Transcriptional regulation of 3,4-dihydroxy-2-butanone 4-phosphate synthase

T. Schlösser1, G. Schmidta,1 and K.-P. Stahmann1

Institut für Biotechnologie 1, Forschungszentrum Jülich GmbH, D-52425 Jülich, Germany1

Author for correspondence: K.-P. Stahmann. Tel: +49 2461 612843. Fax: +49 2461 612710. e-mail: p.stahmann{at}fz-juelich.de


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The filamentous hemiascomycete Ashbya gossypii is a strong riboflavin overproducer. A striking but as yet uninvestigated phenomenon is the fact that the overproduction of this vitamin starts when growth rate declines, which means that most of the riboflavin is produced in the stationary phase, the so-called production phase. The specific activity of 3,4-dihydroxy-2-butanone 4-phosphate (DHBP) synthase, the first enzyme in the biosynthetic pathway for riboflavin, was determined during cultivation and an increase during the production phase was found. Furthermore, an increase of RIB3 mRNA, encoding DHBP synthase, was observed by competitive RT-PCR in the production phase. The mRNAs of two housekeeping genes, ACT1 (encoding actin) and TEF (encoding translation elongation factor-1{alpha}), served as standards in the RT-PCR. Reporter studies with a RIB3 promoter–lacZ fusion showed an increase of ß-galactosidase specific activity in the production phase. This investigation verified that the increase of RIB3 mRNA in the production phase is caused by an induction of promoter activity. These data suggest that the time course of riboflavin overproduction of A. gossypii is correlated with a transcriptional regulation of the DHBP synthase.

Keywords: DHBP synthase, riboflavin, Ashbya gossypii, fungi, gene regulation

Abbreviations: ARS, autonomously replicating sequence; DHBP, 3,4-dihydroxy-2-butanone 4-phosphate

a Present address: Tosoh Biosep GmbH, Zettachring 6, D-70567 Stuttgart, Germany.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The filamentous hemiascomycete Ashbya gossypii (Ashby & Nowell) (Guilliermond, 1928 ) is a plant pathogen and was first isolated from cotton (Ashby & Nowell, 1926 ), but is also able to infect other plants (Phaff & Starmer, 1987 ). What makes the fungus interesting is its ability to overproduce riboflavin (vitamin B2) (Wickerham et al., 1946 ). Yields of up to 15 g l-1 have been reported for strains used in industry for riboflavin production (Bigelis, 1989 ). Nowadays, A. gossypii competes with two other organisms, Candida famata and Bacillus subtilis, for the title of best tool in the biotechnical process (Stahmann et al., 2000 ).

Although the biosynthesis of riboflavin has been elucidated in several organisms (Bacher et al., 2000 ), nothing is known about transcriptional regulation of the corresponding genes in A. gossypii or other eukaryotes. In B. subtilis, a Gram-positive prokaryote, the riboflavin genes are clustered in a single operon which is transcribed into a polycistronic mRNA (Mironov et al., 1994 ). Mironov and coworkers showed that the steady-state level of the transcript is negatively regulated by riboflavin. The assumption that RibC could function as a repressor on the operator region has been refuted (Mack et al., 1998 ). RibC and RibR (Solovieva et al., 1999 ) encode bifunctional and monofunctional flavin kinases, respectively, forming FMN and FAD, which seems to have an effect on the regulation of riboflavin synthesis.

Riboflavin biosynthesis has two roots in the metabolism of the cell. One starts with GTP and the other with ribulose 5-phosphate (Bacher, 1991 ). In parallel reactions GTP is converted by a GTP cyclohydrolase II into 2,5-diamino-6-ribosylamino-4(3H)-pyrimidinone-5-phosphate and ribulose 5-phosphate is converted by 3,4-dihydroxy-2-butanone 4-phosphate (DHBP) synthase into DHBP. With respect to regulation of riboflavin synthesis, DHBP synthase should be of greater importance because this enzyme is needed twice for the formation of one riboflavin molecule while GTP cyclohydrolase II is only required once. Furthermore, some biosynthetic pathways are controlled by transcriptional regulation of the first enzyme. For the mammalian squalene synthase it has been shown that the mRNA level of the corresponding gene, FDFT1, is regulated by cholesterol status and by the cytokines TNF-{alpha} and IL-1ß (Robinson et al., 1993 ; Tansey & Shechter, 2000 ). Transcriptional regulation has also been described for the squalene synthase gene ERG9 in Saccharomyces cerevisiae (Kennedy et al., 1999 ). In Gibberella pulicaris, which produces toxic trichothecene, the first enzyme in the biosynthetic pathway is also transcriptionally regulated (Hohn et al., 1993 ).

In Arabidopsis thaliana (Herz et al., 2000 ) and B. subtilis (Hümbelin et al., 1999 ) DHBP synthase and GTP cyclohydrolase II are part of bifunctional proteins, whereas in Escherichia coli and A. gossypii two genes encode two separate enzymes (Richter et al., 1992 , 1993 ; Revuelta et al., 1995 ). Recently, the first crystallization of a monofunctional DHBP synthase has been reported for Magnaporthe grisea. The amino acid sequence of the M. grisea enzyme is 48·1% identical to the primary structure of the A. gossypii enzyme (Liao et al., 2000 ).

In this study we focused on regulation of DHBP synthase, which is encoded by RIB3 in A. gossypii. First, evidence concerning regulation of riboflavin overproduction was reported by Pfeifer et al. (1950) . They used glucose as the carbon source and found that riboflavin overproduction started after glucose had been exhausted from the medium. To show that catabolite repression by glucose is not the regulation mechanism, we chose fermentation conditions in which glucose was present over the whole cultivation period. A pronounced separation between growth phase and riboflavin production phase was found. We show an increase in specific activity of DHBP synthase in the production phase which correlates with transcriptional regulation of the RIB3 gene.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Strains and cultivation conditions.
A. gossypii strain ATCC 10895 was maintained on solid medium (Stahmann et al., 1994 ). Inocula for batch fermentations were cultivated in liquid medium, consisting of 10 g yeast extract l-1 and 10 g glucose l-1 (HA medium), and incubated at 28 or 30 °C in 500 ml shake flasks, each containing 100 ml medium, on a rotary shaker (Certomat H; B. Braun) at 120 r.p.m. For batch fermentations used to determine DHBP synthase activity, 30 ml of an overnight culture was used to inoculate a 3 l bioreactor (Braun Biostat E; B. Braun) containing 1·5 l HA medium. The initial pH value for both media ranged from 6·4 to 6·7. Fermentations were run at 28 °C with an aeration rate of 1 vol. vol.-1 min-1. The stirrer speed was set to 100 r.p.m. Antifoam Dehysan Z 2611 (Henkel) was added to the medium to a final concentration of 0·2 ml l-1.

For fermenter and shake-flask experiments, used to assay ß-galactosidase specific activity and RIB3 transcript levels, a mineral salt medium (Monschau et al., 1998 ) was used with 1 g yeast extract l-1, 15 mM glycine and 20 g glucose l-1 as carbon source.

For batch cultivations used to assay ß-galactosidase specific activity, a 5 l bioreactor (Labfors System; Infors) was inoculated with 50 ml of an overnight culture containing 5 l mineral salt medium. The gas flow was set to 5 l min-1 and the pO2, controlled by stirring speed (650±350 r.p.m.), was set to a minimum of 80%. Fermentations were run at 30 °C.

Shake-flask experiments were performed at 30 °C under the conditions described for the inocula. For selective growth of transformants the aminoglycoside geneticin (Roche) was used at final concentrations of 400 µg ml-1 for solid media and 50 µg ml-1 for liquid cultures.

Escherichia coli strain DH5{alpha} [supE44 {Delta}lacU169 ({phi}80lacZ{Delta}M15) hsdR17 recA1 endA1 gyrA96 thi-1 relA1] was used for general cloning methods (Hanahan, 1985 ).

DNA techniques.
For cloning, transformation and plasmid isolation standard techniques were used (Sambrook et al., 1989 ). Southern blotting was performed with digoxigenin (DIG)-labelled probes using the DIG-Labelling and Detection Kit as recommended by the supplier (Roche). For the isolation of genomic DNA the DNeasy Plant Maxi Kit (Qiagen) was used. For one preparation 1 g wet mycelium was homogenized in liquid nitrogen and processed as described in the manual.

RT-PCR.
After 8 and 24 h of cultivation the volume of two shake flasks was pooled and the mycelia were used for total RNA isolation. The isolation of total RNA was carried out with the RNeasy Mini Kit (Qiagen). Cell lysis was performed with 0·2 g mycelium, 1 ml glass beads (diam. 0·5 mm) and 1 ml lysis buffer in a bead mill (Type MM2; Retsch) for 5 min at maximum intensity for one probe. Additionally a DNase treatment on the column was performed with the appropriate Qiagen system. Total RNA was eluted with 50 µl H2O.

Total RNA (5 µg) was used for one RT reaction with Super Script II reverse transcriptase (Life Technologies). The reaction was performed at 44 °C with an oligo-dT15 primer. After the RT reaction the preparation was made up with H2O to a volume of 100 µl, of which 2 µl was used as a template in the following PCR.

PCRs were carried out in a volume of 50 µl with 4 U Taq polymerase (Roche), 5 µl reaction buffer, 2 µl each primer (10 pmol µl-1) and 1 µl dNTP mix (10 mM each dNTP). Primers were made according to the RIB3 sequence (EMBL accession no. A46562; RIB3 forw, 5'-AAGCCCATGCACTGATATCG-3'; RIB3 rev, 5'-GCAAGACCGTGCTTCTTGC-3'), the TEF sequence (encoding translation elongation factor-1{alpha}; X73978; TEF forw, 5'-GCCATCTTGATCATTGCTGG-3'; TEF rev, 5'-TTGACTTCAGTGGTGACACC-3') and the ACT1 sequence (encoding actin; AJ131713; ACT forw, 5'-CTTCTACGTGTCCATTCAGG-3'; ACT rev, 5'-AGAGAGGCCAAGATAGAACC-3'). PCR preparations were overlaid with mineral oil (Sigma) and Taq polymerase was added by a hot start procedure in the break mode of the thermal cycler (PTC-100; MJ Research). The amplification protocol was performed as follows: initial denaturation for 1 min at 94 °C; 27 cycles of 30 s at 94 °C, 30 s at 60 °C and 30 s at 72 °C, and a final elongation period for 10 min at 72 °C. After the PCR, 8 µl 6xDNA loading buffer was added. Twenty microlitres of this solution was loaded on a 2% agarose gel for separation of the fragments.

Transformation of A. gossypii and isolation of spores.
The transformation of wild-type mycelia was carried out by the method described in Monschau et al. (1998) . The obtained mycelia were suspended in 1 ml 0·9% NaCl with 10 mg Glucanex ml-1 (Novo) and incubated for 1 h at room temperature. After a centrifugation step (5 min, 3000 g) the pellet was washed three times with 0·9% NaCl. The pellet was resuspended in 500 µl 0·9% NaCl. After the addition of 500 µl paraffin, the emulsion was shaken and stored at 4 °C for phase separation. The paraffin phase, containing the spores, was checked microscopically.

Construction of the reporter plasmid pTS-07.
The lacZ gene was cut out from the plasmid pAG-110 (Kurth et al., 1992 ; Steiner & Philippsen, 1994 ) by BamHI and cloned in the BamHI site of pBR322, leading to plasmid pTS-02. By using primer ts2 (5'-TTTCTCGAggatccGGGCCCGCTAGCGCCGTCGTTTTACAACGTCG-3') an ApaI and an NheI site (double-underlined) were incorporated behind the BamHI site (lower case letters) for directional cloning of the RIB3 promoter. The second primer, ts1 (5'-GGGTACCGAGCTCGAATTCGTAATCATGGTCATAGCTCTCGAGTTT-3') contained, like ts2, an XhoI site (underlined) for religation of the PCR product after XhoI digestion, resulting in pTS-04. Afterwards the BamHI fragment was excised from pTS-04 and cloned in the BamHI site of pUC18, resulting in pTS-05. Genomic A. gossypii wild-type DNA was used for PCR amplification of the RIB3 promoter. Primers ts3 (5'-TTTGGTACCGGGCCCTTCTTGCACGGTCGTTTCTG-AA-3') and ts4 (5'-TTTGGTACCGCTAGCTGTCATGTTGCTTGGTTTGTCG-3') contained, besides the ApaI and the NheI sites (double-underlined), a KpnI site (single-underlined) for religation of the PCR product, resulting in pTS-03. The sequence of the RIB3 gene has been published by Revuelta et al. (1995) . The RIB3 promoter was excised with ApaI and NheI from plasmid pTS-03 and cloned in ApaI- and NheI-digested pTS-05. Finally, the BamHI fragment containing the lacZ gene under the control of the RIB3 promoter was cloned in the BamHI site of plasmid pAG-100 (Kurth et al., 1992 ; Steiner & Philippsen, 1994 ), generating pTS-07. Sequencing of the transition from the promoter to the lacZ gene revealed successful in-frame cloning.

Construction of pTS-10 for chromosomal integration of the RIB3placZ fusion.
In a first step a fragment was amplified by PCR, consisting of the RIB3placZ reporter and the geneticin resistance cassette (Fig. 1). By using PCR, in which the plasmid pTS-07 served as template, BglII sites (underlined) were added at the end of the fragment by the oligonucleotides ts10 and ts11 (ts10, 5'-TTTAGATCTGGATCCGGGCCCTTCTTGCA-3'; ts11, 5'-TTTAGATCTGCCGTCCCGTCAAGTCAGC-3'). The fragment was cloned between the sequence for a serine tRNA and the ICL1 gene, encoding isocitrate lyase in A. gossypii (Maeting et al., 1999 ). A 2·9 kb SphI genomic fragment containing these sequences was cloned into the common cloning vector pUC18. Via PCR a BglII restriction site was added at position 245 (corresponding to EMBL accession no. AJ010727) into the 2·9 kb genomic fragment containing the serine tRNA sequence and the ICL1 gene. The BglII fragment was cloned into this BglII site, generating plasmid pTS-10.



View larger version (16K):
[in this window]
[in a new window]
 
Fig. 1. Chromosomal integration of the ß-galactosidase reporter together with a geneticin resistance marker in front of the ICL1 locus of A. gossypii.

 
Determination of total riboflavin, mycelial dry mass and glucose.
Procedures described by Monschau et al. (1998) were used for the determination of total riboflavin and mycelial dry mass. Instead of lysing enzyme, 100 µl Glucanex (25 mg ml-1) was used for cell disruption. The analysis was carried out by HPLC according to Schmidt et al. (1996) . Glucose was determined in the culture filtrate with a UV test purchased from Roche.

Determination of DHBP synthase activity.
For cell extract preparation mycelia were harvested by filtration (MF membrane filter, 0·8 µm pore size; Millipore) and resuspended in homogenization buffer (100 mM Tris/HCl buffer, pH 7·5, 10 mM MgCl2, 5 mM DTT, 1 mM PMSF, 10 µM antipain, 10 µM leupeptin) at a ratio of 2 ml (g mycelial wet wt)-1. Cells were disrupted in a French press (Aminco) under 19000 p.s.i. (131100 kPa) and the homogenate was centrifuged at 20000 g for 30 min. An aliquot (1·5–2·0 ml) of the supernatant was applied onto a Sephacryl S-200-gel permeation column (40x1·6 cm; Pharmacia). The column was developed using S200 buffer (50 mM Tris/HCl buffer, pH 7·5, 10 mM MgCl2, 5 mM DTT, 150 mM NaCl, 0·02% NaN3) at a flow rate of 18 ml h-1. A 200 µl aliquot of each fraction (1·8 ml) was checked for DHBP synthase activity by adding ribose 5-phosphate (Sigma) and pentose phosphate isomerase (Sigma) to a final concentration of 25 mM and 2 U, respectively, yielding a final volume of 240 µl. Samples were incubated at 37 °C for 4 h. To dephosphorylate the substrate and the formed products, 5 U alkaline phosphatase (Roche) was added with further incubation at 37 °C for 2 h. Finally, assay mixtures were subjected to an ultrafiltration step using Ultrafree-MC 10 centrifugation units (Millipore). The resulting filtrates were analysed for 3,4-dihydroxy-2-butanone by HPLC (Merck) using a Polyspher OAKC column (300x7·8 mm; Merck), which was developed isocratically with 5 mM H2SO4 at a flow rate of 0·4 ml min-1. The column temperature was set to 45 °C. A UV detector ({lambda}=188 nm) and a series-connected RI detector (Merck) were employed to monitor the effluent. One unit of enzyme activity catalyses the formation of 1 nmol DHBP h-1.

Determination of ß-galactosidase activity.
A cell-free crude extract was used to determine ß-galactosidase activity. Mycelia were harvested by filtration and suspended in dissolution buffer consisting of buffer A (5% glycerol, 5 mM Tris, 10 mM KCl, pH 7·5) 30% glycerol and Complete (1 pill per 50 ml; Roche), a protease inhibitor mix. After a French press (Aminco) passage at 20000 p.s.i. (138000 kPa) the crude extract was obtained as supernatant by centrifugation at 20000 g for 15 min at 4 °C. The assay was based on a method described by Miller (1972) . A mixture of 800 µl buffer A, 200 µl ONPG solution (4 mg ONPG ml-1 in 100 mM KH2PO4) was incubated with an appropriate amount of crude extract at 30 °C. The reaction was stopped after 2 min and further time points with 500 µl 1 M Na2CO3 and the absorbance was measured at 420 nm. An extinction coefficient ({epsilon}) of 6·11x105 M-1 cm-1 was used. The amount of total protein was determined by the method of Bradford (1976) with BSA as standard.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Correlation of DHBP synthase specific activity with riboflavin overproduction
Determination of DHBP synthase activity was not possible in crude extracts. Therefore fractionation of proteins was performed with a Sephacryl S-200 column to remove phosphatase activity present in the crude extracts leading to a reduction of the substrate ribose 5-phosphate needed in the assay. DHBP synthase activity was detectable in fractions 25–31 (Fig. 2).



View larger version (16K):
[in this window]
[in a new window]
 
Fig. 2. Elution profile from Sephacryl S-200 chromatography with a crude extract from A. gossypii. DHBP synthase activity ({bullet}) was detected by HPLC analysis. {circ}, Absorption of total protein of each fraction.

 
After establishing the enzyme assay for DHBP synthase, analysis was focused on the time course of DHBP synthase activity during a fermenter cultivation in HA medium. This rich medium was selected to obtain enough biomass for the enzyme assay. During fermenter cultivation over 77·5 h the specific activity of DHBP synthase increased from 2·5 nmol h-1 (mg protein)-1 after 22 h to 156 nmol h-1 (mg protein)-1 at the end of fermentation (77·5 h; Fig. 3). In correlation with the increase in the specific activity of DHBP synthase, production of riboflavin was observed. The production of riboflavin increased from 1·2 mg l-1 after 22·5 h of cultivation to 13 mg l-1 at the end of fermentation.



View larger version (14K):
[in this window]
[in a new window]
 
Fig. 3. DHBP synthase specific activity ({triangleup}) and total riboflavin ({blacktriangleup}) during cultivation of A. gossypii on HA medium. Glucose (not shown), the main carbon source, was exhausted from the medium after 20 h and biomass (not shown) reached a maximum of 4·6 g l-1. The data shown were obtained in a single cultivation. In an independent experiment (data not in graph) enzyme specific activity increased from 12 nmol h-1 (mg protein)-1 before riboflavin production to a maximum of 145 nmol h-1 (mg protein)-1 during riboflavin production.

 
Interestingly, in the growth phase from 0 to 20 h only a basal activity of DHBP synthase was observed. In the first 22·5 h only 9% of the total amount of riboflavin had been formed. The conclusion of this time course experiment is that overproduction of riboflavin in A. gossypii might be caused by regulation of DHBP synthase.

Analysis of RIB3 mRNA levels in growth and production phase
RT-PCR was performed to study whether the increase in specific activity of DHBP synthase in the production phase is due to an increase in the concentration of the RIB3 transcript. The fungus was cultivated for 8 h to obtain mycelium in the growth phase and 24 h to obtain cells in the riboflavin production phase. ACT1 mRNA or TEF mRNA served as a standard in the RT-PCR. The transcript of the ß-actin gene is one of the most frequently used controls in quantitative and semiquantitative PCR assays, but it has been reported that this mRNA is subject to modulation (Suzuki et al., 2000 ). Therefore we decided to use the gene for the translation elongation factor 1{alpha} as a standard. The TEF gene is highly expressed in A. gossypii (Steiner & Philippsen, 1994 ) and, furthermore, it has been shown that TEF has a more homogeneous expression than actin when used as a positive control in in situ hybridization assays (Gruber & Levine, 1997 ). The results of RT-PCR indicated an increase of RIB3 mRNA in the riboflavin production phase after 24 h cultivation (Fig. 4). In such PCRs where RIB3- and TEF- or ACT1-specific oligonucleotides were used in a competitive reaction, the signals of the RIB3 PCR products increased in comparison to the signals of the standards. When the PCRs were performed without RIB3 primers an approximately equal expression level of TEF and ACT1, indicated by equal signal intensities, was observed.



View larger version (90K):
[in this window]
[in a new window]
 
Fig. 4. Relative RIB3 mRNA concentration in comparison to TEF or ACT1 mRNA levels in growth phase (8 h) and production phase (24 h) analysed by RT-PCR. Either one or two pairs of oligonucleotides were used. The upper part of the figure shows the RT-PCR products using TEF and RIB3 oligonucleotides. The RIB3 products have a size of 585 bp and the TEF products a size of 530 bp. The lower part displays the same experiment using ACT1 oligonucleotides. The ACT1 products have a size of 654 bp. Lane M, 100 bp molecular mass marker. The results shown were derived from a typical experiment. Two independent reproductions were performed. One of them was analysed with ACT1, the other with TEF oligonucleotides (results not shown).

 
RIB3 promoter–lacZ reporter studies for RIB3 expression
As well as regulation of gene expression by a controlled promoter activity, regulation may be due to a change in mRNA stability. To study whether the observed increase in RIB3 mRNA in the riboflavin production phase is due to regulation of the promoter activity a reporter plasmid was constructed in which the lacZ gene of E. coli (Kalnins et al., 1983 ) served as the reporter gene. It has been stated that the reporter plasmid pAG-110 is suitable for promoter studies of the TEF gene in A. gossypii (Steiner & Philippsen, 1994 ). In this plasmid the lacZ gene is under the control of the TEF promoter. We used plasmid pAG-110 and replaced the TEF promoter with the RIB3 promoter from A. gossypii, generating pTS-07 (see Methods and Fig. 1). Plasmid pTS-07 was used for transformation of A. gossypii wild-type and several transformants were obtained. The transformants were cultivated for 8 and 24 h to assay ß-galactosidase specific activity (Table 1). Geneticin was used in these cultivations to be sure that the plasmid remained present in the strains. All strains transformed with the plasmid pTS-07 (named AgpTS07-) showed an increase in ß-galactosidase specific activity in the riboflavin production phase. The mean increase of ß-galactosidase specific activity from 8 to 24 h of cultivation was 8·5-fold. The control strain transformed with pAG-110 revealed a weak increase in the production phase in two out of three cases and a decrease in the third. The mean values and standard deviations revealed clearly that there is no significant increase in ß-galactosidase specific activity in the AgpAG-110 strain.


View this table:
[in this window]
[in a new window]
 
Table 1. ß-Galactosidase specific activity of plasmid (AgpTS07-) and site-directed (AgRIB3placZ-) transformants in growth phase (8 h) and riboflavin production phase (24 h)

 
Plasmids pAG-110 and pTS-07 contain autonomously replicating sequence (ARS) elements from S. cerevisiae, which make extrachromosomal replication in A. gossypii possible (Wright & Philippsen, 1991 ). Therefore, we could not rule out the increase in specific activity of ß-galactosidase in the production phase being due to a change in the copy number of pTS-07. In fact, it has been shown that plasmids containing ARS elements are freely replicative in A. gossypii (Wright & Philippsen, 1991 ). By DNA hybridization we found that a single copy of plasmid pTS-07 was integrated into the A. gossypii chromosome in all transformants (data not shown). Only the DNA of AgpTS07-B and AgpTS07-C gave an additional signal for extrachromosomal plasmids. In keeping with the finding that homologous recombination is preferred in A. gossypii (Steiner et al., 1995 ), integration of the plasmid at the RIB3 locus was observed by PCR (data not shown). To confirm that the increase in ß-galactosidase specific activity in the production phase was not due to higher copy numbers of the reporter plasmid caused by a popout, a defined integration of the reporter construction without an ARS element was performed.

Analysis of site-directed mutants carrying the RIB3placZ reporter with respect to genotype and ß-galactosidase specific activity
A linearized form of plasmid pTS-10 was used for transformation of A. gossypii wild-type. A double cross-over event between the entire 2·9 kb SphI homologue sequence should lead to chromosomal integration of the reporter and the resistance marker in front of the ICL1 locus (Fig. 1). Several transformants were obtained and analysed for genomic integration of the reporter and the geneticin resistance cassette. In Fig. 5 the result of a Southern blot experiment is shown for the wild-type and the transformant AgRIB3placZ-1. Chromosomal DNA of both strains was cut with SphI, giving a 2·9 kb fragment for the wild-type and a 4·3 kb fragment for the transformant because of an SphI restriction site between the lacZ gene and the geneticin resistance cassette. The result of Southern blotting clearly revealed the shift from 2·9 kb for the wild-type to 4·3 kb for the transformed strain, AgRIB3placZ-1. Above the 4·3 kb fragment an additional band indicated hybridization of the probe with a second putative serine tRNA locus in the A. gossypii genome (Fig. 5). Besides Southern blotting, the correct integration of the reporter and the geneticin resistance cassette was confirmed by PCR (data not shown).



View larger version (10K):
[in this window]
[in a new window]
 
Fig. 5. Genotype verification of the transformant AgRIB3placZ-1 by Southern blotting. Chromosomal DNA of A. gossypii wild-type (lane W) and AgRIB3placZ-1 (lane T) was digested by SphI and probed with an SphI–BglII fragment derived from pTS-10. Lane M, DIG-labelled molecular mass marker; bands from top to bottom are 23130, 9416 and 6557 bp.

 
Four independently transformed strains with defined genomic integration of the reporter fusion (named AgRIB3placZ-) were cultivated in shake flasks for 8 and 24 h and the specific activity of ß-galactosidase was measured (Table 1). In all strains an increase of ß-galactosidase specific activity in the riboflavin production phase was determined. Taking all measurements together, the mean increase in specific activity was tenfold. The results are comparable with those obtained with strains carrying pTS-07. It was clearly shown that a single integration of the lacZ reporter in the A. gossypii genome is sufficient to see the induction of the RIB3 promoter. Furthermore, it is striking that the induced RIB3 promoter is highly active compared to the activity of the TEF promoter.

Fermenter cultivation of AgRIB3placZ-1
For a better resolution of the growth and riboflavin production phases, strain AgRIB3placZ-1 was cultivated in a 5 l fermenter with the same medium used in shake-flask experiments. The results, shown in Fig. 6(a, lower part) indicated low RIB3 promoter activity in the early and exponential growth phases. The expression of the lacZ reporter was induced in the late growth phase as indicated by an increase in ß-galactosidase specific activity. The specific activity reached a maximum in the riboflavin production phase of about 8 U (mg protein)-1. A correlation between increase in ß-galactosidase specific activity and riboflavin production was observed. After induction of reporter expression, riboflavin production started with a short delay and reached a maximum of about 60 mg (g mycelial dry mass)-1 at the end of fermentation. Glucose depletion as a trigger for riboflavin production could be excluded: at the end of the fermentation, after 73 h, approximately 12 g glucose l-1 was still left in the medium (Fig. 6a, top). Yeast extract in the medium was growth-limiting. The experiment was completed by analysing RIB3 mRNA levels during cultivation in comparison to TEF mRNA levels. The result of the RT-PCR is shown in Fig. 6(b). While the TEF mRNA level remained constant during cultivation, an increase in RIB3 mRNA was observed in the late growth phase. This result was in keeping with the increase in ß-galactosidase specific activity, also detected in the late growth phase. The loss of the RIB3 signal at the end of the cultivation period (after 73 h) could be explained by replacement of dead cells by hyphae germinated from spores. This was visible by microscopical investigation of the culture.



View larger version (32K):
[in this window]
[in a new window]
 
Fig. 6. (a) Fermenter cultivation of the mutant strain AgRIB3placZ-1. Glucose ({triangledown}), CO2 in the exhaust gas ({blacktriangledown}), mycelial dry mass ({square}), ß-galactosidase specific activity ({blacksquare}) and total riboflavin ({blacktriangleup}) during cultivation are shown. The graphs shown represent a typical experiment. This was repeated once, revealing that the kinetics of induction in relation to growth and riboflavin, which is the essential point, were reproducible. In three further experiments ß-galactosidase specific activity was 0·03±0·02 U (mg protein)-1 when CO2 in the exhaust gas was increased to 0·05% and 5·7±1·9 U (mg protein)-1 when maximal riboflavin concentration was reached (means±SD; data not shown in graph). (b) Results of RT-PCR showing RIB3 mRNA concentrations relative to TEF mRNA levels during cultivation. Either one or two pairs of oligonucleotides were used. Arrows indicate RIB3 products with a size of 585 bp and TEF products with a size of 530 bp.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Here we report for the first time the detection of DHBP synthase activity in A. gossypii, although this organism was used to investigate the four-carbon precursor needed for the riboflavin molecule (Plaut, 1954 ; Bacher et al., 1982 , 1983 , 1985 ). The reason for the difficulties in detection is interfering enzymes which have to be removed. Volk & Bacher (1988 , 1990 ) developed the assay in the yeast Candida guilliermondii and DHBP synthase activity has to date also been measured in Arabidopsis thaliana (Herz et al., 2000 ) and the prokaryotes B. subtilis (Hümbelin et al., 1999 ) and E. coli (Richter et al., 1992 ). The range of activity found in A. gossypii is comparable with that reported for the other organisms and, although rather low when compared with specific activities of enzymes of central metabolism, it is sufficient to explain the observed riboflavin production rate.

Until now, regulation of riboflavin synthesis has only been studied at the mRNA level in B. subtilis. We showed regulation of DHBP synthase at the activity and mRNA levels as well as by a ß-galactosidase reporter experiment. Independently, on each level an approximately tenfold induction was shown during the production phase in comparison to the growth phase. Since RIB3 encodes the first enzyme in the quantitatively more important branch of the riboflavin biosynthetic pathway, its regulation might be sufficient to explain riboflavin overproduction. However, the five remaining genes, RIB1, RIB2, RIB4, RIB5 and RIB7, also have to be studied.

The scattering observed in the reporter activities determined was not due to strain instability or copy number effects of the plasmid used. The variability also occurred in strains where the reporter was integrated in the A. gossypii chromosome. The critical point seems to be the use of shake flasks, because the fermenter cultivation, where optimal conditions could be chosen, showed maximum activities which were never reached in a shake flask. In any case, the relative induction is visible despite the scattering of the absolute activities. Furthermore, plasmid-transformed strains can be used for deletion analysis of the RIB3 promoter or reporter constructions of the other RIB genes. In the case of a coordinated induction of several RIB genes, a possible clustering, which has been shown for primary as well as secondary metabolite pathway genes (Keller & Hohn, 1997 ), has to be investigated in A. gossypii.

Glucose repression, which is a general mechanism for adaptation of metabolism to carbon source availability in the environment (see reviews by Ronne, 1995 ; Gancedo, 1998 ; Carlson, 1999 ) in yeast and other fungi, is probably not the regulatory mechanism likely to explain the kinetics of RIB3placZ expression and riboflavin overproduction. The arguments are, first, that more than 10 g glucose l-1 were left in all cases when maximal RIB3placZ expression and riboflavin production were determined. Second, the specific productivity [>60 mg riboflavin (g biomass)-1] was considerably high and cannot be expected to be enhanced by other carbon sources. Third, induced RIB3placZ expression reached the same order of magnitude as the TEFplacZ control. The situation is different in the case of the regulation of secondary metabolism in Penicillium chrysogenum, where the first gene in penicillin biosynthesis, acvA, encoding {delta}-(L-{alpha}-aminoadipyl)-L-cysteinyl-D-valine synthetase, is repressed by glucose (Revilla et al., 1984 , 1986 ). It is striking that in Aspergillus nidulans, acvA is not repressed, but the second gene, ipnA, is repressed by glucose (Brakhage, 1992 ). Recently, the glucose repressor CRE1 was shown to regulate pcbC and cefEF, two genes of cephalosporin synthesis in Acremonium chrysogenum (Jekosch & Kück, 2000 ). Although riboflavin is a primary metabolite, its overproduction has no known function for growth and therefore it can be classified as a product of secondary metabolism. This point of view justifies a comparison with the regulation of antibiotic synthesis pathways. Besides the ICL1 promoter (Maeting et al., 1999 ), the RIB3 promoter is only the second regulated promoter to be described in A. gossypii. It might become a useful tool for the construction of strains overexpressing genes of flux-limiting enzymes in the riboflavin production phase.

An interesting subject for further research is the trigger for riboflavin overproduction. In fermenter cultivations riboflavin overproduction started at declining growth rates. The ideal tool for investigation might be a chemostatic cultivation with the reporter strain AgRIB3placZ-1 at different growth rates.


   ACKNOWLEDGEMENTS
 
We are grateful to H. Sahm for excellent working conditions and helpful discussions. Also, we thank J. L. Revuelta and H. Althöfer for informative discussions during the Ashbya meetings. Many thanks are due to T. Dickhaus and C. Gätgens for technical assistance. Reference substances of 3,4-dihydroxy-2-butanone and DHBP were kindly provided by Dr K. Kis (Department of Organic Chemistry and Biochemistry, Technical University of Munich, Germany).


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Ashby, S. F. & Nowell, N. (1926). The fungi of stigmatomycosis. Ann Botany 40, 69-83.

Bacher, A. (1991). Biosynthesis of flavins. In Chemistry and Biochemistry of Flavoenzymes , pp. 215-259. Edited by F. Müller. Boston:CRC Press.

Bacher, A., Le Van, Q. & Bühler, M. (1982). Biosynthesis of riboflavin. Incorporation of D-[1-13C]ribose. J Am Chem Soc 104, 3754-3755.

Bacher, A., Le Van, Q., Keller, P. J. & Floss, H. G. (1983). Biosynthesis of riboflavin. Incorporation of 13C-labeled precursors into the xylene ring. J Biol Chem 258, 13431-13437.[Abstract/Free Full Text]

Bacher, A., Le Van, Q., Keller, P. J. & Floss, H. G. (1985). Biosynthesis of riboflavin. Incorporation of multiply 13C-labeled precursors into the xylene ring. J Am Chem Soc 107, 6380-6385.

Bacher, A., Eberhardt, S., Fischer, M., Kis, K. & Richter, G. (2000). Biosynthesis of vitamin B2 (Riboflavin). Annu Rev Nutr 20, 153-167.[Medline]

Bigelis, R. (1989). Industrial products of biotechnology: application of gene technology. In Biotechnology , pp. 243. Edited by H. J. Rehm & G. Reed. Weinheim:VCH.

Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding. Anal Biochem 72, 248-254.[Medline]

Brakhage, A. A., Browne, P. & Turner, G. (1992). Regulation of Aspergillus nidulans penicillin biosynthesis and penicillin biosynthesis genes acvA and ipnA by glucose. J Bacteriol 174, 3789-3799.[Abstract]

Carlson, M. (1999). Glucose repression in yeast. Curr Opin Microbiol 2, 202-207.[Medline]

Gancedo, J. M. (1998). Yeast carbon catabolite repression. Microbiol Mol Biol Rev 62, 334-361.[Abstract/Free Full Text]

Gruber, A. D. & Levine, R. A. (1997). In situ assessment of mRNA accessibility in heterogeneous tissue samples using elongation factor-1{alpha} (EF-1{alpha}). Histochem Cell Biol 107, 411-416.[Medline]

Guilliermond, P. (1928). Recherches sur quelques Ascomycetes inferieurs isoles de la stigmatomycose des graines de cotonnier. Essai sur la phylogenie des Ascomycetes. Rev Gen Bot 40, 328-342.

Hanahan, D. (1985). Techniques for transformation of E. coli. In DNA CloningA Practical Approach , pp. 109-135. Edited by D. M. Glover. Oxford:IRL Press.

Herz, S., Eberhardt, S. & Bacher, A. (2000). Biosynthesis of riboflavin in plants. The ribA gene of Arabidopsis thaliana specifies a bifunctional GTP cyclohydrolase II/3,4-dihydroxy-2-butanone 4-phosphate synthase. Phytochemistry 53, 723-731.[Medline]

Hohn, T. M., Desjardins, A. E. & McCormick, S. P. (1993). Analysis of Tox5 gene expression in Gibberella pulicaris strains with different trichothecene production phenotypes. Appl Environ Microbiol 59, 2359-2363.[Abstract]

Hümbelin, M., Griesser, V., Keller, T. & 7 other authors (1999). GTP cyclohydrolase II and 3,4-dihydroxy-2-butanone 4-phosphate synthase are rate limiting enzymes in riboflavin synthesis of an industrial Bacillus subtilis strain used for riboflavin production. J Indust Microbiol Biotechnol 22, 1–7.

Jekosch, K. & Kück, U. (2000). Loss of glucose repression in an Acremonium chrysogenum ß-lactam producer strain and its restoration by multiple copies of the cre1 gene. Appl Microbiol Biotechnol 54, 556-563.[Medline]

Kalnins, A., Otto, K., Rüther, U. & Müller-Hill, B. (1983). Sequence of the lacZ gene of Escherichia coli. EMBO J 2, 593-597.[Medline]

Keller, N. P. & Hohn, T. M. (1997). Metabolic pathway gene clusters in filamentous fungi. Fungal Genet Biol 21, 17-29.[Medline]

Kennedy, M. A., Barbuch, R. & Bard, M. (1999). Transcriptional regulation of the squalene synthase gene (ERG9) in the yeast Saccharomyces cerevisiae. Biochim Biophys Acta 1445, 110-122.[Medline]

Kurth, R., Philippsen, P., Steiner, S. & Wright, M. C. (1992). New promoter region. Patent no. WO 92/00379.

Liao, D.-I., Viitanen, P. V. & Jordan, D. B. (2000). Cloning, expression, purification and crystallization of dihydroxybutanone phosphate synthase from Magnaporthe grisea. Acta Crystallogr D Biol Crystallogr 56, 1495-1497.[Medline]

Mack, M., van Loon, A. P. G. M. & Hohmann, H.-P. (1998). Regulation of riboflavin biosynthesis in Bacillus subtilis is affected by the activity of the flavokinase/flavin adenine dinucleotide synthetase encoded by ribC. J Bacteriol 180, 950-955.[Abstract/Free Full Text]

Maeting, I., Schmidt, G., Sahm, H., Revuelta, J. L., Stierhof, Y.-D. & Stahmann, K.-P. (1999). Isocitrate lyase of Ashbya gossypii – transcriptional regulation and peroxisomal localization. FEBS Lett 444, 15-21.[Medline]

Miller, J. H. (1972). Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Mironov, V. N., Kraev, A. S., Chikindas, M. L., Chernov, B. K., Stepanov, A. I. & Skryabin, K. G. (1994). Functional organization of the riboflavin biosynthesis operon from Bacillus subtilis SHgw. Mol Gen Genet 242, 201-208.[Medline]

Monschau, N., Sahm, H. & Stahmann, K.-P. (1998). Threonine aldolase overexpression plus threonine supplementation enhanced riboflavin production in Ashbya gossypii. Appl Environ Microbiol 64, 4283-4290.[Abstract/Free Full Text]

Pfeifer, V. F., Tanner, F. W., Vojnovich, C. & Traufler, D. H. (1950). Riboflavin by fermentation with Ashbya gossypii. Indust Eng Chem 42, 1776-1781.

Phaff, H. J. & Starmer, W. T. (1987). Yeasts associated with plants, insects and soil. In The Yeasts , pp. 123-174. Edited by A. H. Rose & J. S. Harrison. London:Academic Press.

Plaut, G. W. E. (1954). Biosynthesis of riboflavin. II: Incorporation of 14C-labeled compounds into ring A. J Biol Chem 211, 111-116.[Free Full Text]

Revilla, G., López-Nieto, M. J., Luengo, J. M. & Martín, J. F. (1984). Carbon catabolite repression of penicillin biosynthesis by Penicillium crysogenum. J Antibiot 37, 781-789.[Medline]

Revilla, G., Ramos, F. R., López-Nieto, M. J., Alvarez, E. & Martín, J. F. (1986). Glucose represses formation of {delta}-(L-{alpha}-aminoadipyl)-L-cysteinyl-D-valine and isopenicillin N synthase but not penicillin acetyltransferase in Penicillium chrysogenum. J Bacteriol 168, 947-952.[Medline]

Revuelta, J. L., Buitrago, M. J. & Santos, M. A. (1995). Riboflavin biosynthesis in fungi. Patent no. WO 95/26406.

Richter, G., Volk, R., Krieger, C., Lahm, H.-W., Röthlisberger, U. & Bacher, A. (1992). Biosynthesis of riboflavin: cloning, sequencing, and expression of the gene coding for 3,4-dihydroxy-2-butanone 4-phoshate synthase of Escherichia coli. J Bacteriol 174, 4050-4056.[Abstract]

Richter, G., Ritz, H., Katzenmeier, G., Volk, R., Kohnle, A., Lottspeich, F., Allendorf, D. & Bacher, A. (1993). Biosynthesis of riboflavin: cloning, sequencing, mapping, and expression of the gene coding for GTP cyclohydrolase II in Escherichia coli. J Bacteriol 175, 4045-4051.[Abstract]

Robinson, G. W., Tsay, Y. H., Kienzle, B. K., Smith-Monroy, C. A. & Bishop, R. W. (1993). Conservation between human and fungal squalene synthetases: similarities in structure, function, and regulation. Mol Cell Biol 13, 2706-2717.[Abstract]

Ronne, H. (1995). Glucose repression in fungi. Trends Genet 11, 12-17.[Medline]

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Schmidt, G., Stahmann, K.-P., Kaesler, B. & Sahm, H. (1996). Correlation of isocitrate lyase activity and riboflavin formation in the riboflavin overproducer Ashbya gossypii. Microbiology 142, 419-426.

Solovieva, I. M., Kreneva, R. A., Leak, D. J. & Perumov, D. A. (1999). The ribR gene encodes a monofunctional riboflavin kinase which is involved in regulation of the Bacillus subtilis riboflavin operon. Microbiology 145, 67-73.[Abstract]

Stahmann, K.-P., Kupp, C., Feldmann, S. D. & Sahm, H. (1994). Formation and degradation of lipid bodies found in the riboflavin-producing fungus Ashbya gossypii. Appl Microbiol Biotechnol 42, 121-127.

Stahmann, K.-P., Revuelta, J. L. & Seulberger, H. (2000). Three biotechnical processes using Ashbya gossypii, Candida famata, or Bacillus subtilis compete with chemical riboflavin production. Appl Microbiol Biotechnol 53, 509-516.[Medline]

Steiner, S. & Philippsen, P. (1994). Sequence and promoter analysis of the highly expressed TEF gene of the filamentous fungus Ashbya gossypii. Mol Gen Genet 242, 263-271.[Medline]

Steiner, S., Wendland, J., Wright, M. C. & Philippsen, P. (1995). Homologous recombination as the main mechanism for DNA integration and cause of rearrangements in the filamentous ascomycete Ashbya gossypii. Genetics 140, 973-987.[Abstract/Free Full Text]

Suzuki, T., Higgins, P. J. & Crawford, D. R. (2000). Control selection for RNA quantitation. BioTechniques 29, 332-337.[Medline]

Tansey, T. R. & Shechter, I. (2000). Structure and regulation of mammalian squalene synthase. Biochim Biophys Acta 1529, 49-62.[Medline]

Volk, R. & Bacher, A. (1988). Biosynthesis of riboflavin. The structure of the four-carbon precursor. J Am Chem Soc 110, 3651-3653.

Volk, R. & Bacher, A. (1990). Studies on the 4-carbon precursor in the biosynthesis of riboflavin. J Biol Chem 265, 19479-19485.[Abstract/Free Full Text]

Wickerham, L. J., Flickinger, M. H. & Johnston, R. M. (1946). The production of riboflavin by Ashbya gossypii. Arch Biochem 9, 95-98.

Wright, M. C. & Philippsen, P. (1991). Replicative transformation of the filamentous fungus Ashbya gossypii with plasmids containing Saccharomyces cerevisiae ARS elements. Gene 109, 99-105.[Medline]

Received 1 May 2001; revised 16 July 2001; accepted 30 July 2001.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Schlösser, T.
Articles by Stahmann, K.-P.
Articles citing this Article
PubMed
PubMed Citation
Articles by Schlösser, T.
Articles by Stahmann, K.-P.
Agricola
Articles by Schlösser, T.
Articles by Stahmann, K.-P.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2001 Society for General Microbiology.