Correlation of the Capsular Phenotype in Propionibacterium freudenreichii with the Level of Expression of gtf, a Unique Polysaccharide Synthase-Encoding Gene

Bacterial strains and growth conditions.The 68 strains of P. freudenreichii used in this study either were industrial strains (Standa Laboratoires, Caen, France) or were obtained from the TL collection of our lab (INRA, Rennes, France). All of the strains were routinely cultured in YEL broth (29) at 30°C under microaerophilic conditions. Growth was monitored spectrophotometrically at 650 nm (A₆₅₀). Recombinant strains of P. freudenreichii were grown on solidified YEL medium with 10 μg·ml⁻¹ of chloramphenicol. Lactococcus lactis subsp. lactis IL1403 was grown at 30°C in M17 broth (36) or on SGM17 (M17 medium supplemented with sucrose and glucose) when indicated below. Escherichia coli DH5a cells were grown at 37°C with agitation (150 rpm) in Luria-Bertani medium. Recombinant strains of E. coli DH5α and L. lactis IL1403 were grown in the presence of 10 μg·ml⁻¹ and 5 μg·ml⁻¹ chloramphenicol, respectively.

Immunological detection of capsular EPS.Agglutination tests were performed using an anti-S. pneumoniae serotype 37 antiserum raised against the β-glucan capsule of S. pneumoniae (obtained from the Statens Serum Institute, Hillerød, Denmark). P. freudenreichii strains were grown in YEL medium until the stationary growth phase, and the assays were performed as previously described (40). Some strains were prone to spontaneous agglutination before addition of the antiserum. In these cases, before addition of the antibody, cells were washed with YEL medium, which eliminated aggregation, and they were then tested for agglutination.

Microscopy.Cultures were grown on YEL medium to an A₆₅₀ of 1. For optical microscopy, 10 μl of a culture was pelleted by centrifugation (8,000 × g, 10 min, 4°C) and resuspended in 50 μl of the ultrafiltration (UF) retentate of skim milk (10). The cells were then observed by phase-contrast microscopy. For transmission electron microscopy, bacteria were rinsed with phosphate-buffered saline (PBS) and fixed overnight at 4°C in 0.1 M sodium cacodylate buffer (pH 7.2) containing 2% glutaraldehyde. Fixed bacteria were rinsed and stored at 4°C in cacodylate buffer containing 0.2 M sucrose. They were then postfixed with 1% osmium tetroxide containing 1.5% potassium cyanoferrate and 2% uranyl acetate in water before gradual dehydration in ethanol (30% to 100%) and embedding in Epon. Thin sections (70 nm) were collected on 200-mesh cooper grids and counterstained with lead citrate before examination. For scanning electron microscopy, cultures were treated and observation was performed as previously described (15).

Extraction of genomic DNA and PCR detection of the gtf gene.Genomic DNA was extracted from P. freudenreichii cultures grown in YEL medium to an A₆₅₀ of 1 as described previously (31) with a DNeasy tissue kit (Qiagen, Courtabœuf, France) used according to the instructions of the manufacturer. The sequences of primers used for gtf detection with total genomic DNA as the template are shown in Table S1 in the supplemental material. Each PCR mixture (50 μl) contained Taq polymerase buffer (10 mM Tris-HCl [pH 9.0], 50 mM KCl, 1.5 mM MgCl₂), each deoxynucleoside triphosphate (dNTP) at a concentration of 200 μM, each primer at a concentration of 1 μM, 2.5 U Taq polymerase (Q-BIOgene, Illkirch, France), and 1 μl of template DNA (equivalent to 10 to 20 ng). The cycling conditions were 94°C for 3 min, followed by 30 cycles of 95°C for 40 s, 60°C for 30 s, and 72°C for 40 s and then 72°C for 10 min.

Disruption of the gtf gene in P. freudenreichii and heterologous expression in L. lactis.The chloramphenicol resistance (Cm^r) gene was amplified from the pAMT propionibacterium plasmid (3) using primers NsiI-CmR-F and NsiI-CmR-R (see Table S1 in the supplemental material) and then digested with NsiI and ligated to PstI-digested pUC18. E. coli DH5α was then transformed with the resulting plasmid (pUC:CmR) as described by Hanahan (9). A 551-bp internal fragment of the 1,535-bp gtf open reading frame (ORF) was PCR amplified using primers BamHI-gtf-F and BamHI-gtf-R (see Table S1 in the supplemental material) with strain TL 34 genomic DNA as the template. The PCR fragment was digested with BamHI, purified, and ligated to BamHI-digested plasmid pUC:CmR, which resulted in the suicide vector pUC:Δgtf:CmR. Plasmid DNA isolated from E. coli DH5α transformants was used to transform P. freudenreichii TL 34 by electroporation as previously described (7), with the following modification. For preparation of electrocompetent P. freudenreichii cells, YEL medium was supplemented with 0.5 M sucrose and 1% glycine. After the pulse, cells were diluted in fresh YEL medium and incubated for 16 h at 30°C. P. freudenreichii transformants harboring inserted pUC:Δgtf:CmR were selected on solid YEL medium with 10 μg·ml⁻¹ of chloramphenicol. Disruption of the gtf coding sequence was verified by PCR.

Heterologous expression of gtf was carried out using pMG36c, an E. coli-L. lactis shuttle vector that allows gene expression in L. lactis under transcriptional control of a strong constitutive lactococcal promoter, P32 (37). Primers SalI-gtf-F and PstI-gtf-R (see Table S1 in the supplemental material) were used to amplify the gtf gene with P. freudenreichii TL 176 genomic DNA as the template. The 1.6-kb PCR fragment was digested with SalI and PstI and cloned into SalI-PstI-digested pMG36c. E. coli DH5a was transformed with the resulting plasmid (pMG36c:gtf), which was isolated and then introduced into L. lactis IL1403 competent cells by electroporation.

Total RNA extraction. P. freudenreichii strains were grown on YEL medium until the exponential (A₆₅₀, 0.6) or stationary (A₆₅₀, 1) growth phase. Then 15 ml (A₆₅₀, 0.6) or 3 ml (A₆₅₀, 1) of each culture was harvested and centrifuged (8,000 × g, 10 min, 4°C), and the pellet was stored at −80°C until total RNA was extracted. Pellets were thawed on ice, suspended in 200 μl of lysis buffer (50 mM Tris-HCl, 1 mM EDTA; pH 8.0) containing 20 mg ml⁻¹ lysozyme and 50 U ml⁻¹ mutanolysin, and incubated for 15 min at room temperature. The suspensions were then transferred to 2-ml tubes containing 50 mg of zirconium beads (diameter, 0.1 mm; BioSpec Products, Bartlesville, OK) and 100 μl of SDS (10%). The tubes were shaken twice for 90 s at 30 Hz with a bead beater (MM301; Retsch, Haan, Germany) with chilling on ice for 2 min between the shaking steps. RNA extraction was then performed using an RNeasy minikit (Qiagen) according to the instructions of the manufacturer. RNA was suspended in 50 μl of RNase-free water and treated with DNase (DNA free; Ambion, Cambridgeshire, United Kingdom) used according to the instructions of the supplier, and then 5-μl aliquots were stored at −80°C until they were used. Quantification of RNA was performed and contamination of RNA by proteins was assessed spectrophotometrically using a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Inc., Rockland, DE). The concentrations were between 30 and 200 ng μl⁻¹ in 50 μl. RNA quality was evaluated using an Agilent 2100 bioanalyzer (Agilent Technologies, Santa Clara, CA). The RIN value was calculated as previously described (34). All of the RNA samples had a RIN value greater than 9.4, indicating that the integrity of the rRNA was good. RNA extractions were performed in triplicate. The absence of genomic DNA was confirmed by quantitative PCR (qPCR).

Preparation of cDNA by retrotranscription.Reverse transcription (RT) reactions were carried out using a Bio-Rad iScript RT cDNA synthesis kit (Bio-Rad, Ivry-sur-Seine, France) according to the instructions of the manufacturer. To validate the stability of groL expression with the geNorm application, 100 ng of RNA was retrotranscribed. For subsequent quantification of gene expression, 3 μl of an RNA sample was reverse transcribed. No reduced RT efficiency was observed at a low RNA concentration (data not shown). cDNA was stored at −80°C until it was used.

Quantitative PCR.Primers (see Table S1 in the supplemental material) were designed with Beacon Designer software. Quantitative PCR was carried out using an Opticon2 (Bio-Rad) with the IQ SYBR green supermixture (Bio-Rad) in 96-well plates. Each mixture consisted of 8 μl IQ SYBR green supermixture, 3 μl of a primer pair mixture (1.6 μM each), and 5 μl of DNA or cDNA. The cycling conditions were as follows: initial denaturation at 95°C for 3 min, followed by 40 cycles of 95°C for 15 s and 60°C for 30 s. Fluorescence measurements were recorded during each annealing step. For each run, the absence of DNA contamination in RNA samples was verified by PCR amplification of dilutions of non-reverse-transcribed RNA extract. DNA contamination was never observed. The specificity of the PCR for each primer pair was validated by obtaining a melting curve with an additional step in which the temperature was increased from 60 to 90°C at a rate of 0.05°C s⁻¹. The efficiency of each amplification was controlled by preparing a standard curve with six serial dilutions of genomic DNA whose concentrations were known (from 12.5 ng per well to 12.5 fg per well). The PCR efficiency (E) was calculated as follows: E = [10^(1/−s) − 1] × 100, where s is the slope of the standard curve. The copy numbers for DNA standards were calculated using the following formula: number of copies = (amount·6.022 × 10²³)/(length·1 × 10⁹·650), where the length of the genome is 2.6 Mb. The cycle threshold (C_T) was determined using background fluorescence intensity corresponding to the first C_T of amplification plus 2.

Data analysis. C_T values were converted into copy numbers (i) per 100 ng of total RNA (absolute quantification) and (ii) with geNorm normalization (39). geNorm calculates a gene expression normalization factor for each sample based on the geometric mean for a defined number of reference genes. The stability of groL1 and groL2 was evaluated under the conditions used in the study by geNorm. Then the geNorm application was used to calculate the normalization factor, which was then applied to the raw data (gtf copy number).

Nucleotide sequence accession numbers.All nucleotide sequences have been deposited in the EMBL database (http://www.ebi.ac.uk/embl ). The gtf sequences of P. freudenreichii have been deposited under the following accession numbers: AM850119 for TL 20, AM850120 for TL 34, AM850122 for TL 162, FN557519 for TL 147, AM850123 for TL 176, AM850125 for TL 503, and AM850126 for TL 1348. The sequences of the regions upstream of the gtf ORF have been deposited under the following accession numbers: FN646426 for TL 20, FN646431 for TL 34, FN646429 for TL 162, FN646428 for TL 147, FN646430 for TL 176, and FN646427 for TL 1348.

Correlation of gft gene detection with the β-glucan capsular phenotype in P. freudenreichii.In a previous study, an analysis of the genome of the sequenced strain P. freudenreichii TL 34 revealed the presence of gtf, a gene with a high level of homology to tts of S. pneumoniae, which is responsible for the synthesis of the type 37 capsular polysaccharide, a (1→3,1→2)-β-d-glucan. Sixty-eight strains of P. freudenreichii were screened in this work for the presence of gtf, which is proposed to be involved in capsular EPS production because of its high level of homology with tts. Two PCR primer pairs targeting two distinct parts of the gene were used in order to distinguish incomplete genes. All the strains were positive with both PCR tests, producing amplicons of the expected size (see Table S1 in the supplemental material). Hence, no truncated gene was detected by this approach, and the results obtained suggest that the 68 strains have the same genetic potential. All of the P. freudenreichii strains were assayed further for agglutination as previously described in the presence of the anti-S. pneumoniae serotype 37 antiserum that was raised against (1→3,1→2)-β-d-glucan capsule and that was shown to be very specific for this structure (1, 28). Twenty-four of the 68 P. freudenreichii strains (35%) agglutinated in the presence of the antiserum (data not shown), indicating that there was a (1→3,1→2)-β-d-glucan capsule around the cells. For the agglutination-negative strains, the absence of agglutination could have been due either to the absence of the β-glucan capsule or to an insufficient amount of capsule so that agglutination did not occur.

P. freudenreichii strains were further examined to confirm the presence of a capsule using three microscopy techniques (see Materials and Methods). Using optical microscopy, four agglutination-positive (agglutination⁺) strains and three agglutination negative (agglutination⁻) strains were observed after partial immobilization in a UF retentate of skim milk. A clear halo observed around capsular cells indicates that opaque milk caseins could not contact the cells due to the presence of capsular EPS, as previously shown for other capsular bacteria (11). A halo surrounding the cells was observed for strain TL 176 (agglutination⁺) (Fig. 1B). No halo was observed for agglutination⁺ strains TL 34, TL 147, and TL 162 and agglutination⁻ strains TL 20, TL 503, and TL 1348, as exemplified by strain TL 20 (Fig. 1A). The same results were obtained after India ink staining of the cells (data not shown). The presence of capsule was also examined by transmission electronic microscopy. The agglutination⁺ TL 176 strain appeared to have a hairy surface, which seemed to be loosely bound to the cell wall (Fig. 1D), whereas the agglutination⁻ and capsuleless strain TL 20 had a smooth surface (Fig. 1C). When observed by scanning electron microscopy, strain TL 176 (agglutination⁺) seemed to be trapped in a net-like structure (Fig. 1F). This structure was not observed for strain TL 20 (Fig. 1E). As a capsule was observed by optical microscopy and transmission electron microscopy for the TL 176 strain, the strands observed here could correspond to altered capsular polysaccharides. They might also be polysaccharides released by the cells.

• Open in new tab
• Download powerpoint

FIG. 1.

Microscopic observations of the agglutination-positive TL 176 and agglutination-negative TL 20 strains of P. freudenreichii. (A and B) P. freudenreichii TL 20 (A) and TL 176 (B) suspended in UF retentate of skim milk and observed by light microscopy. The white halo surrounding cells corresponds to capsular EPS. (C and D) Strains TL 20 (C) and TL 176 (D) analyzed by transmission electron microscopy. (E and F) Strains TL 20 (E) and TL 176 (F) examined by scanning electron microscopy.

The discrepancy observed between the PCR test results (which indicated that the gtf gene was present in the 68 strains tested) and the phenotype (positive immunodetection of capsule for 24 strains) may be explained by the inactivation of gtf by evolution of gtf toward a pseudogene in some P. freudenreichii strains. Other food-grade bacteria, such as Lactobacillus bulgaricus, reportedly have a propensity for pseudogene accumulation (38). However, analysis of the gtf sequence in strains TL 34, TL 147, TL 162, and TL 176 (agglutination⁺), as well as in strains TL 20, TL 503, and TL 1348 (agglutination⁻), revealed that gtf is highly conserved; only 6 amino acids vary from one strain to another in the corresponding Gtf proteins. These data indicate that it is possible that a nonfunctional protein could be produced (for example, due to critical changes in amino acids). Alternatively, one can hypothesize that the gtf gene is present but that, unlike the findings for gtf genes in other species, some other genes are required for capsular synthesis and may be missing in capsule-negative strains. In some species, capsular EPS production indeed involves several biosynthetic genes (18). We thus investigated gtf function and expression further.

gtf gene confers the capsular EPS phenotype.The gtf gene was disrupted in capsular strain TL 34 (see Materials and Methods). The immunoagglutination test carried out with wild-type strain TL 34 and the TL 34 Δgtf mutant demonstrated that disruption of gtf resulted in loss of the capsular phenotype (Fig. 2). This functional analysis revealed that gtf is indeed involved in EPS formation, as expected based on sequence similarities with the well-documented S. pneumoniae tts gene. However, in other bacteria (e.g., Staphylococcus aureus and Mycobacterium smegmatis), capsular EPS biosynthesis relies on operons carrying up to 15 different genes. One cannot exclude the possibility that other genes are involved in EPS formation, which could explain the strain dependence of the capsular phenotype. To unambiguously demonstrate the role of the gtf gene in capsule formation, heterologous expression of the gtf gene isolated from the agglutination⁺ strain TL 176 was carried out using the noncapsular strain L. lactis IL1403. Recombinant L. lactis IL1403(pMG36C:gtf) was used in the immunoagglutination test with anti-S. pneumoniae serotype 37 antiserum. L. lactis IL1403(pMG36C:gtf) exhibited clear agglutination, whereas the control strain L. lactis IL1403(pMG36C) and the plasmidless strain L. lactis IL1403 did not agglutinate under the same conditions (Fig. 2). To our knowledge, this is the first time that a gene of a high-G+C-content dairy propionibacterium has been heterologously expressed in L. lactis, the low-G+C-content model LAB species. This result demonstrates that introduction of the P. freudenreichii gtf gene alone is sufficient to enable formation of capsular EPS by L. lactis. Altogether, these results show that gtf is the only structural gene required for the formation of capsular EPS in P. freudenreichii.

• Open in new tab
• Download powerpoint

FIG. 2.

Immunoagglutination test performed with S. pneumoniae type 37-specific antiserum. The agglutination test was performed with P. freudenreichii TL 34 (A) in the absence of the antiserum and (B) in the presence of the antiserum and with the P. freudenreichii TL 34 Δgtf mutant (C) in the presence of the antiserum and (D) in the absence of the antiserum. The agglutination test was also performed with L. lactis IL1403(pMG36C:gtf) before addition of the antiserum (E) and after addition of the antiserum (F) and with L. lactis IL1403(pMG36C) after the addition of the antiserum (G).

A previous study on the probiotic P. freudenreichii strain JS reported production of an EPS with the same structure as S. pneumoniae capsular EPS (30). Some LAB have also been reported to produce EPS with the same structure. All these LAB species harbor a gtf gene with significant homology to S. pneumoniae tts (41). In some of these LAB species harboring gtf, the production of capsular EPS is responsible for the detrimental increase in viscosity in beverages such as wine or cider. In Pediococcus parvulus, the gtf gene conferred a selective advantage as the subsequent production of EPS resulted in greater resistance to stressful conditions (acidic pH and alcohol) and better adhesion capacity (5). In P. freudenreichii, the role of capsular EPS production in probiotic properties, stress resistance, and the ability to modify cheese texture during ripening remains to be investigated.

Level of expression of the gtf gene correlates with capsule formation.Capsule formation was shown to be strain dependent (Table 1), even though all 68 P. freudenreichii strains bore a copy of the gtf gene. Thus, we hypothesized that the noncapsular phenotype might be due to a lack of gtf expression in some strains. In order to test this hypothesis, quantitative RT-PCR (qRT-PCR) assays were performed to quantify the expression of the gtf gene for seven strains. The PCR efficiency was tested for all strains, and the results are shown in Table S2 in the supplementary material. The expression of the gtf gene was analyzed by using absolute quantification and relative expression after geNorm normalization (39). As no data for normalization genes for qRT-PCR in propionibacteria were available, the stability of the groL1 and groL2 genes was evaluated. Although described as stress proteins, GroL1 and GroL2 have a fairly constant rate of synthesis during the exponential and stationary growth phases in YEL medium (26; G. Jan, unpublished results). The levels of expression of groL1 and groL2 after analysis using the geNorm application were stable under the conditions used in this study (see Table S3 in the supplemental material). Thus, groL1 and groL2 were used to normalize the gtf gene expression data.

The absolute quantification and relative expression values were in good agreement for levels of gtf gene expression (Fig. 3). They revealed that gtf was expressed in all seven strains in both the exponential and stationary growth phases, and the minimum rate was 10⁶ copies per 100 ng of RNA. The level of expression of gtf in the exponential growth phase and the level of expression of gtf in the stationary growth phase did not vary significantly. The four agglutination⁺ strains exhibited significantly higher levels of gtf expression than the agglutination⁻ strains (P < 0.005, Student t test) (Table 1). After geNorm normalization, the levels of gtf expression obtained for the agglutination⁺ strains were at least 4.25 ×10⁷gtf mRNA copies per 100 ng of total RNA, and the maximum level was 6.5 × 10⁸gtf mRNA copies per 100 ng for strain TL 176. The levels of expression obtained for all agglutination⁻ strains were less than 2.5 × 10⁶gtf mRNA copies per 100 ng of total RNA. In the stationary growth phase, the levels of gtf expression in the agglutination⁺ strains were 18 to 264 times higher than those in the agglutination⁻ strains. Altogether, these results strongly suggest that the capsular phenotype is correlated with the level of expression of the gtf gene. A level of expression lower than 10⁷gtf mRNA copies per 100 ng of total RNA does not appear to permit biosynthesis of detectable amounts of β-glucan capsule, and the cells did not agglutinate with the specific antiserum. When the level of expression was more than 10⁷gtf mRNA copies per 100 ng of total RNA, biosynthesis of a β-glucan capsule was observed, as demonstrated by a positive agglutination test, but the capsule was not always visible by microscopy. When the level of gtf expression was greater than 6.5 × 10⁸gtf mRNA copies per 100 ng of total RNA, as measured for strain TL 176, the cells agglutinated in the presence of antiserum, and the capsule was visible by microscopy.

• Open in new tab
• Download powerpoint

FIG. 3.

Levels of expression of the gtf gene in agglutination-positive and agglutination-negative strains of P. freudenreichii. The levels of expression in both the exponential and stationary growth phases are indicated by gray and black bars, respectively. Strain numbers are indicated on the x axis, and the agglutination-positive strains are underlined. (A) Absolute levels of expression expressed as copy number for 100 ng of RNA; (B) relative levels of expression after geNorm normalization using groL1 and groL2 as normalization genes.

The presence of an IS sequence is correlated with capsule production.Analysis of the complete sequence of the gtf ORF of agglutination⁺ strains TL 34, TL 147, TL 162, and TL 176 and of agglutination⁻ strains TL 20, TL 503, and TL 1348 did not reveal any frameshifts or insertions. The few point mutations observed resulted in synonymous substitutions or were not in the Gtf active site in the predicted proteins. For each of these strains, there were no data that indicated that the gtf gene was not functional (4).

We therefore sequenced the region upstream of the gtf gene in these seven strains (Fig. 4 and Table 1). The 194 nucleotides upstream of the gtf start codon were 100% conserved in the seven strains, and this sequence exhibited no homology to those in the sequence databases. Upstream of this 194-nucleotide sequence, two ORFs showing strong homology with ORFs encoding transposases were identified in all the agglutination⁺ strains. The predicted protein product of ORF1 aligned best with the tnpB product, a putative transposase (COG 2801) of Mycobacterium avium (IS1601-D; accession number Q9R2Z0), with 74% identity for a 295-amino-acid overlap. The tnpB gene is present at five complete loci in the genome of the sequenced strain P. freudenreichii TL 34 (H. Falentin, unpublished results). The 114-amino-acid protein encoded by ORF2 (tnpD) aligned best with a putative transposase of M. avium (IS1601-C; accession number O68997), with 72% identity for a 100% overlap. This ORF is present at four loci in the TL 34 genome. These two transposase genes are part of a predicted insertion element (IS element) (as determined using IS Finder with the complete TL 34 genome sequence) (data not shown). The nucleotide sequence upstream of ORF2 was found in all of the strains analyzed, and the level of identity was 100% in each case. No homology for this sequence was found in the databases. The presence of such mobile genetic elements was previously demonstrated for Oenococcus oeni and P. parvulus, two Gram-positive species in which gtf is surrounded by two IS elements. In these species, the capsular phenotype was shown to be dependent on the presence of the embedded gtf gene, which might be mobilized from one strain to another (5). In P. freudenreichii, a single IS element was found upstream of the gtf ORF in capsular strains. The presence of a complete outwardly directed promoter in one end of this IS element remains to be demonstrated. However, a similar chromosomal configuration was shown to provide a higher level of tetracycline resistance through a higher level of expression of the tetM gene in Staphylococcus aureus (35). Most interestingly, such phenotype variations (acquisition of a panel of new phenotypes, such as aggregation, biofilm formation, and sliding motility) were observed in M. smegmatis (another high-G+C-content, Gram-positive bacterium), and it was recently demonstrated that transposition of two insertion sequences into the promoter region of the mps operon responsible for glycopeptidolipid biosynthesis leads to overexpression of the mps operon and acquisition of new surface properties (21).

• Open in new tab
• Download powerpoint

FIG. 4.

Organization of the upstream region of the gtf CDS in agglutination-negative (A) and agglutination-positive (B) strains of P. freudenreichii. The gtf gene is indicated by a black arrow. The position of the start codon (ATG) is indicated. Transposase genes are indicated by gray arrows. Regions conserved in all the strains analyzed are indicated by cross-hatched boxes and are linked with lines.

Conclusion.Capsular EPS production has been reported for many food-grade bacteria. However, this production remains poorly studied, although capsular EPS have interesting applications, especially in the dairy industry (11). Screening of a P. freudenreichii strain library revealed that up to 35% of the strains produced capsular β-glucan. We demonstrated that gtf is the only gene responsible for β-glucan capsular EPS production in P. freudenreichii. This report describes the first functional analysis of a P. freudenreichii gene. Interestingly, not all strains possessing a gtf gene exhibited β-glucan capsule production. A qRT-PCR approach was developed to investigate the level of gtf expression, and we report here the first description of normalization genes suitable for analysis of gene expression in P. freudenreichii. This approach showed that the capsular phenotype is correlated with the level of gtf expression; a minimum level of transcription is required for synthesis of a detectable capsule. The presence of an IS element may explain the strain-dependent expression of the capsular phenotype. This work opens avenues for further studies of interstrain phenotypic variability in P. freudenreichii. The involvement of IS element mobility in the acquisition of new phenotypes in P. freudenreichii deserves further investigation.