Introduction

Pinoresinol diglucoside (PDG), (+)-1-pinoresinol 4, 4′-di-O-β-D-glucopyranoside, is a major antihypertensive compound found in Tu-Chung, a traditional herb medicine with excellent efficacy for lowering blood pressure1. PDG also possesses the potential to prevent osteoporosis2. Additionally, in the human intestine, PDG can be converted to enterolignans by intestinal microflora3, and enterolignans have potential to reduce the risk of breast cancer4 and other estrogen-dependent cancers5.

PDG is found primarily in plants as lignans1,6 but yields are very low. Phomopsis sp. XP-8 is an endophytic fungus isolated from the bark of Tu-Chung that was previously found to produce PDG in vitro7, thus, providing an alternative resource to obtain PDG. This is the first report on the capability of a microorganism to synthesize lignan. However, the PDG production by Phomopsis sp. XP-8 was very low, which might be enhanced by regulatory controls based on the biosynthetic pathways. Therefore, it is essential to identify the PDG biosynthetic pathway in this strain.

The lignan biosynthetic pathway has only been reported in plants until now8,9. Synthesis of Pin in plants occurs via oxidative coupling of monolignols, which are synthesized through the phenylpropanoid pathway with Phe, Ca, p-Co, p-coumaroyl-CoA, caffeate, ferulate, feruloy-CoA, coniferylaldehyde, and coniferyl alcohol as intermediates or precursors10,11 (Fig. 1). PMG and PDG are converted from Pin by UDP-glucose-dependent glucosyltransferase8. However, the biosynthesis of PDG from Pin has not been detected in plants and the Pin, PMG, and PDG biosynthetic pathways have not been elucidated in microorganisms.

Figure 1
figure 1

Biosynthetic pathways leading to lignans in plants. The steps which have been detected in plants were shown as solid arrow: synthesis of pinoresinol (Pin) occurs via oxidative coupling of coniferyl alcohol; coniferyl alcohol can be synthesized from glu through the phenylpropanoid pathway10; PMG and PDG are converted from Pin by UDP-glucose-dependent glucosyltransferase8. while the biosynthesis of PDG from Pin has not been detected in plants, so they are connected with dashed arrow. The substrates which have been found in the fermentation medium of Phomopsis sp. XP-8 in our preliminary experiment were labeled as red.

Full size image

We previously reported that Phomopsis sp. XP-8 converts mung bean starch and polysaccharides to Pin, PMG, and PDG. Phe, cinnamic acid, and p-coumaric acid have been detected as products of the bioconversion12,13. Precursor feeding and enzymatic activity measurements indicate that this strain synthesizes PDG via many steps, such as during mass flow of the phenylpropanoid pathway14. Genomic annotation indicates that the phenylpropane pathway exists in this strain15 and some other microorganisms16. However, the functions of the denoted genes have not been verified until now. Therefore, it is necessary to verify the entire PDG biosynthetic pathway in Phomopsis sp. XP-8.

Using stable or radioactive isotope-labeled compounds is an efficient and reliable strategy to verify the mass flow of unknown biosynthetic pathways by tracing the isotope-labeled compounds from substrates to products17. 13C-labeled substrates have been used to shed light on the biodegradation pathways of organic pollutants18. Isotope labeling combined with high-resolution mass spectrometry have also been used to track the abiotic transformation of pollutants in aqueous mixtures19. In recent years, liquid chromatography-mass spectrometry (LC–MS) and ultra-high-performance liquid chromatography (UPLC) systems have been developed to facilitate the analysis of many substances at the same time with high sensitivity and selectivity20. Stable isotope-labeled compounds have also been employed in several areas of biomedical research21. The combination of stable isotope-labeling techniques with MS has allowed rapid acquisition and interpretation of data and has been used in many fields, including distribution, metabolism, food, and excretion studies22,23,24. The biochemical pathway of the aromatic compounds in tea has been also been revealed using the stable isotope labeling method25.

In this study, we applied stable isotopes labeling and MS to trace the PDG biosynthetic pathway. Stable isotope-labeled 13C6 glu and 13C6 Phe were used as the substrates and ultra-high-performance liquid chromatography-quadrupole time of flight mass spectrometry (UPLC-Q-TOF-MS/MS) was used to identify the products.

Materials and Methods

Microorganism and chemicals

Phomopsis sp. XP-8 previously isolated from the bark of Tu-Chung and stored at the China Center for Type Culture Collection (Wuhan, China) (code: Phomopsis sp. CCTCC M 209291) was used in this study.

Phe (purity ≥98%, Sigma, St. Louis, MO, USA), Ca and p-Co (purity ≥98%; Aladdin, Shanghai, China), PDG, PMG, and Pin (purity ≥99%; National Institutes for Food and Drug Control, Beijing, China) were used as the standards (dissolved in methanol) for the structural analysis and product identification. [13C6]-Labeled Phe and glu were purchased from the Qingdao TrachinoidCo (≥99%; Qingdao, China). The purity of the [13C6]-labeled Phe and glu was 99%. Methanol (HPLC grade) was purchased from Fisher Scientific (Fairlawn, NJ, USA). The water used in the experiment was purified using a Milli-Q water purification system (18.5 M) (Millipore Corp., Bedford, MA, USA). Other reagents and chemicals were of analytical grade.

Preparation of Phomopsis sp. XP-8 cells

Phomopsis sp. XP-8 was grown at 28 °C on potato dextrose agar plates for 5 days. Then, three pieces of mycelia (5 mm in diameter) were inoculated into 100 mL liquid potato dextrose broth in a 250-mL flask and cultivated at 28 °C on a rotary shaker (180 rpm). After 4 days, the cells were collected by centrifugation at 4 °C (1,136 × g for10 min) using a centrifuge (HC-3018R, Anhui USTC Zonkia Scientific Instruments Co., Ltd., Anhui, China). The cells were washed twice with sterile water and used for bioconversion according to the experimental design.

Bioconversion systems

The bioconversion with unlabeled glu as the sole substrate was carried out in a 250-mL flask containing 100 mL of ultrapure water (pH 7), 5 g/L glu, and the prepared Phomopsis sp. XP-8 cell set a ratio of 10 g cells (wet weight) per 100 mL medium. To track the mass flow from glu to PDG, glu was changed to 5 g/L [13C6]-labeled glu in the above medium and the same conditions were used for bioconversion.

Bioconversion with Phe as the sole substrate was carried out in medium without glu, 7 mM [13C6]-labeled phenylalanine, and the prepared Phomopsis sp. XP-8 cells at a ratio of 10 g wet cells per 100 mL medium.

All bioconversions were carried out for 48 h at 28 °C and 180 rpm. At the end of bioconversion, the broth was collected and filtered through an intermediate speed qualitative filter paper before the products were detected.

Identification of the accumulated products during bioconversion

The products were extracted from the vacuum-evaporated (0.09 MPa, 50 °C) bioconversion broth with methanol and adjusted to 4 mL for the UPLC measurements after filtration through a membrane (0.45 µm, 13 mm diameter; Millipore, Billerica, MA, USA). The UPLC analysis was performed on a Waters Acquity UPLC system (Waters Corp., Milford, MA, USA), equipped with a binary pump, a thermostatically controlled column compartment, and a UV detector. Gradient elution was performed on an Acquity UPLCTM BEH C18 column (50 mm × 2.1 mm I.D., 1.7 m; Waters) and the column temperature was maintained at 30 °C, while sample temperature was 10 °C13.

The MS analysis of the products was carried out on a Q-TOF PremierTM with an ESI source (Waters Corp.) at the optimized parameters of: capillary voltage, 2.8 kV; sampling cone voltage, 20 V; extractor voltage, 4 V; source temperature, 100 °C; desolvation temperature, 250 °C, and flow rate of the desolvation gas (N2), 400 L/h. The collision cell parameters for the Q-TOF-MS/MS analysis were: collision gas (Argon) flow rate, 0.45 L/h; collision energy, 15–35 eV. The mass spectra were recorded using full scan mode over a mass range of m/z 100–800 in negative ion mode. The MS acquisition rate was set to 1.0 s, with a 0.02 s interscan delay. The Q-TOF-MS/MS experiments were carried out by setting the quadrupole to allow ions of interest to pass prior to fragmentation in the collision cell.

Accurate mass measurements were obtained by means of a lock mass that introduces a low flow rate (3 L/min) of a chrysophanol (253.0499) calibrating solution in the ESI-Q-TOF-MS and ESI-Q-TOF-MS/MS. All operations and acquisition and data analyses were controlled by Masslynx V4.1 software (Waters Corp.).

Data processing

Peak detection, alignment, and identification of the detected compounds were performed using Masslynx V4.1 software (Waters Corp.). The MS/MS fragmentation patterns were used for informative non-targeted metabolic profiling of the LC-MS data, and the acquired LC-MS/MS spectrum was identified after comparison with spectra proposed by the Mass bank database (www.massbank.jp), the KEGG database, and related reports.

Results

Detection of products converted from unlabeled glu

Production of PDG, PMG, Pin, Phe, p-Co, and Ca were detected in bioconversion systems using glu as the sole substrate. Data in Figs. 2–5 show the mass spectra of these compounds accumulated in the bioconversion systems and the corresponding standards.

Figure 2
figure 2

Total ion current chromatogram and mass spectrum of phenylalanine (A), cinnamic acid (B), p-coumaric acid (C). (A-1, B-1, and C-1 show the total ion current chromatogram of standard phenylalanine, cinnamic acid and p-coumaric acid respectively; A-2, B-2, and C-2 show mass spectrum of standard phenylalanine, cinnamic acid and p-coumaric acid respectively; A-3, B-3, and C-3 show the total ion current chromatogram of phenylalanine, cinnamic acid and p-coumaric acid in the samples, respectively; A-4, B-4, and C-4 show mass spectrum of phenylalanine, cinnamic acid and p-coumaric acid in the samples, respectively. Ion reaction were set to m/z = 164, m/z = 147 and m/z = 163 respectively.)

Full size image
Figure 3
figure 3

Total ion current chromatogram and mass spectrum of PDG (A–C are the total ion current chromatogram, precursor ions, and daughter ions of standard PDG, respectively; D–F are the total ion current chromatogram, precursor ions, and daughter ions of the samples, respectively).

Full size image
Figure 4
figure 4

Total ion current chromatogram and mass spectrum of PMG (A,B show the total ion current chromatogram and mass spectrum of standardpinoresinol-4-O-β-D-glucopyranoside, respectively; D–F show the total ion current chromatogram and the mass spectrum ofpinoresinol-4-O-β-D-glucopyranosideinsamples, respectively. Ion reaction was set to m/z = 518.5–519.5).

Full size image
Figure 5
figure 5

Total ion current chromatogram and mass spectrum of Pin (A–C show the total ion current chromatogram, precursor ions, and daughter ions of the Pin standard, respectively; D–F show the total ion current chromatogram, precursor ions, and daughter ions of Pin in the samples, respectively. Ion reaction was set to m/z = 356.5–357.5).

Full size image

Production of Phe was detected as m/z = 164.08, and m/z = 147.06 (Fig. 2A-4), which was consistent with the data obtained from the corresponding standards (Fig. 2A-2). Similarly, production of PDG, PMG, Pin, p-Co, and Ca was also detected in the bioconversion system, indicating that glu was converted to these products by Phomopsis sp. XP-8, as only glu was provided in the bioconversion system.

Identification of products converted from [13C6]-labeled Phe

The phenylpropanoid pathway in plants starts with Phe and ends with p-Co. The same mass flow was previously detected during PDG biofrom glu by Phomopsis sp. XP-813. To verify this finding and the role of the Phe pathway in the biosynthesis of PDG, PMG, and Pin, [13C6]-labeled Phe was used as the sole substrate in the bioconversion system without glu (mainly used as the glucoside donor). As results, 13C labeled Pin, Phe, p-Co, and Ca were successfully detected (Fig. 6). The products were successfully detected at the same RT of their corresponding unlabeled standard substrates. All 13C-labeled product data and their corresponding standard substrates are summarized in Table S1 (Supporting information).

Figure 6
figure 6

Mass spectrum of phenylalanine (A), cinnamic acid (B), p-coumaric acid (C), and Pin (D) in the resting cell system using [13C6]-labeled phenylalanine as the substrate.

Full size image

As shown in Fig. 6B, 13C-labeled Ca was detected as m/z = 153.07, indicating that six 13C from [13C6]-labeled Phe were incorporated into Ca. A daughter ion of 13C-labeled Ca was obtained at m/z = 109.08, indicating six 13C referring to the standard Ca (m/z = 103.06). The structure of 13C-labeled Ca without -COO− was observed at m/z = 109.08. Therefore, it was deduced that the six 13C were incorporated into the benzene ring of Ca not into –COO−.

P-Co produced in the conversion system was detected as m/z = 169.05 and revealed six 13C by consulting the p-Co standard (Fig. 6C). A daughter ion of 13C-labeled p-Co was obtained at m/z = 125.07, indicating 6 Da mass shift than p-Co standard (m/z = 119.06). The structure of 13C-labeled p-Co without –COO− (44 Da lost) was observed at m/z = 125.07. Therefore, it was deduced that the six 13C might be distributed in the benzene ring.

13C-labeled Pin was detected (Fig. 6D-1) and compared with the mass spectra of the Pin standard (C20H22O6, RT = 9.736 min, detected as m/z = 357.13 and m/z = 151.04 respectively) (Table S2, Supporting information). 13C-labeled Pin was detected as m/z = 369.05, indicating 12 Da mass shift than Pin standard (m/z = 357.13). A daughter ion of 13C-labeled Pin was observed at m/z = 157.06, which showed a mass increase of 6 Da than Pin standard (m/z = 151.04). The structure of 13C-labeled Pin with loss of a benzene ring was identified as the major daughter ion of m/z = 157.06 (Fig. 6D-1). This result confirmed that the six13C were distributed in a benzene ring, whereas the other six13C might be in a symmetrical benzene ring. Therefore, we deduced that the Pin with 12 13C was bio-converted from the [13C6]-labeled Phe, Ca, or/and p-Co. This finding also confirmed that the benzene ring in Pin came from Phe, which is consistent with that of the lignan biosynthetic pathway in plants.

Identification of products converted from [13C6]-labeled glucose

To explore where Phe originated from the Pin biosynthetic pathway, [13C6]-labeled glu was supplied as the sole substrate in the bioconversion system with Phomopsis sp. XP-8 cells. As results, 13C labeled PDG, PMG, Pin, Phe, p-Co, and Ca were detected (Figs. 7 and 8).

Figure 7
figure 7

Mass spectrum of phenylalanine (A), cinnamic acid (B), p-coumaric acid (C) and PDG (D) in the resting cell system using [13C6]-labeled glucose as the substrate. The increase of m/z owing to incorporation of 13C is shown in red.

Full size image
Figure 8
figure 8

Mass spectrum of PMG (A) and Pin (B) in the resting cell system using [13C6]-labeled glucose as the substrate. The increase of m/z owing to incorporation of 13C is shown in red.

Full size image

The isotopic patterns observed in the MS and MS/MS spectra suggest the 13C from 13C6-labeled glu were incorporated into the products of Phe (Fig. 7A), Ca (Fig. 7B) p-Co (Fig. 7C), PDG (Fig. 7D), PMG (Fig. 8A), Pin (Fig. 8B), respectively. The observed mass shifts, indicating the number of incorporated 13C, were shown in the spectra. The detailed information on the products and possible positions of 13C in the products are summarized in Table S2 (Supporting information).

Interestingly, the analysis revealed that 13C6-labeled glu were incorporated into the core structure of PDG and PMG, and their glycosides. Additionally, the maximum of 16 13C was detected in the formed Pin (C20H22O6), indicating the [13C6]-labeled glu partly contributed to the formation of Pin. The possible positions of 13C in the structures are summarized in Table S2 (Supporting information).

Taken together, the mass flow from [13C6]- Phe to [13C6]-Ca, [13C6]-p-Co, and [13C12]-Pin was verified by the experiments using [13C6]-labeled Phe as the sole substrate (Fig. 9A). The mass flow from [13C6]- glu to [13C]-Phe, [13C]-Ca, [13C]-p-Co, [13C]-Pin, [13C]-PMG, and[13C]-PDG was verified by the data obtained using [13C6]- glu as the sole substrate (Fig. 9B).

Figure 9
figure 9

Pin and PDG bioconversion pathway in Phomopsis sp. XP-8. (A: Pin biosynthesis scheme from 13C stable isotope labeled phenylalanine; B: PDG biosynthesis scheme from 13C stable isotope labeled glucose). The abbreviations indicate phenylalanine ammonia-lyase (PAL), trans-cinnamate 4-hydroxylase (C4H), 4-coumarate-CoA ligase (4CL), p-coumarate 3-hydroxylase (C3H), caffeic acid 3-O-methyltransferase (COMT), cinnamoyl-CoA reductase (CCR), carbamyl phosphate synthetase (CAD), dirigent protein (Dir). 13C stable isotopes were signed by coloured dots. Among the dots, red dots mean that the 13C stable isotopes were converted from 13C isotopes labeled glucose through Phosphoenolpyruvate (PEP); blue dots mean that the 13C isotopes were converted from 13C isotopes labeled glucose through Enthrose 4-phosphate; green dots means that the 13C stable isotopes were converted from another 13C isotopes labeled glucose through the intermediate substances of PEP.

Full size image

Possible pathways for biosynthesis of PDG and PMG

The evidences for the possible biosynthetic pathways of PDG, PMG, and Pin are summarized in Figs. 9 and 10. The pathway from Phe to Pin, glu to Phe, Pin, PMG and PDG was verified (Fig. 9A,B, Supporting information Table S1 and Table S1). In addition, the bioconversion between PDG and PMG in Phomopsis sp. XP-8 was reported for the first time, and the analysis was showed in Fig. 10.

Figure 10
figure 10

Bioconversion between PDG and PMG in Phomopsis sp. XP-8. Glu with red box means 13C6 labeled glycoside, Glu without red box means unlabeled glycoside. “×”means the pathway does not exist in Phomopsis sp. XP-8.

Full size image

As shown in Fig. 10, two structures of PMG were detected: one was [13C12]-PMG with two benzene rings converted from 13C-labeled glu and an unlabeled glycoside (PMG m/z 531.29), and the other was [13C18]-PMG with both benzene ring structures converted and a glucoside from 13C-labeled glu (PMG m/z 537.33). Similarly, two PDG structures were detected: one was [13C18]-PDG with a two benzene ring structure and one glycoside converted from 13C-labeled glu (PDG m/z 699.27); the other one was [13C24]- PDG with two benzene rings and two glycosides from 13C-labeled glu (PDG m/z 705.26).

If PMG was the direct precursor of PDG, PMG m/z 531.29 would be converted to PDG m/z 699.27 by bonding one [13C6]-labeled glycoside through glycosylation; PMG m/z 537.33 could also be converted to PDG m/z 699.27 by bonding one unlabeled glycoside through glycosylation and to PDG m/z 705.26 by bonding one [13C6]-labeled glycoside. If this is true, PDG m/z 699.27 would have two glucoside sources, whereas PDG m/z 705.26 would have only one glucoside source. Therefore, the concentration of PDG m/z 705.26 should be lower than PDG m/z 699.27. However, the data show that the relative abundance of PDG m/z 705.26 was much higher than that of PDG m/z 699.27 (Fig. 7D-3). Therefore, PMG was not the precursor of PDG.

In contrast, if PDG was the direct precursor of PMG, PDG m/z 699.27 would be converted to PMG m/z 531.29 by hydrolyzation of one [13C6]-labeled glycoside and to PMG m/z 537.33 by hydrolyzation of one unlabeled glycoside; PDG m/z 705.26 would be converted to PMG m/z 537.33 by hydrolyzation of one [13C6]-labeled glycoside. If this is true, PMG m/z 537.33 would have two glycoside sources, whereas PMG m/z 531.29 would have only one source. The concentration of PMG m/z 531.29 should be lower than PMG m/z 537.33. However, the data show that relative abundance of m/z = 531.29 was higher than that of m/z = 537.33 (Fig. 8A-3). Therefore, PDG was not the precursor of PMG.

Discussion

The 13C stable isotope labeling method was successfully used in this study to verify the phenylpropanoid-pinoresinol and biosynthetic pathway of its glycosides in Phomopsis sp. XP-8 during mass flow. This lignan biosynthetic pathway was only reported in plants until now8,9, so it was very significant to verify the occurrence of this pathway in microorganisms. Stable Isotope-assisted metabolomics is an efficient way to trace and identify bio-transformed products and the metabolic pathways involved in their formation, such as understanding the fate of organic pollutants in environmental samples17. It was the first time to use this method to verify the Phenylpropanoid-pinoresinol in a microorganism. In our previous studies, many methods such as precursor feeding13, detection of enzyme activity14, and genomic annotation15 have been used to analyze the Phenylpropanoid-pinoresinol biosynthetic pathway in Phomopsis sp. XP-8. Through these studies, the precursors, enzymes activity and genes of PDG biosynthetic pathway have been found. The 13C stable isotope labeling method gave further verification on the occurrence of lignan biosynthetic pathway in microorganisms by now. In addition to this, it is the first time that differences between the PDG and PMG biosynthetic pathways have been verified.

The results obtained in this study verify the existence of the phenylpropanoid-lignan metabolic pathway in Phomopsis sp. XP-8. Genomic annotation is an efficient way to discover the pathways that are normally difficult to reveal by metabolic and enzymatic evidence due to low intermediate accumulation, low end-product, and silent gene expression under normal conditions. This method has been successfully used to identify the existence of a phenylpropanoid metabolic pathway in Aspergillus oryzae26, and the molecular genetics of naringenin biosynthesis, a typical plant secondary metabolite in Streptomyces clavuligerus27, and the occurrence of the phenylpropanoid-lignan pathway in Phomopsis sp. XP-815. This study reports the existence of the phenylpropanoid-lignan pathway in Phomopsis sp. XP-8 during mass flow and identified the metabolites.

Additional studies should illustrate the origin of the genes in the phenylpropanoid-lignan pathway of Phomopsis sp. XP-8. Horizontal gene transfer (HGT) has long been recognized as an important force in the evolution of organisms28. HGT occurs among different bacteria and plays important roles in the adaptation of microorganisms to different hosts or environmental conditions29. More and more evidence for gene transfer between distantly related eukaryotic groups has been presented28.Therefore, we cannot exclude the possibility that XP-8 may have acquired the genes related to the lignan biosynthetic pathway from its host plant by HGT during long-term symbiosis and evolution. However, further evidence is still needed to verify this proposed process.

The results obtained in this study provide useful information on the biosynthesis of lignans and their glycosides via microbial fermentation. Biosynthesis of lignans is of great interest to organic chemists as it provides a model for biomimetic chemistry and has extensive applications30. Improvement has been made in the techniques to biosynthesize lignan products by regulating the lignan biosynthetic pathway in trees through genetic modifications31. However, the lignan biosynthetic pathway has rarely been reported. More importantly, the bioconversion sequence from Pin to PDG and the direct precursor of PDG have remained unclear until now. In previous studies on Phomopsis sp. XP-8, the highest production of PDG and PMG did not occur simultaneously12 and PMG was not the precursor of PDG because PDG production decreased and/or disappeared when PMG yield increased13. The present study demonstrated that PMG was not the precursor of PDG, and PDG was not the precursor of PMG, indicating that Pin might be converted to PMG and PDG via two different pathways in Phomopsis sp. XP-8, which has not been revealed in plants.

Furthermore, this study revealed that the bioconversion of Pin, PMG, and PDG from glu occurred simultaneously as that from Phe. We found that the benzene ring structure of Phe did not open throughout the entire Pin bioconversion process in Phomopsis sp. XP-8 when Phe was used as the sole substrate, indicating that the Pin benzene ring originated from Phe. Glu was converted to Phe and was the sole glycoside donor for PDG biosynthesis. Therefore, glu not only participated in the formation of glycosides in PDG, but also provided the PDG benzene ring structure. This is different from that found in plants, indicating there might be some other different pathways to produce these products in Phomopsis sp. XP-8.

Not all intermediates in the KEGG-identified plant-lignan biosynthetic pathway related to Pin, PMG, and PDG formation were found in Phomopsis sp. XP-8, such as caffeic acid, ferulic acid, and coniferyl alcohol (Fig. 1). This may be because the pathways after p-Co are different in XP-8 from those in plants, or the accumulation of these intermediates was too low to be detected. Further studies are needed to verify this hypothesis.

In conclusion, the capability of Phomopsis sp. XP-8 to biosynthesize Pin, PMG and PDG from [13C6]-Phe and [13C6]-glu was verified. The study illustrated the phenylpropanoid-pinoresinol biosynthetic pathway in microorganism by using stable isotope assisted UPLC-Q-TOF-MS/MS, thus, demonstrating a completely new way to produce Pin, PMG and PDG by bioconversion process. In the further studies, Phomopsis sp. XP-8 could be used in producing these lignans and their derivatives by microbial fermentation or enzymatic reaction. In addition, the microbial fermentation production of Pin, PMG and PDG could be enhanced by regulatory controls based on the biosynthetic pathways proved in this study.