Differential aluminum tolerance and absorption characteristics in Pinus massoniana seedlings colonized with ectomycorrhizal fungi of Lactarius deliciosus and Pisolithus tinctorius

Plant tolerance to aluminum (Al) toxicity can be enhanced by an ectomycorrhizal (ECM) fungus through biological filtering or physical blockage. To understand the roles of ECM colonization in Al absorption with regard to Al tolerance, Pinus massoniana seedlings were inoculated with either Lactarius deliciosus (L.:Fr.) Gray isolate 2 or Pisolithus tinctorius (Pers.) Coker et Couch isolate 715 and cultivated in an acid yellow soil with or without 1.0 mM Al3+ irrigation for 10 weeks. Biomass production, Al bioaccumulation and transport in seedlings colonized by the two ECM fungi were compared, and the three absorption kinetics (pseudo-first order, pseudo-second order and intraparticle diffusion) models used to evaluate variances in root Al3+ absorption capacity. Results show that both fungi increased aboveground biomass and Al tolerance of P. massoniana seedlings, but L. deliciosus 2 was more effective than P. tinctorius 715. Lower Al absorption capacity, fewer available active sites and decreased affinity and boundary layer thickness for Al3+, and higher Al accumulation and translocation contributed to the increased Al tolerance in the ECM-inoculated seedlings. These results advance our understanding of the mechanisms and strategies in plant Al-tolerance conferred by ECM fungi and show that inoculation with L. deliciosus will better enhance Al tolerance in P. massoniana seedlings used for forest plantation and ecosystem restoration in acidic soils, particularly in Southwest China and similar soils worldwide.


Introduction
Aluminum (Al), the most abundant metal in the lithosphere, can be released from soil minerals as active Al when the soil becomes acidic (Godbold et al. 1988). The level of active Al is toxic to plants when it reaches a 0.04-0.15 mM threshold in a soil solution (Gu et al. 2005). For example, the growth of Pinus massoniana was inhibited by ≥ 0.15 mM Al and greatly inhibited by ≥ 1.0 mM Al (Liu and Liu 1995). Not surprisingly, the stock volume of masson pine has been declining in yellow soil that contains ~ 1.0 mM Al 3+ in China (Du and Tian 1996;Huang and Qu 1996). Al dissolution resulting from acidic soil may be aggravated by numerous factors, including climate warming, acid deposition (NO x , SO x ) and fertilization (Mayor et al. 2015). In central Europe, North America and southern China, activated Al has become a hazardous factor to forest health and caused a new type of forest decline (Gu et al. 2019;Jentschke and Godbold 2000).
A number of ectomycorrhizal (ECM) fungi can enhance plant Al tolerance by immobilizing active Al (1) in soil Abstract Plant tolerance to aluminum (Al) toxicity can be enhanced by an ectomycorrhizal (ECM) fungus through biological filtering or physical blockage. To understand the roles of ECM colonization in Al absorption with regard to Al tolerance, Pinus massoniana seedlings were inoculated with either Lactarius deliciosus (L.:Fr.) Gray isolate 2 or Pisolithus tinctorius (Pers.) Coker et Couch isolate 715 and cultivated in an acid yellow soil with or without 1.0 mM Al 3+ irrigation for 10 weeks. Biomass production, Al bioaccumulation and transport in seedlings colonized by the two ECM fungi were compared, and the three absorption kinetics (pseudo-first order, pseudo-second order and intraparticle diffusion) models used to evaluate variances in root Al 3+ absorption capacity. Results show that both fungi increased aboveground biomass and Al tolerance of P. massoniana seedlings, but L. deliciosus 2 was more effective than P. tinctorius 715. Lower Al absorption capacity, fewer available active sites and decreased affinity and boundary through chelation with secretions such as organic acids, amino acids, and polysaccharides; (2) adsorption to the cell wall of external hyphae, mycelial cords, fungal mantle and Hartig net; and (3) sequestration in the vacuole of hyphal cells (Gu et al. 2005(Gu et al. , 2019Luo et al. 2014;Schlunk et al. 2015;Gaitnieks et al. 2016;Shi et al. 2019). Through these pathways, activated Al can be converted to non-toxic Al or be sequestered by the fungus. Thus, ECM fungi can decrease active Al concentrations in the soil and/or limit Al 3+ transport into root cells.
Studies have demonstrated that ECM fungi in vitro can bioconcentrate Al in fungal cells during + Al exposure. For example, Pisolithus tinctorius and Lactarius deliciosus cultured in vitro accumulated 2.07-and 4.43-fold more Al 3+ when exposed to 1.0 mM Al 3+ and 2.0 mM Al 3+ , respectively, and P. tinctorius had significantly higher Al 3+ in its cell walls than in L. deliciosus (Gu et al. 2021). Piloderma fallax and Tricholoma matsutake concentrated 106.4-and 27.7-fold more Al 3+ when they were cultured on ammonium tartrate-glucose medium amended with Shiro rock fragments than in control conditions without the Shiro rock fragments (Vaario et al. 2015). Mycelia of Laccaria bicolor 270, L. bicolor S238A, and L. bicolor S238N in vitro had 1.95-, 5.26-and 4.80-fold more Al 3+ , respectively when exposed to 1.0 mM compared with 0.2 mM Al 3+ (Gu and Huang 2010). In addition, ECM fungi contribute to Al 3+ immobilization in roots of its host while inhibiting Al translocation upward. For example, the ECM fungus Tricholoma vaccinum restricts Al 3+ transport from the mycelium into host roots (Schlunk et al. 2015). Moyer-Henry et al. (2005) found that the fungal mantle and Hartig net accumulated far more Al than in the cells of lateral roots in Pinus taeda seedlings inoculated with P. tinctorius. In addition, an ECM fungus with greater tolerance to a toxic metal contributes greater tolerance to the metal in the host plant compared with an ECM fungus with less tolerance, and the fungus is cultivated easily, grows rapidly, and readily forms the ECM association with its host (Shi et al. 2019).
These aforementioned studies thus indicate that mycelial immobilization or sequestration of Al 3+ is beneficial to ECM symbiotic Al tolerance. Our previous in vitro study (Gu et al. 2021) verified that Al tolerance was positively related to Al accumulation in ECM fungus and to significantly greater mycelial Al accumulation and biomass production conferred by L. deliciosus compared with P. tinctorius. In addition, Gu et al. (2021) also found that Al tolerance in an ECM fungus was correlated with active site numbers for Al 3+ , boundary layer thickness, cation exchange capacity (CEC) and extracellular Al accumulation in fungal mycelia. In addition, the higher Al bioaccumulation capacity in the ECM fungus in vitro was associated with greater Al tolerance in the ECM-inoculated seedlings (Gu and Huang 2010;Gu et al. 2019). However, the translocation factors for Al in seedlings showed that L. bicolor 270 and L. bicolor S238A facilitated the transport of Al to aboveground parts, whilst L. bicolor S238N did not significantly affect Al translocation. Although seedlings inoculated with L. bicolor S238A were tolerant to Al, seedlings inoculated with L. bicolor 270 or L. bicolor S238N were sensitive to Al (Gu et al. 2019), whereas these three isolates were Al sensitive when exposed to 1.0 mM Al 3+ in vitro (Gu and Huang 2010). Thompson and Medve (1984) also reported that the field responses of three isolates of P. tinctorius to Al treatment differed from their responses to Al in vitro. Thus, we hypothesized that (1) Al bio-accumulation is involved in Al tolerance in ECM seedlings and that (2) the symbiosis could alter the process of Al immobilization or deposition in roots of ECM-colonized seedlings. We thus set up pot experiments and studied the kinetics of Al 3+ absorption to examine (1) whether the Al-tolerant ECM fungus enhanced host Al tolerance, (2) whether the process of Al immobilization in ECM seedlings differed from that in the ECM fungus in vitro, and (3) whether the effect of the ECM fungus on Al immobilization in ECM seedlings is related to Al tolerance.

Site description
The experiment was conducted at a field site (29°48′46″ N, 106°24′39″ E, 260 m a.s.l.) at Southwest University, Beibei, Chongqing, China. This area has a subtropical humid monsoon climate with an annual mean rainfall of 1152 mm, annual mean temperature of 18.9 °C, mean monthly coldest temperature in January and February of 8.7 °C, and mean warmest temperature in July and August of 28.6 °C. The soil used in the present study is a yellow soil (Haplic Alisols), which developed from sandstones of the Jurassic Period, derived from a topsoil (0-20 cm depth, pH 4.23) in a P. massoniana stand (29°50′13″ N, 106°24′25″ E, 741 m a.s.l.) on Jinyun Mountain, Beibei, Chongqing, China. Soil organic matter was 35.8 g kg -1 , total N was 0.37 g kg -1 , available N 50.5 mg kg -1 , TP 0.38 g kg -1 , available P 2.48 mg kg -1 , total K 7.02 g kg -1 , available K 46.3 mg kg -1 , total Al 30.8 g kg -1 , and labile Al was 0.73 g kg -1 . Sterilized seeds were sown in 3 kg soil (sterilized with 2% v/v formaldehyde) in a plastic pot (19 × 14 × 17 cm, upper diameter × lower diameter × depth), then the pots were moved to a sterilized nursery room at 25 °C, 12 h light/12 h dark, 62 μmol m -2 ·s -1 light intensity at the Lab of Silviculture at Southwest University, Chongqing, China.

ECM fungi and growth of seedlings after inoculation with ECM fungi
ECM fungal isolates of L. deliciosus 2 and P. tinctorius 715 (see Gu et al. 2021) used in the present experiment were originally isolated from a P. massoniana stand in Chongqing and an eucalyptus stand in Sichuan, southwestern China, respectively (Wang 2020). The mycelia were grown on Pachlewski's agar medium (pH 5.8) at 25 °C for 21 days in the dark at the Lab of Silviculture at Southwest University.
Six agar discs of mycelia (5 mm in diameter) from the Pachlewski's agar medium were subcultured statically for inoculate respective 250-mL flasks with 100 mL Pachlewski's broth (pH 5.8) at 25 °C in the dark for 21 days. After the removal of growth medium with sterile deionized water, the mycelia were filtered and mixed into a slurry using a motorized stirrer (Philips HR 2024, Philips Inc., Tokyo, Japan) for 1.0 min. The mycelial mixture from five flasks were used to inoculate 50 seeds of P. massoniana (supplied by the Chongqing Tree Seeds and Seedlings Service, Chongqing, China), which had been disinfested for 20 min using 10% v/v H 2 O 2 , washed using deionized water, soaked in water at 30 °C for 24 h, and allowed to germinate on moistened perlite at 25 °C in the dark for 24 h. The inoculated seeds were placed on the potted soil described above and covered with sterilized fine sand (0.5 cm thick). The non-ECM control seedlings were grown with sterilized mycelia.

Al treatment
After 12 weeks, seedlings in each pot were irrigated weekly for 10 weeks with 100 mL distilled water (pH 3.8) containing 0.0 (non-Al) or 1.0 mM Al 3+ (+ Al) as Al 2 (SO 4 ) 3 ·7H 2 O, then grown and irrigated for 8 more weeks with distilled water only. The total amount of added Al 3+ was 0.0 for the non-Al treatment and 81 mg pot −1 for the + Al treatment. Seedlings in pots treated with or without ECM fungus or Al were randomly arranged, with six replicates for a total of 36 pots (3 ECM treatments × 2 Al additions × 6 replicates). These seedlings were grown for a total of 30 weeks from 12 March to 8 October 2019, at 25 °C, 12 h light/12 h dark, 62 μmol m -2 ·s -1 light intensity in a nursery room at the Lab of Silviculture at Southwest University.

Sampling and analyses
After 30 weeks, the seedlings were randomly harvested, and roots were examined for ECM to determine the colonization rate with a dissecting microscope (XSZ-Hs7, Chongqing Photoelectric Instrument Co., Ltd., Chongqing, China) based on the morphology of absorbing roots and safraninfast green (Meng 1996). Visible root ECM colonization was to 95% in all + ECM seedlings, but not in any non-ECM seedlings. The ECM species were distinguished by root morphology and structure (see Wang 2020). Shoots and roots were separately harvested, dried to a constant mass in a 70 °C oven, weighed, and then ground at 25,000 rpm for 1.0 min in a grinder (BJ-150, Baijie, Inc., Zhejiang, China). As described by Gu et al. (2021) and Bao (2000), 1.0 g sampling powder was digested in 5 parts HNO 3 to 1 part HClO 4 (v/v) to measure Al 3+ concentration spectrometrically using the aluminon method (Bao 2000). Accumulation of Al 3+ in roots or shoots was determined as Al concentration × biomass of roots or shoots.
The ratio of shoot Al to root Al is the Al translocation factor (Yuan et al. 2019).

Al absorption kinetics study
Roots of 30-week-old non-ECM and + ECM seedlings that received no Al were rinsed with sterilized deionized water, then oven-dried to constant mass and adjusted to 0.5 g dry mass. Seedlings remained in sterilized deionized water for 48 h to increase its Al uptake ability, then each one was placed in a 500-mL flask with 200 mL Al 2 (SO 4 ) 3 ·18H 2 O solution (pH 3.8) and 0.1 mM Al 3+ in the dark at 25 ± 1 °C for 60 h, with shaking at 150 rpm. After 0.5, 1, 2, 4, 8, 15, 25, 49, 54, and 58 h, a 10-mL sample was removed to determine the Al 3+ concentration as described above. During the incubation, the pH of the solution was adjusted to 3.8 using 0.1 M HCl or 0.1 M NaOH, and sterilized deionized water was added to the incubation solution to maintain a volume of 200 mL.
The amount of Al (Q e , mg g. -1 ) absorbed by the root was calculated as (Eq. 1) where C 0 , C e and C t (mg L -1 ) are the solution concentration of Al at time 0, at equilibrium, and at sampling time t, respectively, M is the root biomass (mg, dry mass), 0.20 and 0.01 are the incubation and sampling solution volume (L), respectively.
Root absorption kinetics of Al were studied using the three models (pseudo-first order, pseudo-second order and intraparticle diffusion model) (Lin et al. 2012;Alvani et al. 2019). Briefly, three assumptions were expressed by the three kinetics models to elucidate the mechanism of metal ion absorption by the absorbent in a liquid absorption system. The pseudo-first order model assumes that the absorption rate is proportional to the effective number of absorption sites and can be described as (Eq. 2) where Q t (mg g -1 ) is the amount of Al adsorbed onto the root at time t, K 1 (h -1 ) is the rate constant and can be calculated from the slope of ln (Q e -Q t ) versus t plot. The pseudosecond order model assumes that available number of active sites in the absorbent directly influences the absorption capacity and can be calculated as (Eq. 3), where K 2 (g mg -1 h. -1 ) is the rate constant and can be measured from the slope of the line plotted from t/Q t versus t. The intraparticle diffusion model assumes that either one or more than one process, including diffusion on the film, across the pore, on the surface and adsorption onto the pore surfaces, control the absorption process in the absorbent and can be described using Eq. (4) where K i (mg g -1 h -1 ) is the rate constant and C is the intercept of the line plotted from Q t versus t. The rate of Al. 3+ absorption at the initial phase (h) can be calculated from Eq. (5) In addition, the second linear section's slope ( K � i ) and intercept (C′) of the line plotted from Q t versus t represent, respectively, the absorbent's affinity and boundary layer thickness for metal ions. The fitting efficacy of the three models were measured using determination coefficients (R 2 ) and non-linear Chi-square (χ. 2 ), which can be calculated as (Eq. 6), and (Eq. 7) where Q e , Q � e and Q � e are the experimental, model estimated and the average model fitted amount of Al absorbed by the root at equilibrium time, respectively, n is the measurement number. According to Boparai et al. (2011), the larger the R 2 and the smaller the χ 2 , the better the model fit.
Concentrations of Al in extracted and digested solutions were measured spectrometrically using the aluminon method (Bao 2000). Briefly, 5-10 mL sample solution, 20 mL H 2 O and 1 drop of p-nitrophenol indicator were added to a 50-mL volumetric flask, then adjusted to yellow using 2 M NaOH and then to colorless using 0.5 M H 2 SO 4 ; 2 mL of 100 mg L -1 HONH 2 HCl was added, and after 5 min of shaking, (4) Q � e 1 mL CH 3 COONH 4 -CH 3 COOH buffer (pH 4.5) was added with shaking, then 2 mL of 1 g L -1 aluminon was added with shaking and then distilled water was added to dilute the reaction solution to 50 mL. After 30 min, absorbance of the solution in 2-cm cuvettes was measured with a spectrophotometer at 520 nm and compared with values based on a 5 µg Al /mL Al 2 (SO 4 ) 3 ·18H 2 O standard solution.

Statistical analyses
All data (means ± SD, n = 6) were tested for normal distribution and variance homogeneity using Shapiro-Wilk's test and Levene's test, respectively. Then a two-way analysis of variance was used to test for treatment (ECM fungus inoculation, Al treatment, and ECM fungus inoculation × Al 3+ treatment) effects on dry biomass production and Al accumulation in the seedlings. Fisher's least significant difference post hoc test was used to test for significant differences between treatments at P ≤ 0.05. SPSS v.23.0 software (IBM, Armonk, NY, USA) was used for all analyses. GraphPad Prism version 8.3.0 (GraphPad Software, San Diego, CA, USA) was used to produce graphs.

Biomass production of seedlings inoculated with ECM fungus
Compared to the non-ECM inoculation, L. deliciosus 2 and P. tinctorius 715 inoculation significantly enhanced root biomass production by 38.2% and 20.6% in 30-week-old seedlings receiving 1.0 mM Al 3+ (+ Al) (Fig. 1a). In contrast, for seedlings that received no Al irrigation (non-Al), root biomass of seedlings without ECM and inoculated with L. deliciosus 2 was similar, but differed significantly from that of seedlings inoculated with P. tinctorius 715. In addition, root biomass production was significantly lower (16.9%) for seedlings in the + Al treatment than for non-ECM seedlings and seedlings inoculated with P. tinctorius 715 in the non-Al treatment, but was similar for seedlings inoculated with L. deliciosus 2.
Shoot biomass production in the seedlings exposed to + Al was significantly higher when colonized by L. deliciosus 2 or P. tinctorius 715 compared to seedlings without ECM, and was significantly higher with L. deliciosus 2 (61.6%) than with P. tinctorius 715 (29.4%) (Fig. 1b). Comparatively, shoot biomass production in the seedlings notexposed to Al was also significantly higher with L. deliciosus 2 or with P. tinctorius 715 and higher with L. deliciosus 2 (33.1%) than with P. tinctorius 715 (16.9%). In addition, shoot biomass was similar between + Al and non-Al treatments for seedlings with the same ECM.
The shoot to root ratio in seedlings exposed to + Al was significantly higher with the L. deliciosus 2 or P. tinctorius 715 inoculation compared to the non-ECM treatment, was higher with L. deliciosus 2 (16.5%) than with P. tinctorius 715 (6.94%) (Fig. 1c). In contrast, the shoot to root ratio of seedlings without Al addition was significantly higher by 24.9% when colonized by L. deliciosus 2 but was similar to that of seedlings colonized by P. tinctorius 715. In addition, the shoot to root ratio was significantly higher with + Al than without for seedlings without ECM and for those with P. tinctorius 715, but was similar to that of seedlings inoculated with L. deliciosus 2 with and without Al addition.

Al bioaccumulation and transport in seedlings inoculated with ECM fungus
Compared to the root Al content in seedlings without ECM, root Al content in the seedlings exposed to + Al was 43.1% significantly higher in those colonized by L. deliciosus 2, but was similar in those with P. tinctorius 715 (Fig. 2a). In contrast, root Al content in seedlings without Al addition was 33.9% significantly lower than in those colonized by L. deliciosus 2, but 14.4% significantly higher than in those with P. tinctorius 715. Root Al content was 44.7% significantly lower in seedlings irrigated with + Al than in those without Al either without ECM or with P. tinctorius upper-or lower-case letters above bars indicate significant differences, respectively, between different ECM seedlings exposed to the same concentration of Al or between different Al exposures for the same ECM seedlings in Fisher's LSD test at P ≤ 0.05 Different upper-or lower-case letters above bars indicate significant differences, respectively, between different ECM seedlings exposed to the same concentration of Al or between different Al exposures for the same ECM seedlings in Fisher's LSD test at P ≤ 0.05 715, but was 21.5% significantly higher in seedlings with L. deliciosus 2 without Al.
After L. deliciosus 2 colonization, shoot Al content in the seedlings exposed to + Al was 61.9% significantly higher than in the non-ECM treatment, but was similar to that after the P. tinctorius 715 colonization (Fig. 2b). In contrast, under non-Al, although L. deliciosus 2 yielded 32.4% significantly higher shoot Al accumulation compared with P. tinctorius 715. In addition, shoot Al accumulation was 25.1%, yet significantly higher when seedlings were exposed to + Al than to non-Al after L. deliciosus 2 colonization, but was similar for non-ECM seedlings and P. tinctorius 715 with either no Al or with Al treatments.
The Al translocation factor in seedlings under + Al was significantly 12.2% higher when colonized by L. deliciosus 2 compared to the non-ECM inoculation, but was similar to that when colonized by P. tinctorius 715 (Fig. 2c). In contrast, the Al translocation factor without Al addition was 75.4% significantly higher with L. deliciosus 2 colonization, but 23.4% significantly lower with P. tinctorius 715. In addition, the Al translocation factor was significantly higher under + Al than under non-Al for seedlings without ECM or with P. tinctorius 715, but or seedlings colonized by L. deliciosus 2 had similar Al translocation factors with and without Al addition.

Al absorption capacity by seedlings with ECM
During the 58-h period, the absorption capacity on Al by roots of non-ECM, L. deliciosus 2 or P. tinctorius 715 inoculated seedlings showed a trend of increase before 49 h and reached an equilibrium between 49 and 58 h (Fig. 3). The process of Al absorption had three stages: a rapid rise in the first 2 h with an average slope of 0.1115, a relatively slower rise with an average slope of 0.0119 from 2 to 49 h, and an equilibrium phase from 49 h on. Compared to the Al 3+ absorption by non-ECM seedlings, the absorption of Al 3+ in seedlings with L. deliciosus 2 or P. tinctorius 715 was slower and lower in the first and second phases. In addition, Al 3+ absorption by seedlings colonized with L. deliciosus 2 was more rapid and higher in the second phase compared with absorption by seedlings colonized with P. tinctorius 715.

Al absorption kinetics by seedlings inoculated with ECM fungus
In the simulation of the Al absorption kinetics of seedlings without ECM or colonized by L. deliciosus 2 or P. tinctorius 715 using the three models ( Fig. 4 and Table 1), the pseudosecond order best fit the observed data (Fig. 4). The highest R 2 and lowest χ 2 between the data fitted by models and observed in the experiment were found in the pseudo-second order model in the seedlings without ECM or colonized by L. deliciosus 2 or P. tinctorius 715. The pseudo-second order model provided an estimate of Al 3+ sorption capacity at equilibrium ( Q � e ) that was closest to the observed Al 3+ sorption capacity at equilibrium (Q e ) than the pseudo-first order model for the seedlings without ECM or inoculated with P. tinctorius 715, but not for those with L. deliciosus 2 (Table 1). In addition, P. tinctorius 715 had lower Q e and Q � e values in either the pseudo-first or second order model than L. deliciosus 2 inoculated seedlings. The + ECM seedlings also had lower Q e and Q � e values than the non-ECM seedlings. The non-ECM seedlings had higher initial rate (h) than the + ECM seedlings for the calculated data from the pseudo-second order model (Table 1) and for experimental data (Fig. 3).
The plots of Q t versus t 1/2 had three segments with different slopes (Fig. 4c). The first segment rose sharply, followed by a relatively slowly rise and ending with an equilibrium. The rate constant ( K � i ) of Al 3+ sorption in the second linear section from the intraparticle diffusion model was slightly higher in the seedlings without ECM compared to those inoculated with L. deliciosus 2, and was lowest in those inoculated with P. tinctorius 715 (Table 2). In addition, intercept values (C') were greater in seedlings inoculated with L. deliciosus 2 and P. tinctorius 715 than in the non-ECM inoculated seedlings (Table 2).

Al tolerance in seedlings with ECM
The tolerance of P. massoniana seedlings in response to 1.0 mM Al 3+ weekly irrigations was enhanced by both tested ECM fungi, and seedlings colonized by L. deliciosus 2 had higher Al tolerance than those colonized by P. tinctorius 715 (Fig. 1). The present result was in accordance with the finding that L. deliciosus 2 had higher Al tolerance than P. tinctorius 715 when exposed in vitro to 1.0 or 2.0 mM Al 3+ (Gu et al. 2021). Egerton-Warburton (2015) also found that in pseudo-first order, pseudo-second order and intraparticle diffusion model, respectively. R 2 and χ 2 are the determination coefficients and non-linear Chi-square between the data fitted by the model and the observed data from the experiment Seedlings Pseudo-first order model Pseudo-second order model Non-ECM  Table 2 Al 3+ absorption rate constant ( K � i ), intercept values (C′) and coefficient of determination (R 2 ′) in the intraparticle diffusion model for 30-week-old Pinus massoniana seedlings without ECM (non-ECM) or inoculated with Lactarius deliciosus 2 or Pisolithus tinctorius 715, exposed to 0.1 mM Al 3+ solution at pH 3.8, 25 °C K � i and C′ are the rate constant and intercept, respectively, in the second linear section from the intraparticle diffusion model. R 2 ′, coefficient of determination in the second linear section between the data fitted by model and observed in experiment Al tolerance was similar between three Pisolithus ecotypes in vitro and Eucalyptus seedlings that were inoculated with the three ecotypes. Thompson and Medve (1984), however, found that the response of P. tinctorius to Al in vitro was not consistent with that in ECM seedlings in the field. Gu and Huang (2010) and Gu et al. (2019) also observed that Al tolerance of ECM fungi in vitro was not similar to Al tolerance in ECM-inoculated seedlings. Three L. bicolor isolates (270, S238A and S238N) were sensitive to Al in vitro (Gu and Huang 2010), but seedlings of P. massoniana that were inoculated with L. bicolor S238A were tolerant to Al in the field (Gu et al. 2019). These results indicate that the response of ECM-inoculated seedlings to Al can differ from the in vitro response of the ECM fungi, although metaltolerant ECM fungi can enhance a plant's metal tolerance (Shi et al. 2019). Thus, field tests of the metal tolerance of ECM-inoculated plants are needed to determine the efficacy of any ECM applications. In our previous in vitro study, L. deliciosus 2 had produced significantly more biomass than P. tinctorius 715 when exposed to 1.0 or 2.0 mM Al 3+ (Gu et al. 2021), and in our present study, seedlings colonized by L. deliciosus 2 grew significantly better than those colonized by P. tinctorius 715 when treated with 1.0 mM Al 3+ (Fig. 1). Thus, Al-tolerant L. deliciosus 2 can be used to improve Al tolerance in P. massoniana seedlings. We therefore recommend inoculating P. massoniana seedlings with L. deliciosus 2 to enhance plant Al tolerance during afforestation and ecological restoration in acidic soil in Southwest China or in similar soils. ECM inoculation facilitated more aboveground biomass production than belowground biomass production in P. massoniana seedlings after the non-Al and the + Al treatments, with greater effects conferred by L. deliciosus 2 than by P. tinctorius 715 (Fig. 1). L. deliciosus 2 also produced more biomass in vitro than P. tinctorius 715 (Gu et al. 2021) as did seedlings (Fig. 1) inoculated with L. deliciousus 2 in the treatments with and without Al. These results suggested that the higher the Al tolerance of the ECM fungus in vitro and ECM-colonized seedlings, the greater was aboveground growth. Gu et al. (2019) also found that the aboveground biomass production was significantly higher in P. massoniana seedlings inoculated with Al-tolerant ECM fungus (L. bicolor S238A) than with Al-sensitive ECM (L. bicolor 270 and S238N) when exposed to 1.0 mM Al 3+ . The growth of shoots but not roots of P. strobus seedlings growing in sand treated with 0-100 mg L -1 Al at pH 3.8 was also significantly improved by P. tinctorius (Schier and McQuattie 1995). Other studies (Gu et al. 2019(Gu et al. , 2020 showed that ECM fungus mobilized inorganic P in rhizosphere soil and facilitated nutrient uptake and transport; thus, both of the fungal species tested in the present study might have similar effects on nutrient mobilization and transport, which improved P. massoniana seedlings' aboveground growth.
Further studies are needed to understand the underlying mechanisms.

Bioaccumulation of Al in seedlings inoculated with ECM fungus
L. deliciosus 2 but not P. tinctorius 715 significantly increased Al bio-accumulation in the roots and shoot of P. massoniana seedlings that were irrigated with 1.0 mM Al 3+ (Fig. 2). In addition, L. deliciosus 2 facilitated Al translocation to aboveground parts (Fig. 2c). The capacity for bio-accumulating metals (Zn, Cr, Cu, Hg, or Pb) is closely related to the tolerance to the metals in Paxillus ammoniavirescens and Pisolithus sp. 1 (Fernández-Fuego et al. 2017;Sedlakova-Kadukova et al. 2019;Shi et al. 2019). Similarly, Gu et al. (2021) reported that Al 3+ accumulation was positively related to the level of Al tolerance of L. deliciosus 2 and P. tinctorius 715. However, differing results have been reported for ECM-inoculated seedlings. For example, Schlunk et al. (2015) and Fernández-Fuego et al. (2017) indicated that Al 3+ accumulation might contribute to Al tolerance in host plants, and Al absorption in roots and translocation to aboveground parts might be inhibited through biological filtering or physical blocking in fungal mycelia. However, Gu et al. (2019) found that inoculation with L. bicolor 270 and L. bicolor S238A did not significantly influence Al accumulation while improving upward Al transport, whereas seedlings inoculated with L. bicolor S238N accumulated less Al but did not significantly influence Al translocation. However, all three isolates of L. bicolor improved the growth of P. massoniana seedlings that were exposed to 1.0 mM Al 3+ . In the present study, neither the accumulation nor the translocation of Al in P. massoniana seedlings receiving + Al were significantly influenced by P. tinctorius 715 inoculation (Fig. 2), although they produced more biomass (Fig. 1). Aforementioned results indicate that (1) Al bio-accumulation involved in the tolerance of P. massoniana seedlings to Al was enhanced by L. deliciosus 2 inoculation and that (2) L. deliciosus 2 and P. tinctorius 715 did not significantly inhibit the transport of Al to aboveground tissues in the seedlings. Nevertheless, the process and mechanisms of Al 3+ translocation from the ECM fungus to the roots and shoot need further study.

Al absorption characteristics in ECM inoculated seedlings
Al 3+ is absorbed by P. massoniana seedlings inoculated with or without ECM fungus in three stages, fitted best by the pseudo-second order model (R 2 = 0.988, 0.998 and 0.974 for non-ECM, L. deliciosus 2 and P. tinctorius 715 inoculated seedlings, respectively. Figures 3 and 4, Table 1). In addition, ECM colonization decreased the initial rate of Al 3+ absorption and capacity in the seedlings. Similar to our results from in vitro Al 3+ absorption experiments (Gu et al. 2021), L. deliciosus 2 conferred higher initial rate and capacity for Al 3+ absorption than P. tinctorius 715 did. According to the assumption for the pseudo-second order model that the number of available active sites in the absorbent is related to the absorption capacity (Lin et al. 2012;Alvani et al. 2019), the present results indicate that inoculation with L. deliciosus 2 and P. tinctorius 715 decreased the number of available active sites for Al 3+ in the roots, and more active sites were available in seedlings inoculated with L. deliciosus 2 than with P. tinctorius 715 (Fig. 4, Table 1). Similar results were obtained for both fungi in vitro (Gu et al. 2021). Shi et al. (2019) speculated that the Cr-tolerant ECM fungus Pisolithus sp. Sp.1 might have a stronger intracellular binding capacity or more electronegative sites to bind Cr than the Cr-sensitive fungus Hebeloma vinosophyllum. Results from the present pot studies and our in vitro experiments (Gu et al. 2021) confirmed that the more Al-tolerant ECM fungus L. deliciosus 2 and seedlings colonized with that fungus had more active sites to absorb Al than the less Al-tolerant ECM fungus P. tinctorius 715 and the seedlings it colonized. Compared to P. tinctorius 715, L. deliciosus 2, an indigenous ECM fungus originally from a P. massoniana stand grown in an acid yellow soil (pH 3.9-4.3) in China, produced more mycelial biomass (Gu et al. 2021) and biomass in inoculated seedlings in the + Al treatment (Fig. 1). In addition, higher CEC (Gu et al. 2021) and functional groups such as carboxyl, carbonyl, hydroxyl, phosphate and amino groups, could contribute to the increase in the number of active sites (Lin et al. 2012;Sedlakova-Kadukova et al. 2019;Zhu et al. 2019). Moreover, Gu et al. (2019) found that inoculation with L. bicolor 270, L. bicolor S238A or L. bicolor S238N facilitated nutrient (P, Ca and Mg) absorption and transport in P. massoniana seedlings. Greater absorption and translocation for Ca and Mg in seedlings inoculated with an ECM fungus might decrease the number of available active sites for Al 3+ because of competition between Ca and Mg. More information is therefore needed about differences in structure and nutrient (e.g., Ca, Mg) absorption among the non-ECM, L. deliciosus 2-and P. tinctorius 715-inoculated seedlings.
The intraparticle diffusion model plots (Fig. 4c) indicated that three steps controlled the process of Al 3+ absorption in P. massoniana seedlings with and without ECM fungus (Alvani et al. 2019;Sedlakova-Kadukova et al. 2019;Gu et al. 2021): (1) diffusion on film or externally, which transport Al 3+ from the incubation solution to the outer surface, (2) diffusion of the particle or across the pore, which drives Al 3+ into the interior sites, and (3) internal absorption. The affinity and boundary layer thickness of absorbents for metals is provided by the second linear section's slope and intercept in the intraparticle diffusion model, respectively (Alvani et al. 2019;Sedlakova-Kadukova et al. 2019). Results in the present study (Fig. 4c, Table 2) indicated that colonization by either of the ECM decreased the affinity for Al 3+ absorption but increased the boundary layer thickness for Al 3+ movement into the roots. The presence of fewer available active sites for Al 3+ in + ECM than in non-ECM seedlings (Fig. 4, Table 1) confirmed that + ECM colonization decreased the affinity for Al 3+ adsorption on the external surface. Thicker boundary layers in + ECM seedlings might come from the thick fungal mantle on the fine roots and the Hartig net that cover cortical cells (Wang 2020). Al 3+ absorption by the ECM fungal cell could also thicken the boundary layer (Gu et al. 2021). Contrary to the thicker boundary layer found in vitro for the mycelium of L. deliciosus 2 compared with P. tinctorius 715 (Gu et al. 2021), the present result for inoculated seedlings showed that L. deliciosus 2 had a thinner boundary layer than P. tinctorius 715 did (Fig. 4c, Table 2). L. deliciosus 2 is a native ECM fungus isolated from a Chinese stand of P. massoniana, which develops a better symbiotic association with P. massoniana seedlings and higher Al tolerance than P. tinctorius 715, which was isolated from a eucalyptus stand in Sichuan Province, China (Fig. 1). In addition, L. deliciosus 2 colonization facilitated Al 3+ translocation upward (Fig. 2c). The abovementioned results suggest that more of the Al 3+ absorbed by L. deliciosus 2 is transported into the roots compared with the case with P. tinctorius 715. Regardless, the greater boundary layer thickness in the + ECM seedlings lowered the amount of Al 3+ that entered the roots, especially in P. tinctorius 715-colonized seedlings (Fig. 3, Table 1).

Conclusions
Our study demonstrated that inoculation with either L. deliciosus 2 or P. tinctorius 715 improved Al tolerance and facilitated aboveground biomass production of P. massoniana seedlings in response to Al exposure; such positive effects on plant performance were greater with L. deliciosus 2 than with P. tinctorius 715. In addition, after L. deliciosus 2 inoculation, Al was translocated to aboveground parts of seedlings after + Al irrigation. Al 3+ was absorbed by non-ECM, L. deliciosus 2, and P. tinctorius 715-inoculated seedlings in a three-stage pattern, which was best fit by the pseudo-second order model. Greater Al accumulation and boundary layer thickness and lower affinity and fewer available active site numbers for Al 3+ contributed to Al tolerance in the inoculated P. massoniana seedlings. The present results demonstrated that L. deliciosus 2 most enhanced Al tolerance of the seedlings and is thus preferred ECM fungus for inoculating P. massoniana seedlings for forest plantation and ecosystem restoration in acidic soils in Southwest China and similar soils worldwide.