Fermentation of xylose to produce ethanol by recombinant Saccharomyces cerevisiae strain containing

ARTICLESchinese Science Bulletin 2005 Vol. 50 No. 7 652-657allowing xylose to be converted to ethanol. 3. A majorFermentation of xylose todrawback with these recombinant strains is that the Km ofXR for NADPH is an order of magnitude lower than thatproduce ethanol by recombi-for NADH. This leads to accumulation of xylitol andNADH in the pathway, especially under anaerobic condi-nant Saccharomyces cerevisiae tions, In order to solve this problem, anvil not requiringstrain containing XYLA anderal bacteria, such as Actinoplanes missouriensis, Bacillussubtilis, Clostridium thermosulfurogenes, Escherichia coliXKSIand Lactobacillus pentosus had been cloned in S.cerevisiae. but the resulting recombinant strains could notLIU Xiaolin, JIANG Ning, HE Peng, LU Dajgrow on xylose as a sole carbon source 5.6)SHEN AnXylulokinase(XK), encoded by XKSI, is the third enInstitute of Microbiology, Chinese Academy of Sciences, Beijing 100080,zyme in the xylose metabolic pathway. It catalyses theconversion of xylulose to xylulose-5-phosphate. SignifiCorrespondence should be addressed to Jiang Ning(email: jang@ cant amount of xylulose observed in xylose fermentationbroth of XYLA harboring S. cerevisiae indicated thatproduce ethanol using lignocellulosic biomass would make activity in S. cerevisiae although xylose Co el of XKAbstract Fermentation of the pentose sugar xylose to xylose assimilation was limited by the native levbioethanol production economically more competitive. Sac- verted to xylulose. Owing to the low activity of encharomyce cerevisise, an efficient ethanol producer, cannot dogenous xylulokinase in the yeast 9, it is essential to in-utilize xylose because it lacks the ability to convert xylose to troduce an endogenous or a heterologous XKSl to increaseits isomer xylulose. In this study, XYLA gene encoding xylose xK activity to an optimal level for xylose uptake rate andisomerase (Xi from Thermoa naerobacter tengcongensisMB4T and XKSI gene encoding xylulokinase (XK)from subsequent ethanol fermentation 10-12)Pichia stipitis were cloned and functionally coexpressedThis study tries to elucidate the effect from co-expresSaccharomyces cerevisiae EF-326 to construct a recombinant sion of xylose isomerase gene from T. tengcongensixylose-utilizing strain. The resulting strain S. cerevisiae EF MB4T and xylulokinase gene from P. stipitis in S.cere1014 not only grew on xylose as sole carbon source, but also visite on the fermentation of xylose to produce ethanolproduced ethanol under anaerobic conditions. Fermentations End-product concentrations and activities of the recombitemperatures demonstrated that the highest ethanol produc- identify vmes under various conditions were analyzed toperformed with different xylose concentrations at different nant ettivity was 0.11 g/g xylose when xylose concentration was propossible strategy for the further improvement ofvided at 50 g/L. Under this condition, 28. 4% of xylose wasthe strainconsumed and 1.54 gL xylitolformed. An increasing 1 Materials and methodsfermentation temperature from 30C to 37 did not im- 1.1 Strains and plasmidsprove ethanol yieldKeywords: xylose isomerase, xylulokinase, co-expression, xylose,E. coli dh5a and s cerevisiae ef 326 were used asimyces cerevisiaehosts for plasmid constructions and yeast transformation,DOI:10.1360982004668espectively. T. tengcongensis MB4T, a thermophile iso-lated from a hot spring in western China, was chosen asXylose utilization is of commercial interest for efficient the source of XYLA. Its complete genome sequence isonversion of abundant plant material to produce ethanol. available in Genbank(Accession No. AE008691)3].TheIn xylose-fermenting yeasts, xylose is first reduced to natural xylose-utilzing yeast P. stipllts was used as thexylitol by xylose reductase(XR) and then oxidized to xy- source of XKSI. Plasmids pURL129 and pVgb-EX2 wereulose by xylitol dehydrogenase (XDh). In bacteria,available in the labxylose is directly isomerized to xylulose by xylose isom- 1.2 Media and culture conditionserase (Xi) before entering pentose phosphate pathwaySaccharomyces cerevisiae is the most important ethanE. coli was grown at 37C in LB medium suppleproducer so far for its high alcohol tolerance, productivity mented with 100 ug/mL ampicillin when necessary. Yeastand rapid fermentation rate using glucose. But it is unable strains were中国煤化士t2% glucose1 %o yeast exto utilize xylose due to the absence of XR and XDH, the tract, 2%first two enzymes of xylose metabolism.(YEPD). YeCNMH Gected on YEPDo Previous attempts to introduce P stipitis NAD(P)H-de- containing 0.6 mmol/L formaldehyde. For growth ex-dent XR, and NAD-dependent XDH into S cerevisiae periment, yeasts were cultivated on YP medium contain-have yielded a new metabolic pathway allowing xylose to ing 2% xylose For enzymes assays, YP medium suppleChinese Science Bulletin Vol 50 No. 7 April 2005ARTICLESmented with 0.5%o glucose and 2%0 xylose was usedtions of each primer, 0. 1 ug of template, and 2.5 U of Pfirpolymerase enzyme in a final volume of 50 uL. The ther-ycler(Eppendorf 5530, German) was used undeNucleic acid manipulations were performed by stan- following conditions: 94C for 10 min; 30 cycles of 94Cdard techniques". Plasmids were prepared by the Ding- for 40s. 50C for 40 s. and 72C for 80s: 10 min at 72Cguo mini plasmid purification kit(Dingguo BiotechnologyBeijing, China). Restriction enzymes and other modifyThen the mixture was chilled to 4C. For the amplificationenzymes were purchased from Takara (Takara, Dalian, of XKS1, denaturalization temperature was 55C. and ex-China). Transformation of S. cerevisiae was performed tension time was 120 saccording to the procedure described by Adams et al. 51The PCR products of XYLA and XKS/ were insertedand the calcium chloride method was used in the plasmid into plasmid pRULI29 sequentially to obtain plasmidtranformation of E. coli[141pRUX2. To replace the selectable marker URA3 withThe genomic DNA of T. tengcongensis MB4T was SFAl, the latter encodes formaldehyde dehydrogenase andused as template for the amplification of XYLA by PCR. results in formaldehyde resistance; pRUX2 and pGbThe primers were Pl(5-CGTCTCTAGAATTTGTGGAT- EX2 were digested with Sac I and Kpn I, respectively. TheTCCCTGGC-3, Xba I is underlined)and P2(5'-CGCA- PCR products obtained from pRUX2 and the selectableTGATCAGCTTTCTACCCCTCCTCG-3, Bcli is un- marker SFAl fragment from pVgb-EX2 were ligated, e-derlined ) To amplify XKSl, the genomic DNA of P stipi- sulting in pRX2(Fig. 1). Plasmid pRX2 was used totis was used as template, and the primers were Pl (5'-. transform S. cerevisiae EF 326AATAGATCTTCACGTAGTTGACACTCAC-3, Bglll 1.4 Enzyme assaysis underlined) and P2 (5-TGAAGCTTGTCTATCGTGATATTCGCAC-3, Hind lI is underlined ) Both genesYeast cells were grown at 30C in YEPD medium towere cloned with their native promoters and terminators. stationary phase. The cells were harvested by centrifugaAll primers were designed according to the published se- tion at 3000 g for 5 min and washed once in 0.9%NaClquences in Genbank using Primer Primer 5.0 softwarehe pellet was resuspended in disintegration buffer(100For the amplification of XYLA, Pfu polymerase(Promega, mmol/L triethanolamine pH 7.0, 1.0 mmol/L PhenylUSA)was used. The PCR mix contained PCr buffer with methanesulfonyl fluoride) and vortexed twice for 5 min at2 mmol/L MgSO4, 2 mmol/L dNTP, 0.5 umol concentra- 4C with an equal volume of glass beads(0.5 mm in diURA3XYLACEN6/ARSH4Hind llPGPDI9000bpYKSICEN6ARSH4SFA八CYClXYLABcl I9400cYc17YKSHind lllPGPDIBgl llCEN/ARSHXba ICEN6/ARSH4pRUL129pVgb-EX2CyCl000bp7200bpIcYCGPDI SFA1∥Bgll中国煤化工VgbCNMHGKba BcIFig. 1. Strategy of plasmid pRX2 construction. Restriction enzymes sites used in plasmid construction are shownC9数鸦 ce Bulletin Vo.50No.7Apn2005ARTICLESameter). The disintegrated cell mixture was centrifuged at Transformants were selected on YPD agar plates contain000 g for 5 min at 4C, and the supernatant was stored at ing 0.6 mmol/L formaldehydeaccording to the method of Bradford ls s were determined d. ce y previous attempts to express XI from bacteria in20C until analyzed for protein concentration and enzyme activities. Protein concentrationvisiae were unsuccessful because most XIs produced by the recombinant S. cerevisiae are inactive. ImAssays of the recombinant XI were performed by a proper protein folding, posttranslational modifications,nethod according to Kuyper et al. 7. XK activity was as- inter- and intra-molecular disulfide bridge formation andsayed as described by Flanagan!l6. Specific activities of the internal ph of yeast have been thought of as possibleenzymes were expressed as Units/mg protein. One unit is reasons for the inactive proteins produced. Both Waldefined as the amount of enzyme required to convert 1 fridsson et al. and Bao et al. o speculated that the key toumol of substrate per minute under the assay conditions. constructing an efficient recombinant S. cerevisiae strainActivities in cell extracts were determined spectropexpressing the active XYLA gene was to use a close relatedtometrically at 340 nmgene donor in taxonomy. So far most functional expressedActivities of the recombinant xi were measuredheretologous XIs are from Thermus thermophilus.Invarious temperatures between 30 and 100C, pH7.0. Asthis paper, we demonstrated that the XYLa gene from Tay mixture contained Mgso4. 7H,o 10 mmol/L, trietha-tengcongensis MB4T could be efficiently expressed. Thisprovided a new choice of XYLA Source for engineeredtol dehydragenase(Sigma)0.2 U and cell-free extract 0.2 Xylose-fermenting strainmL, adjusted to a final volume of 1.0 mL. The reactionsThe P. stipitis xylulokinase( Genbank Accession Nowere started by adding 250 mmol/L D-xylose ReactionAF127802) is the only xylulokinase characterized frontime was 30 mineukaryote that is helpful for converting xylose to ethanolXK assay mixture consists of Tris 50 mmol/L, glycine The heterologous expression of the P. stipitis gene for50 mmol/L, potassium chloride 50 mmol/L, EDTA I xylulokinase enhances xylulose fermentation in S. ceremmol/L, potassium cyanide 10 mmol/L, sodium fluoride visiaelo. The ORF of P. stipitis XKSI is 1872 bp.It10 mmol/L, phosphoenolpyruvate 1 mmol/L, magne- showed 97.76% identity to the published sequence. Thesium chloride 5 mmolL. ATP 0.5 mmol/L NAdH 0.3 amino acid sequences deduced from their ORFs displayedmmol/L, lactate dehydrogenase and pyruvate kinase 25 U, 99.83%0 homology. P. stipitis D-xylulokinase differs fromand cell-free extract 0. 1 mL, in a final volume of 0.5 mL. S. cerevisiae endogenous Xk in that it does not possessReactions were started by adding 1 mmol/L D-xyluloseD-ribulokinase activity and hence is more specific for1.5 Fermentation conditions and analytical methodsFermentations were carried out in 250 mL shake-flask2.2 Enzyme activitiesfilled with 100 mL of YP medium supplemented with xyStationary phase cells were harvested, and the specificlose as sole carbon source. The flasks were plugged with activities of XI and XK of EF-326 and EF1014 wererubber stoppers, and a gas outlet was secured by inserting measured. The specific activities of XI and Xk in cell-freea cannula. Fermentations were performed at 30C or extracts of the recombinant strain EF 1014 at 30C were37C, at 100 rpm shaker for 72 h. The initial cell concen- 0.12 and 1. 2 U/mg protein, respectively. At 37C,recomration was I g(dry weight)/L. All the experiments were binant Xl activity increased slightly to O14 U/mg protecarried out in duplicwhile XK activity decreased drastically to 0.07 U/mg proGrowth rate of yeast cell was determined by cell dry tein. Cell-free extract of the parent strain, S. cerevisiaeweight(CDW, g/L). A known volume of culture sample EF-326, lacked detectable activities of XI and XK at anywas centrifuged 3000 g at 4C for 5 min, and washed temperaturetwice with distilled water. The pellet was dried overnight XI activities were also measured at different temperaat 105C on pre-weighed dry weight cups and then tures and pHs. The optimum temperature of the recombiweighed. Xylitol concentration was determined by chemi- nant XI was 80C. At 80C, the activity of XI reached 1.0cal methodl/. Ethanol and xylose concentrations were U/mg protein. The activity decreased drastically at temdetermined using a biosensor(SAB-400peratures below 60C and were only 12% and 15% of the2 Results and discussionmaximum at 30C and 40C, respectively. This result2. 1 Expression of XYLA and XKSI in S cerevisiae中国煤化工 was still thermoXYLA and XSLfused to the gpditaple metHCNMHGm PH 6.0 to 7.0,CyCI terminator in expressing vector pRX2 with SFA/ asFurthermore. the recombinant strain exhibited stable Xiselectable marker. S. cerevisiae eF-326 was transformed and XK activities through 20 generations of continuouswith plasmid pRX2, resulting in S. cerevisiae EF-1014. cultivationChinese Science Bulletin Vol 50 No. 7 April 2005ARTICLESTemperature/Cig.2. The relative activity at different temperatures and pH of the transformed isomerase2.3 Effect of expressing XYLA and XKSl on the growth The low efficiency of the xylose transport system and theof recombinant strainlow XI activity at the growth temperature may contributeto the slow growth of S. cerevisiae EF-1014 on xyloseGrowth experiments were carried out both in liquid cul- Xylose is transported by the facilitated glucose transport-res and on plates. The results showed that the cell diing system in S. cerevisiae cells, which has a 200-foldweight of the recombinant strain EF-1014 increased sig- lower affinity for xylose than for glucosel9,20.Additionnificantly after 72 h culture in the medium with xylose as ally, the S cerevisiae cytosolic environment is not optimalsole carbon source while the growth of the parent strainEF 326 was hardly detectable(Fig 3). EF 326 and EF for XI activity. The optimum temperature for XI is 80c1014 both grew quickly on glucose, but only EF 1014 while the experimental growth temperature is only 30Cgrew on xylose(Fig 4). Compared to cultivation on glu- 2.4 Effect of expressing XYLA and XKSI on anaerobiccose, the growth rate of EF 1014 on xylose was still low. xylose fermentationEF-32EF 1014 fermentations were performed with 50, 10010150, 200 g/L xylose as carbon source, respectively. Etha-nol and xylitol formation were measured. The results aresummarized in Table 1. Xylose consumption increased0.6slightly, although the xylose concentration increased drastically. Xylitol yield increased in response to an increasein xyloseption. Ethanol yield was highest whenxylose concentration was 100 g/L. But the highest ethanoproductivity and xylose utilization rate were observed atth of EF-326 and EF-1014 with xylose as the sole carbon tivity was O Il g/g xylose, and 28.4%0 of xylose was utilource in shake-flask cultureized. Thus comparison of EF 326 and EF 1014 was per-中国煤化工CNMHGFig 4. Growth of EF 326(the left)and EF 1014( the right)on xylose and glucose. YP medium with 20 g sugar per liter was used. Theplates were incubated at 30C for 3 dC閂数据 nce Bulletin Vol.50No.7Apmn2005ARTICLESTable 1 Fermentation results of eF-1014 with different xylose concentration at 30CXylose concentrationXYlitol yielEthanol yieldEthanol productivity Residue xylose Xylose conversion rateg·Lg·( xylose)1.60.111.680.1016132.71.781.550.09182.3formed at this xylose concentration, ef 326 metabolized microbiology, CAS) for the gift of Thermoanaerobacter tengcongensis3.3 g/L of xylose, and produced 0.91 g/L of xylitol with- genome DNA. This research was supported by the National Basic reout ethanol formation. Ow did not increase after fer- Knowledge Innovation Project from the Chinese Academy of Sciencesmentation for 72 h. EF 1014 utilized 14.2 g/L of xylose, Referencesylitol and ethanol were produced at 1.54 and 1.61 g/Lrespectively. CDW also increased more than two-fold in 1. Traff, K L, Otero Cordero, R.R., van Zyl, W. H et al., Deletion ofthe end of fermentation. The amount of xylose decreasinthe gRE3 aldose-reductase gene and its influence on xylose neslightly in EF 326 fermentation medium and xylitol for-abolism in recombinant strains of Saccharomyces cerevisiae exmation in both ef 326 and EF 1014 fermentation brothpressing the xylA and xksl genes, Appl. Environ. Microbiol., 2001ht be due to a putative aldo-keto reductase(Ar)en-67:5668-5674coded by the GRE3 gene/2. Deletion of the GRE3 gene in 2. Walfridsson, M, Hallborn, J, Penttila, M et al., Xylose-metaboScerevisiae is efficient in reducing xylitol formation22lizing Saccharomyces cerevisiae strains overexpression TKLI andAlternatively, an endogenous XDH2 could cause theTALl genes encodingtose phosphate pathway enzymeequilibrium of this reaction to favor xylitol formationtransketolase and transaldolase, Appl. Environ. Microbiol., 1995.Xylitol has a dual effect on ethanol formation. It inhibits61:4184-4190.XIZ+, reducing the rate of xylose uptake and diverts car3. Sedlak, M, Ho. N. wbon flow from ethanol-producing metabolism. EF 1014biomass hydrolysates using genetically engineered Saccharomycesfermentations with different xylose concentrations wereyeast capable of cofermenting glucose and xylose, Appl. Biochemlso carried out at 37C, but the results(data not shown)Biotech4. Kuyper, M, winkler, A. A, van Dijken, J. P. et al., Minimal metahad no obvious difference with the results at 30Cbolic engineering of Saccharomyces cerevisiae for efficient anSo far all the recombinant S. cerevisiae strains expressaerobic xylose fermentation: A proof of principle, FEMS Yeasting XYLA only can hardly grow on xylose as sole carbones2004,4:655-664.sourcel4-b. 21. Both cell growth and ethanol formation wereWalfridsson, M. Bao.X. M. Anderlund. M. et al. ethanolicshown in a recombinant strain TMB3102 which expressedmentation of xylose with Saccharomyces cerevisiae harboring theXYLA but deleted GRE3I. Xylitol was formed but noThermus thermophilus xylA gene, which expresses an active xyloseethanol was detected by the strain TMB 3103 expressin(glucose)isomerase, Appl. Environ Microbiol., 1996,62:XYLA plus XKS/LIncreasing of xylose consumption rate and decreasing 6. Bao, X.M., Gao, D Wang, Z.N., Expression of xylose isomeraseof xylitol accumulation should be the key measures togene(xylA)in Saccharomyces cerevisiae from Clostridium Ther-improve ethanol production from xylose fermentationmohydrosulfiricum(in Chinese), Acta Microbiologica Sinica, 1999,Using a multicopy plasmid is efficient in improving XIhich should lead in more efficient xylulose7. Kuyper, M, Harhangi, H. R, Stave, A. K, High-level functionalmetabolism. So that xylulose is not available for the enexpression of a fungal xylose isomerase: The key to efficient etha-dogenous XDH and xylitol formation decreases. Metanolic fermentation of xylose by Saccharomyces cerevisiae? FEMSbolic modeling has suggested that a constitutively overYeast res2003,1574:1-10produced kinase at the beginning of a metabolic pathway 8. Jin, Y S, Jeffries, T.w., Changing flux of xylose metabolites bydepletes the intracellular ATP pool and causes intracellularaltering expression of xylose reductase and xylitol dehydrogenaseaccumulation of sugar phosphates 26. Growth inhibitionin recombinant Saccharomyces cerevisiae, Appl. Biochem. Bicincreased and ethanol yields from xylose decreased withtechnol,2003,105-108:277-286increasing XK activit 81. These suggest that levels of in- 9. Xue, X. D, Ho, N. W.Y., Xylulokinase activity ineaststroduced enzyme ativity should be designed in concertnces cerevisiae containkinase gelwith the capacity of the surrounding metabolic network中国煤化工∞0While strain improvement will probably continue for several years, ethanol production rates and yields are becom-10. Jin.Y SCNMH Gular cloning of XYL3ing practicable for some commercial applications/271(D-xylulokinase) from Pichia stipitis and characterization of itsphysiological function, Appl. Environ. Microbiol, 2002,Acknowledgments The authors are grateful to Dr. Xiang(Institute of1232-1239Chinese Science Bulletin Vol 50 No. 7 April 2005ARTICLES11. Johansson, B, Christensson, C, Hobley, T. et al, Xylulokinase Chinese Science Bulletin 2005 Vol50 No.7657-664pressing xylose reductase and xylitol dehydrogenase and its effect Paleomagnetic dating of theon fermentation of xylose and lignocellulosic hydrolysate, ApplEnviron. Microbiol. 2001. 67: 4249-4255topmost terrace in Koumaanalysis of xylose metabolism in recombinant Saccharomyces cer Henan and its indication toevisiae using continuous culture, Metabolic Engineering, 200316-31the Yellow River s running13. Bao, Q, Tian, Y, Li, w. et al., A complete sequence of the Ttengcongensis genome, Genome Res, 2002, 12: 689-700through sanmen Gorges14. Sambrook, J. Fritsch, E. F, Maniatis, T, Molecular CloningLaboratory Manual, 2nd ed, New York: Cold Spring Harbor Labo- PAN Baotian, WANG Junping, GAO Hongshantory Press, 1989GUAN Qingyu, WANG Yong, SU Huai15.Adams,A,Gottschling,DE.Kaiser, C. A et al., Methods in LI Bingyuan& LI JijunYeasts Genetics, Cold Spring Harbor Laboratory Course ManualNew York: Cold Spring Harbor Laboratory Press, 19971. Key Laboratory of Western Chinas Environmental Systems, Ministry16. Flanagan, T, Waites, M. J, Purification and characterization ofof Education, and Department of Geography, Lanzhou University,D-xylulokinase from the pentose- fermenting yeast PichiaLanzhou 730000 China:NCYC-1541, Enzyme Microb. Technol., 1992, 14: 975--979of Geographic Sciences and Natural Resources ResearchChinese Academy of Sciences (CAS), Beijing 100117. Bok, S. H, Demain, A. L, An improved colorime tric assay for Correspondence should be addressed to Pan Baotian(email: panbt@lzupolyols, Anal. Biochem., 1977, 81: 18-208. Amore, R, Wilhelm, M, Hollenberg, C. P, The fermentation ofxylose-An analysis of the expression of Bacillus and antinoAbstract In the east of Xiaolangdi, many river terraceslanes xylose isomerase genes in yeast, Appl Microbiol. Biotech- are developed at the exit of the Yellow River Gorges. Amongnol1989.30:351-357them the terraces in Kouma yanshi of henan province areSedlak, M. Ho. N. w.Y., Characterization of the effectiveness ofmost typical, where the Yellow River developed three stair.exose transporters for transporting xylose during glucose and xycase terraces, among which the altitude of gravel stratum oflose co-fermentation by a recombinant Saccharomyces yeast, Yeast, he topmost terrace is 30-35 m higher than the river level.2004,21:671-684The top of the gravel stratum was covered by 60 m eolian20. Eliasson, A. Christensson, C. Wahlbom, C. F. et al., Anaerobicloess deposits which have many brownish-red paleosol stripsxylose fermentation by recombinant Saccharomyces cerevisiaeAnd the paleosol S1 is at its bottom. Research on systematicrrying XYLl, XYL2 and XKSI in mineral medium chemostat cumagnetostratigraphy and paleosol-loess matching indicatestures, Appl. Environ Microbiol., 2000, 66: 3381-3386that the bottom age of the loess on the topmost terrace is1165 ka. Therefore. it can be concluded that the terrace deNADPH-dependent aldose reductase activity influences product velops no later than 1 165 Ma and the situation that the Yel-formation during xylose consumption in recombinant Saccharmvces cerevisiae. Yeast. 2004. 21: 141-150East China Sea happened at least before 1.165 Ma.22. Traff,KL,Lonn,A, Otero Cordera, R.R. et al., xylose isom- Keywords: Yellow River, terrace, paleomagne tism, dating, run-erase activity influences xylose fermentation with recombinantSaccharomyces cerevisiae strains expressing mutated xylA from DOI: 10.1360/03wd0290e microbHow did the Yellow River develop and evolve? When23. Richard, P, Toivari, M. H, Penttila, M, Evidence that the gene did it form? The geoscientists have concerned these m-YLRO7Oc of Saccharomyces cerevisiae encodes a xylitol dehydro-portant questions for a long time. They have done a lot ofgenase, FEBS Lett., 1999, 457: 135-13824. Yamanaka, K, Inhibition of D-xylose isomerase by pentitols andresearches on these issues and gotten a lot of valuableD-lyxose, Arch. Biochem Biophys., 1969, 107: 179-197results, while there are still many controversies. In general,it mainly includes several points as follows. According to25. Moes, C.J., Pretorius, L.S., van Zyl, W H, Cloning and expression the transition of the fluviolacustrine strata, biological fos-of the Clostridium thermosulfurogenes Dxylose isomerase gene(xylA)in Saccharomyces cerevisiae, Biotech. Lett, 1996, 3: 269sils and geologic structure in Lanzhou and Yinchuan, Linmade a conclusion that the yellow river flowed into26. Teusink, B, Walsh, MC, van Dam, K et al, The danger of Weihe River in Eocene, and developed its square bendmetabolic pathways with turbo design, Trends Biochem. Sci., 1998, around the qocene23:162-169timen.The中国煤化工 revealed that the27. Jeffries, T. w. Jin, Y S, Metabolic engineering for improved fer- Yellow Rivermentation of pentoses by yeasts, Appli. Microbiol. Biotechnol, Sanmen gorCNMHGd shannxi and2004,63:495-509searchers considered that the middle and lower reaches of(Received November 23, 2004; accepted March 2, 2005) the Yellow River had formed between Early-PleistoceneChinese Science Bulletin Vol 50 No. 7 April 2005