Seeds collection and germination
We collected T. sebifera seeds by hand from populations in both the introduced (US—16 populations in total) and native (China—14 populations in total) ranges (for details see Table S1). At each population, we haphazardly selected 5–10 trees, and harvested thousands of seeds from each tree. In the laboratory, we removed the waxy coats around these seeds by hand after immersing them in a mixture of water and laundry detergent (10 g/L) for 24 h [29]. Then, we rinsed, surface sterilized (10% bleach), and dried them. In order to improve germination, we put these seeds in wet sand and stored them in the refrigerator (4 °C) for at least 30 days. In spring, we sowed these seeds in greenhouse trays (50 holes/tray) which were filled with sterilized (autoclave at 121 °C for 30 min) commercial potting soil, and then kept them in an open-sided greenhouse at Henan University in Kaifeng, Henan, China (34°49′13′′ N, 114°18′18′′ E) or unheated greenhouse at Rice University, Houston, TX USA (29°43′08′′ N 95°24′11′′ W). After seeds germinated and seedlings reached the 4 true leaf stage, we selected similar size seedlings to conduct the following experiments.
Common garden experiment—differences in AM fungal colonization and plant growth
To investigate the differences in AM fungal colonization and growth between plants from introduced (US) and native populations (CH), we carried out a common garden experiment at Henan University. We collected soil in a corn field, which includes most common AM fungal species based on previous reports [33, 34]. It was a sandy soil with total nitrogen and total phosphorus of 1.9 g/kg (DW) and 0.6 g/kg (DW), respectively, and pH of 7.68. We removed surface litter before collecting topsoil (10–15 cm depth) and then combined equal parts of soil and fine sand in 132 pots (21 cm × 16 cm, ~3 kg of soil mix each) after they were passed through a 1-cm mesh screen. We planted seedlings from 22 populations (12 native and 10 introduced populations, 6 seedlings of each population, Table S1) individually in these prepared pots and placed them in the open-sided greenhouse. We protected them from herbivores with nylon mesh (16 openings/cm) cages during the experiment. After 60 and 90 days, we harvested 3 seedlings from each population as 3 reps each time and carefully washed their whole roots from the soil. We collected ~30 fresh fine roots (>1 cm/segment) from each plant root to measure AM fungal colonization. In brief, we cleared (in 10% KOH), bleached, acidified, and stained (trypan blue) these samples then slide mounted 30 one cm long pieces of fine root for each plant [7]. AM fungal colonization was determined by the gridline intersect method with 300 intersection points per plant [35]. We dried and weighed the roots and shoots.
Collection of root exudates and flavonoids analysis for root exudates
Our previous study found higher concentrations of flavonoids but lower concentrations of tannins in roots of introduced populations of T. sebifera than in native populations [17] with quercetin and quercitrin being the main flavonoids [28, 30]. In our pilot experiment, we only detected quercetin and quercitrin in root exudates but no other flavonoids. Therefore, in this study we focused on quercetin and quercitrin in root exudates and their functions. We determined their amounts in root exudates from native (China) and introduced (US) populations at Henan University. We filled 132 glass beakers (1000 ml) with Hoagland’s solution [36] and covered the opening with a foam board with a hole in its center. We took 6 seedlings from each of 22 populations (12 native, 10 introduced, Table S1) and carefully washed the soil from their roots with tap water, then transplanted them individually into the beakers (1 seedling per beaker) and fixed them with a sponge. Because of mortality, only 80 plants of 17 populations (9 native, 8 introduced) survived until exudate collection. The odds of a plant dying did not depend on population origin (F1,20 = 3.7, P = 0.0679) or population (Z = 1.3, P = 0.0937). We checked these glass beakers and filled them with Hoagland’s solution every day.
After these plants grew for 57 or 87 days in an open-sided greenhouse with a typical temperature range of 18 °C (night) to 28 °C (day) and 13–14 h of natural daylight, we put DI water into these beakers instead of Hoagland’s solution to minimize the effects of environments on root exudates. Three days later (i.e., at 60 and 90 days) these plants were harvested to obtain their dry root mass. The root exudates were dried at 40 °C under vacuum by rotary evaporators. Then we extracted the chemicals from these concentrates in 3 ml of methanol solution with 0.4% phosphoric acid water (48:52, v:v) and filtered them through 0.22 μm hydrophobic membranes. The concentrations of quercetin and quercitrin were assessed by high-performance liquid chromatography [30]. In brief, 20 μl of extract was injected into an HPLC with a ZORBAX Eclipse C18 column (4.6 × 250 mm, 5 μm; Agilent, Santa Clara, CA, USA) with the following flow: 1.0 mL min−1 with a 100% methanol (B) and 0.4% phosphoric acid in water (A) as the mobile phase. The gradient was as follows: 0–10 min 52:48 (A:B); 10–24 min 48:52 (A:B). UV absorbance spectra were recorded at 254 nm. The concentrations of flavonoid compounds were calculated and standardized by peak areas of standards of known concentrations.
Root exudate addition experiment—effects of different populations on AM fungal colonization
In order to investigate the role that root exudates play in the interactions between AM fungi and plants, we conducted an experiment in which exudates were collected from plants in liquid (donor) and applied to the soils of other plants (target). The exudate donor plants were grown in 1080 (two venues: 540 seedlings at Rice University and 540 seedlings at Henan University) containers, each with 1000 ml of Hoagland’s solution, that each had a foam board top with a hole and a bottom drain tube that could be regulated. At each venue, we washed the soil from ~500 sets of plants (US = 465, China = 504) from native (8 populations for venue US and 7 populations for venue CH) or introduced (13 populations for venue US and 12 populations for venue CH) populations and secured them (3 plants per container) in the containers using sponges (details in Table S1). The remaining containers were left as plant-free controls. We started the application experiment after 7 days.
For exudate target plants, we collected the soil from different sites in the introduced or native ranges (See Table S1). At each site, we collected soil under the canopy of a T. sebifera tree (Home soil) and that more than 3 meters away from the canopy of a T. sebifera tree (Away soil). We collected the topsoil to a depth of 15 cm after removing the surface litter, air dried them, and screened them (1 cm mesh). These soils were mixed with vermiculite (1:2 volume). Then we used these mixes to fill 1080 pots (15 cm × 12 cm; 540/venue). Each pot in China received a mixed soil from a site in China and each pot in US received soil from a single small area within a site in the US. We transplanted a seedling from a native (12 populations for venue US and 3 populations for venue CH, See Table S1) or introduced (13 populations for venue US and 5 populations for venue CH, See the Table S1) population into each pot (270 of each per venue). We randomly assigned a target plant to each set of donor plants or water only controls.
Every 4 days we changed the Hoagland’s solution to DI water for 3 days to collect root exudates from donor plants. Then we applied this water solution from a donor set to its target plant. After 70 days, we harvested the target seedlings, kept a fine root sample for AM fungal colonization determination, then dried and weighed leaves, stems, and roots.
Chemical addition experiment—quercetin and quercitrin effects on AM fungal colonization
We transplanted 391 seedlings from 8 native populations (CH) and 9 introduced populations (US) into 391 pots with field soil (1.3 kg/pot) in nylon mesh cages at Henan University. To test the effect of quercetin and quercitrin on AM fungal colonization, we prepared solution of quercetin or quercitrin in acetone (10 mg/mL) (acetone did not affect AM fungal colonization based on our preliminary experiment). Then these solutions were diluted in water to 2 concentrations (1 mg/L and 10 mg/L) based on the result of chemical analyses of root exudates and the 0.1% of acetone in water as control. We watered 15 ml of solution (5 reps per population) or water (3 reps per population) around the base of seedling stems every 3 days (16 times in total). Four plants died (3 in quercitrin application treatment, 1 in quercetin application treatment). After 70 days, we collected seedlings by cutting at ground level and collected fine roots to test AM fungal colonization.
Activated carbon experiment—AM fungal colonization with inactivated chemicals
In order to verify the chemicals in root exudates play a key role in the relationship between AM fungi and plant roots, we conducted an experiment at Henan University with activated carbon (AC) addition to block bioactivity of root exudate chemicals. We filled plastic pots in mesh cages at Henan University with either 1.3 kg of field soil (N = 78) or field soils amended with activated carbon (N = 78, Sinopharm Chemical Reagent Co., Ltd, Beijing, China) added as 1:500 v:v. We transplanted seedlings from 13 populations (6 native and 7 introduced, Table S1) into the pots with 6 seedlings for each population. Eighteen seedlings died during this experiment (12 seedlings from AC treatment, 6 seedlings from control). After 70 days, we harvested plants and used a fine root sample to determine AM fungal colonization.
Field survey of AM fungal assemblages
We collected rhizosphere soil from 3 sites in China (Dawu, Hubei, 31°28′N, 114°16′E; Wuhan, Hubei, 30°32′N, 114°25′E; Guilin, Guangxi, 25°04′N, 110°18′E) for AM fungal species identification via high-throughput sequencing. At each of these sites, we selected 3 T. sebifera trees per site and dug the soil close to the tree trunk until its root branch was found. We collected soils from 3 roots per plant. We removed the bulk soil from these roots by shaking, and then collected the soil remaining on these roots using brushes (1 new brush per tree). The rhizosphere soils on the roots from same tree were mixed fully. About 5 g of fresh rhizosphere soil from one tree was collected and stored in dry ice and ultra-low temperature freezer (−80 °C) until they were used to test the AM fungi abundance based on high-throughput sequencing [37, 38].
For DNA extraction, microbial DNA was extracted from the prepared samples (0.25 g soil per sample) using the E.Z.N.A.® soil DNA Kit (Omega Bio-tek, Norcross, GA, U.S.) according to the manufacturer’s protocols. The DNA concentration and purification were determined by NanoDrop 2000 UV-vis spectrophotometer (Thermo Scientific, Wilmington, USA), and DNA quality was checked by 1% agarose gel electrophoresis [39].
For the PCR amplification, nested PCR was conducted to amplify the AM fungi 18S rRNA. The primer pairs AML1 (5′-ATCAACTTTCGATGGTAGGATAGA-3′) and AML1 (5′-GAACCCAAACACTTTGGTTTCC-3′) were used in the first run. The primer pairs AMV4.5NF (5′-AAGCTCGTAGTTGAATTTCG-3′) and AMDGR (5′-CCCAACTATCCCTATTAATCAT-3′) were used in the second run in the thermocycler PCR system (GeneAmp 9700, ABI, USA). The PCR reactions were conducted using the program according to Xiao et al. [39].
For each sample, purified amplicons were pooled in equimolar and paired-end sequenced (2 × 300) on an Illumina MiSeq platform (Illumina, San Diego, USA) according to the standard protocols of Majorbio Bio-Pharm Technology Co. Ltd. (Shanghai, China). The raw fastq files were quality-filtered by Trimmomatic and merged by FLASH with the following criteria: (i) the reads were truncated at any site receiving an average quality score <20 over a 50 bp sliding window. (ii) Sequences whose overlap being longer than 10 bp were merged according to their overlap with mismatch no more than 2 bp. (iii) Sequences of each sample were separated according to barcodes (exactly matching) and Primers (allowing 2 nucleotide mismatching), and reads containing ambiguous bases were removed. Operational taxonomic units (OTUs) were clustered with 97% similarity cutoff using UPARSE (version 7.1 http://drive5.com/uparse/) with a novel “greedy” algorithm that performs chimera filtering and out clustering simultaneously. The taxonomy of each 16S rRNA gene sequence was analyzed by RDP Classifier algorithm (http://rdp.cme.msu.edu/) against the Silva (SSU123) 16S rRNA database using a confidence threshold of 70%. Finally, we estimated the abundances of genera as the read numbers of OTUs.
Spore germination experiment—effects of exudates on germination
We examined the effect of crude root exudates on AM fungal spore germination by the paper disk diffusion method according to Akiyama [40]. Because the field survey indicated that Glomus was dominant and Glomus mosseae has been reported as one the most common species [40, 41], we tested this species in our study. We obtained the strain from Bank of Glomales in China (BGC), Institute of Plant Nutrition and Resources, Beijing Academy of Agriculture and Forestry Sciences. We propagated G. mosseae AM fungal spores according to published methods [42].. We collected crude root exudates from seedlings which were planted in soil or Hoagland’s solution. For the seedlings planted in soil, we selected 75 similar size seedlings from 15 populations (9 CH and 6 US). After they grew for 27, 57, or 87 days, we washed them from soil and transferred them individually into 1000 ml of DI water. We collected root exudates after 3 days. For seedlings planted in liquid, we selected 30 of similar size from 10 populations (4 CH and 6 US) washed soil from their roots and put them individually into beakers with 1000 ml of Hoagland’s solution. We pumped air into the solutions with air pump every day to supply enough oxygen to roots and added new Hoagland’s solution every day to provide them enough nutrition. After 87 days of growth, we replaced the solution with DI water for 3 days to collect their root exudates.
To test the spore germination, we first put a paper disk (6 mm in diameter), which was soaked with 20 μl of crude exudate or water, on the center of a dish that contained 0.2% Phytagel and 3 mM MgSO4. We inoculated ten G. mosseae spores in the dish (spaced 15 mm apart). The control was 20 μl of water. Each dish was treated as a replicate, with five replicates per treatment (control and root exudates from every seedling). We placed dishes in an incubator at 28 °C and recorded the total number of germinated spores in each dish after 5, 10, 15, and 20 days.
Spore germination experiment—effects of quercetin and quercitrin on germination
In order to evaluate the bioactivity of quercetin and quercitrin on AM fungi, we tested the effect of different doses of the two chemicals on spore germination. In this experiment, the tested doses of quercetin or quercitrin were 0.02, 0.2, and 2 mg in 20 ml of culture medium in a petri dish after the two chemicals were dissolved in 1 ml of acetone. The culture medium included 0.2% Phytagel and 3 mM MgSO4. We mixed quercetin or quercitrin in culture medium before it became solid (~35 °C). After 12 h, we inoculated ten G. mosseae spores in each petri dish and cultured them at 28 °C in an incubator [43]. The positive control and negative control were 1 ml of water and acetone, respectively. There were five replicates (i.e., dishes) per treatment. We recorded the total number of germinated spores in each dish after 5, 10, 15, and 20 days.
Statistical analyses
We performed mixed model ANOVAs (proc mixed) to examine the dependence of AM fungal colonization or plant mass on the fixed effects population origin, plant age (60 or 90 days), and their interaction and the random effect population (origin) in the common garden experiment. We used partial difference contrast tests to distinguish among means for significant fixed effect factors with more than two levels in this and other ANOVAs.
We used a mixed model ANOVA with the fixed effects plant population origin (US vs. China), plant age (60 or 90 days), and their interactions and the random effect plant population nested in origin to determine their effects on the root mass specific concentrations of flavonoids (quercetin or quercitrin) in root exudates in the root exudate production experiment. These data were log transformed to meet assumptions of ANOVA.
For the exudate application experiment, we used mixed model ANOVAs to test the dependence of mycorrhizal colonization or plant mass on the fixed effects population origin of target recipient plant (“T”, China or US), venue (“V”, China or US experimental block), population origin of exudate donor plant (“E”, China or US), home vs. away soil collection (“H”), and their interactions and the random effects target population nested in target origin, exudate population nested in exudate origin, and soil site (“S”) nested in venue. We fit a family of models that included all combinations of the factors T, V, E, and H. For each set of factors, we first fit full-interaction models then all lower level interaction levels (to main effects only). We used AIC values to select the best fitting models for mycorrhizal colonization and plant mass. Model selection indicated the best fit model for mycorrhizal colonization was the full-interaction model with all factors (Table S2). For mass, the best fitting model was one with only target origin, venue, and their interaction as fixed effects (Table S2).
For the chemical addition experiment, we fit mixed model ANOVAs to test the dependence of mycorrhizal colonization on the fixed effects population origin of target recipient plant (China or US), chemical addition (control, quercetin, quercitrin), concentration nested in chemical addition (0, 0.24 or 2.4 mg), their interaction and the random effect population (origin).
For the activated carbon experiment, we fit a mixed model ANOVA to test the dependence of mycorrhizal colonization on the fixed effects population origin of target recipient plant (China or US), charcoal addition (control or addition), their interaction and the random effect population (origin).
For the exudate and spore germination experiment, we used an ANOVA to test the dependence of AM fungal spore germination odds (ln[number of germinating/number of not germinating]) on root exudate plant origin (US or China), growth medium (soil or liquid), plant age (30, 60, 90 days), days after start of germination (5, 10, 15, 20 days) and their interactions as fixed effects along with the random term population nested in origin along with terms to account for the non-independence from exudates from a single plant being applied to multiple petri dishes and germination being measured repeatedly on the same petri dish.
For the chemical and spore germination experiment, we used ANOVA to test the effect of different concentrations of quercetin and quercitrin along with water or acetone controls (“treatment”) on AM fungal spore germination in two experimental blocks over 4 time periods (categorical variable: 5, 10, 15, or 20 days).
Source: Ecology - nature.com
