concentrations in shoot of these treatments were alike (Fig. 5a, b). When R. irregularis and Pb were applied in different root compartments, the Pb concentration and content in shoot were also increased, and this indicated an improvement of Pb transfer from non-colonized root to shoot by AM symbiosis (Chen et al. 2005;Xin et al. 2017).
In root, Pb application increased the Pb concentration and content (Fig. 5a, b) and this was consistent with previous study (Xin et al. 2017). The increment of Pb concentration and content in colonized and non-colonized M. truncatula roots by R. irregularis (comparing TO-RAP and SE-LNP with OP-RNP) (Fig. 5a, b) was in accordance with previous study that AM plants accumulated more Pb in root than nonmycorrhizal plants (Sudova and Vosatka 2007). The increased root Pb concentration and content by R. irregularis might be due to the increased water and nutrient uptake of root, which was proven by the higher shoot biomass and lower root biomass of treatment SE and TO than those of treatment OP (Fig.  2a) and the positive correlations of the relative expression of MtPT4 and root Pb content. Moreover, the increment of root Pb content by R. irregularis in colonized root was higher than that in non-colonized root (comparing TO-RAP with SE-LNP) (Fig. 5a). This indicated an increased Pb accumulation capability of AM fungi colonized root, which have two nutrient and water uptake pathways (Ferrol et al. 2016;Smith and Smith 2011) and may have the mycorrhizal pathway delivers Pb to plant root (Sudova and Vosatka 2007).
The Pb accumulation capability of different roots was compared through the ratio of root Pb content to root surface area. The ratio in treatment TO-RAP was higher than that in treatment SE-LNP (Fig. 5c) and was positive correlated with the relative expression of MtPT4 and MtBCP1. This result further con rmed that the AM fungi supply Pb to colonized root besides increase Pb accumulation capability of plant root. However, the Pb concentration and content in shoot of treatment TO was lower than those of treatment SE (Fig. 5a, b), and this indicated that the Pb supplied by mycorrhizal pathway to plant root have lower mobility than Pb absorbed from growth substrate by plant root itself (Fig. 8). The retention of Pb by AM fungi might be the results of sequestration by their cell wall and proteins, and compartmentation of vacuoles (Ferrol et al. 2016;Salazar et al. 2018).
The nutrient delivery of mycorrhizal pathway was ascertained to follow the water ow (Cooper and Tinker 1981;Kikuchi et al. 2016), which involves participation of aquaporins (Maurel et al. 2015). The relative expression of gene encoding MtPIP2, which was suggested to have higher water permeability than PIP1 and form heterotetramer with PIP1 (Jozefkowicz et al. 2016), was up-regulated in root compartment TO-RAP (Fig. 7) and was positive correlated with the Pb content and concentration in root and shoot, and the ratio of root content to root surface area. This result tted the hypothesis that the Pb uptake by plant root follows water ow (Fig. 8). The speci c role of MtPIP2 in Pb uptake is under study.
To summarize, inoculation of R. irregularis had a bene cial effect on M. truncatula and could alleviate the Pb toxicity. The AM symbiosis increased the Pb uptake in both colonized and non-colonized plant root, while the AM fungi transferred extra Pb to the colonized root section. The Pb transferred from soil to plant root by the mycorrhizal pathway had low mobility move from root to shoot, and might be sequestrated and compartmented by AM fungi. The Pb uptake of plant root might follow water ow that facilitated by the aquaporin MtPIP2. Further researches will quantify the Pb that directly transfer of from R.irregularis to plant root, and decipher the role of MtPIP2 in root Pb uptake.