Enhanced production of glycyrrhetic acid 3-O-mono-β-D-glucuronide by fed-batch fermentation using pH and dissolved oxygen as feedback parameters☆

Bo Lü,Xiaogang Yang,Xudong Feng*,Chun Li

School of Life Science,Beijing Institute of Technology,Beijing 100081,China

1.Introduction

Glycyrrhizin(GL),which is a major natural compound extracted from the traditional Chinese medicine licorice root[1],harbors a broad spectra of bene ficial pharmacological activities,such as antiviral[2],antiulcerative[3],antitumor[4]and anti-in flammatory activities[5].Due to its sweet taste and pleasant smell,GL is also widely used in many food products.Glycyrrhetic acid 3-O-mono-β-D-glucuronide(GAMG)is an importantderivative ofGL through hydrolysisofone molecule of distalglucuronic acid(Fig.1)[6].Compared with GL,GAMG has stronger biological activities due to its proper polarity and solubility[7].In addition,GAMG possesses a higher sweetness with an extremely low caloric value(5 times higher than GL,950 times higher than sucrose)[8].Moreover,the LD50value of GAMG(5000 mg·kg-1)is much higher than that of GL(805 mg·kg-1),demonstrating its higher safeness[9].Given these properties,GAMG which is considered to be a potential and better substitute of GL has inspired global interest in pharmacy and food industry[10].

Currently,most commercial GAMG is obtained through chemical synthesis with heavy metals as catalysts,and the overall yield was only 2.2%,which has largely limited its bulk industrial production[11].Recently,biosynthesis of GAMG has drawn much attention due to high chemical bond selectivity,high reaction rate,mild reaction conditions and eco-friendliness.β-Glucuronidase(GUS,EC 3.2.1.31)which can hydrolyze glycosidic bonds has been applied to produce GAMG with particular concern to enzyme activity and substrate selectivity[12],widely ranging from bacterial and fungi to plants and animals[13-15].However,most of them suffer from very low hydrolytic selectivity,which show preference for hydrolyzing GL to glycyrrhetic acid(GA)[16,17].To date,only two β-glucuronidases from yeast Cryptococcus magnus MG 27 and liver of domestic duck have been used for the biotransformation of GL[18,19].However,such enzymes exhibit low activity,high cost and unsustainability.Therefore,an efficient biotransformation process for GAMG is still highly desirable.

Within this context,we previously isolated a filamentous fungal strain named P.purpurogenum Li-3,and it can directly hydrolyze GL into GAMG by its β-glucuronidase(pgus,GenBank Accession NO.EU095019)with very few by-products GA[6].However,the amount of enzyme produced by P.purpurogenum Li-3 was quite low and it was unable to meet the high requirement of GAMG production.In order to decrease GAMG production costs,optimizing fermentation conditions is another effective strategy,such as suitable pH,temperature and other environmental tolerances.Besides,GL functions as the sole carbon source and inducer due to the catabolite repression,simultaneously achieving high biomass and relatively high β-glucuronidase production.Normally,most researches mainly investigate the effects ofinitialpHon biomass[20-22],butlittle attention has been paid to develop a pH control strategy by feeding GL to obtain both high cell growth rate and GAMG production rate.Therefore,it is quite necessary to establish a suitable substrate feeding strategy during fermentation to maintain a high-yield GAMG production.

Fig.1.The hydrolysis of GL into GAMG by Penicillium purpurogenum Li-3[6].

In this article,we aim to evaluate the production of GAMG through P.purpurogenum Li-3 cultivation performed through both batch and fed-batch processes in bioreactors.In batch process,the effect of pH,temperature,and agitation intensity on biomass and GAMG yield was thoroughly investigated.Then,a fine substrate feeding strategy for fed-batch process was established by using pH and DO as feedback parameters to develop an efficient and high-yield process for GAMG production.Finally,a separation process based on D101 resin absorption and preparative HPLC was developed to purify GAMGfromfermentation broth.This study provides a new insight into the industrial bioprocess of GAMG production.

2.Materials and Methods

2.1.Materials

GL monoammonium salt and glycyrrhetic acid(GA)as standards were purchased from Sigma Chemical Co.(USA).GL monoammonium salt with 75%purity for fermentation was purchased from Xinjiang Tianshan Pharmaceutical Ind.Co.,Ltd.Glycyrrhetic acid 3-O-mono-β-D-glucuronide(GAMG)was generously donated by Nanjing University of Technology(China).D101 macroporous resins were purchased from Nankai Hecheng S&T Co.,Ltd.(Tianjin,China).All other reagents were of analytical grade and used without any pretreatment.

2.2.Microorganism and media culture medium

The GAMG-producing strain P.purpurogenum Li-3 was screened and isolated from soil of licorice plantation in Xinjiang,China and was preserved by the lab of biotransformation and microecology(Beijing Institute of Technology,China).Optimization studies on media components for maximum production of GAMG have been performed[6,23].The activation medium used for seed activation contained 6 g·L-1glucose,3 g·L-1NaNO3,1 g·L-1K2HPO4,0.5 g·L-1KCl,0.5 g·L-1MgSO4·7H2O,0.01 g·L-1FeSO4·7H2O.The fermentation mediumused for GAMGproduction contained 6 g·L-1GL monoammonium salt,3 g·L-1NaNO3,1 g·L-1K2HPO4,0.5 g·L-1KCl,0.5 g·L-1MgSO4·7H2O,0.01 g·L-1FeSO4·7H2O.All the media were finally adjusted to pH 4.8-5.0,and then sterilized at 121°C for 15 min.

2.3.Preparation of seed culture

Conditions of seed culture like plate age,inoculum age and percentage of inoculum were optimized previously in our laboratory.The strain from PDAagarplatespre-cultured for5-7 d in an incubatorat30°C,was transferred to 100 ml sterile seed medium in each 250 ml Erlenmeyer flask.The pH of the medium was adjusted to 5.0 and the culture was incubated on a rotary shaker at 170 r·min-1and 32 °C.The seed culture was allowed to grow for 72 h and then 5%of the seed culture was transferred to 200 ml sterile seed medium again in each 500 ml Erlenmeyer flask,followed by incubation at 170 r·min-1and 32 °C for 24 h.At least two generations of activation cultures were required before fermentation.The resultant culture was used as the seed culture.

2.4.Batch and fed-batch fermentation of P.purpurogenum Li-3

A 150 ml aliquot of seed culture was inoculated into a 2.5 L stirred bioreactor(Infors,Switzerland)with 1.5 L working volume.The bioreactor was equipped with a DO analyzer and an automatic pH controller controlled by the software.During the fermentation,samples were collected every 12 h for the determination of biomass,GL conversion and GAMG yield.Various parameters like pH(4.4-5.8),temperature(28-38 °C),agitation(100-300 r·min-1)and aeration(0.5-1.5 vvm)were studied in order to achieve the optimal operation.Fed-batch was performed by the addition of 100 ml 90 g·L-1GL each time as carbon source and substrate to the medium at specific fermentation time.Samples were withdrawn after specified time intervals(6 h)fromthe bioreactor for the routine assay.

2.5.Analytical methods

Biomass was estimated based on dry cell mass.The mycelia from samples were collected by centrifugation at 4 °C and 8000 r·min-1,and washed twice with 10 mldistilled water underthe same conditions.The resulting mycelia were dried under vacuum at 80°C to constant mass.For GAMG analysis,the sample of fermentation broth was centrifuged at 4 °C and 8000 r·min-1,and the supernatant was filtered(0.22 μm pore size membrane)for further analysis.The concentrations of GL,GAMG and GA were determined by HPLC equipped with SHIMADZU Shim-pack VP-ODS(4.6 mm × 150 mm,5 μm).The isocratic elution was used with methanol(solvent A)and water/0.6%acetic acid(solvent B)at 1 ml·min-1and A/B ratio of 81:19.GL,GAMG and GA were detected at 254 nm and 40°C.

2.6.Kinetic parameters calculation

The specific growth rate(μ)and specific GAMG production rate(qp)can be calculated according to Eqs.(1-2),respectively.

where X is the dry cell mass concentration(g·L-1).

The GL conversion yield(XS)and GAMGproduction yield(ηp)can be calculated according to Eqs.(3-4),respectively.

where CS0is the initial substrate concentration,CStis the substrate concentration at time t,CPtis the product concentration at time t,and MSand MPare the molecular weight of substrate and product.

2.7.The separation and purification strategy

Ten grams of D101 resins(dry mass)was soaked in 95%ethanol for 24 h,and then packed into a 15 ml glass column(10 cm in length)followed by washing with 95%ethanol until no white precipitate was observed.Then,column was sequentially equilibrated with 5%HCl,deionized water,5%NaOH,deionized water for 5 bed volumes individually.For GAMG separation,the fermentation broth was pumped through the column at flow rate of 0.5-1.5 ml·min-1and the dynamic adsorption capacity was determined.The elute in the outlet was analyzed periodically with HPLC as described in Section 2.5.The breakthrough pointwas defined as the pointwhen the effluentconcentration reached 1%of the feed concentration.Before eluting the GAMG,the column was first washed with water for two bed volumes to remove the absorbed contaminants such as mineral salts,contaminant protein,mycelia and some other metabolites.Then,the target product GAMG was eluted with 70%ethanol at two different flow rates of 0.5 and 1 ml·min-1.The resultant GAMG was further purified with a preparative HPLC equipped with an Agela Promosil C18 column(5 μm,30 mm × 25 mm).The operational parameters were extensively optimized as follows: flow rate of 10 ml·min-1,mobile phase composition of mixture of methanol and water(87.5:12.5,v/v),GAMG loading of 250 mg.

3.Results and Discussion

3.1.Optimization of pH and temperature for GAMG production

pH is one important factor for fungal submerged fermentation within the various conditions since it can profoundly affect cell growth,enzyme production,product biosynthesis,mycelia morphology and cell lyses[24].In order to investigate the effect of initial medium pH on GAMG biosynthesis,batch experiments were performed in the fermentor by varying initial pH from 4.4 to 5.8.As shown in Fig.2,the result indicated that the initial pH of medium severely affected the GAMG production.The maximal GAMG concentration of 3.55 g·L-1was obtained atpH 5.0.When pHwas lower or higher than 5.0,the GAMGproduction slightly decreased.For pH lower than 5.0,considering our previous work,pH affected the ionization of GL further in fluencing its solubility in fermentation medium[25].For pH higher than 5.0,the stability and activity ofβ-glucuronidase decreased which was notfavorable for GL biotransformation process,leading to low concentration of GAMG[6].Therefore,the optimal pH for GAMG production by P.purpurogenum Li-3 was found to be 5.0.

Fig.2.The effect of pH on GAMG production at temperature 45°C and agitation speed 200 r·min-1.

Temperature is another vital operating parameter affecting the fermentation performance[26].Fig.3 shows the time-course of GL consumption,GAMG production and biomass in a typical batch process at different temperatures from 28 to 38°C.The dry cell mass increased as the increase of temperature from 28 to 32°C,and then decreased as the further increase of temperature,indicating 32°C is the optimal temperature for P.purpurogenum growth.Moreover,dramatic GAMG biosynthesis occurred after 36 h fermentation at 30-32°C.However,at other temperatures from 34 to 38°C,GAMG biosynthesis started at 48 h,which was 12 h later than the former,which meant that it was the prerequisite for achieving efficient β-glucuronidase production and mycelia growth at 32°C.Furthermore,we found that although the lag phase for GAMG production was longer and the biomass were much less at higher temperature(34-38°C),it took a very short period to biotransform GL into GAMGand led to similar final GAMGconcentration compared to that at 32 °C.This may be due to that β-glucuronidase had higher activity at higher temperature within the investigated range(28-38°C),and it has been reported that the optimum temperature for the activity of β-glucuronidase from P.purpurogenum Li-3 in vitro is 45 °C[27,28].Therefore,we set 32 °C as the optimal temperature for GAMG production by fermentation of P.purpurogenum Li-3 due to the high biomass and low required energy.

3.2.Effect of agitation intensity on biomass and GAMG production

Oxygen is often a limiting component in filamentous fungi fermentation due to its complex structure with considerably higher biomass than other microbial thus causing a higher viscosity of culture and limiting the oxygen transfer[24,29].Therefore,a high agitation rate is necessary to provide enough mixing and mass transfer,but mechanical forces can also cause mycelia damage,so a compromise needs to be made between enough mass transfer and the least cells damage.In this study,the effect of different agitation speeds(100,200 and 300 r·min-1)on biomass and GAMG production were investigated.As can be seen from Fig.4,compared to 100 and 200 r·min-1,higher agitation speed of 300 r·min-1can prompt the growth of cells,leading to 6 h faster to reach steady stage.This may be due to the high oxygen in the broth increasing the specific growth rate.However,the finalDCWof 300 r·min-1was lower than that of 100 and 200 r·min-1,which may be caused by the damage of cells by high shear.In addition,high agitation speed was not favorable for GAMG production:a similar final concentration of around 3.3 g·L-1GAMG was obtained at low agitation speeds(100,200 r·min-1)at 60 h,which was higher than that of high agitation speed of 300 r·min-1(2.64 g·L-1).As show in Table 1,the higher conversion rate,yield and biomass were obtained at 100 and 200 r·min-1at 60 h,which was also higher than that of 300 r·min-1.Considering that higher speed would cause more sheer stress on mycelia and more foam,the agitation speed of 100 r·min-1was selected for GAMG production.

Aeration rate is another factoraffecting the level ofdissolved oxygen(DO).We also investigated the effectofaeration rate on the biomass and GAMG production.There was no significant difference found between the aeration rates from 0.5 vvm to 1.5 vvm.We selected 1.0 vvm as the optimal aeration rates for GAMG production.

Fig.3.The effect of temperature on GL concentration(a),GAMG concentration(b),dry cell mass(c),conversion and yield(d).The pH was 5.0 and agitation speed was 200 r·min-1.

3.3.Fed-batch fermentation combined with the pH and DO control

A fed-batch fermentation process was developed to further increase GAMG production.In this study,pH and DO were chosen as the feedback parameters for feeding GL based on the following reasons:1)pH was closely relevant to GL consumption,when GL was gradually transformed to GAMG during fermentation,the pH would increase rapidly,therefore,the fermentation can be easily monitored according to the change of pH;2)DO was also well related to GL consumption,since GL was used as the carbon sources,foraerobic fermentation,the oxygen consumption was related to carbon source utilization,so DO was used as parameter to monitor the fermentation process.

Fig.4.The effectofagitation on GAMGconcentration(a),specific growth rateμ(b),dry cellmass(c),and specific GAMGproduction rate q p(d).The temperature was 45°Cand pHwas 5.0.

Table 1 Effect of agitation speed on μmax,q pmax,GL conversion and GAMG yield

To verify the above points,the fermentation profiles of P.purpurogenum Li-3 were further investigated in a 5 L fermentor under natural pH and three independent DO control schemes,ranging from 100 to 300 r·min-1.As shown in Fig.5,at all investigated agitation speeds,the pH profile of the whole fermentation process displayed a very interesting“V”shape trend.The pH of the fermentation slightly decreased to a minimum of 4.8 first,and remained relatively stable for 12 h,and then rebounded up to 5.1 after 48 h of cultivation.It seems that the slight decrease of pH at initial stage may be due to the produced organic acids caused by the rapid consumption of glucose from inoculum,and this phenomenon has also been observed in the relevant study[30].The increase of pH at later stage was due to the hydrolysis of GL by the secreted β-glucuronidase by the cells,and the glucuronic acid was subsequently consumed by the cells,resulting in the decrease of acids and thus the increase of pH.This corresponded well to the consumption of GL and production of GAMG.DO showed a similar changing tendency as pH.The initial decrease of DO was due to the residual glucose in the medium for cell growth.Then DO became stable during the induced production of β-glucuronidase.Following that,DO decreased again because of the hydrolysis of GL.Therefore,it was reasonable to choose pH and DO as in situ signals for the feedback of feeding.

On the basis of the above results,the maximal GAMG content and productivity were obtained at pH 5.0 and temperature 32°C.The fedbatch operation with pH and DO control was performed to obtain a high amount of biomass with high GAMG content.The first feeding point was set as 48 h,and then 100 ml 90 g·L-1GL was fed each time whenever pH increased to above 5.0 and DO was increased to above 80%(each cycle was approximately 24 h,so feeding time was practically 48,72,96 and 120 h during fermentation).As can be seen from Fig.6,the conversion of GL and the yield of GAMG were greatly improved by feeding GL during fermentation.Each cycle after feeding,the pH decreased significantly due to the provided acidic GL,and DO also decreased due to both the consumption of oxygen by metabolism of GL by cells,and the increased viscosity in the broth due to the provided GL.When pH and DO became stable,the next feeding was performed.In addition,the enzyme activity and the dry cell mass were also monitored during the feeding of GL each time(data not shown).The enzyme activity did not show significant change and maintained constant between pH 4.6 to 5.2,which corresponded well to GAMG productivity being relatively stable during the fed-batch process.The dry cell mass was also not greatly affected by feeding GL since other nutrients such as nitrogen and minerals were not provided.After the fourth feeding,the viscosity ofthe broth was very high due to the large amountofaccumulated GL and GAMG,the non-Newton fluid was observed and too much anti-foaming agent was used due to the foam.This situation was no longerproperforthe growth ofcells,so the feeding was stopped.The achieved GL conversion was 95.34%with GAMGyield of95.15%,and GAMG concentration was 16.62 g·L-1which was more than 5 times higherthan the batch.These results indicate thatfeeding batch-wise addition of GL at appropriate time could significantly improve the transformation efficiency of GL.

3.4.Preliminary separation of GAMG by D101 resin packed column

In this study,D101 resins were packed into a column and used to crudely separate GAMG from fermentation broth.The dynamic absorption capacity ofGAMGwas investigated atthree loading flow rate:0.5,1 and 1.5 ml·min-1.As can be seen from Fig.7(a),the leaked amount of GAMG in the outlet increased as the increase of flow rate.This result was reasonable since the flow rate directly affected the diffusion of GAMG into the internal surface of the resin.When flow rate was too high,the contact time between GAMG and D101 was too short so that GAMG cannot efficiently diffuse into the internal space of the resin.At flow rate of 0.5 ml·min-1,the maximum absorption capacity achieved to be 76.1 mg·g-1resin,indicating thatthe loading flow rate should not be too high in order for preventing the leak of GAMG(Table 2).Therefore,0.5 ml·min-1was chosen as the optimal loading flow rate.The absorption capacity reported here was higher than the references.For example,Guo et al.used D101 resin to separate puerarin from Puerariae Lobatae Radix with absorption capacity of 51.4 mg·g-1[31].D101 resin has also been applied in the separation and purification of amygdalin from bayberry,and the absorption capacity was around 33.0 mg·g-1[32].The difference in absorption capacity should be attributed to the different characteristics of the target compounds,and comparison also indicated that D101 resin was suitable for the preliminary separation of GAMG.

Before eluting GAMG,the column was first washed with water for two bed volumes to remove the absorbed contaminants such as mineral salts,contaminant protein,mycelia and some other metabolites.Then,the target product GAMG was eluted with 70%ethanol at two different flow rates of 0.5 and 1 ml·min-1,as shown in Fig.7(b).For both flow rates,three bed volumes were required to completely wash off GAMG,and 1 ml·min-1was adopted for elution due to the high efficiency.After the preliminary separation,the purity of GAMG was 41.53%(Table 3).

Fig.5.The variation profile of pH(a)and DO(b)during fermentation.

Fig.6.The changing profile of GL(a),GAMG(b),pH(c)and DO(d)in fed-batch fermentation.

Fig.7.(a)The breakthrough curve ofGAMGatdifferentloading flow rates on the absorption ofD101 macroporous resins,(b)The elution ofGAMGwith 70%ethanolatdifferent flowrates.

3.5.Fine separation of GAMG by preparative HPLC

To obtain GAMG with higher purity,the crude GAMG from last step was further purified with preparative HPLC as described in Section 2.7.The chromatogram was shown in Fig.8,and the major peak was identified to be GAMG.The purity of GAMG after each purification step is listed in Table 3.After the separation of D101 resin,the purity of GAMG was 41.53%,and recovery yield was 90.24%.After the firstpurification by chromatography,the purity was 79.27%,when it was repeated again by chromatography,the final purity of GAMG was 95.79%with recovery yield of 53.45%.

Table 2 Effect of loading flow rate on the absorption yield(E a)and the maximum absorption capacity(Q max)of D101 resin

4.Conclusions

In this study,a fine substrate feeding strategy for fed-batch process was established by using pH and DO as feedback signals for the highlevel production of GAMG through fermentation of P.purpurogenum Li-3.The achieved GL conversion was 95.34%with GAMG yield of 95.15%,and GAMG concentration was 16.62 g·L-1which was more than 5 times higher than that of batch.Then,a two-step separation strategy was established to separate GAMG from fermentation broth by crude extraction of 15 ml column packed with D101 resin followedby fine purification with preparative C18 chromatography.The obtained GAMGpurity was 95.79%.The results indicate thatthe fermentation and downstream process developed in this study is bene ficial to the industrial production of GAMG.

Table 3 Purity and recovery yield of GAMG in different processes of purification

Fig.8.The chromatogram of the purification of GAMG by preparative HPLC with 87.5%methanol elution, flow rate of 10 ml·min-1 and GAMG loading of 250 mg.

References

[1]T.C.Kao,C.H.Wu,G.C.Yen,Bioactivity and potential health bene fits of licorice,J.Agric.Food Chem.62(2014)542-553.

[2]R.Pompei,O.Flore,M.A.Marccialis,A.Pani,B.Loddo,Glycyrrhizic acid inhibits virus growth and inactivates virus-particles,Nature 281(1979)689-690.

[3]R.Doll,I.D.Hill,C.Hutton,D.J.V.Underwood,Clinical trial of a triterpenoid liquorice compound in gastric and duodenal ulcer,Lancet 2(1962)793-796.

[4]G.Shiota,K.Harada,M.Ishida,Y.Tomie,M.Okubo,S.Katayama,H.Ito,H.Kawasaki,Inhibition of hepatocellular carcinoma by glycyrrhizin in diethylnitrosamine-treated mice,Carcinogenesis 20(1999)59-63.

[5]S.Matsui,H.Matsumoto,Y.Sonoda,K.Ando,E.Aizu Yokota,T.Sato,T.Kasahara,Glycyrrhizin and related compounds down-regulate production of in flammatory chemokines IL-8 and eotaxin 1 in a human lung fibroblast cell line,Int.Immunopharmacol.4(2004)1633-1644.

[6]S.J.Feng,C.Li,X.L.Xu,X.Y.Wang,Screening strains for directed biosynthesis of beta-D-mono-glucuronide-glycyrrhizin and kinetics of enzyme production,J.Mol.Catal.B Enzym.43(2006)63-67.

[7]W.J.Tang,Y.A.Yang,H.Xu,J.B.Shi,X.H.Liu,Synthesis and discovery of 18 alpha-GAMG as anticancer agent in vitro and in vivo via down expression of protein p65,Sci.Rep.4(2014)5.

[8]K.Mizutani,T.Kuramoto,Y.Tamura,N.Ohtake,S.Doi,M.Nakaura,O.Tanaka,Sweetness of glycyrrhetic acid 3-O-beta-D-monoglucuronide and the related glycosides,Biosci.Biotechnol.Biochem.58(1994)554-555.

[9]K.Mizutani,T.Kambara,H.Masuda,Y.Tamura,T.Ikeda,O.Tanaka,H.Tokuda,H.Nishino,M.Kozuka,T.Konoshima,Glycyrrhetic acid monoglucuronide(MGGR):biological activities,International congress series1998 225-235.

[10]H.Seki,K.Ohyama,S.Sawai,M.Mizutani,T.Ohnishi,H.Sudo,T.Akashi,T.Aoki,K.Saito,T.Muranaka,Licorice beta-amyrin 11-oxidase,a cytochrome P450 with a key role in the biosynthesis of the triterpene sweetener glycyrrhizin,Proc.Natl.Acad.Sci.U.S.A.105(2008)14204-14209.

[11]M.Kanaoka,S.Yano,H.Kato,T.Nakada,Synthesis and separation of 18-betaglycyrrhetyl monoglucuronide from serum of a patient with glycyrrhizin-induced pseudo-aldosteronism,Chem.Pharm.Bull.34(1986)4978-4983.

[12]D.-M.He,I.Kaleem,S.-Y.Qin,D.-Z.Dai,G.-Y.Liu,C.Li,Biosynthesis of glycyrrhetic acid 3-O-mono-beta-D-glucuronide catalyzed by beta-D-glucuronidase with enhanced bond selectivity in an ionic liquid/buffer biphasic system,Process Biochem.45(2010)1916-1922.

[13]H.Sakurama,S.Kishino,Y.Uchibori,Y.Yonejima,H.Ashida,K.Kita,S.Takahashi,J.Ogawa,beta-Glucuronidase from Lactobacillus brevis useful for baicalin hydrolysis belongs to glycoside hydrolase family 30,Appl.Microbiol.Biotechnol.98(2014)4021-4032.

[14]B.D.Wallace,H.W.Wang,K.T.Lane,J.E.Scott,J.Orans,J.S.Koo,M.Venkatesh,C.Jobin,L.A.Yeh,S.Mani,M.R.Redinbo,Alleviating cancer drug toxicity by inhibiting a bacterial enzyme,Science 330(2010)831-835.

[15]X.Song,Z.Jiang,L.Li,H.Wu,Immobilization of β-glucuronidase in lysozymeinduced biosilica particles to improve its stability,Front.Chem.Sci.Eng.8(2014)353-361.

[16]D.H.Kim,S.W.Lee,M.J.Han,Biotransformation of glycyrrhizin to 18 betaglycyrrhetinic acid-3-O-beta-D-glucuronide by Streptococcus LJ-22,a human intestinal bacterium,Biol.Pharm.Bull.22(1999)320-322.

[17]T.Akao,Hydrolysis of glycyrrhetyl mono-glucuronide to glycyrrhetic acid by glycyrrhetyl mono-glucuronide beta-D-glucuronidase of Eubacterium sp GLH,Biol.Pharm.Bull.20(1997)1245-1249.

[18]T.Kuramoto,Y.Ito,M.Oda,Y.Tamura,S.Kitahara,Microbial production of glycyrrhetic acid 3-O-mono-β-D-glucuronide from glycyrrhizin by Cryptococcus magnus MG-27,Biosci.Biotechnol.Biochem.58(1994)455-458.

[19]D.Q.Lu,H.Li,Y.Dai,P.K.Ouyang,Biocatalytic properties of a novel crude glycyrrhizin hydrolase from the liver of the domestic duck,J.Mol.Catal.B Enzym.43(2006)148-152.

[20]D.Dionisi,J.A.Anderson,F.Aulenta,A.McCue,G.Paton,The potential of microbial processes for lignocellulosic biomass conversion to ethanol:a review,J.Chem.Technol.Biotechnol.90(2015)366-383.

[21]C.Flores,A.Diaz-Barrera,F.Martinez,E.Galindo,C.Pena,Role of oxygen in the polymerization and de-polymerization of alginate produced by Azotobacter vinelandii,J.Chem.Technol.Biotechnol.90(2015)356-365.

[22]S.Pandit,G.Balachandar,D.Das,Improved energy recovery from dark fermented cane molasses using microbial fuel cells,Front.Chem.Sci.Eng.8(2014)43-54.

[23]C.Wang,X.X.Guo,X.Y.Wang,F.Qi,S.J.Feng,C.Li,X.H.Zhou,Isolation and characterization of three fungi with the potential of transforming glycyrrhizin,World J.Microbiol.Biotechnol.29(2013)781-788.

[24]M.Papagianni,Fungal morphology and metabolite production in submerged mycelial processes,Biotechnol.Adv.22(2004)189-259.

[25]S.Jin,S.Fu,J.Han,S.Jin,Q.Lv,Y.Lu,J.Qi,W.Wu,H.Yuan,Improvement of oral bioavailability of glycyrrhizin by sodium deoxycholate/phospholipid-mixed nanomicelles,J.Drug Target.20(2012)615-622.

[26]D.Karadag,O.E.Koroglu,B.Ozkaya,M.Cakmakci,S.Heaven,C.Banks,A review on fermentative hydrogen production from dairy industry wastewater,J.Chem.Technol.Biotechnol.89(2014)1627-1636.

[27]S.P.Zou,G.Y.Liu,I.Kaleem,C.Li,Purification and characterization of a highly selective glycyrrhizin-hydrolyzing beta-glucuronidase from Penicillium purpurogenum Li-3,Process Biochem.48(2013)358-363.

[28]S.P.Zou,S.Y.Guo,I.Kaleem,C.Li,Purification,characterization and comparison of Penicillium purpurogenum beta-glucuronidases expressed in Escherichia coli and Pichia pastoris,J.Chem.Technol.Biotechnol.88(2013)1913-1919.

[29]A.E.Posch,C.Herwig,O.Spadiut,Science-based bioprocess design for filamentous fungi,Trends Biotechnol.31(2013)37-44.

[30]X.Li,Y.Lin,M.Chang,Q.Jin,X.Wang,Ef ficient production of arachidonic acid by Mortierella alpina through integrating fed-batch culture with a two-stage pH control strategy,Bioresour.Technol.18(2015)275-282.

[31]H.D.Guo,Q.F.Zhang,J.G.Chen,X.C.Shangguang,Y.X.Guo,Large scale purification of puerarin from Puerariae Lobatae Radix through resins adsorption and acid hydrolysis,J.Chromatogr.B 980(2015)8-15.

[32]T.Wang,S.Lu,Q.Xia,Z.Fang,S.Johnson,Separation and purification of amygdalin from thinned bayberry kernels by macroporous adsorption resins,J.Chromatogr.B 975(2015)52-58.