Does cryptic microbiota mitigate pine resistance to an invasive beetle-fungus complex? Implications for invasion potential

Microbial symbionts are known to assist exotic pests in their colonization of new host plants. However, there has been little evidence linking symbiotic invasion success to mechanisms for mitigation of native plant resistance. The red turpentine beetle (RTB) was introduced with a fungus, Leptographium procerum, to China from the United States and became a destructively invasive symbiotic complex in natural Pinus tabuliformis forests. Here, we report that three Chinese-resident fungi, newly acquired by RTB in China, induce high levels of a phenolic defensive chemical, naringenin, in pines. This invasive beetle-fungus complex is suppressed by elevated levels of naringenin. However, cryptic microbiotas in RTB galleries strongly degrade naringenin, and pinitol, the main soluble carbohydrate of P. tabuliformis, is retained in L. procerum-infected phloem and facilitate naringenin biodegradation by the microbiotas. These results demonstrate that cryptic microbiota mitigates native host plant phenolic resistance to an invasive symbiotic complex, suggesting a putative mechanism for reduced biotic resistance to symbiotic invasion.

b "T" indicates type strain. Ecologically related strains were given information on isolation source.
c Percentage of similarity of sequences was pairwise compared from Genbank.

Experiment 1
Inoculation procedures: Fungal inoculation was conducted by making a wound on each P. tabuliformis seedling with a sterile 5-mm-diameter cork borer in the main stem 2 cm above the soil line to expose the cambium layer (one seedling with one inoculation point). Eleven fungal associates of RTB in China (categorized into Chinese-invasive, shared, and Chinese-resident groups; Supplementary Table S1 online) were inoculated on seedlings (n = 3 to 5 per treatment) following procedures described above. After three weeks, control tissues and treatment tissues with infected areas (between 5 mm above and below inoculation point) were excised from seedlings for chemical extraction. Fungal isolates were re-isolated from the inoculation areas to confirm that there had been no cross-contamination following methods described 1 .

Experiment 2
We inoculated 4-5-yr-old P. tabuliformis seedlings with one of the four fungal species, including the Chinese-invasive L. procerum and three Chinese-resident H.
pinicola, L. truncatum and L. sinoprocerum following inoculation procedures as described above. Seven to ten seedlings per treatment were uprooted at 3, 6, 9, 12, 18 and 24 d after inoculation. Phloem from 5mm above and below the inoculation points with distinct necrotic areas formation was excised and flash-frozen in liquid nitrogen for chemical extraction. Mock-inoculation seedlings at each time point were sampled in the same manner. Phenolic compounds extraction and identification were followed procedures as described in Experiment 1. The polymeric phenolic compound, lignin, was detected by spectrophotometer (Beckman Coulter, USA) and quantified by standard calibration curves with five different concentrations of lignin (TCI, Japan) according to the methods described before 4 .

Statistics:
The responses of the eight phenolic compounds were tested using two-way ANOVAs (time and isolate as fixed factors). As we sampled tissues on independent seedlings at each time point, we assumed the responses of phenolic compounds to be independent. We then tested difference in concentrations of phenolic compounds among isolates at each time point using one-way ANOVA followed by Bonferroni approach for pair-wise comparisons when isolate in two-way ANOVA was significant. For all ANOVA analyses, we tested the normal distribution

Experiment 4
The boring performance of female RTB to naringenin was evaluated by tunnelling length test according to the protocol 5 with minor modifications. Phloem medium was prepared as described 6  Statistics: In this bioassay, data were analyzed by one-way ANOVAs with treatment as a fixed factor, followed by Bonferroni post hoc tests for pair-wise comparisons.
For ANOVA analyses, we tested the normal distribution (normality diagnostics) and homogeneity (Levene's test) of the variances. When data did not conform to equal variance, we performed one-way Brown-Forsythe's ANOVA 9 , followed by pair-wise comparisons using Dunnett's T3 test 10 .

Experiment 7
We (subtract remaining quantity of naringenin caused by gallery or healthy phloem from initial quantity of naringenin) divided by gram of dry weight of gallery or healthy phloem tissue. HPLC system that was applied here was identical to Experiment 1.
To further determine that this degrading activity was conferred by microbial symbionts distributed in RTB galleries, we used another 17 galleries to test this hypothesis. Gallery tissue was soaked in 1 ml of 10% PBS buffer followed by subsequent procedures as shown before, and then 0. Statistics: Pinitol is a main soluble carbohydrate in P. tabuliformis phloem; we thus focused on it for further studies. The change in pinitol levels was tested using two-way ANOVA (time and isolate as fixed factors). As we sampled tissues on independent seedlings at each time point, we assumed the change of pinitol to be independent. Levels of pinitol in pine phloem among isolates at each time point was further compared using one-way ANOVA, followed by S-N-K post-hoc test for pair-wise comparisons. For ANOVA analysis, we tested the normal distribution (normality diagnostics) and homogeneity (Levene's test) of the variances. When data did not conform to conditions of parametric test, we performed nonparametric Kruskal-Wallis one-way ANOVAs, followed by pair-wise comparisons by Mann-Whitney U tests using Bonferroni correction to adjust the probability (α = 0.05/10 = 0.005). We used SPSS 20 for the statistical procedures.

Experiment 9
To compare the naringenin biodegrading activity and the growth of gallery microbiota in a mixture of naringenin and pinitol with those in naringenin alone, two kinds of inorganic culture solution, one containing only 1 mM naringenin, the other one containing 1 mM naringenin and 5 mM pinitol, were prepared. Eighteen RTB galleries were macerated with plastic pestles in 0.5 ml 10% PBS buffer and then 40 μl of crude extract from the same gallery was transferred into 1 ml of those two kinds of solutions, respectively. After incubation for 72 h with shaking, the reaction was stopped by adding 1 ml of methanol. Using the extraction method described in Experiment 7, 1ml methanol supernatant containing remaining naringenin was obtained for analysis by HPLC using the same conditions as in Experiment 1. Lower quantities of remaining naringenin indicated greater naringenin degradation capacity.
Pellets from centrifugation were kept for measurement of growth of microbiota.
Pellets were washed three times by sterile ddH 2 O and re-suspended in 1 ml sterile ddH 2 O for detection at OD 600 by the microplate reader.

Statistics:
We performed a nonparametric two-related samples test (Wilcoxon signed ranks test) to compare quantity of remaining naringenin caused by microbiota in culture solutions with or without pinitol. Growth of the microbiota in naringenin with or without pinitol was compared by the parametric paired-samples T test.

Relationships between stem diameter and pinitol concentration of pine seedlings inoculated with RTB fungal associates
To test the effect of tree size on induction of retention of pinitol by each of the four fungal species (L. procerum, H. pinicola, L. truncatum and L. sinoprocerum), P.
tabuliformis seedlings with ascending diameters across pine ages were inoculated following procedures as described in Experiment 3. Eight to ten seedlings per treatment in each pine age were uprooted at 24 d after inoculation. We excised the phloem tissues near the inoculation points and then flash froze them in liquid nitrogen for pinitol extraction and quantification following procedures as described in Experiment 8. The PCR products were used for sequencing after verification by 1% agarose gel electrophoresis. Consensus sequences were assembled in MEGA 5 with manual editions according to chromatograms visualized in Chromas. The accession numbers of sequences registered in the GenBank database were listed in Supplementary Table   S3 and S4 online.

Naringenin biodegradation by bacteria and yeasts isolated from RTB galleries
Representative strains of each species identified in the above experiment were transferred to tubes with 2 ml of TSB (for bacterial strains) or YMEB (0.3% yeast extract, 0.3% malt extract, 1% glucose, 0.5% tryptone and 2% agar; for yeast strains), and incubated at 30 ºC for 24 h. We adjusted the OD 600 values of culture solutions for all strains to 0.5, and then 10 μl of each adjusted solution was added into 1 ml inorganic culture solutions containing 1 mM naringenin with and without 5 mM pinitol. In addition, 1 ml of 1 mM naringenin alone was set as control. Five replicates for control and strains (treatments) were applied. After 30 ºC incubation for 72 h with shaking (150 rmp), remaining naringenin was extracted from each culture solution following steps in Experiment 7, re-dissolved in 1 ml methanol and then analyzed by HPLC using identical conditions as in Experiment 1.
Statistics: Quantity of remaining naringenin among control and treatments were performed by one-way Brown-Forsythe's ANOVA for yeast group and bacterial group, respectively, allowing unequal variances. Post hoc pair-wise comparisons were done using Dunnett's T3 test. Quantity of remaining naringenin caused by individual strain with or without pinitol was compared by the parametric independent-samples T test.

Field survey for the Chinese-invasive fungus and its association with naringenin-biodegrading microbes in RTB galleries
The remaining portions of 111 RTB galleries in Experiment 7 were performed for isolation and identification of L. procerum. Each gallery phloem was cut into small pieces using sterile surgical scissors, and then directly placed on the surface of media selective for ophiostomatoid fungi (2% MEA amended with 0.05% cycloheximide and 0.04% streptomycin) in Petri dishes, which were sealed with parafilm and incubated at 25 ºC. We checked daily and carefully transferred presumed hyphae of L. procerum onto new media to obtain a pure culture when observed. Further identification of isolated fungal strains was carried out by comparing morphological characteristics with the stock isolate (CMW 25626) on 2% malt extract agar medium (2 g malt extract, 2 g agar, and 100 ml double distilled water), 2% agar medium (2 g agar, 100 ml double distilled water), pine twig medium (2 g malt extract, 2 g agar medium, 100 ml double distilled water and one pine twig), and 1.5% oat agar medium (1.5 g oat powder, 2 g agar and 100 ml double distilled water). The combined information of naringenin-biodegrading microbes in each gallery with the isolation results of L. procerum here, led us to investigate the closeness of their association in the field.