Isolation and Characterization Of

5/26/2018 Isolation and Characterization Of

1/12

Microbiological Research 163 (2008) 255266

Isolation and characterization ofCandidamembranifaciens subsp. flavinogenie W14-3,a novel riboflavin-producing marine yeast

Lin Wang, Zhenmin Chi, Xianghong Wang, Liang Ju, Zhe Chi, Ning Guo

Unesco Chinese Center of Marine Biotechnology, Ocean University of China, Yushan Road, No. 5, Qingdao,

Shandong 266003, China

Received 17 June 2007; received in revised form 3 December 2007; accepted 9 December 2007

KEYWORDS

Riboflavin;Marine yeasts;C. membranifaciens

subsp. flavinogenie;Molecular

identification

SummaryWe found that the marine yeast strain W14-3 isolated from seawater of China EasternSea could produce riboflavin. It is interesting to observe that the marine yeast strainproduced a large amount of riboflavin in the medium containing xylose, sucrose,galactose and maltose under the conditions of vigorous shaking. The yeast strain wasfound to belong to Candida membranifaciens subsp. flavinogenie based on theresults of routine and molecular identification. The protein sequences deduced fromthe partial genes encoding GTP cyclohydrolase II and 3,4-dihydroxy-2-butanone-4-phosphate synthase in the yeast exhibited high identity with those of thecorresponding enzymes for riboflavin biosynthesis in other yeasts. Fe3+ available inthe medium repressed riboflavin production and expression of the genes responsiblefor riboflavin biosynthesis in the yeast. The results have evidenced that a riboflavinsynthesis pathway indeed existed in the yeast. This is the first study to report thatC. membranifacienssubsp. flavinogenieW14-3 from the marine environment couldproduce riboflavin.&2008 Elsevier GmbH. All rights reserved.

Introduction

Riboflavin, a yellow, water-soluble vitaminhas many physiological roles in human and animals.The best-known biochemically active coenzymes

formed from riboflavin are mononucleotide (FMN)and flavin adenine dinucleotide (FAD), needed aselectron acceptors in oxidoreductases (Stahmannet al., 2000). In order to avoid deficiency symptomslike dermatitis, a nutritional requirement of0.31.8mg riboflavin/day for humans and 14 mgriboflavin/kg diet for animals is recommended.Riboflavin can also be used in soft drinks and yogurt

ARTICLE IN PRESS

www.elsevier.de/micres

0944-5013/$ - see front matter&2008 Elsevier GmbH. All rights reserved.doi:10.1016/j.micres.2007.12.001

Corresponding author. Tel/fax: +86 532 82032266.E-mail address:[email protected] (Z. Chi).

5/26/2018 Isolation and Characterization Of

2/12

(Stahmann et al., 2000). Chemical production stillaccounts for the major part of industrial riboflavinsynthesis. However, the disadvantages of thechemical synthesis are a maximum yield of about60%, thus generating a lot of waste, requiringorganic solvents and 25% more energy in compar-ison to the fermentation process (Stahmann et al.,

2000). In recent years, the commercial fermenta-tions for riboflavin production have been estab-lished and it was found that riboflavin productionby fermentation has many merits over the chemicalprocess. For example, mild conditions are used,less energy is needed, riboflavin is more easilyrecovered and less wastes are produced during thefermentation. So far the microorganisms that havebeen used for riboflavin biosynthesis on a largescale include the hemiascomycetes Ashbya gossypii(Stahmann et al., 2000; Wendland and Walther,2005), a filamentous fungus, Candida famata

(Stahmann et al., 2000), a yeast, Pichia guillier-mondii (Fayura et al., 2007) and the geneticallyengineeredBacillus subtilis, which requires at leastthe deregulation of purine synthesis and a mutationin a flavokinase/FAD synthetase (Stahmann et al.,2000).

During the isolation and identification of themarine yeasts obtained in this laboratory, we foundthat the culture of the marine yeast strain W14-3isolated from seawater of China Eastern Sea turnedyellow when it was grown in the medium containingxylose, sucrose, galactose and maltose, at 25 1C for24 h. Therefore, the main purposes of the present

study are to identify and characterize the marineyeast strain W14-3. We also purify and identify theyellow substance produced by the marine yeaststrain W14-3.

Experimental

Yeast strains

Different samples (200 m depth, 8 1C, pH 8.1 and2.89% salinity) of seawater and sediments of China

Eastern Sea were collected during winter 2006. Twograms of the sediments and 2.0 ml of the seawaterwere suspended in 20.0ml of YPD medium contain-ing 2.0% glucose, 2.0% polypeptone and 1.0% yeastextract and supplemented with 0.05% chloramphe-nicol (to inhibit bacterial growth) immediatelyafter sampling and cultivated at natural tempera-ture for 5 days. Suitable dilutions were preparedand plated out on YPD agar plates with 0.05%chloramphenicol and the plates were incubated at2025 1C for 5 days. Different colonies from theplates were transferred to the fresh YPD plates.

One isolate, strain W14-2, isolated from seawaterof China Eastern Sea, was selected for further studybased on its production of a yellow substance. Theyeast strain was identified to be Candida membra-nifaciens subsp. flavinogenie according to theresults of routine yeast identification and molecu-lar methods as described below (MCCC No.

2E00233).C. membranifaciensPYCC2727T, the typeyeast strain, was kindly supplied by Dr. IsabelSpencer-Martins from Centro de Recursos Micro-biologicos (CREM) Biotechnology Unit, Faculty ofScience and Technology, New University of Lisbon,Portugal. The yeast strains were maintained in YPDmedium at 4 1C.

Riboflavin production

One loop of the cells of the yeast strain wastransferred to 50.0 ml of YPD medium preparedwith distilled water in 250 ml flasks and aerobicallycultivated at 251C for 24 h. At a point whenthe culture reached a high cell density(OD600nm 20.0), 0.2 ml of the culture was trans-ferred to 50.0 ml of the production medium thatcontained 2.0% of different sugars (xylose, maltose,galactose, sucrose, glucose, trehalose, raffinoseand cellobiose), 0.5% of (NH4)2SO4, 0.1% of KH2PO4,0.05% of MgSO4 7H2O, 0.01% of CaCl2 2H2O, 0.01%of NaCl and 0.2% of yeast extract and grown byshaking at 170 rpm and 25 1C for 5 days. In order todetermine effects of different carbon sources on

riboflavin production by the yeast strain, thedifferent carbon sources were added to theproduction medium. The culture was centrifugedat 14,000gand 4 1C for 10 min and the supernatantobtained was used as the crude riboflavin solutionfor quantitative determination of riboflavin.

Effects of iron in the production medium onriboflavin production in the yeast

In order to study the effects of iron in theproduction medium on riboflavin production by the

marine yeast used in this study, iron was removedfrom the medium with 8-hydroxyquinoline asdescribed by Cowart et al. (1980). The iron-supplemented medium contained 0.005% FeCl3.

Quantitative determination of riboflavin

The amount of riboflavin in the supernatant wasmeasured quantitatively at 440 nm by using aspectrophotometer, and riboflavin from Sigmaserved as standard.

ARTICLE IN PRESS

L. Wang et al.256

5/26/2018 Isolation and Characterization Of

3/12

Partial purification of riboflavin

Fifty milliliters of 0.1mol/l hydrochloric acid wasadded to the crude riboflavin solution (50ml) in a250ml conical flask. The solution was placed in awater bath at 100 1C for 30 min. After being allowedto cool, it was adjusted to pH 4.5 with 2.5mol/l

sodium acetate (Ndaw et al., 2000) and filteredthrough filter paper. The filtrate obtained wasapplied to an MTcleanup and concentration column(TEDA FUJI Chromatogram Developing Corp). Theriboflavin absorbed on the column was eluted byusing pure methanol. The eluate was filteredthrough a MILLEXsHV filter (0.45 mm). The filtratewas used for high-performance liquid chromato-graphy (HPLC) analysis of riboflavin.

Chromatographic analysis

The riboflavin in the filtrate was analyzed byusing a HPLC system (Waters, USA). The systemconsisted of a pump Waters Delta 600, a detectorWaters 996 and a column YMC-pack ODS-(A),2504.6 mm, 5mm. The elution was performedby a methanol:water solution 40:60 and the flowrate was 1.0 ml/min. Other conditions used werethose described by Arella et al. (1996) for thedetermination of riboflavin. Chromatographicpeaks were quantified using a Star Chromato-graphic integrator. The riboflavin standard waspurchased from Sigma.

DNA extraction, PCR and DNA sequencing

DNA extraction and PCR techniques for amplifi-cation of D1/D2 26S rDNA in the yeast wereperformed according to the methods described byChi et al. (2007). The common primers foramplification of the D1/D2 26S rDNA sequence inthe yeast were used, the forward primer NL-1:50-GCATATCAATAAGCGGAGGAAAAG and the reverseprimer NL-4: 50-GGTCCGTGTTTCAAGACGG (Sugitaet al., 2003). The D1/D2 26S rDNA fragments

inserted on the vector (pMD-19T) were sequencedby Shanghai Sangon Company.

Phylogenetic analysis of the yeast

The sequences obtained above were analyzed forsimilarity by using Clustal X 1.83. For comparisonwith currently available sequences, sequenceswere retrieved with over 98% similarity belongingto different genera from NCBI (http://www.ncbi.nlm.nih.gov ). The phylogenetic treewas constructed by using PHYLIP software package

version 3.56 (Felsenstein, 1995). Distance matriceswere generated by the DNADIST program, based onKimuras two-parameter model (Kimura, 1980).Neighbor-joining analysis of the data sets wascarried out with the program Neighbor of thePHYLIP package.

Metabolic characterization of the yeast

The yeast strain was also identified by usingBIOLOG system TM (Biolog MicroStation with Micro-log System, Release 4.20, Biolog, Hayward, CA,USA), Biolog Universal Yeast Agar (Biolog) andBiolog Yeast microplate (Biolog) according to theprocedures offered by the manufacturer (Praphai-long et al., 1997). The routine identification of theyeast was performed by using the methods asdescribed byKurtzman and Fell (2000).

PCR-based cloning of partial GTPcyclohydrolase II gene and 3,4-dihydroxy-2-butanone-4-phosphate synthase gene

PCR was used for partial GTP cyclohydrolaseII and 3,4-dihydroxy-2-butanone-4-phosphate synthaseDNA amplification. Genomic DNA was preparedas described above. The conserved motifswere usually used to design degenerate primersto clone these homologs. In this case, amino acidsequences of GTP cyclohydrolase II and 3,4-dihy-droxy-2-butanone-4-phosphate synthase from dif-

ferent species of eukaryotic microorganisms weredownloaded from GenBank (http://www.ncbi.nlm.nih.gov/) and aligned. The degenerate senseprimers and antisense primers for amplification ofpartial GTP cyclohydrolase II DNA were rib1_1_2f1:CAACAGGGAACACTTGGCAAT and rib1_1_4r2:CGTTCACCACAATCACATCGT, and the degeneratesense primers and antisense primers for amplifica-tion of partial 3,4-dihydroxy-2-butanone-4-pho-sphate synthase DNA were rib3_0_1f: GAACGTG-AAAATGAAGGTGAT and rib3_3_3r:CGACACCGACCT-CAGTGT. PCR was performed on a GeneAmp PCR

System 2400 made by Perkin

Elmer using a programof 94 1C for 1 min, 52 1C for 1 min and 72 1C for 2minfor 30 cycles, followed by extension for 8 min at72 1C. PCR products were separated by agarose gelelectrophoresis and recovered by using UNIQ-col-umn DNA gel recovery kits (BIOASIA, Shanghai). Therecovered PCR products were ligated into pMD19-Tand transformed into competent cells of Escher-ichia coli DH5a. The transformants were selectedon LB plates with ampicillin. The plasmids in thetransformant cells were extracted by using themethods as described by Sambrook et al. (1989).

ARTICLE IN PRESS

Isolation and characterization ofC. membranifacienssubsp. flavinogenieW14-3 257

5/26/2018 Isolation and Characterization Of

4/12

The cloned DNA fragments inserted on the vectorwere sequenced by Shanghai Sangon Company. Theamino acid sequences of the cloned DNA frag-ments were deduced and protein sequences werealigned using CLUSTAL X program (Thompson et al.,1994).

Total RNA isolation and RT-PCR

Total RNA was isolated from strain W14-3 grownin the production medium without and with ironsupplementation using RNAiso Reagent (TaKaRa,Japan) according to the manufacturers instruc-tions. Total RNA in the sample was determined byspectrophotometry at 260 nm. The same amount oftotal RNA (1.0 mg) was used to analyze mRNA levelsby RT-PCR. To synthesize the first cDNA strand ofthe two cloned genes mentioned above, reverse

transcriptase M-MLV (RNase H

) was used accordingto the protocols provided by the manufacturer. PCRamplification was performed using Taq DNA poly-merase from Promega. Two sets of the specificprimers (rib1_rt_f1: TATTCCCTGGTGGATTACAAGC,rib1_rt_r1: CACAATCACATCGGGCACT, rib3_rt_f1:GCGGAATCAATCACCCAAG and rib3_rt_r1: GCGTA-GTCGCAAGTTATCGTAT designed according to thesequences of the partial genes cloned above)were used for RT-PCR. One set of specific primersfor amplification of the 18S rRNA gene (18s_rt_f1:TACAGTGAAACTGCGAATGGC and 18s_rt_r1: AGCA-CAAGGTCATGCGATT) was designed according

to the sequence of the 18S rRNA gene (accessionnumber: EF362753) in this strain. The conditionsfor RT-PCR amplification were as follows: initialdenaturation at 94 1C for 5 min, denaturation at94 1C for 30s, annealing temperature at 511Cfor 30s, extension at 721C for 1 min and finalextension at 72 1C for 10 min. RT-PCR was run for 32cycles. 18S rRNA gene was used as an internalstandard.

Results

Isolation of yeast strain W14-3 and yellowsubstance production

A total of 100 yeast strains from seawater andsediments in China Eastern Sea were obtained. Wefound that the culture of the marine yeast strainW14-3 turned into yellow color when it wasaerobically grown in the medium containing mal-tose at 25 1C for 24h (Figure 1B). The yeast strainwas isolated from China Eastern Sea. However, thetype yeast strain C. membranifaciens PYCC2727T

did not produce such a yellow substance in thesame medium under the same conditions (Figure 1A).In order to know if the yeast strain could producethe yellow substance when it was grown onother carbon sources, different carbon sources(glucose, sucrose, fructose, maltose, trehalose,

xylose, D-raffinose, D-galactose) were added tothe production medium. The results in Table 1indicate that the yeast strain produced more yellowsubstance only in the medium containing xylose,sucrose, galactose and maltose.

Analysis of the yellow substance by HPLC

The supernatant from the culture with a yellowcolor was treated as described in Materials and

ARTICLE IN PRESS

Figure 1. The culture color occurred in the cultures of the marine yeast strain W14-3 (B) and type yeast strainPYCC2727T (A) within 72 h. The sugar concentrations were 2.0%.

Table 1. The yellow substance production from differ-

ent carbon sources

A B C D E F G H

+ ++ ++++ ++ ++++ + +++ +++

A: D-trehalose; B: D-glucose; C: D-xylose; D: fructose; E: sucrose;F: D -raffinose; G: D -galactose; H: maltose. The sugar concentra-tions were 2.0%.++++: the most colorful solution; +++: more colorful solution;++: the colorful solution; +: trace yellow color in the solution.

L. Wang et al.258

5/26/2018 Isolation and Characterization Of

5/12

ARTICLE IN PRESS

200.00

220.00 240.00 260.00 280.00 300.00 320.00 340.00 360.00 380.00 400.00 420.00 440.00 460.00 480.00

5.75

440.7

364.7

256.8

218.8

5.75

440.7

364.7

265.8

218.8

nm

200.00

220.00 240.00 260.00 280.00 300.00 320.00 340.00 360.00 380.00 400.00 420.00 440.00 460.00 480.00

nm

0.003

0.002

0.001

0.000

-0.001

AU

2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00

Minutes

Minutes

0.012

0.010

0.008

0.006

0.004

0.002

0.000

AU

2.00 4.00 6.00 8.00 10.00 12.00 14.00

Figure 2. HPLC spectra of riboflavin standard (A) and the partially purified yellow substance (B).

Isolation and characterization ofC. membranifacienssubsp. flavinogenieW14-3 259

5/26/2018 Isolation and Characterization Of

6/12

methods. The treated supernatant was applied tothe column and the yellow substance in the eluatewas analyzed by using HPLC. The HPLC spectra ofthe partially purified yellow substance show thatthere was only one main peak that was identical tothat of the riboflavin standard (Figure 2A and B). Itcan also be seen from the results in Figure 2 that

ultraviolet spectra of the riboflavin standardand the sample were the same (upper panel ofFigure 2A and B). These results demonstrate thatthe yellow substance produced by the marine yeaststrain W14-3 was riboflavin. However, further work,using analytical techniques (such as NMR, MS, etc.)is needed to obtain structural characterization that

will ultimately provide conclusive evidence thatthe compound of interest is indeed a riboflavin, or ariboflavin-like compound.

Cloning of the partial genes encoding GTPcyclohydrolase II and 3,4-dihydroxy-2-

butanone-4-phosphate synthase

In order to confirm that theRIB1gene encoding GTPcyclohydrolase II and the RIB3 gene encoding 3,4-dihydroxy-2-butanone-4-phosphate synthase exist instrain W14-3 and type yeast strainC. membranifaciensPYCC2727T used in this study, the conserved motifs

ARTICLE IN PRESS

Figure 3. Multiple alignment of protein sequences encoding yeast GTP cyclohydrolase II. The GTP cyclohydrolase II arereferenced by their databases. Multiple sequence alignment of proteins was carried out using the method of Clustral X1.83 based on amino acid sequences of GTP cyclohydrolase II from Candida albicans-EAK96830 (1), Candida famata-CAH17652 (2), Candida glabrata-XP_446157 (3), Debaryomyces hansenii-XP_456888 (4), Kluyveromyces lactis-XP_452081 (5), Pichia guilliermondii-CAA88916 (6), Pichia stipitis-XP_001383311 (7), Saccharomyces cerevisiae-NP_009520 (8) and amino acid sequence (9) deduced from the cloned partial DNA fragment of the yeast strain used inthis study.

L. Wang et al.260

5/26/2018 Isolation and Characterization Of

7/12

were used to design degenerate primers to clone thepartial genes encoding GTP cyclohydrolase II and 3,4-dihydroxy-2-butanone-4-phosphate synthase in theyeast strain W14-3 and C. membranifaciens

PYCC2727T, respectively. In this case, amino acidsequences of GTP cyclohydrolase II and 3,4-dihy-droxy-2-butanone-4-phosphate synthase from differ-ent species of eukaryotic microorganisms weredownloaded from GenBank (http://www.ncbi.nlm.nih.gov) and aligned. The PCR-generated fragmentsobtained from genomic DNA of the yeast strain W14-3were sequenced. The alignment and comparison ofthe protein (GTP cyclohydrolase II and 3,4-dihydrox-y-2-butanone-4-phosphate synthase) sequences de-duced from the partial genes (accession numberswere EU239888 and EU275163) with sequences in the

protein databases using BLAST program are shown inFigures 3 and 4, respectively. The deduced proteinsshowed very high identity with GTP cyclohydrolase IIand 3,4-dihydroxy-2-butanone-4-phosphate synthasefrom other yeasts (Figures 3 and 4).

Routine identification of the marine yeaststrain W14-3

On yeast extract and malt (YM) medium thecolonies were white to chalky, dull, powdery, dry,

and wrinkled with a rough border (Figure 5A).Single cells were oval, producing daughter cellsby single polar budding in liquid YM medium

(Figure 5B). Pseudomycelia occurred, but no sexualreproduction was seen. It is very interesting to notethat the colonies of the marine yeast strain W14-3on the plate containing maltose are white,although it is common to all other microorganismsthat riboflavin production is recognizable by theyellow color of the colonies. We found that theyeast strain produced riboflavin only by vigorousshaking (Figure 1B). The yeast strain could notferment maltose, galactose, lactose and raffinose,but could ferment glucose and sucrose (data notshown). Biolog analysis shows that it could assimilate

ARTICLE IN PRESS

Figure 4. Multiple alignments of protein sequences of yeast 3,4-dihydroxy-2-butanone-4-phosphate synthases. Theyare referenced by their databases. Multiple sequence alignment of proteins was carried out using the method of ClustralX 1.83 based on amino acid sequences of of 3,4-dihydroxy-2-butanone-4-phosphate synthase from Saccharomycescerevisiae-AAB64927 (1), Pichia stipitis-XM_001386886 (2), Kluyveromyces lactis-XP_451875 (3), Saccharomycescerevisiae-Z21619 (4), Yarrowia lipolytica-XM_500663 (5), Candida glabrata-CAG60514 (6), Schizosaccharomyces

pombe-CAA18874 (7) and amino acid sequence (8) deduced from the cloned partial DNA fragment of the yeast strainused in this study.

Figure 5. Photographs of colonies (A) and microphoto-graph of vegetable cells (B) of the marine yeast strainW14-3. Medium: YPD medium; incubation temperature:25 1C; time: 2 days.

Isolation and characterization ofC. membranifacienssubsp. flavinogenieW14-3 261

5/26/2018 Isolation and Characterization Of

8/12

glucose, galactose, L-sorbose, sucrose, maltose,cellobiose, trehalose, melibiose, raffinose, inulin,D-xylose, L-arabinose, D-arabinose and L-rhamnose,but could not assimilate L-rhamnose (Table 2).Based on the fermentation spectra and carbonsource assimilation spectra of the marine yeast andthose of the type strain (C. membranifaciens) listed

in The yeast: a taxonomic study, 4th revised andenlarged edition (Kurtzman and Fell, 2000), wefound that the yeast strain W14-3 was closelyrelated to C. membranifaciens.

Phylogenetic analysis of partial sequences ofthe D1/D2 26S rDNA sequence

According to Kurtzman and Fell (2000), tradi-tional and routine identification methods, which

depend on phenotype, usually lead to uncertainand inaccurate interpretations of species interac-tion. Sequence analysis of phylogeny for microbialtaxonomy is a more accurate method for determin-ing inter- and intra-specific relationships. There-fore, a partial sequence of D1/D2 26S rDNA of theyeast strain was determined and aligned by using

BLAST analysis (http://www.ncbi.nlm.nih.gov/BLAST). A phylogenetic tree was constructed byusing PHYLIP software package version 3.56(Felsenstein, 1995). Distance matrices were gener-ated by the DNADIST program based on Kimurastwo-parameter model (Kimura, 1980). Neighbor-joining analysis of the data sets was carried outwith the program Neighbor of the PHYLIP package.Ricciocarpos natans was used as the out-groupduring the construction of a consensus tree of the

ARTICLE IN PRESS

Table 2. Assimilation of carbon sources of strain W14-3 by Biolog analysis

Carbon sources Results Carbon sources Results

Acetic acid D-Galactose +Formic acid D-Psicose +Propionic acid L-Rhamnose Succinic acid L-Sorbose +Succinic acid mono-methyl ester a-Methyl-D-glucoside +L-Aspartic acid b-Methyl-D-glucoside +L-Glutamic acid Amygdalin L-Proline w Arbutin +D-Gluconic acid + Salicin +Dextrin Maltitol Inulin + D-Mannitol +

Fumaric acid w D-Sorbitol +L-Malic acid w Adonitol +Succinic acid mono-methyl ester Lactose +Bromosuccinic acid Amidulin +L-Glutamic acid + D-Arabitol +g-Aminobutyric acid w Xylitol +a-Ketoglutaric acid i-Erythritol +2-Keto-gluconic acid + Glycerol +D-Gluconic acid + Tween 80 Dextrin L-Arabinose +D-Cellobiose + D-Arabinose +Gentiobiose + D-Ribose Maltose + D-Xylose +Maltotriose + Succinic acid mono-methyl ester plus D-xylose wD-Melezitose + N-acetyl-L-glutamic acid plus D-xylose

D-Melibiose + Quinic acid plus D-xylose Palatinose + D-Glucuronic acid plus D-xylose D-Raffinose + Dextrin plus D-xylose Stachyose + a-D-Lactose plus D-xylose Sucrose + D-Melibiose plus D-xylose +D-Trehalose + D-Galactose plus D-xylose +Turanose + m-Inositol plus D-xylose N-Acetyl-D-glucosamine + 1,2-Propanediol plus D-xylose D-Glucosamine W Acetoin plus D-xylose a-D-glucose +

W: weak; +: positive; : negative.

L. Wang et al.262

5/26/2018 Isolation and Characterization Of

9/12

ARTICLE IN PRESS

Figure 6. Consensus tree of the isolate based on D1/D2 26S rDNAs obtained in this study and 23 previously publishedsequences obtained from GenBank. The outgroup we used was Ricciocarpos natans. Numbers on tree branches indicatethe percentages of bootstrap samplings derived from 1000 samples that supported the internal branches by 50% orhigher. All the strains shown are type strains.

Isolation and characterization ofC. membranifacienssubsp. flavinogenieW14-3 263

5/26/2018 Isolation and Characterization Of

10/12

isolate based on the D1/D2 26S rDNA sequence. Thesearch for the similarity between the D1/D2 26SrDNA sequence of the isolate and that in the NCBIdatabase shows that many phylogenetically relatedyeast species were similar to the marine yeaststrain obtained in this study. Phylogenetic relation-ships of the D1/D2 26S rDNA sequence of the marine

yeast strain are shown inFigure 6and its GenBankaccession number is EF362752. The topology of thephylogram inFigure 6confirms that 100% bootstrapvalues were detected between the D1/D2 26S rDNAsequence of strain W14-3 and that of C. membra-nifaciens. Therefore, the yeast strain W14-3 wasclosely related to C. membranifaciens.

Discussion

As shown in Figures 1 and 2 and Table 1, the

marine yeast strain W14-3 produced a large amountof riboflavin, whereas the type yeast strainC. membranifaciens PYCC2727T did not produceany riboflavin under the same conditions. Theresults (Figure 7) also show that the marine yeaststrain W14-3 also produced a large amount ofriboflavin in the production medium prepared withseawater, but Fe3+ available in the mediumcompletely repressed riboflavin production. Finally,the yeast strain W14-3 was identified to beC. membranifaciens subsp. flavinogenie (Table 2,Figure 6), meaning that it was different from thetype yeast strain C. membranifaciens due to its

production of riboflavin and the presence of genesresponsible for riboflavin biosynthesis (Figures 3and 4). In addition, the yeast strain W14-3 grewwell in the medium containing 60% glucose, while

C. membranifaciensPYCC2727T grew poorly in thesame medium (data not shown).

The biosynthetic pathway of riboflavin has beenstudied in considerable detail in bacteria, fungi andyeasts (Humbelin et al., 1999; Stahmann et al.,2000; Karos et al., 2004). It has been welldocumented that GTP cyclohydrolase II is encoded

by the RIB1 gene, while the 3,4-dihydroxy-2-butanone-4-phosphate synthase is encoded by theRIB3gene (Humbelin et al., 1999;Stahmann et al.,2000;Karos et al., 2004). The deduced amino acidsequences from the partial genes encoding GTPcyclohydrolase II and 3,4-dihydroxy-2-butanone-4-phosphate synthase in the marine yeast strainW14-3 showed very high identity with those ofGTP cyclohydrolase II and 3,4-dihydroxy-2-buta-none-4-phosphate synthase from different eucar-yotic microorganisms (Figures 3 and 4), suggestingthat a riboflavin synthesis pathway indeed existed

in the yeast.Among the terrestrial microorganisms,A. gossy-pii, a filamentous fungus, C. famata, a yeast, andthe genetically engineered B. subtilis have beencommercially used to produce riboflavin by fer-mentation (Stahmann et al., 2000; Wendland andWalther, 2005). In addition,Sabry et al. (1989)usedC. guilliermondii Wickerham to produce riboflavinandBuzzini and Rossi (1998)used the immobilizedCandida tropicalis cells to produce riboflavin.Eremothecium ashbyii, a fungus was also found tohave the ability to produce a large amount ofriboflavin (Kalingan and Liao, 2002). It has been

reported that P. guilliermondii is capable ofriboflavin overproduction under iron deficiency(Fayura et al., 2007). Therefore, this is the firststudy to report that C. membranifaciens subsp.flavinogenie W14-3 isolated from the marineenvironment can produce riboflavin. Usually, ironin the medium will repress riboflavin productionby the terrestrial yeasts and fungi. The results inTable 3also show that added iron in the productionmedium completely repressed riboflavin productionby the marine yeast. However, removal of iron onlyweakly affected riboflavin production by the

marine yeast. Therefore, it is more suitable andeconomical to apply the marine yeast strain toriboflavin production on a large scale than anyother riboflavin producers. The results in Figure 8

ARTICLE IN PRESS

Table 3. Effect of iron in the production medium on riboflavin production by the marine yeast

Media Medium treated with8-hydroxyquinoline

Production mediumwithout any treatment

Medium supplementedwith 0.005% FeCl3

Riboflavin yield (mg/ml) 16.370.1 14.770.3 070.1

Figure 7. Riboflavin production in the production media.1: The production medium prepared with distilled water;2: the production medium supplemented with iron; 3: theproduction medium prepared with seawater.

L. Wang et al.264

5/26/2018 Isolation and Characterization Of

11/12

indicate that added iron in the medium repressedexpression of RIB1 and RIB3 genes in the yeastcells. This means that iron also negatively affectsriboflavin production by the marine yeast strainW14-3 at the transcriptional level. In order to

enhance riboflavin production, optimization of themedium and cultivation conditions for riboflavinproduction by C. membranifaciens subsp. flavino-genieW14-3 is being undertaken in the laboratory.

Acknowledgments

This research was supported by the Hi-TechResearch and Development Program of China(863), and the grant number is 2006AA09Z403.

References

Arella F, Lahely S, Bourguignon J, Hasselmann C. Liquidchromatographic determination of vitamins B1 and B2in foods: a collaborative study. Food Chem 1996;56:816.

Buzzini P, Rossi J. Semi-continuous and continuousriboflavin production byCandida tropicalisin concen-trated rectified grate calcium-alginate-immobil-ized must. World J Microbiol Biotechnol 1998;14:37781.

Chi ZM, Ma CL, Wang P, Li HF. Optimization of medium andcultivation conditions for alkaline protease productionby the marine yeast Aureobasidium pullulans. Bior-esour Technol 2007;98:5348.

Cowart RE, Marquardt MP, Foster BG. The removalof iron and other trace elements from a complexbacteriological medium. Microbiol Lett 1980;13:

117

22.Felsenstein J. PHYLIP (Phylogenetic Inference Package),Version 3.75. Distributed by author. Seattle, WA:Department of Genetics, University of Washington;1995.

Fayura LR, Fedorovych DV, Prokopiv TM, Boretsky YR,Sibirny AA. The pleiotropic nature ofrib80, hit1, andred6mutations affecting riboflavin biosynthesis in theyeast Pichia guilliermondii. Microbiology 2007;76:559.

Humbelin M, Griesser V, Keller T, Schurter W, Haiker M,Hohmann HP, et al. GTP cyclohydrolase II and 3,4-dihydroxy-2-butanone 4-phosphate synthase are rate-limiting enzymes in riboflavin Bacillus subtilis strain

used for synthesis of an industrial riboflavin produc-tion. J Ind Microbiol Biotechnol 1999;22:17.

Kalingan AE, Liao CM. Influence of type and concentra-tion of flavinogenic factors on production riboflavin byEremothecium ashbyii NRRL 1363. Bioresour Technol2002;82:21924.

Karos M, Vilarino C, Bollschweiler C, Revuelta JL. Agenome-wide transcription analysis of a fungal ribo-flavin overproducer. J Biotechnol 2004;113:6976.

Kimura M. A simple method for estimating evolutionaryrate of base substitutions through comparative stu-dies on nucleotide sequences. J Mol Evol 1980;2:8790.

Kurtzman CP, Fell JW. In: The yeasts: a taxonomic study,4th revised and enlarged edition. Amsterdam: Else-vier; 2000. p. 1525.

Ndaw S, Bergaertzle M, Aoude-Werner D, Hasselmann C.Extraction procedures for the liquid chromatographicdetermination of thiamin, riboflavin and vitamin B infoodstuffs. Food Chem 2000;71:12938.

Praphailong W, Gestel MV, Fleet GH, Heard GM. Evalua-tion of the Biolog system for the identification of foodand beverage yeasts. Lett Appl Microbiol 1997;24:4559.

Sabry SA, El-Refal AH, Gamati SY. Physiological study onriboflavin production by a hydrocarbon-utilizing cul-

ture of Candida guilliermondii Wickerham. J IslamicAcad Sci 1989;2:2730.Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a

laboratory manual, 2nd ed. Beijing: Cold SpringHarbor Laboratory Press; 1989. p. 36770 (Chinesetranslated ed.).

Stahmann KP, Revuelta JL, Seulberger H. Three biotech-nical processes Ashbya gossypii, Candida famata orBacillus subtilis compete with using chemical ribo-flavin production. Appl Microbiol Biotechnol 2000;53:50916.

Sugita T, Takashima M, Kodama M, Tsuboi R, Nishikawa A.Description of a new yeast speciesMalassezia japonica,

ARTICLE IN PRESS

Figure 8. The changes of the amount of mRNAs encodingGTP cyclohydrolase II and 3,4-dihydroxy-2-butanone-4-phosphate synthase in the yeast cells grown in differentmedia. 1: The amount of 18S rRNA in the yeast cellsgrown in the production medium. 2: The amount of 18SrRNA in the yeast cells grown in the production mediumsupplemented with iron. 3: The amount of mRNAencoding GTP cyclohydrolase II in the yeast cells grownin the production medium. 4: The amount of mRNAencoding GTP cyclohydrolase II in the yeast cells grown inthe production medium supplemented with iron. 5: Theamount of mRNA encoding 3,4-dihydroxy-2-butanone-4-

phosphate synthase in the yeast cells grown in theproduction medium. 6: The amount of mRNA encoding3,4-dihydroxy-2-butanone-4-phosphate synthase in theyeast cells grown in the production medium supplemen-ted with iron. M: DNA markers (the DNA bands frombottom to top are 0.1, 0.25, 0.5, 0.75, 1.0 and 2.0 kb).

Isolation and characterization ofC. membranifacienssubsp. flavinogenieW14-3 265

5/26/2018 Isolation and Characterization Of

12/12

and its detection in patients with atopic dermatitisand healthy subjects. J Clin Microbiol 2003;41:46959.

Thompson JD, Higgins DD, Gibson JJ. CLUSTAL W:improving the sensitivity of progressive multiplesequence alignment through sequence weighting,

positions specific gap penalties and weight matrixchoice. Nucleic Acids Res 1994;22:467388.

Wendland J, Walther A. Ashbya gossypii: a model forfungal development biology. Nat Rev Microbiol2005;5:4219.

ARTICLE IN PRESS

L. Wang et al.266