By continuing to browse this site you agree to us using cookies as described in About Cookies
Wiley Online Library is migrating to a new platform powered by Atypon, the leading provider of scholarly publishing platforms. The new Wiley Online Library will be migrated over the weekend of February 24 & 25 and will be live on February 26, 2018. For more information, please visit our migration page: http://www.wileyactual.com/WOLMigration/
M.X.L. and H.Y.Z. contributed equally to this work.
Dong-Hui Yang, State Key Laboratory of Natural and Biomimetic Drugs, School of Pharmaceutical Sciences, Peking University, 38 Xueyuan Road, Beijing 100191, China.E-mail: email@example.com
Shu-Lin Liu, Genomics Research Center, Harbin Medical University, 157 Baojian Road, Harbin 150081, China. E-mail: firstname.lastname@example.org
Secoisolariciresinol (SECO) is increasingly recognized for potential clinical application because of its preventive effects against breast and colon cancers, atherosclerosis and diabetes, and its production through biotransformation has been attempted. However, previously reported bacteria all required stringent anaerobic culture conditions, precluding large-scale production. Here, we report the isolation and characterization of bacteria that produce SECO under less stringent anaerobic culture conditions.
Methods and Results
Using defatted flaxseed as raw material, we isolated a facultative anaerobic bacterium from human faeces that hydrolysed secoisolariciresinol diglucoside-3-hydroxy-3-methyl glutaric acid (SDG-HMGA) oligomers in flaxseed to produce SECO. Both conventional assays and 16S rRNA gene sequence analysis demonstrated its close relatedness with Bacteroides uniformis. The transformation efficiency of SDG in defatted flaxseed to SECO was more than 80% by this bacterial strain. We investigated factors that might influence fermentation, such as redox potential and pH, for large-scale fermentation of defatted flaxseed to produce SECO.
The method to produce SECO through biotransformation of defatted flaxseed with this bacterial strain is highly efficient and economic.
Significance and Impact of the Study
This bacterial strain can transform SDG to SECO under less stringent anaerobic culture conditions, which will greatly facilitate industry-scale production of SECO.
Lignans have phyto-oestrogenic activities and are effective in the prevention of breast and colon cancers, atherosclerosis and diabetes (Thompson 1998; Prasad 1999, 2001; Prasad et al. 2000). Flaxseed is a rich source of lignans, containing 75–800 times more secoisolariciresinol diglucoside (SDG) than other food materials (Thompson et al. 1991; Mazura et al. 1996). However, SDG in flaxseed is present as part of an oligomeric structure composed of five SDG residues interconnected by four 3-hydroxy-3-methyl glutaric acid (HMGA) residues (Kamal-Eldin et al. 2001) (Fig. 1). Extraction of the lignans in flaxseed involves multiple steps with organic solvent, including dissolving the SDG oligomer and liberating SDG monomers by base hydrolysis and the release of secoisolariciresinol (SECO) by acid or β-glucuronidase treatment (Obermeyer et al. 1995; Mazur and Adlercreutz 1998; Setchell et al. 1999; Johnsson et al. 2000; Eliasson et al. 2003), which, however, may not completely hydrolyse the lignans. More importantly, the stability of the products from acid hydrolysis has recently been questioned, because SECO may be partially converted to its anhydrous form anhydrosecoisolariciresinol (AHS) under acidic conditions (Liggins et al. 2000; Charlet et al. 2002). Therefore, better methods of SECO production, such as those by bio-processes, would be preferred.
SDG could be metabolized to enterodiol (END) and enterolactone (ENL), which are two well-known lignan phytoestrogens, by the intestinal bacteria in vivo and in vitro, and SECO is an important intermediate in the conversion from SDG to END and ENL (Axelson et al. 1982; Rickard et al. 1996; Rickard and Thompson 1998, 2000; Wang et al. 2000; Clavel et al. 2006a,b). END and ENL possess a variety of biological activities, such as oestrogenic and anti-oestrogenic activities, antioxidant and antitumour effects, and roles in the prevention of cardiovascular diseases (Adlercreutz et al. 1992; Mousavi and Adlercreutz 1992; Adlercreutz 1995; Martin et al. 1995; Sung et al. 1998; Kitts et al. 1999; Demark-Wahnefried et al. 2001; Wang 2002). Of great significance, it is not until recently that SECO was found to have a broad range of important biological activities, such as effects in anti-depression and the prevention of breast and colon cancers (Wang et al. 2012; Lehraiki et al. 2010), atherosclerosis, and diabetes; meanwhile, SECO is a potent antioxidant (Yamauchi et al. 2007, 2008). Therefore, efforts have been made to produce SECO from raw materials through biotransformation by bacteria (Clavel et al. 2006a,b, 2007). However, previously reported bacteria that can produce SECO all required stringent anaerobic culture conditions, so large-scale production cannot be easily achieved. In this study, we used defatted flaxseed as the raw material to select human intestinal bacteria that can produce SECO through biotransformation under less stringent anaerobic culture conditions. We isolated one active bacterial strain that can transform SDG in flaxseed lignans to SECO. We determined the efficiency of transformation from SDG in defatted flaxseed to SECO by this bacterial strain and investigated factors that might influence fermentation, such as redox potential and pH, for large-scale fermentation of defatted flaxseed to produce SECO.
Materials and methods
Chemicals and reagents
High-performance liquid chromatography (HPLC)-grade acetonitrile was purchased from Merck KGaA Co. Ltd (Darmstadt, German), and purified water was provided by Hangzhou Wahaha Co. Ltd (Zhejiang, China). Analytical-grade methanol, n-butanol, petroleum ether, ethanol, KH2PO4 and K2HPO4 were purchased from Beijing Chemical Reagents Co. Ltd (Beijing, China). Secoisolariciresinol standard was purchased from Sigma Chemical Co. (St Louis, MO, USA). Amberlite XAD-2 macroporous resin (20–60 mesh size, 330 m2 g−1 average surface area) was purchased from Supelco, Sigma-Aldrich Co. Ltd (Bellefonte, PA, USA). Optical rotations were measured in MeOH solutions with an Autopol III automatic polarimeter (Rudolph Research Analytical, Hackettstown, NJ, USA) at 25°C. 1H and 13C NMR spectra were recorded on Varian INOVA-500. Redox potential was measured by the redox potential electrode (PT4805-DPAS-SC-K8S/200, Mettler Toledo M300, Mettler company). pH was measured using FE20 Five Easy pH instrument (Mettler Toledo, Shanghai, China).
SDG was extracted from defatted flaxseed according to a modified method (Degenhardt et al. 2002; Pihlava et al. 2004). Briefly, the method involves steps of 70% ethanol–water (70 : 30, v/v) extraction, base hydrolysis (0·25 mol l−1 NaOH) for 3 h at room temperature, isolation and purification through XAD-2 macroporous resin, ODS-C18, and Sephadex LH-20 column chromatography. Purity of the isolated SDG was at least 96% according to analysis by HPLC. The purified compound was identified by 1H and 13C NMR, and the spectra were in agreement with data of the pure SDG obtained in CD3OD solvent (Qiu et al. 1999; Knust et al. 2006). The isolated SDG was confirmed as (+)-SDG by polarimetry + 3·6 (c = 0·25, MeOH) (Saarinen et al. 2002; Knust et al. 2006).
Flaxseed samples were collected from Bei-An County of Heilongjiang Province, China, and were identified as the dried seeds of Linum usitatissimum L. Voucher specimens (sample no. 071024) were deposited in the herbarium of pharmacognosy research group, School of Pharmaceutical Sciences, Peking University. They were ground into powder (pass 40 mesh sieve) and then defatted by petroleum ether prior to use.
Collection and processing of faecal samples
Initially, fresh faecal specimens (c. 4·0 g each), obtained from 10 healthy young subjects (four women and six men, 21–30 years old), were suspended in 20 ml sterile phosphate buffer saline (PBS, 2·6 g l−1 KH2PO4, 1·85 g l−1 K2HPO4, pH 7·4) and 2 ml such faecal suspension was transferred to 20 ml culture medium (see below).
Culture media and bacterial culture
Anaerobic broth (Culture medium I) and anaerobic agar (Culture medium II) were purchased from Beijing Land Bridge Technology Co. Ltd (Beijing, China). Anaerobic broth (g l−1) consisted of peptone 15·0, yeast 5·0, soybean peptone 5·0, beef powder 5·0, glucose 5·0, sodium chloride 5·0, soluble starch 3·0, cysteine 0·5, potassium dihydrogen phosphate 2·5, hemin 0·005, and vitamin K1 0·001. Anaerobic agar contained the same ingredients and also agar 15·0 g l−1. Carbon-free medium (Culture medium III) was designed to lack carbon source in the medium, which contained the following components (in 1 l): NaCl 3 g, KH2PO4 2·6 g, K2HPO4 1·85 g, 1% (v/v) reducing solution (30 g l−1 l-aminothiopropionic acid and 30 g l−1 sodium hyposulphite, dissolved in PBS), and 1 g NH4Cl.
The bacterial culture procedure was as follows: 0·5 g of defatted flaxseed (substrate, also acted as carbon source) was added into each of tubes containing 5 ml of Carbon-free medium, then sealed with liquid paraffin and autoclaved at 121°C for 15 min. Into the culture system, 2 ml faecal suspension was added and incubated at 37°C for 72 h. The supernatant of the culture was then analysed for the appearance of SECO by HPLC. The faecal specimen that we used for SECO production was from a 23-year-old woman.
Bacterial identification by the bacterial 16S rRNA gene and the API 20A system
A single colony of the bacteria was incubated at 37°C in anaerobic medium for 24 h, and the bacteria were collected by centrifugation at 12 000 g for 2 min. Total DNA was extracted with a TIANamp Bacteria DNA Kit (Tiangen, Beijing, China) as instructed by the manufacturer. The 16S rRNA gene was amplified by polymerase chain reaction (PCR) with universal bacterial primers 8F (5′-AGAGTTTGATCCTGGCTCAG-3′) and 1541R (5′-AAGGAGGTGATCCAGCCGCA-3′). Amplification proceeded in a reaction mixture containing 0·2 mmol l−1 of each primer, 25 μl of 2× Taq PCR Master Mix (Tiangen), and 21 μl of template DNA. The PCR programme was as follows: initial denaturation at 94°C for 3 min; 30 cycles of 94°C for 30 s, 53°C for 30 s, and 72°C for 1 min; and a final extension step at 72°C for 5 min. PCR products were sent to Sangon (Beijing, China) for sequencing.
API 20A (BioMérieux Industry, Shanghai, China) consists of a plastic panel with 20 cupules containing dehydrated substrates, which allows for the determination of 21 biochemical reactions. Included are tests for indole, urease, catalase, gelatin liquefaction, esculin hydrolysis, and fermentation of glucose, mannitol, lactose, sucrose, maltose, salicin, xylose, arabinose, glycerol, cellobiose, mannose, melezitose, raffinose, sorbitol, rhamnose and trehalose.
A bacterial suspension (density ≥ 3 McFarland) was prepared in API anaerobe basal medium from pure growth on a fresh anaerobic agar plate. The panel was inoculated and then incubated at 37°C in an anaerobe jar for 24 h. Necessary reagents were added, and reactions were interpreted as instructed by the manufacturer. A seven-digit profile number was generated to assign an identification by the software (atb, ver. 4·0).
High-performance liquid chromatography
The HPLC system consisted of Agilent 1200 series HPLC apparatus (Agilent Technologies, Santa Clara, CA, USA), including high-pressure binary-gradient solvent-delivery pump, DAD (diode-array detector), autosampler, thermostat column compartment and chemstation (9·01 edition). Zorbax SB-C18 column (4·6 × 250 mm, 5 μm) was used to analyse the samples. Mobile phase consisted of water (A) and acetonitrile (B) in a linear gradient change from 100% A to 50% A and 50% B in 30 min. Detection wavelength was 280 nm, and the temperature of the column oven was 25°C with a flow rate of 1·0 ml min−1.
Calibration of the SDG and SECO curves
The stock solutions of SDG standard (200·0 μg ml−1) and SECO standard (264·0 μg ml−1) were prepared by accurately weighing and transferring each of them into a volumetric flask (5 ml) and dissolving it in methanol. Solutions for SDG calibration (2·5–200·0 μg ml−1) and SECO calibration (1·65–264·0 μg ml−1) were prepared by dilution of the stock solutions with methanol. For each calibration curve, independent dilutions were analysed. The calibration equation of SDG was obtained by plotting HPLC peak areas (y) vs the concentration of calibrators (x, mg ml−1), which was as follows: y = 10·25x −5·892 (R2 = 0·999), with a good linearity over the range from 2·5 to 200 μg ml−1, and the calibration equation of SECO was obtained by plotting HPLC peak areas (y) vs the concentration of calibrators (x, μg ml−1), which was as follows: y = 14·25x + 7·515 (R2 = 0·9999), with a good linearity over the range from 1·65 to 264·0 μg ml−1.
Limits of detection and quantification
Stock solutions of SDG and SECO standards were separately diluted to make a series of solutions with methanol and analysed by HPLC. On the basis of signal-to-noise ratio (S/N), the limits of detection (LOD) and quantification (LOQ) of SDG standard were determined to be 1·25 μg ml−1 (S/N = 3) and 2·5 μg ml−1 (S/N = 10), respectively. The LOD and LOQ of SECO standard were measured to be 0·33 μg ml−1 (S/N = 3) and 0·66 μg ml−1 (S/N = 10), respectively.
Evaluation of SDG content in defatted flaxseed
The method of SDG extraction from defatted flaxseed was developed according to the method described by Zhang (2007). Portions (1·0 g) of defatted flaxseed were extracted with 15 ml of 50% ethanol–water (v/v) in test tubes by sonication (ultrasonic power: 200 W) for 30 min in a 40°C water bath. After centrifugation (2000 g, 15 min), the residue was extracted again by the same method. The supernatants were merged and subjected to alkaline hydrolysis for 3 h using 0·25 mol l−1 aqueous sodium hydroxide at room temperature. After hydrolysis, the samples were acidified to pH 4–6 using 6 mol l−1 hydrochloric acid, and their volume was adjusted to 50 ml in volumetric flasks.
Sampling of the cultures
A volume of 200 μl of culture was sampled every 24 h and extracted with 400 μl n-butanol saturated with water. A portion of n-butanol extract (320 μl) was transferred to a centrifuge tube and evaporated to dryness by N2. The residue was dissolved in 200 μl methanol and centrifuged for 3 min (8700 g), and then, 20 μl of the filtered supernatant was analysed by HPLC.
Selection and phylogenetic characterization of SECO-producing bacteria
A bacterial suspension from fresh faeces of a healthy volunteer was repeatedly cultured in 5 ml of carbon-free culture medium, containing defatted flaxseed as a substrate at 37°C in an anaerobic jar. A portion of the culture was enriched and then 10-fold diluted for isolation. A 10-μl aliquot of each of serial dilutions (10−1~10−6) was spread on an anaerobic agar plate separately and incubated for 48 h at 37°C in the sealed jar with an anaerobic pack. According to the colonial morphology, such as size, colour and shape, twelve single colonies well separated from one another on the agar plates were picked up and cultured in anaerobic broth and then screened for their activity of transforming flaxseeds to SECO. SECO-producing bacteria were characterized and identified by physical and chemical tests and 16S rRNA gene sequencing as previously reported (Wang et al. 2010).
A Gram-negative anaerobic bacillus with no spore that could produce SECO from defatted flaxseed was isolated and designated ZL1. Using the API 20A system (BioMérieux Industry), we identified ZL1 to be Bacteroides uniformis, with a 97·9% of identification rate. Strain ZL1 could hydrolyse esculin and produce indole with the substrate tryptophan. In addition, ZL1 fermented most of the saccharides tested to produce acid, including glucose, lactose, maltose, xylose, saccharose, cellobiose, etc.
Comparison of 16S rRNA gene sequence of strain ZL1 (GenBank accession no. JN873344) with those of sequenced bacteria in GenBank also revealed close phylogenetic relatedness of strain ZL1 to Bact. uniformis strains (Fig. 2), with the closest being Bact. uniformis mat-344 (AB215084).
Transformation efficiency of pure SDG into SECO by strain ZL1
To verify that strain ZL1 produced SECO from defatted flaxseed through the deglycosylation of SDG, we added 2 ml overnight culture of strain ZL1 into 25 ml anaerobic medium containing 5 mg purified SDG and incubated the bacteria at 37°C. After 15 h of incubation, an intermediate, P1 (m/z = 524), was produced which corresponds to mono-deglycosylated secoisolariciresinol diglucoside as determined by the technique of HPLC-MS (Clavel et al. 2006a) (Fig. 3). At the same time, SECO was detected and its amount increased gradually and steadily, the yield reaching the maximum after about 40 h of incubation (Fig. 4). The transformation efficiency of SDG to SECO was 82·6 ± 2·7% (transformation efficiency = (M′SDG/MSDG) × 100%, M′SDG = (CSECO × V/362·2) × 686·3, MSDG: the original amount of pure SDG or SDG in the defatted flaxseed; CSECO: concentration of SECO in the culture; V: volume of culture).
Transformation efficiency of SDG in defatted flaxseed into SECO by strain ZL1
First, a 0·5 ml aliquot of an overnight ZL1 culture was transferred to 5 ml anaerobic broth and incubated at 37°C for 24 h. Then, 0·5 ml of this culture was added into 5·0 ml of carbon-free culture medium (culture medium III) containing 0·1 g defatted flaxseed, with 0·5 ml of liquid paraffin added on top of it. The culture was maintained at 37°C for 10 days and sampled regularly for analysis by HPLC. SECO was first detected from the culture 9 h after the initiation of incubation. After 24 h of incubation, the concentration of SECO reached 29·2 ± 1·1 mg l−1. When the incubation continued, the amount of SECO was gradually increased. After 4 days of incubation, 56% of SDG in defatted flaxseed was converted to SECO (40·0 ± 1·3 mg l−1). The maximum concentration of SECO reached 48·3 ± 1·4 mg l−1, 7 days after incubation, with about 70% SDG converted to SECO. Then, the yield declined a little and remained relatively stable for an additional 2 days (Fig. 5).
SECO production by strain ZL1 in scaled-up culture systems
For large-scale production of SECO, we increased the volume of the carbon-free culture medium from 5 ml to 2 l. First, we cultured strain ZL1 in 200 ml anaerobic broth with 20 ml liquid paraffin added on top at 37°C for 24 h. We then added this 200 ml culture into 2 l carbon-free culture medium in a 3l Erlenmeyer flask containing 40 g defatted flaxseed and continued incubation at 37°C. During the incubation, supernatant of the culture was sampled regularly for HPLC analysis. The content–time course of SECO production in such scaled-up culture systems is shown in Fig. 5. After 12 h of incubation, SECO was detected in the culture. During the second 12 h of incubation, SECO concentration increased rapidly, reaching 28·4 ± 3·5 mg l−1, which corresponds to more than 40% of the SDG in the added defatted flaxseed converted. With the culture continuing, the yield of SECO in the culture kept increasing and reached high concentration (48·0 ± 1·4 mg l−1) on day 4, with about 70% of SDG transformed to SECO. Finally, 80·5% of the SDG in defatted flaxseed was converted to SECO (56·9 ± 1·2 mg l−1) on day 7.
To optimize the fermentation conditions, we determined the redox potential, using a redox potential probe (PT4805-DPAS-SC-K8S/200, Mettler Toledo M300), and pH in the 2 l culture system. The redox potential probe was inserted into the middle of the flask and was fixed there. The content-time curve of the redox potential in the culture during fermentation is shown in Fig. 6. During the first 24 h of incubation, the redox potential almost declined in a straight line trend from 170 to –526 mV. After 24 h of incubation, the redox potential nearly remained stable. The pH value was measured using a FE20 Five Easy pH instrument (Mettler Toledo). Similar to redox potential, the pH significantly decreased during the first 24 h of incubation and then kept relatively stable (pH 5·6) after 48 h.
Enrichment of SECO from the cultures
We treated the cultures for 7 days with twofold volumes of 95% ethanol (4 l) to terminate the culture and to precipitate the macromolecular substances in the culture. We then filtered the culture and evaporated the filtered liquid at 50°C under reduced pressure to about 220 ml concentrated solution. The filtrate was chromatographed on XAD-2 macroporous resin column (230 g, 5 × 35 cm). The column was washed with H2O (300 ml) and then eluted sequentially with 25% ethanol-aqueous solution (600 ml), 30% ethanol-aqueous solution (600 ml) and 45% ethanol-aqueous solution (1800 ml). As shown in Fig. 7, SECO was mainly present in 45% ethanol-aqueous eluent and the production of SECO reached up to 2·1 mg g−1 in it, which was taken as yield of the weight of defatted flaxseed. According to the polarimetric data + 27·3° (c = 0·33, MeOH), the produced SECO was of the (+)-form (Knust et al. 2006).
Flaxseed is the seed of L. usitatissimum L., which is widely distributed in northern China, with an annual output of 420 000 tons (ranking fourth in the world). Usually, flaxseeds are used to extract oils, and the residues are discarded. The defatted flour of flaxseeds contains abundant SDG, with a content of around 11·7–24·1 mg g−1 in the dry matter (Johnsson et al. 2000). However, because of the fact that SDG in flaxseeds exists as part of an oligomeric form, the utilization of defatted flaxseed resources has been limited. In this study, we isolated a bacterial strain that could readily transform the (+) SDG oligomers in defatted flaxseeds to produce valuable active constituents (+) SECO. The method we report here is highly efficient and completely eco-friendly.
The SECO producer that we isolated from the microflora of a healthy volunteer was identified to be a bacterial strain closely related to Bact. uniformis based on both API 20A characterization and full-length 16S rRNA gene sequence analysis. Bacteroides are commonly found in the human intestine, making up the most substantial portion of the mammalian gastrointestinal flora, where they assist the host in breaking down food and producing valuable nutrients and energy that the body needs (Rigottier-Goisa et al. 2003; Eckburg et al. 2005; Wexler 2007). To our knowledge, this is the first report that Bact. uniformis-related bacteria can transform SDG in defatted flaxseed to SECO. The conversion ratio of SDG to SECO by ZL1 was more than 80% with either defatted flaxseed or pure SDG, demonstrating that the method of bacterial biotransformation we report here is highly efficient for producing SECO from defatted flaxseed. As the effects of SECO production in 5 ml and 2 l culture systems were similar, the biotransformation method may be optimized for possible industrial production.
In practice, redox potential and pH are widely used in the control and monitoring of fermentation processes, because numerous biological processes and reactions, such as degradation of organics, are essentially oxidation–reduction processes, as has been demonstrated on redox potential (Koch and Oldham 1985; Chang et al. 1994) and on pH (Al-Ghusain et al. 1994; Kim and Hao 2001; Kishida et al. 2003; Akin and Ugurlu 2005; Hua et al. 2005). The results of redox potential and pH values during the fermentation processes in our culture systems were consistent with those of the previously reports (Peddie et al. 1990). The initial rapid drop in both redox potential and pH during the first 24 h of incubation and then gradual stabilization are in good correlation with the biotransformation process of SDG in the defatted flaxseed to produce SECO. Therefore, redox potential and pH may be useful indicators of the biotransformation processes. To date, numerous researchers have reported linear relationships between redox potential and the log of dissolved oxygen at low-oxygen levels (Peddie et al. 1990; Janssen et al. 1998; Ndegwa et al. 2007), but dissolved oxygen cannot always be monitored accurately because of the low sensitivity of commercial dissolved oxygen probes (Fuerhacker et al. 2000; Ndegwa et al. 2007). As such, redox potential may indirectly reflect the relative amount of dissolved oxygen.
In this study, the configuration of the produced SECO was identified to be of the (+)-form, which is consistent with the results previously published (Knust et al. 2006). As (+)-SDG was used in this study, it is reasonable to conclude that the bacterial strain ZL1 could convert (+)-SDG to (+)-SECO.
We previously isolated bacterial strain Klebsiella S1 that also produces SECO from defatted flaxseed (Wang et al. 2010); strain Bacteroides ZL1 had similar productivity of SECO under same culture conditions, although they may use different pathways to produce SECO. We recently found that the biotransformation activity of S1 declined after preservation at low temperature (−60°C) for 3 months; we are in the process to determine whether Bacteroides ZL1 may behave similarly after preservation at low temperature (−60°C) for a certain period of time. In any case, to maintain the activity and stability of the isolated SECO-producing bacteria, we need to identify functional genes and obtain genetic engineering bacteria.
In conclusion, biotransformation by Bacteroides ZL1 will be a very efficient and environmentally friendly way of mass-producing (+)-SECO from defatted flaxseed.
We gratefully acknowledge Dr Qi-De Han for support and encouragement throughout this project. This work was supported by grants from the National Natural Science Foundation of China (NSFC, no. 30672622) and National Science and Technology Major Projects for Major New Drugs Innovation and Development (no. 2011ZX09102-011-06) of China to D.H.Y., an NSFC grant (no. 30970078) and a grant of Natural Science Foundation of Heilongjiang Province of China to G.R.L.; and NSFC grants (nos. 30370774, 30870098, 30970119), and a 985 Project grant of Peking University Health Science Center to S.L.L.