• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of aemPermissionsJournals.ASM.orgJournalAEM ArticleJournal InfoAuthorsReviewers
Appl Environ Microbiol. Jan 2007; 73(2): 545–553.
Published online Nov 10, 2006. doi:  10.1128/AEM.01496-06
PMCID: PMC1796965

Identification and Characterization of a Fructose Phosphotransferase System in Bifidobacterium breve UCC2003[down-pointing small open triangle]


In silico analysis of the Bifidobacterium breve UCC2003 genome allowed identification of four genetic loci, each of which specifies a putative enzyme II (EII) protein of a phosphoenolpyruvate:sugar phosphotransferase system. The EII encoded by fruA, a clear homologue of the unique EIIBCA enzyme encoded by the Bifidobacterium longum NCC2705 genome, was studied in more detail. The fruA gene is part of an operon which contains fruT, which is predicted to encode a homologue of the Bacillus subtilis antiterminator LicT. Transcriptional analysis showed that the fru operon is induced by fructose. The genetic structure, complementation studies, and the observed transcription pattern of the fru operon suggest that the EII encoded in B. breve is involved in fructose transport and that its expression is controlled by an antiterminator mechanism. Biochemical studies unequivocally demonstrated that FruA phosphorylates fructose at the C-6 position.

Members of the genus Bifidobacterium, first described in 1900 by Tissier (34), are gram-positive, high-G+C-content bacteria commonly present in human and animal intestines. The potential health-promoting or probiotic activities of bifidobacteria are well known and include reduction of symptoms of irritable bowel disease (5), stimulation of the immune response (37), and inhibition or competitive exclusion of pathogenic bacteria (16). Bifidobacteria are saccharolytic organisms and are capable of fermenting a wide variety of oligosaccharides (4). However, relatively little is known about the transport of sugars in this genus.

It has been established that glucose transport in Bifidobacterium breve is mediated via a phosphoenolpyruvate (PEP):sugar phosphotransferase system(s) (PTS) (9). The PTS acts through concomitant internalization and phosphorylation of carbohydrates (25). The transfer of phosphate from PEP to the incoming sugar is mediated via a phosphorylation chain, which involves enzyme I (EI), histidine-containing protein (HPr), and enzyme II (EII). EI and HPr, which are cytoplasmic proteins, are collectively called the general element of the PTS because they are shared components of PTS-mediated sugar transport systems in a given bacterium (25). EII proteins specifically promote the transport of one or more sugars. Seven EII families have been described, and members of each family exhibit at least 25% sequence identity (2). A specific EII may occur as a single protein containing (usually) three distinct domains (EIIA, EIIB, and EIIC), although one or more of the EII domains may also exist as separate polypeptides. The EIIA and EIIB cytoplasmic domains are involved in phosphoryl transfer, while the membrane-spanning EIIC domain (in conjunction with EIID for the mannose PTS family) is responsible for sugar translocation. Many bacteria encode a multitude of different EII systems (e.g., 21 and 15 systems in Escherichia coli [33] and Bacillus subtilis [27], respectively). Genome analysis of Bifidobacterium longum (31) showed that this bacterium contains only one gene, ptsG, encoding EIIBCA belonging to the glucose family, which more precisely belongs to the β-glucoside subclass. Genes encoding an EII responsible for the transport of a particular sugar usually form an operon, whose expression is induced by the presence of this sugar (10). For numerous genes, this regulation has been shown to be mediated by transcription activators or antiterminators (AT), which contain regulatory domains referred to as PTS regulation domains (PRDs) (10, 32).

The first ATs described were BglG and SacY of E. coli and B. subtilis, respectively (29), which consist of an RNA-binding domain followed by two PRDs, PRDI and PRDII (29). A rho-independent terminator structure is present in the 5′ untranslated region of genes and operons regulated by ATs. Partially overlapping this terminator is a sequence referred to as an RNA antiterminator (RAT) (36), which can form a secondary structure. This RAT structure, when stabilized through binding of the corresponding AT, prevents the formation of the terminator and thereby counteracts premature transcription termination (32). Phosphorylation of the PRD(s) modifies the ability of the AT to bind to the RAT region (7, 35).

We show here that in B. breve fructose is the main substrate for a β-glucoside PTS. This PTS phosphorylates fructose at the C-6 position. This is in complete contrast to what happens in another actinobacterium, Streptomyces coelicolor, which transports fructose by a PTS belonging to the fructose family and phosphorylates this substrate at the C-1 position (19). We also demonstrate that the expression of the gene encoding the B. breve PTS is regulated by an antitermination mechanism.


Bacterial strains, plasmids, media, and growth conditions.

Bacterial strains and plasmids used in this study are listed in Table Table1.1. B. breve UCC2003 and derivatives of this strain were routinely inoculated from stock cultures into 10 ml of de Man-Rogosa-Sharpe medium (MRS) (Oxoid Ltd., Basingstoke, Hampshire, United Kingdom) supplemented with cysteine-HCl (0.05%) and were incubated anaerobically at 37°C in a chamber using the Anaerocult oxygen-depleting system (Merck, Darmstadt, Germany). Following 36 h of growth the culture was diluted (1:100) in 10 ml of modified MRS medium (MFR), which contained a specific sugar at a concentration of 1% (wt/vol) or, when a combination of two sugars was used, each sugar at a concentration of 0.5% (wt/vol). When necessary, chloramphenicol was added to a final concentration of 2 μg ml−1.

Strains and plasmids used in this study

The E. coli strains used as hosts for cloning experiments were routinely grown in Luria-Bertani medium, and when necessary, ampicillin was added at a final concentration of 100 μg ml−1 or chloramphenicol was added at a final concentration of 20 μg ml−1.

E. coli strains were transformed by electroporation with a Gene Pulser apparatus (Bio-Rad Laboratories, Hercules, CA) used as recommended by the manufacturer, and B. breve UCC2003 was transformed as previously described (20). In short, 30 ml of MRS was incubated with 2.4 ml of an overnight culture. When the culture reached an optical density at 600 nm of approximately 0.6, bacteria were incubated on ice for 20 min. The cells were then collected by centrifugation for 10 min at 4,500 rpm, and the pellet was washed twice with buffer A (1 mM citrate buffer [pH 5.8], 0.5 M sucrose). The cells were subsequently resuspended in 100 μl of buffer A. Fifty microliters of the cell suspension was used for electrotransformation (25 μF, 200 Ω, 2 kV). After transformation, the cells were incubated for 2 h in reinforced clostridial medium (BD, Franklin Lakes, NJ) before cells were plated at an appropriate dilution on reinforced clostridial agar (BD, Franklin Lakes, NJ) containing the appropriate antibiotic.

General molecular biological methods.

Chromosomal DNA of B. breve UCC2003 was extracted using a previously described protocol (21). Briefly, 40 ml of an overnight culture was subjected to centrifugation, and the resulting pellet was resuspended in 12 ml of TSE buffer (6.7% sucrose, 50 mM Tris-HCl [pH 8], 1 mM EDTA, 20 mg ml−1 lysozyme, 0.1 U ml−1 mutanolysin). The cells were incubated for 30 min at 50°C, after which 400 μl of proteinase K (20 mg ml−1) and 600 μl of 10% sodium dodecyl sulfate were added; this was followed by incubation at 60°C for 1 h. Chromosomal DNA was purified further by phenol-chloroform extraction. Plasmids and PCR products were purified using a JETquick kit (GENOMED Gmbh, Lohne, Germany). Restriction endonucleases, alkaline phosphatase, and T4 DNA ligase were used according to the manufacturer's instructions (Roche Diagnostic GmbH, Mannheim, Germany). Standard PCRs were carried out using Taq PCR MasterMix (QIAGEN, Santa Clara, CA), while high-fidelity PCR was carried out using KOD polymerase (Novagen, Darmstadt, Germany).

Construction of pNZ272-derived plasmids.

The promoter region of the fru operon was amplified using primers pfruEc and pfruBm containing EcoRI and BamHI restriction sites, respectively (Table (Table2).2). The PCR product was cleaved by EcoRI and BamHI and inserted into plasmid pNZ272 cut with the same restriction enzymes in order to create a transcriptional fusion with the promoterless gusA gene (24). The ligated product was transformed into E. coli NM522, and constructs containing the correct promoter insert size were screened by colony PCR. A recombinant pNZ272 derivative containing the expected insert, as verified by DNA sequencing, was designated pNZfru. pNZ272 and pNZfru were transformed into B. breve UCC2003 to generate strains NZ272 and NZfru, respectively.

Primers used in this study

To obtain a derivative of the fru promoter containing a partially deleted RAT or terminator element, a three-step PCR approach was used (8). The primers used for the PCRs are listed in Table Table2.2. The PCR product corresponding to the fru promoter region with a deleted RAT and the PCR product corresponding to the terminator element were inserted into pNZ272, generating pNZRAT and pNZTerm, respectively. The correct sequence of the fru promoter regions in the plasmids was confirmed by DNA sequencing. The B. breve UCC2003 derivatives containing plasmids pNZRAT and pNZTerm were designated NZRAT and NZTerm, respectively.

Transcriptional fusion assays with the different constructs were carried out as previously described (24).

Construction of pCR2.1 TOPO-derived plasmids.

The fruA gene was amplified using primers fruAT01 and fruAT02 with KOD polymerase. The PCR product was subsequently incubated for 15 min at 72°C with Taq PCR MasterMix to allow addition of an A overhang at the extremity of the PCR product in order to facilitate insertion of the amplicon in the TA cloning plasmid pCR2.1 TOPO (Invitrogen, Paisley, United Kingdom). The ligated product was transformed into E. coli 499 ZSC112L, and constructs containing fruA in the orientation that allowed its transcription by the lac promoter present on the plasmid were screened by colony PCR. A recombinant pCR2.1 TOPO derivative containing the fruA gene in its desired orientation, as verified by DNA sequencing, was designated pTOPOfruA. A 500-bp region of B. breve noncoding DNA was cloned in pCR2.1 TOPO to generate control plasmid pTOPOX.

Determination of the site of FruA-mediated fructose phosphorylation.

To carry out FruA-mediated fructose phosphorylation, we prepared membrane fragments from E. coli 499 ZSC112L harboring pTOPOfruA. This strain and the parental strain without a plasmid were grown in 500 ml of Luria-Bertani medium (in the absence of isopropyl-β-d-thiogalactopyranoside [IPTG]). When the optical density reached 0.7, the cells were centrifuged for 8 min at 5,000 × g and resuspended in 10 ml of buffer A (50 mM Tris-HCl [pH 7.4], 1 mM dithiothreitol, 1 mM EDTA) containing 1 mg ml−1 lysozyme. Cells were disrupted by sonication, and cell debris was removed by low-speed centrifugation (10 min at 10,000 × g). Membranes were isolated by ultracentrifugation for 1 h at 120,000 × g and 4°C. To wash the membranes, the pellet was resuspended in 20 ml buffer A and centrifuged for another hour at 120,000 × g. The final pellet was resuspended and homogenized in 2 ml of buffer A.

In the first reaction step, fructose phosphorylation was carried out in 100 μl of 50 mM Tris-HCl buffer (pH 7.4) containing 1 mM dithiothreitol, 1 mM EDTA, 10 mM MgCl2, 5 mM fructose, 10 μg B. subtilis EI, and 2 μg B. subtilis HPr (the two B. subtilis proteins were purified as described by Galinier et al. [11]), and 50 μl of the membrane suspension described above by incubating the assay mixtures for 30 min at 37°C in the presence or absence of 10 mM PEP. The reactions were stopped by heating the solutions for 10 min at 80°C, and denatured proteins were subsequently removed by centrifugation for 10 min at 10,000 × g.

By carrying out a coupled colorimetric assay, we tried to determine whether fructose-6-phosphate or fructose-1-phosphate had been formed during PEP-dependent FruA-mediated fructose phosphorylation. To do this, we prepared 500-μl reaction mixtures which contained 100 μl of the supernatants described above obtained after heat treatment, 50 mM Tris-HCl (pH 7.4), 0.16 mM NADH, 10 mM ATP, 1 U of fructose-1,6-bisphosphate aldolase, 90 U of triose phosphate isomerase, and 3 U of glycerol-3-phosphate dehydrogenase. The reaction was started by adding either 24 μg of E. coli fructose-1-phosphate kinase or 5 U of rabbit fructose-6-phosphate kinase, and the decrease in the optical density at 334 nm was determined to monitor glycerol-3-phosphate-dependent conversion of NADH to NAD+.

Most enzymes were purchased from Sigma. The only exception was E. coli fructose-1-phosphate kinase (FruK), which was purified in our laboratory. To overproduce His-tagged FruK, the fruK gene of E. coli was amplified by PCR with Pyrobest polymerase (Takara Bio Inc., Shiga, Japan) and primers fruK1 and fruK2 (Table (Table2),2), which contain BamHI and PstI restriction sites, respectively. The PCR fragment was cut with BamHI and PstI and inserted into the His tag expression vector pQE30 (QIAGEN) cleaved with the same enzymes. The resulting plasmid, pQEfruK, was used to transform E. coli NM522, and the correct sequence of the insert was confirmed by DNA sequencing. FruK carrying an N-terminal His6 tag was synthesized and purified on Ni-nitrilotriacetic acid columns by using the standard protocol of QIAGEN.

RNA purification and mapping of the transcription start site.

Total RNA was isolated from B. breve UCC2003 grown to the mid-exponential growth phase using the Macaloid method (14). The 5′ ends of the mRNA of the fru operon were mapped using the 5′ rapid amplification of cDNA ends (RACE) technique performed with a 3′/5′ RACE kit (Roche Diagnostic GmbH, Mannheim, Germany). The oligonucleotides used for cDNA synthesis (pts3), for PCR after poly(A) tailing (pts2), and for sequencing (pts1) are listed in Table Table22.


Five micrograms of RNA extracted from B. breve grown in MRS containing 1% fructose was treated with DNase (Roche) and used as a template in a 100-μl reaction mixture containing 20 ng of random primers, each deoxyribonucleotide triphosphate at a concentration of 0.125 mM, and the Superscript enzyme (Invitrogen, Paisley, United Kingdom), which was used according the manufacturer's instructions to produce cDNA. The cDNA generated was used as a template for reverse transcription (RT)-PCRs performed with primers RT-PCR1 and RT-PCR2.

Nucleotide sequence analysis.

Sequence data were obtained from Artemis-mediated genome annotations of the B. breve UCC2003 sequencing project (S. Leahy, M. O'Connell-Motherway, J. A. Moreno-Munoz, D. Higgins, G. Fitzgerald, and D. van Sinderen, unpublished data). Database searches were performed with BLAST (1) and at the National Center for Biotechnology Information internet site (http://www.ncbi.nlm.nih.gov). Sequence alignment was performed using the SEQMAN program of the DNASTAR package. Sequence analysis was performed using the Macvector program.

Nucleotide sequence accession number.

The sequence of the fru operon has been deposited in the EMBL data library under accession number AM055642.


Identification of a gene encoding an EIIBCAGlc-like PTS component.

The B. breve UCC2003 genome sequencing project revealed that this organism contains genes for the general PTS proteins ptsH and ptsI and at least four different putative EII-encoding loci on the basis of their similarity to EII-encoding genes from B. subtilis (S. Leahy, M. O'Connell-Motherway, J. Moreno Munoz, G. F. Fitzgerald, D. Higgins, and D. van Sinderen, unpublished results). This result was somewhat surprising because analysis of the genome of B. longum NCC2705 indicated that this bacterium has just a single EII, encoded by ptsG (31).

Of the four B. breve UCC2003 EII-encoding loci, fruA turned out to be the closest homologue of ptsG of B. longum NCC2705 (44% identity between the encoded proteins) (Fig. (Fig.1A).1A). In addition to the sequence similarity, the gene organizations around fruA and ptsG of B. longum are nearly identical, and the only difference is that glcP is present only in B. longum (Fig. (Fig.2A).2A). FruA contains all three sugar-specific PTS domains in the order EIIB-EIIC-EIIA from the N terminus to the C terminus, identical to the organization found in one of the members of the glucose PTS family (Fig. (Fig.1B).1B). Database searches indicated that FruA is similar to various EII proteins, all of which are members of the PTS glucose family and more specifically the sucrose/β-glucoside subfamily (2) (Fig. (Fig.1A1A).

FIG. 2.
(A) Comparison of bacterial chromosomal regions encoding homologues of FruA of B. breve UCC2003 and PtsG of B. longum NCC2705. fruA and ptsG encode an EIIBCA, licT and fruT encode an antiterminator, pgm encodes a phosphoglucomutase, glcP encodes a glucose ...

Analysis of the FruA sequence showed that the linkers separating the EIIB domain from the EIIC domain and the EIIC domain from the EIIA domain are 35 and 50 amino acids long, respectively, which is 25 amino acids longer than the sequences usually observed in EIIBCA proteins (15). The increased size of the linkers connecting the EII domains appears to be specific for high-G+C-content bacteria, as linkers that are similar sizes have been identified in EIIBCA proteins from Corynebacterium diphtheriae and Propionibacterium acnes (Fig. (Fig.1A).1A). Interestingly, the linker separating the EIIB domain from the EIIC domain of B. longum ptsG is around 100 amino acids long, and this is the largest linker detected so far in an EII.

Immediately downstream from fruA is a second gene designated fruT, whose deduced product exhibits significant similarity to AT proteins from a variety of bacteria. Sequence analysis of FruT showed that the histidine residues of the predicted PRDI and PRDII domains, which are subject to phosphorylation by P~EIIB and P~His-HPr in homologous ATs (10), are conserved in FruT, suggesting that these histidines are also subject to phosphorylation in order to modulate FruT activity. RT-PCR using primers RT-PCR1 and RT-PCR2, which spanned the intergenic region between fruA and fruT, generated specific products (data not shown), indicating that fruA and fruT constitute an operon, designated the fru operon here. Upstream from fruA there is a gene transcribed in the opposite direction. This gene, pgm, encodes a protein similar to predicted or characterized phosphoglucomutases.

The organization of genes surrounding fruA is very similar to the organization of the ptsG locus in B. longum NCC2705, except that B. longum NCC2705 contains an additional gene, glcP, encoding a putative glucose/proton symporter, located between ptsG and pgm (Fig. (Fig.2A).2A). The levels of identity of FruA and FruT proteins from different actinobacteria are also indicated in Fig. Fig.2A2A.

Mapping of the transcription start site of the fru operon.

In order to locate the transcription initiation site of the fru operon, 5′ RACE PCR experiments were carried out using total RNA extracted from B. breve UCC2003 grown in MFR supplemented with 1% (wt/vol) fructose and harvested during exponential growth. The transcription start site of the fru operon was identified as either a C located 207 nucleotides upstream of the start codon, a T located 208 nucleotides upstream of the start codon, or a T located 209 nucleotides upstream of the start codon. These bases are preceded by sequences that resemble −10 (TATAAT) and −35 (TTGCCG) hexamers, which are located 212 and 236 bp upstream of the start codon of fruA, respectively (Fig. (Fig.2B2B).

Sequence analysis of the 5′ untranslated region of fruA identified an inverted repeat sequence, which potentially acts as a terminator (Fig. (Fig.2B),2B), as well as a second hairpin loop structure which overlapped the presumptive terminator by 6 bases and which resembles RAT sequences described for other bacteria (Fig. (Fig.2C)2C) (36), including B. longum and C. diphtheriae (22, 23).

Regulation of the fru operon.

Expression of the gene(s) encoding a particular EII is usually induced by the sugar transported by the EII in question (10). In order to investigate whether the expression of the fru operon could be induced by a particular sugar, strains NZ272 and NZfru were constructed, which are B. breve UCC2003 derivatives harboring plasmid pNZ272 with a promoterless gusA and harboring plasmid pNZfru containing a fru-gusA transcriptional fusion, respectively. These two strains were grown in MFR supplemented with 1% (wt/vol) glucose, fructose, sucrose, ribose, cellobiose, or lactose, and β-glucuronidase activities were measured as described in Materials and Methods (Fig. (Fig.3).3). Glucose, sucrose, and the β-glucoside cellobiose were chosen because in many other bacteria these sugars are transported by a PTS belonging to the β-glucoside subfamily. The two other β-glucosides frequently utilized by bacteria, salicin and arbutin, were not included in this experiment because they do not support growth of B. breve (unpublished data). Fructose, lactose, and ribose were included in this experiment as potential PTS sugars. The level of β-glucuronidase activity in B. breve strain NZ272 (negative control) was very low irrespective of growth on the sugars mentioned above (only the activity detected in the presence of glucose is shown in Fig. Fig.3,3, bar 1). Measurement of the β-glucuronidase activity in the B. breve NZfru strain (which carried the fru-gusA fusion) revealed that growth on ribose resulted in the lowest expression of the fru-gusA fusion and that growth on fructose resulted in the highest expression of the fru-gusA fusion (Fig. (Fig.3,3, bars 5 and 3), suggesting that fructose is the principal sugar transported by this EII system. It was noted that in the presence of fructose B. breve NZfru grows more slowly than B. breve NZ272, suggesting that a high level of β-glucuronidase expression may be toxic. The presence of glucose, sucrose, cellobiose, or lactose in the growth medium of this strain also induced the fru promoter, although the expression level observed in the presence of these sugars was about 2.5-fold lower than that observed with fructose (Fig. (Fig.3),3), suggesting that FruA might also transport glucose, sucrose, lactose, and cellobiose.

FIG. 3.
β-Glucuronidase activity in NZfru grown in various sugars. The Gus activity in strain NZfru was measured during exponential growth in the presence of the following sugars at a concentration of 1% (wt/vol): glucose, fructose, sucrose, ribose, lactose, ...

Previous experiments with B. breve UCC2003 established that glucose can repress the expression of genes involved in the metabolism of sucrose (30). In order to test if glucose could also affect the expression of the fru operon, we determined the β-glucuronidase activity of NZfru in the presence of both glucose and fructose (each at a concentration of 0.5% [wt/vol]). The β-glucuronidase activity observed in this case was similar to the activity observed in the presence of glucose alone (Fig. (Fig.3,3, bars 2 and 8), suggesting that glucose does indeed repress the expression of the fru operon.

FruA restores growth of an E. coli strain deficient in fructose and glucose transport.

The finding that fructose induces the expression of the fru operon suggested that FruA is involved in fructose transport in B. breve UCC2003. In order to confirm this, E. coli strain 499 ZCS112L, which has mutations in genes required for the transport of glucose and fructose (ptsG, manXYZ, fruA, and glkA), was used in a heterocomplementation assay. Plasmids pTOPOfruA and pTOPOX (pCR2.1 TOPO containing fruA and a noncoding region, respectively) were introduced into E. coli 499 ZCS112L, and growth of the resulting strains in minimal medium M9 (in the absence of IPTG) was monitored. Growth on galactose was used as a reference. Only the strain containing pTOPOfruA was capable of growth in the presence of fructose and glucose, indicating that FruA is capable of transporting either of these sugars (Fig. (Fig.4)4) and that in B. breve UCC2003 FruA transports glucose and fructose. The fact that FruA is a member of the sucrose/β-glucoside PTS subfamily suggested that FruA might also transport this type of sugar. However, the absence of growth of E. coli 499 ZCS112L harboring pTOPOfruA in the presence of the β-glucosides salicin, arbutin, and cellobiose (note that only cellobiose supports growth of B. breve) indicates that FruA does not possess β-glucoside transport activity (data not show). If a β-glucoside is transported by FruA, it would be metabolized by the constitutively expressed enzyme E. coli phospho-β-glucosidase A (26) and therefore support growth.

FIG. 4.
Heterocomplementation of E. coli strain 499 ZSC112L. The cell density was measured by determining the absorbance at 600 nm for E. coli strain 499 ZSC112L without a plasmid (open bars) or containing pTOPOX (cross-hatched bars) or pTOPOfruA (bars with dots) ...

FruA phosphorylates fructose at C-6.

The expression of the fru-gusA fusion and the results of the heterocomplementation experiments suggested that FruA transports fructose. Fructose transported by a PTS belonging to the glucose family usually becomes phoshorylated at the C-1 position (13). However, B. breve lacks a fructose-1-phosphate kinase, which is normally necessary for the metabolism of fructose-1-phosphate. In addition, bifidobacteria have a unique hexose metabolism that occurs via a phosphoketolase pathway (3), and entry into this pathway happens at fructose-6-phosphate. There were two possibilities for how B. breve could respond to these requirements. Either this organism uses a mutase enzyme activity to convert fructose-1-phosphate into fructose-6-phosphate, or phosphorylation mediated by EIIBCA leads directly to the formation of fructose-6-phosphate. In order to examine the latter possibility, we tried to phosphorylate fructose in the presence of PEP, EI, HPr, and E. coli membrane fragments with or without B. breve FruA (prepared from strain 499 ZSC112L with or without plasmid pTOPOfruA). In a second reaction step we determined whether fructose-1-phosphate or fructose-6-phosphate had been formed. To do this, aliquots of the PEP-dependent fructose phosphorylation assay mixture were heated and subsequently incubated with ATP and either His-tagged E. coli fructose-1-phosphate kinase or fructose-6-phosphate kinase. The formation of fructose-1-6-bisphosphate was detected by inclusion of NADH and the enzymes fructose-1,6-bisphosphate aldolase, triose phosphate isomerase, and glycerol-3-phosphate dehydrogenase in the assay mixture. Experiments with purchased fructose-1-phosphate and fructose-6-phosphate established that owing to the high specificity of the two phosphofructokinases this method allowed a clear distinction between the two phospho forms of fructose (data not shown). Control experiments, in which the first phosphorylation step was attempted in the absence of PEP, revealed that no detectable amounts of fructose-1-phosphate or fructose-6-phosphate had been formed. Similarly, neither of the fructose phosphates was obtained when the first phosphorylation reaction was carried out in the presence of PEP and with membrane fragments prepared from E. coli 499 ZSC112L without the pTOPOfruA plasmid (Fig. (Fig.5),5), confirming that this strain has no residual fructose PTS activity. When the FruA-mediated phosphorylation reaction was carried out in the presence of PEP and with membrane fragments prepared from E. coli 499 ZSC112L containing the pTOPOfruA plasmid, a decrease in the NADH concentration (optical density at 334 nm) was observed only when the second reaction was carried out in the presence of fructose-6-phosphate kinase, while no decrease in the NADH concentration was detected with fructose-1-phosphate kinase (Fig. (Fig.5).5). These results clearly establish that the B. breve FruA-mediated fructose phosphorylation leads exclusively to the formation of fructose-6-phosphate and thus allows direct entry of this molecule into the bifid shunt via the fructose-6-phosphate phosphoketolase reaction. This result was surprising, as other characterized fructose-specific EIIs, which have the three domains arranged in a single protein, have been shown to have the domain order ABC and to phosphorylate fructose at the C-1 position (13). To the best of our knowledge, the results described above establish for the first time that a PTS belonging to the β-glucoside family (EII domain order, BCA) can phosphorylate fructose at the C-6 position.

FIG. 5.
Determination of the phosphorylated form of fructose (fructose-6-phosphate or fructose-1-phosphate) generated by FruA. The coupled photometric assay is described in Materials and Methods. The decrease in the NADH concentration was monitored at 334 nm. ...

Interestingly, two operons, fru1 and fru2 encoding fructose-phosphorylating EIIs, appear to be present in the genome of Lactobacillus casei ATCC 334 (unpublished observation). While fru1 resembles fru operons of most other bacteria and is composed of the three genes fruRKA1, which encode a transcription regulator of the DeoR family, a fructose-1-phosphate kinase, and an EIIABC, respectively, the fru2 operon contains only the fruKA2 genes. Nevertheless, a divergently oriented gene (fruR2) encoding a sugar phosphate-controlled transcription regulator precedes fru2. Interestingly, fruA2 encodes an EII resembling FruA of B. breve, as it also has the domain order BCA instead of ABC. It is therefore likely that similar to FruA of B. breve, the fruA2-encoded L. casei EIIBCA phosphorylates fructose. However, the L. casei FruA2 probably transfers the phosphoryl group to the C-1 position rather than the C-6 position of fructose, as fruK2, like fruK1, is predicted to specify a fructose-1-phosphate kinase. Genes strongly resembling fruR2 and fruKA2 of L. casei (encoding FruA2 with the domain order BCA) are present in the same orientation in Lactobacillus johnsonii (LJ0144 to LJ0146, respectively) (unpublished observation).

RAT and the terminator sequence present in the promoter of fruA are involved in transcriptional regulation.

Analysis of the sequence upstream of the start codon of fruA allowed us to find a potential rho-independent terminator sequence preceded by a potential RAT sequence (Fig. (Fig.2B).2B). To determine if these elements of the promoter region are implicated in its regulation, we determined the β-glucuronidase activities in strains NZRAT and NZTerm, which are B. breve UCC2003 derivatives harboring the promoter probe vector pNZ272 containing the fru-gusA fusion with the fru promoter carrying a partial deletion of the RAT element and a partial deletion of the terminator, respectively (Fig. (Fig.2B).2B). When these two strains were grown in MFR supplemented with glucose, fructose, or a mixture of glucose and fructose, only a very low level of β-glucuronidase activity was detected in NZRAT (Fig. (Fig.6,6, bars 3, 6, and 9), which was similar to the activity observed for the negative control strain, strain NZ272 (Fig. (Fig.6,6, bar 1), indicating that the RAT element is required for transcription of the fru operon. On the other hand, the β-glucuronidase activity detected in strain NZTerm in the presence of fructose was similar to the activity detected for the wild-type promoter (Fig. (Fig.6,6, bars 5 and 7). The activity detected in strain NZTerm grown in the presence of glucose or the combination of glucose and fructose was nearly twice as high as the activity observed in strain NZfru (Fig. (Fig.6,6, bars 2, 4, 8, and 10). This result suggests that the repressive effect of glucose is mediated via the antiterminator FruT in a manner similar to that described for LicT of B. subtilis (17).

FIG. 6.
Deletion analysis of the fru promoter region. β-Glucuronidase activity was determined for B. breve strains grown to the mid-exponential phase in the presence of 1% (wt/vol) glucose (bars 1 to 4) or fructose (bars 5 to 7) or in the presence of ...

If FruT and LicT are indeed subject to the same regulatory mechanism, this means that phosphorylation of PRDII is required for FruT binding to the RAT sequence, while phosphorylation of PRDI leads to the loss of this binding activity (17). In the presence of a repressing sugar (such as glucose), P~His-HPr is not expected to phosphorylate the histidine residues conserved in the PRDII domain of FruT, and the resulting inactive FruT is unable to prevent premature termination of fru transcription. In the absence of both glucose and fructose (a noninducing condition), P~His-HPr probably phosphorylates the conserved histidines of PRDII, but the conserved histidines of the PRDI domain of FruT are phosphorylated by FruA, thereby keeping FruT inactive. In the presence of fructose (and in the absence of glucose), FruA preferentially phosphorylates the incoming sugar (fructose) rather than the PRDI domain of FruT. In the latter case, FruT is phosphorylated at its PRDII domain, while its PRDI domain remains unphosphorylated, a situation that enables FruT to bind to the RAT sequence of the fru transcript, thereby preventing the formation of the terminator structure and allowing transcription of the complete fru operon.

If glucose is present in addition to fructose, it partially inhibits FruT activity. Nevertheless, growth on glucose also stimulates FruT activity compared to growth on ribose. It is therefore likely that glucose inhibits its own induction of fruA. This kind of autorepression has been observed for other PTS, including the fructose-transporting lev-PTS of B. subtilis (18). In the case of fruA, this inhibition is probably due to diminished phosphorylation of PRDII in FruT. It is possible that other PTS sugars have a similar effect on PRDII phosphorylation.

The β-glucuronidase activity detected in strain NZTerm in the presence of fructose is higher than the β-glucuronidase activity detected in the presence of glucose (Fig. (Fig.6,6, bars 4, 7, and 10), suggesting that the fru operon is regulated by a second mechanism or that binding of FruT to the RAT element might stabilize the fru mRNA.


Previous work has indicated that there is carbon catabolite repression (CCR) in bifidobacteria (30). The absence of a B. breve gene encoding a homologue of either the catabolite repressor CcpA of B. subtilis or cyclic AMP receptor protein of E. coli (Leahy et al., unpublished results) suggests that the mechanism by which catabolite repression operates in bifidobacteria is different from the mechanism described for many other bacteria. The expected regulatory linkage between glucose transport and CCR (6) together with our finding that fructose and glucose are probable PTS sugars suggests that the PTS must be involved in CCR in this bacterium. This indicates that although CCR in the high-G+C-content bacterium B. breve UCC2003 occurs by a mechanism different from the mechanism operating in many gram-positive and gram-negative bacteria, it nevertheless involves one or more components of the PTS.


This research was financially supported by Marie Curie Development Host Fellowship HPMD-2000-00027 and by the Science Foundation Ireland Alimentary Pharmabiotic Centre located at University College Cork.

We thank B. Erni for providing E. coli strain 499 ASC112L and N. Sauvageot and P. Curley for valuable discussions.


[down-pointing small open triangle]Published ahead of print on 10 November 2006.


1. Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic local alignment search tool. J. Mol. Biol. 215:403-410. [PubMed]
2. Barabote, R. D., and M. H. Saier, Jr. 2005. Comparative genomic analyses of the bacterial phosphotransferase system. Microbiol. Mol. Biol. Rev. 69:608-634. [PMC free article] [PubMed]
3. Biavati, B., B. Sgorbati, and V. Scardovi. 1991. The genus Bifidobacterium, p. 816-833. In A. Balows, H. G. Trüper, M. Dworkin, W. Harder, and K.-H. Schleifer (ed.), The prokaryotes, 2nd ed. Springer-Verlag, New York, NY.
4. Bouhnik, Y., B. Flourie, L. D'Agay-Abensour, P. Pochart, G. Gramet, M. Durand, and J. C. Rambaud. 1997. Administration of transgalacto-oligosaccharides increases fecal bifidobacteria and modifies colonic fermentation metabolism in healthy humans. J. Nutr. 127:444-448. [PubMed]
5. Brigidi, P., B. Vitali, E. Swennen, G. Bazzocchi, and D. Matteuzzi. 2001. Effects of probiotic administration upon the composition and enzymatic activity of human fecal microbiota in patients with irritable bowel syndrome or functional diarrhea. Res. Microbiol. 152:735-741. [PubMed]
6. Brückner, R., and F. Titgemeyer. 2002. Carbon catabolite repression in bacteria: choice of the carbon source and autoregulatory limitation of sugar utilization. FEMS Microbiol. Lett. 209:141-148. [PubMed]
7. Chen, Q., J. C. Arents, R. Bader, P. W. Postma, and O. Amster-Choder. 1997. BglF, the sensor of the E. coli bgl system, uses the same site to phosphorylate both a sugar and a regulatory protein. EMBO J. 16:4617-4627. [PMC free article] [PubMed]
8. Darbon, E., P. Servant, S. Poncet, and J. Deutscher. 2002. Antitermination by GlpP, catabolite repression via CcpA, and inducer exclusion elicited by P~GlpK dephosphorylation control Bacillus subtilis glpFK expression. Mol. Microbiol. 43:1039-1052. [PubMed]
9. Degnan, B. A., and G. T. Macfarlane. 1993. Transport and metabolism of glucose and arabinose in Bifidobacterium breve. Arch. Microbiol. 160:144-151. [PubMed]
10. Deutscher, J., A. Galinier, and I. Martin-Verstraete. 2002. Carbohydrate uptake and metabolism, p. 129-150. In A. L. Sonenshein, J. A. Hoch, and R. Losick (ed.), Bacillus subtilis and its closest relatives: from genes to cells. ASM Press, Washington, DC.
11. Galinier, A., J. Haiech, M. C. Kilhoffer, M. Jaquinod, J. Stülke, J. Deutscher, and I. Martin-Verstraete. 1997. The Bacillus subtilis crh gene encodes a HPr-like protein involved in carbon catabolite repression. Proc. Natl. Acad. Sci. USA 94:8439-8444. [PMC free article] [PubMed]
12. Gough, J. A., and N. E. Murray. 1983. Sequence diversity among related genes for recognition of specific targets in DNA molecules. J. Mol. Biol. 166:1-19. [PubMed]
13. Kornberg, H. L. 2001. Routes for fructose utilization by Escherichia coli. J. Mol. Microbiol. Biotechnol. 3:355-359. [PubMed]
14. Kuipers, O. P., M. M. Beerthuyzen, R. J. Siezen, and W. M. De Vos. 1993. Characterization of the nisin gene cluster nisABTCIPR of Lactococcus lactis. Requirement of expression of the nisA and nisI genes for development of immunity. Eur. J. Biochem. 216:281-291. [PubMed]
15. Lengeler, J. W., K. Jahreis, and U. F. Wehmeier. 1994. Enzymes II of the phosphoenolpyruvate-dependent phosphotransferase systems: their structure and function in carbohydrate transport. Biochim. Biophys. Acta 1188:1-28. [PubMed]
16. Lievin, V., I. Peiffer, S. Hudault, F. Rochat, D. Brassart, J. R. Neeser, and A. L. Servin. 2000. Bifidobacterium strains from resident infant human gastrointestinal microflora exert antimicrobial activity. Gut 47:646-652. [PMC free article] [PubMed]
17. Lindner, C., M. Hecker, D. Le Coq, and J. Deutscher. 2002. Bacillus subtilis mutant LicT antiterminators exhibiting enzyme I- and HPr-independent antitermination affect catabolite repression of the bglPH operon. J. Bacteriol. 184:4819-4828. [PMC free article] [PubMed]
18. Martin-Verstraete, I., J. Stulke, A. Klier, and G. Rapoport. 1995. Two different mechanisms mediate catabolite repression of the Bacillus subtilis levanase operon. J. Bacteriol. 177:6919-6927. [PMC free article] [PubMed]
19. Nothaft, H., S. Parche, A. Kamionka, and F. Titgemeyer. 2003. In vivo analysis of HPr reveals a fructose-specific phosphotransferase system that confers high-affinity uptake in Streptomyces coelicolor. J. Bacteriol. 185:929-937. [PMC free article] [PubMed]
20. O'Mahony, C. 2002. Development of genetic technologies for the manipulation of Bifidobacterium species. M.Sc. thesis. Department of Microbiology, National University of Ireland, Cork, Ireland.
21. O'Riordan, K. 1998. Studies on antimicrobial activity and genetic diversity of Bifidobacterium species: molecular characterisation of a 5.75 kb plasmid and a chromosomally encoded recA gene homologue from Bifidobacterium breve. Ph.D. thesis, Department of Microbiology, National University of Ireland, Cork, Ireland.
22. Parche, S., M. Beleut, E. Rezzonico, D. Jacobs, F. Arigoni, F. Titgemeyer, and I. Jankovic. 2006. Lactose-over-glucose preference in Bifidobacterium longum NCC2705: glcP, encoding a glucose transporter, is subject to lactose repression. J. Bacteriol. 188:1260-1265. [PMC free article] [PubMed]
23. Parche, S., A. W. Thomae, M. Schlicht, and F. Titgemeyer. 2001. Corynebacterium diphtheriae: a PTS view to the genome. J. Mol. Microbiol. Biotechnol. 3:415-422. [PubMed]
24. Platteeuw, C., G. Simons, and W. M. de Vos. 1994. Use of the Escherichia coli β-glucuronidase (gusA) gene as a reporter gene for analyzing promoters in lactic acid bacteria. Appl. Environ. Microbiol. 60:587-593. [PMC free article] [PubMed]
25. Postma, P. W., J. W. Lengeler, and G. R. Jacobson. 1993. Phosphoenolpyruvate:carbohydrate phosphotransferase systems of bacteria. Microbiol. Rev. 57:543-594. [PMC free article] [PubMed]
26. Prasad, I., B. Young, and S. Schaefler. 1973. Genetic determination of the constitutive biosynthesis of phospho-β-glucosidase A in Escherichia coli K-12. J. Bacteriol. 114:909-915. [PMC free article] [PubMed]
27. Reizer, J., S. Bachem, A. Reizer, M. Arnaud, M. H. Saier, Jr., and J. Stülke. 1999. Novel phosphotransferase system genes revealed by genome analysis—the complete complement of PTS proteins encoded within the genome of Bacillus subtilis. Microbiology. 145:3419-3429. [PubMed]
28. Reuter, G. 1963. Vergleichenden Untersuchung uber die Bifidus-Flora im Sauglings und Erwachsenenstuhl. Zentbl. Bakteriol. Parasitenkd. Infektionskr. Hyg. Abt. 1 Orig. 191:486-507.
29. Rutberg, B. 1997. Antitermination of transcription of catabolic operons. Mol. Microbiol. 23:413-421. [PubMed]
30. Ryan, S. M., G. F. Fitzgerald, and D. van Sinderen. 2005. Transcriptional regulation and characterization of a novel β-fructofuranosidase-encoding gene from Bifidobacterium breve UCC2003. Appl. Environ Microbiol. 71:3475-3482. [PMC free article] [PubMed]
31. Schell, M. A., M. Karmirantzou, B. Snel, D. Vilanova, B. Berger, G. Pessi, M. C. Zwahlen, F. Desiere, P. Bork, M. Delley, R. D. Pridmore, and F. Arigoni. 2002. The genome sequence of Bifidobacterium longum reflects its adaptation to the human gastrointestinal tract. Proc. Natl. Acad. Sci. USA 99:14422-14427. [PMC free article] [PubMed]
32. Stülke, J., M. Arnaud, G. Rapoport, and I. Martin-Verstraete. 1998. PRD—a protein domain involved in PTS-dependent induction and carbon catabolite repression of catabolic operons in bacteria. Mol. Microbiol. 28:865-874. [PubMed]
33. Tchieu, J. H., V. Norris, J. S. Edwards, and M. H. Saier, Jr. 2001. The complete phosphotranferase system in Escherichia coli. J. Mol. Microbiol. Biotechnol. 3:329-346. [PubMed]
34. Tissier, H. 1900. Recherches sur la flore intestinale des nourrissons (état normal et pathologie). Ph.D. thesis. University of Paris, Paris, France.
35. Tortosa, P., S. Aymerich, C. Lindner, M. H. Saier, Jr., J. Reizer, and D. Le Coq. 1997. Multiple phosphorylation of SacY, a Bacillus subtilis transcriptional antiterminator negatively controlled by the phosphotransferase system. J. Biol. Chem. 272:17230-17237. [PubMed]
36. Yang, Y., N. Declerck, X. Manival, S. Aymerich, and M. Kochoyan. 2002. Solution structure of the LicT-RNA antitermination complex: CAT clamping RAT. EMBO J. 21:1987-1997. [PMC free article] [PubMed]
37. Young, S. L., M. A. Simon, M. A. Baird, G. W. Tannock, R. Bibiloni, K. Spencely, J. M. Lane, P. Fitzharris, J. Crane, I. Town, E. Addo-Yobo, C. S. Murray, and A. Woodcock. 2004. Bifidobacterial species differentially affect expression of cell surface markers and cytokines of dendritic cells harvested from cord blood. Clin. Diagn. Lab. Immunol. 11:686-690. [PMC free article] [PubMed]

Articles from Applied and Environmental Microbiology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...