Relationship between Expression of Chalcone Synthase Genes and Chromones in Artificial Agarwood induced by Formic Acid Stimulation Combined with Fusarium sp. A2 Inoculation

Agarwood (gaharu) is a fragrant resin produced in the heartwood of resinous Gyrinops and Aquilaria species. Artificial agarwood samples were obtained from Aquilaria sinensis (Lour.) Gilg using formic acid (FA) stimulation combined with Fusarium sp. A2 inoculation. The relationship between the expression of chalcone synthase genes (CHS) and dynamic changes in chromone content was explored in resin-deposited parts of the trunks of A. sinensis. CHS gene expression levels were detected by qRT-PCR analysis. The chemical composition of agarwood obtained from the heartwood of A. sinensis before and within 1 year after induction was determined by GC-MS. After induction with FA stimulation combined with F. sp. A2 inoculation, the CHS1 gene showed relatively high expression, whereas the CHS2 gene showed low expression. The relative gene expression level of CHS1 peaked at 12 months, with a 153.1-fold increase, and the dominant period of the CHS2 gene expression was 10 months with a 14.13-fold increase. Moreover, chromones were not detected until after 2 months, and a large proportion of chromone compounds were detected after 4 months. Chromone content increased with time and peaked at 12 months. CHS1 gene expression was significantly correlated with 6-hydroxy-2-(2-phenylethyl)chromone accumulation, and CHS2 gene expression was significantly correlated with 5-hydroxy-6-methoxy-2-(2-phenylethyl)chromone accumulation. CHS gene expression was extremely sensitive to FA stimulation combined with F. sp. A2 inoculation and responded to late-onset injury. CHS genes expression also preceded the chromone accumulation. This work laid the foundation for studies on the mechanism by which genes regulate chromone biosynthesis pathways during the formation of agarwood resin in A. sinensis.


Introduction
Agarwood (gaharu) is a fragrant resin collected from resinous wood of Gyrinops and Aquilaria species. Agarwood is widely used as an effective traditional Chinese medicine and used to produce production in a 6-year-old A. sinensis [14,15,19]. The heartwoods of A. sinensis with or without resin were extracted before induction and within 1 year after induction, including seven time points. The relationship between CHS gene expression level and chromone content was tested and evaluated, and the results lay the foundation for studies on the mechanism regulating the chromone biosynthetic pathway during formation of agarwood resin in A. sinensis.

Analysis of the Accumulation of Secondary Metabolite Compounds
The three parallel samples collected at 0, 2, 4, 6, 8, 10, and 12 months after induction were analyzed by GC-MS, and the total ion chromatograms showed the changes in the composition and accumulation of compounds in artificial agarwood induced through FA stimulation combined with F. sp. A2 inoculation (Figure 1). the foundation for studies on the mechanism regulating the chromone biosynthetic pathway during formation of agarwood resin in A. sinensis.

Analysis of the Accumulation of Secondary Metabolite Compounds
The three parallel samples collected at 0, 2, 4, 6, 8, 10, and 12 months after induction were analyzed by GC-MS, and the total ion chromatograms showed the changes in the composition and accumulation of compounds in artificial agarwood induced through FA stimulation combined with F. sp. A2 inoculation ( Figure 1).
The components of 21 samples eluted within the total ion chromatogram were extracted in the Automatic Mass Spectral Deconvolution and Identification System (AMDIS), and sesquiterpenes and chromone compounds were identified by comparing the resolved mass spectra with those of the standards in the National Institute of Standards and Technology (NIST) Mass Spectral Library (05) (Tables 1 and 2). The percentages of the total ion current were determined using the area normalization method. The main component was sesquiterpenes before 100 min and chromones after 169 min [33]. Overlapping GC-MS chromatogram for A. sinensis at pre-and post-induction by formic acid (FA) stimulation combined with F. sp. A2 inoculation at different time points. S1-S3: Before induction; S4-S6: 2 months after induction; S7-S9: 4 months after induction; S10-S12: 6 months after induction; S13-S15: 8 months after induction; S16-S18: 10 months after induction; S19-S21: 12 months after induction.
The components of 21 samples eluted within the total ion chromatogram were extracted in the Automatic Mass Spectral Deconvolution and Identification System (AMDIS), and sesquiterpenes and chromone compounds were identified by comparing the resolved mass spectra with those of the standards in the National Institute of Standards and Technology (NIST) Mass Spectral Library (05) ( Tables 1 and 2). The percentages of the total ion current were determined using the area normalization method. The main component was sesquiterpenes before 100 min and chromones after 169 min [33].  As shown in Tables 1 and 2, 19 chromones and 16 sesquiterpenes were detected throughout the 12-month observation period. Chromones and sesquiterpenes were not identified in the artificial agarwood samples until 2 months after induction. A maximum of 23 secondary metabolite components with a relative percentage of 27.32% were identified in the samples after four months. The major chrome compounds (6.67%) included chromones 2-(2-phenylethyl) (2.20%), 6,7-dimethoxy-2-(2-phenylethyl) (1.95%), and 6-methoxy-2-(2-phenylethyl) (1.85%). A total of 29 secondary metabolite compounds were identified in the samples after six months and comprised 22.84% of chromones and 8.89% of sesquiterpenes. Up to 29 secondary metabolite components were identified in artificial agarwood samples induced by FA stimulation combined with F. sp. A2 after eight months, consisting of 42.29% chromones and 8.763% sesquiterpenes. A total of 13 chromones and 12 sesquiterpenes were identified in the artificial agarwood samples induced by FA stimulation combined with F. sp. A2 after 10 months. The accumulation of chromones reached the maximum in the agarwood samples artificially treated for 12 months, with a relative percentage of 44.87%.

Relative Expression of Candidate Genes in Agarwood Samples Induced by FA Stimulation Combined with F. sp. A2 Inoculation
To investigate the expression patterns of the genes related to chromone metabolism, we analyzed the expression level of CHS1 and CHS2 genes as shown in Figure 2. In the first 10 months, the relative expression level of CHS1 gene in the treated groups was below 25-fold. At the 12-month time point, the relative expression level increased markedly to 153.1-fold. The relative expression levels of CHS2 gene were low in the artificial agarwood induced by FA stimulation combined with F. sp. A2 inoculation from 0 to 8 months, ranging from 0.4585-fold to 2.676-fold. The largest increase in the expression levels of CHS2 gene was observed at 10 months with a 14.13-fold value. However, the relative expression level decreased to 1.437 after 12 months. Tables 1 and 2, 19 chromones and 16 sesquiterpenes were detected throughout the 12-month observation period. Chromones and sesquiterpenes were not identified in the artificial agarwood samples until 2 months after induction. A maximum of 23 secondary metabolite components with a relative percentage of 27.32% were identified in the samples after four months. The major chrome compounds (6.67%) included chromones 2-(2-phenylethyl) (2.20%), 6,7-dimethoxy-2-(2-phenylethyl) (1.95%), and 6-methoxy-2-(2-phenylethyl) (1.85%). A total of 29 secondary metabolite compounds were identified in the samples after six months and comprised 22.84% of chromones and 8.89% of sesquiterpenes. Up to 29 secondary metabolite components were identified in artificial agarwood samples induced by FA stimulation combined with F. sp. A2 after eight months, consisting of 42.29% chromones and 8.763% sesquiterpenes. A total of 13 chromones and 12 sesquiterpenes were identified in the artificial agarwood samples induced by FA stimulation combined with F. sp. A2 after 10 months. The accumulation of chromones reached the maximum in the agarwood samples artificially treated for 12 months, with a relative percentage of 44.87%.

Relative Expression of Candidate Genes in Agarwood Samples Induced by FA Stimulation Combined with F. sp. A2 Inoculation
To investigate the expression patterns of the genes related to chromone metabolism, we analyzed the expression level of CHS1 and CHS2 genes as shown in Figure 2. In the first 10 months, the relative expression level of CHS1 gene in the treated groups was below 25-fold. At the 12-month time point, the relative expression level increased markedly to 153.1-fold. The relative expression levels of CHS2 gene were low in the artificial agarwood induced by FA stimulation combined with F. sp. A2 inoculation from 0 to 8 months, ranging from 0.4585-fold to 2.676-fold. The largest increase in the expression levels of CHS2 gene was observed at 10 months with a 14.13-fold value. However, the relative expression level decreased to 1.437 after 12 months. After being induced through FA stimulation combined with F. sp. A2 inoculation, the CHS1 gene showed a relatively high expression level, whereas that of the CHS2 gene was low. On the basis of CHS After being induced through FA stimulation combined with F. sp. A2 inoculation, the CHS1 gene showed a relatively high expression level, whereas that of the CHS2 gene was low. On the basis of CHS gene expression level, we speculated that the CHS1 gene was specifically expressed in phenylalanine pathways after induction through FA stimulation combined with F. sp. A2 inoculation.
The orthogonal partial least squares discriminant analysis (OPLS-DA) (R 2 X = 0.652, R 2 Y = 0.922, Q 2 (cum) = 0.622) score plot ( Figure 3A) showed that artificial agarwood was evidently separated into three groups according to its induction period: samples obtained 0, 2, 4, and 6 months after induction, those obtained 8 and 10 months after induction, and those obtained 12 months after induction.

Discussion
In this study, we used the FA stimulation combined with F. sp. A2 inoculation to produce artificial agarwood in the field. GC-MS combined with AMDIS and retention index (RI) correction index were used to study the variations in chemical components during agarwood formation. There is a clear trend in the approximate timing of chromone and sesquiterpene compounds increase, however, there seems to be a large difference in the GC-MS data. This may be due to individual variation caused by uncontrollable factors from field experiments. Chromone metabolites were not detected in the trunks in the first 2 months of induction; however, they were detected in the other four time points, and showed an increasing tendency. 6,7-dimethoxy-2-(2-phenylethyl)chromone showed the highest frequency. Lin et al. [34] detected 2-(2-phenylethyl)chromone compounds in agarwood after 1 year of artificial induction through fungal inoculation (Melanotus flavolivens) compared with that after a half-year induction. Qi et al. [24] found that chromone contents increased with time after artificial induction by M. flavolivens; the relative content of chromone in artificial agarwood also increased but not significantly.
The CHS gene is an inducible expression gene and can be induced after a plant is subjected to injury. CHS activation will be significantly enhanced and then CHS genes will be actively transcribed,

Discussion
In this study, we used the FA stimulation combined with F. sp. A2 inoculation to produce artificial agarwood in the field. GC-MS combined with AMDIS and retention index (RI) correction index were used to study the variations in chemical components during agarwood formation. There is a clear trend in the approximate timing of chromone and sesquiterpene compounds increase, however, there seems to be a large difference in the GC-MS data. This may be due to individual variation caused by uncontrollable factors from field experiments. Chromone metabolites were not detected in the trunks in the first 2 months of induction; however, they were detected in the other four time points, and showed an increasing tendency. 6,7-dimethoxy-2-(2-phenylethyl)chromone showed the highest frequency. Lin et al. [34] detected 2-(2-phenylethyl)chromone compounds in agarwood after 1 year of artificial induction through fungal inoculation (Melanotus flavolivens) compared with that after a half-year induction. Qi et al. [24] found that chromone contents increased with time after artificial induction by M. flavolivens; the relative content of chromone in artificial agarwood also increased but not significantly.
The CHS gene is an inducible expression gene and can be induced after a plant is subjected to injury. CHS activation will be significantly enhanced and then CHS genes will be actively transcribed, transmitted, and amplified in vivo. CHS is regulated by a phenylalanine metabolism pathway to promote synthesis and release of chromone compounds [29]. Wang et al. [31] observed that chalcone synthase may regulate the biosynthesis of 2-(2-phenylethyl)chromones in response to salinity stress involved in A. sinensis calli. Ye et al. [32] performed the transcriptome analysis of different parts of artificial agarwood induced by formic acid. They found that the expression level of chalcone synthase in the agarwood part was much higher than that in the white wood part. After being induced by FA stimulation combined with F. sp. A2 inoculation, CHS gene expression level initially increased, then decreased, and finally increased. Our result showed that CHS1 gene was extremely sensitive to FA stimulation combined with F. sp. A2 inoculation, and CHS1 and CHS2 genes both responded to late-onset injury. Several studies showed that flavonoid accumulation paralleled the transcription level of change in the CHS gene [35][36][37]. Our study also showed the same trend in the agarwood samples induced by a chemical method plus fungal pathogen infection before 10 months. However, the CHS1 gene had the highest level of transcription at the last time point, while chromone accumulation reached a platform period. This suggests that the CHS1 gene may be turned into another biosynthesis pathway when the specific accumulation of chromone compounds reaches a steady state. Zhang et al. [38] demonstrated that the whole anthocyanin pathway was diverted to the synthesis of chlorogenic acid and its complexes when the expression of the CHS gene was severely repressed. Hoffman et al. [39] showed that the flavonoid pathway was shunted to the phenylpropanoid pathway when the CHS gene activity was reduced.

Source of Plant Materials
Materials were obtained from 6-year-old A. sinensis trees growing in a farm in Xinyi suburban district in Guangdong Province, China. These trees were identified as A. sinensis by Prof. Yan (College of Traditional Medicine, Guangdong Pharmaceutical University). Fusarium sp. A2 (EU781659) [40], provided and identified as F. sp. by Prof. Zhang (Guangdong Institute of Microbiology), was used to inoculate A. sinensis trees. Artificial agarwood from A. sinensis was induced through FA stimulation combined with F. sp. A2 inoculation through pinhole instillation according to the method of Gao et al. [33].
Induction was performed on the 24 June 2012, and wood chips containing embedded black resin were collected. The acquisition timeline was 0 month before induction and 2, 4, 6, 8, 10, and 12 months after induction. Three parallel samples were collected at each time point. Physical damage caused by manual removal of resin can induce agarwood production, so samples of different time points came from different A. sinensis trees individually. Samples from the same tree were divided into two portions. One portion was immediately wrapped in tinfoil and stored in liquid nitrogen for quantitative gene expression analysis, whereas the other portion was directly dried for GC-MS analysis.

Gas Chromatography Mass Spectrometry Analysis
All dried heartwood samples were cut into small pieces and filtered through 40-mesh sieves. The powder samples (0.5 g) were extracted in chloroform (10 mL, 24 h) at room temperature. The solvent was evaporated in a water bath (80 • C), then reconstituted to 2 mL chloroform and stored in a dark and air tight vial at 4 • C.
GC-MS analysis was performed using a GCMS QP-2010E (Shimadzu, Kyoto, Japan) equipped with a Rtx-5MS (Restek Corp., Bellefonte, PA, USA) capillary fused silica column (30 m × 0.25 mm I.D. × 0.25 µm film thickness), and operated in the electron ionization (EI) mode (70 eV). Helium was the carrier gas, with the flow rate of 1 mL/min. The operating parameters were the temperature program of 90 • C for 4 min, ramp of 2.5 • C/min up to 160 • C (5 min), then increased to 180 • C with a 0.3 • C/min heating vamp, kept for 5 min, and then ramp of 2.0 • C/min up to 200 • C. Subsequently it was increased to 230 • C with a 1 • C/min heating vamp, and kept at 230 • C for 120 min. A 1 µL sample solution was injected. The injections were performed in a 1:30 split ratio at 230 • C. The m/z values were recorded in the range of 50-500 amu. A 1 µL C 10 -C 31 sample was injected separately and was run in the same program as the heartwood samples.

Quantitative Real-Time PCR (qRT-PCR)
To analyze gene expression levels, total ribonucleic acid (RNA) was isolated from heart wood samples as previously described [41]. Reverse transcription was performed using a ReverTra Ace qCR RT kit Master Mix with gDNA Remover (TOYOBO, Osaka, Japan) with specific primers (Supplementary Table S1). The specificity of the oligonucleotide sequences, in relation to its annealing efficiencies, was evaluated using the Primer 5.0 program in advance. A fragment of the glyceraldehyde-3-phosphate dehydrogenase (GADPH) gene was also amplified as a blank control [42]. The qPCR analysis was performed using an ABI 7500 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA), using a THUNDERBIRD SYBR qPCR Mix (Toyobo, Osaka, Japan). The Ct (threshold cycle) was used to measure the starting copy numbers of each target gene, and was detected by the ABI 7500 Real-Time PCR System. Relative quantitation of each target gene expression level was performed using the comparative 2 −∆∆Ct method [43]. All experiments were conducted in triplicate.

Data Processing and Statistics Analysis
The components were identified based on the comparison of their retention indices and mass spectra as previously described [33,42]; moreover, we used NIST MS search 2.0 with the database of NIST 05 after elution within the total ion chromatogram (TIC) extracted in the AMDIS. Retention indices were calculated using a series of n-alkanes (C 10 -C 31 ). Correlations between chemical components and expression of the CHS gene were analyzed by the Pearson's correlation analysis of SPSS (20.0 for Windows). Peak-area percentage of chromone compounds and relative expression of CHS were examined by OPLS-DA using SIMCA-P 12.0 software.

Conclusions
2-(2-Phenylethyl)chromone compounds, which are the main characteristic components of agarwood, are closely related to the plant self-defense response. Defensive substances, such as sesquiterpenes and chromones with bacteriostatic activity, will be produced when a plant is subjected to injury or fungal infection. These defensive substances can protect A. sinensis from further injury. Recent investigations have revealed that in the early stage of injury, a plant mainly releases a large proportion of volatile compounds derived from lipoxygenase and then releases chromone compounds later. Chromone content increases with continuous injury [44]. In this study, we used an agarwood inducing method involving FA stimulation combined with F. sp. A2 inoculation to induce agarwood production in individual A. sinensis trees. The relationship between the expression of chalcone synthase gene (CHS) and dynamic changes in chromone content was explored by using qRT-PCR and GC-MS analysis. The present study showed that CHS gene expression preceded the chromone accumulation after induction through FA stimulation combined with F. sp. A2 inoculation. Chromone compounds were not detected in the artificially induced trunk until after 2 months, even when CHS gene expression had already begun. Chromone compounds were detected 4 months after induction. The release of chromone compounds was closely related to CHS expression in agarwood formation. Therefore, we can regulate CHS expression to produce chromone compounds in the future, which would lay the foundation for highly efficient artificial agarwood production and elucidation of the mechanism of agarwood formation in A. sinensis.
Supplementary Materials: The supplementary materials are available online.